ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 167 by ph10, Wed May 9 15:53:54 2007 UTC revision 1486 by ph10, Wed Jun 18 16:48:57 2014 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
4  Version 7.2 01-May-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".
47    9.  Give a compile-time error for \o{} (as Perl does) and for \x{} (which Perl
48        doesn't).
50    10. Change 8.34/15 introduced a bug that caused the amount of memory needed
51        to hold a pattern to be incorrectly computed (too small) when there were
52        named back references to duplicated names. This could cause "internal
53        error: code overflow" or "double free or corruption" or other memory
54        handling errors.
56    11. When named subpatterns had the same prefixes, back references could be
57        confused. For example, in this pattern:
59          /(?P<Name>a)?(?P<Name2>b)?(?(<Name>)c|d)*l/
61        the reference to 'Name' was incorrectly treated as a reference to a
62        duplicate name.
64    12. A pattern such as /^s?c/mi8 where the optional character has more than
65        one "other case" was incorrectly compiled such that it would only try to
66        match starting at "c".
68    13. When a pattern starting with \s was studied, VT was not included in the
69        list of possible starting characters; this should have been part of the
70        8.34/18 patch.
73    Version 8.35 04-April-2014
74    --------------------------
76    1.  A new flag is set, when property checks are present in an XCLASS.
77        When this flag is not set, PCRE can perform certain optimizations
78        such as studying these XCLASS-es.
80    2.  The auto-possessification of character sets were improved: a normal
81        and an extended character set can be compared now. Furthermore
82        the JIT compiler optimizes more character set checks.
84    3.  Got rid of some compiler warnings for potentially uninitialized variables
85        that show up only when compiled with -O2.
87    4.  A pattern such as (?=ab\K) that uses \K in an assertion can set the start
88        of a match later then the end of the match. The pcretest program was not
89        handling the case sensibly - it was outputting from the start to the next
90        binary zero. It now reports this situation in a message, and outputs the
91        text from the end to the start.
93    5.  Fast forward search is improved in JIT. Instead of the first three
94        characters, any three characters with fixed position can be searched.
95        Search order: first, last, middle.
97    6.  Improve character range checks in JIT. Characters are read by an inprecise
98        function now, which returns with an unknown value if the character code is
99        above a certain treshold (e.g: 256). The only limitation is that the value
100        must be bigger than the treshold as well. This function is useful, when
101        the characters above the treshold are handled in the same way.
103    7.  The macros whose names start with RAWUCHAR are placeholders for a future
104        mode in which only the bottom 21 bits of 32-bit data items are used. To
105        make this more memorable for those maintaining the code, the names have
106        been changed to start with UCHAR21, and an extensive comment has been added
107        to their definition.
109    8.  Add missing (new) files sljitNativeTILEGX.c and sljitNativeTILEGX-encoder.c
110        to the export list in Makefile.am (they were accidentally omitted from the
111        8.34 tarball).
113    9.  The informational output from pcretest used the phrase "starting byte set"
114        which is inappropriate for the 16-bit and 32-bit libraries. As the output
115        for "first char" and "need char" really means "non-UTF-char", I've changed
116        "byte" to "char", and slightly reworded the output. The documentation about
117        these values has also been (I hope) clarified.
119    10. Another JIT related optimization: use table jumps for selecting the correct
120        backtracking path, when more than four alternatives are present inside a
121        bracket.
123    11. Empty match is not possible, when the minimum length is greater than zero,
124        and there is no \K in the pattern. JIT should avoid empty match checks in
125        such cases.
127    12. In a caseless character class with UCP support, when a character with more
128        than one alternative case was not the first character of a range, not all
129        the alternative cases were added to the class. For example, s and \x{17f}
130        are both alternative cases for S: the class [RST] was handled correctly,
131        but [R-T] was not.
133    13. The configure.ac file always checked for pthread support when JIT was
134        enabled. This is not used in Windows, so I have put this test inside a
135        check for the presence of windows.h (which was already tested for).
137    14. Improve pattern prefix search by a simplified Boyer-Moore algorithm in JIT.
138        The algorithm provides a way to skip certain starting offsets, and usually
139        faster than linear prefix searches.
141    15. Change 13 for 8.20 updated RunTest to check for the 'fr' locale as well
142        as for 'fr_FR' and 'french'. For some reason, however, it then used the
143        Windows-specific input and output files, which have 'french' screwed in.
144        So this could never have worked. One of the problems with locales is that
145        they aren't always the same. I have now updated RunTest so that it checks
146        the output of the locale test (test 3) against three different output
147        files, and it allows the test to pass if any one of them matches. With luck
148        this should make the test pass on some versions of Solaris where it was
149        failing. Because of the uncertainty, the script did not used to stop if
150        test 3 failed; it now does. If further versions of a French locale ever
151        come to light, they can now easily be added.
153    16. If --with-pcregrep-bufsize was given a non-integer value such as "50K",
154        there was a message during ./configure, but it did not stop. This now
155        provokes an error. The invalid example in README has been corrected.
156        If a value less than the minimum is given, the minimum value has always
157        been used, but now a warning is given.
159    17. If --enable-bsr-anycrlf was set, the special 16/32-bit test failed. This
160        was a bug in the test system, which is now fixed. Also, the list of various
161        configurations that are tested for each release did not have one with both
162        16/32 bits and --enable-bar-anycrlf. It now does.
164    18. pcretest was missing "-C bsr" for displaying the \R default setting.
166    19. Little endian PowerPC systems are supported now by the JIT compiler.
168    20. The fast forward newline mechanism could enter to an infinite loop on
169        certain invalid UTF-8 input. Although we don't support these cases
170        this issue can be fixed by a performance optimization.
172    21. Change 33 of 8.34 is not sufficient to ensure stack safety because it does
173        not take account if existing stack usage. There is now a new global
174        variable called pcre_stack_guard that can be set to point to an external
175        function to check stack availability. It is called at the start of
176        processing every parenthesized group.
178    22. A typo in the code meant that in ungreedy mode the max/min qualifier
179        behaved like a min-possessive qualifier, and, for example, /a{1,3}b/U did
180        not match "ab".
182    23. When UTF was disabled, the JIT program reported some incorrect compile
183        errors. These messages are silenced now.
185    24. Experimental support for ARM-64 and MIPS-64 has been added to the JIT
186        compiler.
188    25. Change all the temporary files used in RunGrepTest to be different to those
189        used by RunTest so that the tests can be run simultaneously, for example by
190        "make -j check".
193    Version 8.34 15-December-2013
194    -----------------------------
196    1.  Add pcre[16|32]_jit_free_unused_memory to forcibly free unused JIT
197        executable memory. Patch inspired by Carsten Klein.
199    2.  ./configure --enable-coverage defined SUPPORT_GCOV in config.h, although
200        this macro is never tested and has no effect, because the work to support
201        coverage involves only compiling and linking options and special targets in
202        the Makefile. The comment in config.h implied that defining the macro would
203        enable coverage support, which is totally false. There was also support for
204        setting this macro in the CMake files (my fault, I just copied it from
205        configure). SUPPORT_GCOV has now been removed.
207    3.  Make a small performance improvement in strlen16() and strlen32() in
208        pcretest.
210    4.  Change 36 for 8.33 left some unreachable statements in pcre_exec.c,
211        detected by the Solaris compiler (gcc doesn't seem to be able to diagnose
212        these cases). There was also one in pcretest.c.
214    5.  Cleaned up a "may be uninitialized" compiler warning in pcre_exec.c.
216    6.  In UTF mode, the code for checking whether a group could match an empty
217        string (which is used for indefinitely repeated groups to allow for
218        breaking an infinite loop) was broken when the group contained a repeated
219        negated single-character class with a character that occupied more than one
220        data item and had a minimum repetition of zero (for example, [^\x{100}]* in
221        UTF-8 mode). The effect was undefined: the group might or might not be
222        deemed as matching an empty string, or the program might have crashed.
224    7.  The code for checking whether a group could match an empty string was not
225        recognizing that \h, \H, \v, \V, and \R must match a character.
227    8.  Implemented PCRE_INFO_MATCH_EMPTY, which yields 1 if the pattern can match
228        an empty string. If it can, pcretest shows this in its information output.
230    9.  Fixed two related bugs that applied to Unicode extended grapheme clusters
231        that were repeated with a maximizing qualifier (e.g. \X* or \X{2,5}) when
232        matched by pcre_exec() without using JIT:
234        (a) If the rest of the pattern did not match after a maximal run of
235            grapheme clusters, the code for backing up to try with fewer of them
236            did not always back up over a full grapheme when characters that do not
237            have the modifier quality were involved, e.g. Hangul syllables.
239        (b) If the match point in a subject started with modifier character, and
240            there was no match, the code could incorrectly back up beyond the match
241            point, and potentially beyond the first character in the subject,
242            leading to a segfault or an incorrect match result.
244    10. A conditional group with an assertion condition could lead to PCRE
245        recording an incorrect first data item for a match if no other first data
246        item was recorded. For example, the pattern (?(?=ab)ab) recorded "a" as a
247        first data item, and therefore matched "ca" after "c" instead of at the
248        start.
250    11. Change 40 for 8.33 (allowing pcregrep to find empty strings) showed up a
251        bug that caused the command "echo a | ./pcregrep -M '|a'" to loop.
253    12. The source of pcregrep now includes z/OS-specific code so that it can be
254        compiled for z/OS as part of the special z/OS distribution.
256    13. Added the -T and -TM options to pcretest.
258    14. The code in pcre_compile.c for creating the table of named capturing groups
259        has been refactored. Instead of creating the table dynamically during the
260        actual compiling pass, the information is remembered during the pre-compile
261        pass (on the stack unless there are more than 20 named groups, in which
262        case malloc() is used) and the whole table is created before the actual
263        compile happens. This has simplified the code (it is now nearly 150 lines
264        shorter) and prepared the way for better handling of references to groups
265        with duplicate names.
267    15. A back reference to a named subpattern when there is more than one of the
268        same name now checks them in the order in which they appear in the pattern.
269        The first one that is set is used for the reference. Previously only the
270        first one was inspected. This change makes PCRE more compatible with Perl.
272    16. Unicode character properties were updated from Unicode 6.3.0.
274    17. The compile-time code for auto-possessification has been refactored, based
275        on a patch by Zoltan Herczeg. It now happens after instead of during
276        compilation. The code is cleaner, and more cases are handled. The option
277        PCRE_NO_AUTO_POSSESS is added for testing purposes, and the -O and /O
278        options in pcretest are provided to set it. It can also be set by
279        (*NO_AUTO_POSSESS) at the start of a pattern.
281    18. The character VT has been added to the default ("C" locale) set of
282        characters that match \s and are generally treated as white space,
283        following this same change in Perl 5.18. There is now no difference between
284        "Perl space" and "POSIX space". Whether VT is treated as white space in
285        other locales depends on the locale.
287    19. The code for checking named groups as conditions, either for being set or
288        for being recursed, has been refactored (this is related to 14 and 15
289        above). Processing unduplicated named groups should now be as fast at
290        numerical groups, and processing duplicated groups should be faster than
291        before.
293    20. Two patches to the CMake build system, by Alexander Barkov:
295          (1) Replace the "source" command by "." in CMakeLists.txt because
296              "source" is a bash-ism.
298          (2) Add missing HAVE_STDINT_H and HAVE_INTTYPES_H to config-cmake.h.in;
299              without these the CMake build does not work on Solaris.
301    21. Perl has changed its handling of \8 and \9. If there is no previously
302        encountered capturing group of those numbers, they are treated as the
303        literal characters 8 and 9 instead of a binary zero followed by the
304        literals. PCRE now does the same.
306    22. Following Perl, added \o{} to specify codepoints in octal, making it
307        possible to specify values greater than 0777 and also making them
308        unambiguous.
310    23. Perl now gives an error for missing closing braces after \x{... instead of
311        treating the string as literal. PCRE now does the same.
313    24. RunTest used to grumble if an inappropriate test was selected explicitly,
314        but just skip it when running all tests. This make it awkward to run ranges
315        of tests when one of them was inappropriate. Now it just skips any
316        inappropriate tests, as it always did when running all tests.
318    25. If PCRE_AUTO_CALLOUT and PCRE_UCP were set for a pattern that contained
319        character types such as \d or \w, too many callouts were inserted, and the
320        data that they returned was rubbish.
322    26. In UCP mode, \s was not matching two of the characters that Perl matches,
323        namely NEL (U+0085) and MONGOLIAN VOWEL SEPARATOR (U+180E), though they
324        were matched by \h. The code has now been refactored so that the lists of
325        the horizontal and vertical whitespace characters used for \h and \v (which
326        are defined only in one place) are now also used for \s.
328    27. Add JIT support for the 64 bit TileGX architecture.
329        Patch by Jiong Wang (Tilera Corporation).
331    28. Possessive quantifiers for classes (both explicit and automatically
332        generated) now use special opcodes instead of wrapping in ONCE brackets.
334    29. Whereas an item such as A{4}+ ignored the possessivenes of the quantifier
335        (because it's meaningless), this was not happening when PCRE_CASELESS was
336        set. Not wrong, but inefficient.
338    30. Updated perltest.pl to add /u (force Unicode mode) when /W (use Unicode
339        properties for \w, \d, etc) is present in a test regex. Otherwise if the
340        test contains no characters greater than 255, Perl doesn't realise it
341        should be using Unicode semantics.
343    31. Upgraded the handling of the POSIX classes [:graph:], [:print:], and
344        [:punct:] when PCRE_UCP is set so as to include the same characters as Perl
345        does in Unicode mode.
347    32. Added the "forbid" facility to pcretest so that putting tests into the
348        wrong test files can sometimes be quickly detected.
350    33. There is now a limit (default 250) on the depth of nesting of parentheses.
351        This limit is imposed to control the amount of system stack used at compile
352        time. It can be changed at build time by --with-parens-nest-limit=xxx or
353        the equivalent in CMake.
355    34. Character classes such as [A-\d] or [a-[:digit:]] now cause compile-time
356        errors. Perl warns for these when in warning mode, but PCRE has no facility
357        for giving warnings.
359    35. Change 34 for 8.13 allowed quantifiers on assertions, because Perl does.
360        However, this was not working for (?!) because it is optimized to (*FAIL),
361        for which PCRE does not allow quantifiers. The optimization is now disabled
362        when a quantifier follows (?!). I can't see any use for this, but it makes
363        things uniform.
365    36. Perl no longer allows group names to start with digits, so I have made this
366        change also in PCRE. It simplifies the code a bit.
368    37. In extended mode, Perl ignores spaces before a + that indicates a
369        possessive quantifier. PCRE allowed a space before the quantifier, but not
370        before the possessive +. It now does.
372    38. The use of \K (reset reported match start) within a repeated possessive
373        group such as (a\Kb)*+ was not working.
375    40. Document that the same character tables must be used at compile time and
376        run time, and that the facility to pass tables to pcre_exec() and
377        pcre_dfa_exec() is for use only with saved/restored patterns.
379    41. Applied Jeff Trawick's patch CMakeLists.txt, which "provides two new
380        features for Builds with MSVC:
382        1. Support pcre.rc and/or pcreposix.rc (as is already done for MinGW
383           builds). The .rc files can be used to set FileDescription and many other
384           attributes.
386        2. Add an option (-DINSTALL_MSVC_PDB) to enable installation of .pdb files.
387           This allows higher-level build scripts which want .pdb files to avoid
388           hard-coding the exact files needed."
390    42. Added support for [[:<:]] and [[:>:]] as used in the BSD POSIX library to
391        mean "start of word" and "end of word", respectively, as a transition aid.
393    43. A minimizing repeat of a class containing codepoints greater than 255 in
394        non-UTF 16-bit or 32-bit modes caused an internal error when PCRE was
395        compiled to use the heap for recursion.
397    44. Got rid of some compiler warnings for unused variables when UTF but not UCP
398        is configured.
401    Version 8.33 28-May-2013
402    ------------------------
404    1.  Added 'U' to some constants that are compared to unsigned integers, to
405        avoid compiler signed/unsigned warnings. Added (int) casts to unsigned
406        variables that are added to signed variables, to ensure the result is
407        signed and can be negated.
409    2.  Applied patch by Daniel Richard G for quashing MSVC warnings to the
410        CMake config files.
412    3.  Revise the creation of config.h.generic so that all boolean macros are
413        #undefined, whereas non-boolean macros are #ifndef/#endif-ed. This makes
414        overriding via -D on the command line possible.
416    4.  Changing the definition of the variable "op" in pcre_exec.c from pcre_uchar
417        to unsigned int is reported to make a quite noticeable speed difference in
418        a specific Windows environment. Testing on Linux did also appear to show
419        some benefit (and it is clearly not harmful). Also fixed the definition of
420        Xop which should be unsigned.
422    5.  Related to (4), changing the definition of the intermediate variable cc
423        in repeated character loops from pcre_uchar to pcre_uint32 also gave speed
424        improvements.
426    6.  Fix forward search in JIT when link size is 3 or greater. Also removed some
427        unnecessary spaces.
429    7.  Adjust autogen.sh and configure.ac to lose warnings given by automake 1.12
430        and later.
432    8.  Fix two buffer over read issues in 16 and 32 bit modes. Affects JIT only.
434    9.  Optimizing fast_forward_start_bits in JIT.
436    10. Adding support for callouts in JIT, and fixing some issues revealed
437        during this work. Namely:
439        (a) Unoptimized capturing brackets incorrectly reset on backtrack.
441        (b) Minimum length was not checked before the matching is started.
443    11. The value of capture_last that is passed to callouts was incorrect in some
444        cases when there was a capture on one path that was subsequently abandoned
445        after a backtrack. Also, the capture_last value is now reset after a
446        recursion, since all captures are also reset in this case.
448    12. The interpreter no longer returns the "too many substrings" error in the
449        case when an overflowing capture is in a branch that is subsequently
450        abandoned after a backtrack.
452    13. In the pathological case when an offset vector of size 2 is used, pcretest
453        now prints out the matched string after a yield of 0 or 1.
455    14. Inlining subpatterns in recursions, when certain conditions are fulfilled.
456        Only supported by the JIT compiler at the moment.
458    15. JIT compiler now supports 32 bit Macs thanks to Lawrence Velazquez.
460    16. Partial matches now set offsets[2] to the "bumpalong" value, that is, the
461        offset of the starting point of the matching process, provided the offsets
462        vector is large enough.
464    17. The \A escape now records a lookbehind value of 1, though its execution
465        does not actually inspect the previous character. This is to ensure that,
466        in partial multi-segment matching, at least one character from the old
467        segment is retained when a new segment is processed. Otherwise, if there
468        are no lookbehinds in the pattern, \A might match incorrectly at the start
469        of a new segment.
471    18. Added some #ifdef __VMS code into pcretest.c to help VMS implementations.
473    19. Redefined some pcre_uchar variables in pcre_exec.c as pcre_uint32; this
474        gives some modest performance improvement in 8-bit mode.
476    20. Added the PCRE-specific property \p{Xuc} for matching characters that can
477        be expressed in certain programming languages using Universal Character
478        Names.
480    21. Unicode validation has been updated in the light of Unicode Corrigendum #9,
481        which points out that "non characters" are not "characters that may not
482        appear in Unicode strings" but rather "characters that are reserved for
483        internal use and have only local meaning".
485    22. When a pattern was compiled with automatic callouts (PCRE_AUTO_CALLOUT) and
486        there was a conditional group that depended on an assertion, if the
487        assertion was false, the callout that immediately followed the alternation
488        in the condition was skipped when pcre_exec() was used for matching.
490    23. Allow an explicit callout to be inserted before an assertion that is the
491        condition for a conditional group, for compatibility with automatic
492        callouts, which always insert a callout at this point.
494    24. In 8.31, (*COMMIT) was confined to within a recursive subpattern. Perl also
495        confines (*SKIP) and (*PRUNE) in the same way, and this has now been done.
497    25. (*PRUNE) is now supported by the JIT compiler.
499    26. Fix infinite loop when /(?<=(*SKIP)ac)a/ is matched against aa.
501    27. Fix the case where there are two or more SKIPs with arguments that may be
502        ignored.
504    28. (*SKIP) is now supported by the JIT compiler.
506    29. (*THEN) is now supported by the JIT compiler.
508    30. Update RunTest with additional test selector options.
510    31. The way PCRE handles backtracking verbs has been changed in two ways.
512        (1) Previously, in something like (*COMMIT)(*SKIP), COMMIT would override
513        SKIP. Now, PCRE acts on whichever backtracking verb is reached first by
514        backtracking. In some cases this makes it more Perl-compatible, but Perl's
515        rather obscure rules do not always do the same thing.
517        (2) Previously, backtracking verbs were confined within assertions. This is
518        no longer the case for positive assertions, except for (*ACCEPT). Again,
519        this sometimes improves Perl compatibility, and sometimes does not.
521    32. A number of tests that were in test 2 because Perl did things differently
522        have been moved to test 1, because either Perl or PCRE has changed, and
523        these tests are now compatible.
525    32. Backtracking control verbs are now handled in the same way in JIT and
526        interpreter.
528    33. An opening parenthesis in a MARK/PRUNE/SKIP/THEN name in a pattern that
529        contained a forward subroutine reference caused a compile error.
531    34. Auto-detect and optimize limited repetitions in JIT.
533    35. Implement PCRE_NEVER_UTF to lock out the use of UTF, in particular,
534        blocking (*UTF) etc.
536    36. In the interpreter, maximizing pattern repetitions for characters and
537        character types now use tail recursion, which reduces stack usage.
539    37. The value of the max lookbehind was not correctly preserved if a compiled
540        and saved regex was reloaded on a host of different endianness.
542    38. Implemented (*LIMIT_MATCH) and (*LIMIT_RECURSION). As part of the extension
543        of the compiled pattern block, expand the flags field from 16 to 32 bits
544        because it was almost full.
546    39. Try madvise first before posix_madvise.
548    40. Change 7 for PCRE 7.9 made it impossible for pcregrep to find empty lines
549        with a pattern such as ^$. It has taken 4 years for anybody to notice! The
550        original change locked out all matches of empty strings. This has been
551        changed so that one match of an empty string per line is recognized.
552        Subsequent searches on the same line (for colouring or for --only-matching,
553        for example) do not recognize empty strings.
555    41. Applied a user patch to fix a number of spelling mistakes in comments.
557    42. Data lines longer than 65536 caused pcretest to crash.
559    43. Clarified the data type for length and startoffset arguments for pcre_exec
560        and pcre_dfa_exec in the function-specific man pages, where they were
561        explicitly stated to be in bytes, never having been updated. I also added
562        some clarification to the pcreapi man page.
564    44. A call to pcre_dfa_exec() with an output vector size less than 2 caused
565        a segmentation fault.
568    Version 8.32 30-November-2012
569    -----------------------------
571    1.  Improved JIT compiler optimizations for first character search and single
572        character iterators.
574    2.  Supporting IBM XL C compilers for PPC architectures in the JIT compiler.
575        Patch by Daniel Richard G.
577    3.  Single character iterator optimizations in the JIT compiler.
579    4.  Improved JIT compiler optimizations for character ranges.
581    5.  Rename the "leave" variable names to "quit" to improve WinCE compatibility.
582        Reported by Giuseppe D'Angelo.
584    6.  The PCRE_STARTLINE bit, indicating that a match can occur only at the start
585        of a line, was being set incorrectly in cases where .* appeared inside
586        atomic brackets at the start of a pattern, or where there was a subsequent
587        *PRUNE or *SKIP.
589    7.  Improved instruction cache flush for POWER/PowerPC.
590        Patch by Daniel Richard G.
592    8.  Fixed a number of issues in pcregrep, making it more compatible with GNU
593        grep:
595        (a) There is now no limit to the number of patterns to be matched.
597        (b) An error is given if a pattern is too long.
599        (c) Multiple uses of --exclude, --exclude-dir, --include, and --include-dir
600            are now supported.
602        (d) --exclude-from and --include-from (multiple use) have been added.
604        (e) Exclusions and inclusions now apply to all files and directories, not
605            just to those obtained from scanning a directory recursively.
607        (f) Multiple uses of -f and --file-list are now supported.
609        (g) In a Windows environment, the default for -d has been changed from
610            "read" (the GNU grep default) to "skip", because otherwise the presence
611            of a directory in the file list provokes an error.
613        (h) The documentation has been revised and clarified in places.
615    9.  Improve the matching speed of capturing brackets.
617    10. Changed the meaning of \X so that it now matches a Unicode extended
618        grapheme cluster.
620    11. Patch by Daniel Richard G to the autoconf files to add a macro for sorting
621        out POSIX threads when JIT support is configured.
623    12. Added support for PCRE_STUDY_EXTRA_NEEDED.
625    13. In the POSIX wrapper regcomp() function, setting re_nsub field in the preg
626        structure could go wrong in environments where size_t is not the same size
627        as int.
629    14. Applied user-supplied patch to pcrecpp.cc to allow PCRE_NO_UTF8_CHECK to be
630        set.
632    15. The EBCDIC support had decayed; later updates to the code had included
633        explicit references to (e.g.) \x0a instead of CHAR_LF. There has been a
634        general tidy up of EBCDIC-related issues, and the documentation was also
635        not quite right. There is now a test that can be run on ASCII systems to
636        check some of the EBCDIC-related things (but is it not a full test).
638    16. The new PCRE_STUDY_EXTRA_NEEDED option is now used by pcregrep, resulting
639        in a small tidy to the code.
641    17. Fix JIT tests when UTF is disabled and both 8 and 16 bit mode are enabled.
643    18. If the --only-matching (-o) option in pcregrep is specified multiple
644        times, each one causes appropriate output. For example, -o1 -o2 outputs the
645        substrings matched by the 1st and 2nd capturing parentheses. A separating
646        string can be specified by --om-separator (default empty).
648    19. Improving the first n character searches.
650    20. Turn case lists for horizontal and vertical white space into macros so that
651        they are defined only once.
653    21. This set of changes together give more compatible Unicode case-folding
654        behaviour for characters that have more than one other case when UCP
655        support is available.
657        (a) The Unicode property table now has offsets into a new table of sets of
658            three or more characters that are case-equivalent. The MultiStage2.py
659            script that generates these tables (the pcre_ucd.c file) now scans
660            CaseFolding.txt instead of UnicodeData.txt for character case
661            information.
663        (b) The code for adding characters or ranges of characters to a character
664            class has been abstracted into a generalized function that also handles
665            case-independence. In UTF-mode with UCP support, this uses the new data
666            to handle characters with more than one other case.
668        (c) A bug that is fixed as a result of (b) is that codepoints less than 256
669            whose other case is greater than 256 are now correctly matched
670            caselessly. Previously, the high codepoint matched the low one, but not
671            vice versa.
673        (d) The processing of \h, \H, \v, and \ in character classes now makes use
674            of the new class addition function, using character lists defined as
675            macros alongside the case definitions of 20 above.
677        (e) Caseless back references now work with characters that have more than
678            one other case.
680        (f) General caseless matching of characters with more than one other case
681            is supported.
683    22. Unicode character properties were updated from Unicode 6.2.0
685    23. Improved CMake support under Windows. Patch by Daniel Richard G.
687    24. Add support for 32-bit character strings, and UTF-32
689    25. Major JIT compiler update (code refactoring and bugfixing).
690        Experimental Sparc 32 support is added.
692    26. Applied a modified version of Daniel Richard G's patch to create
693        pcre.h.generic and config.h.generic by "make" instead of in the
694        PrepareRelease script.
696    27. Added a definition for CHAR_NULL (helpful for the z/OS port), and use it in
697        pcre_compile.c when checking for a zero character.
699    28. Introducing a native interface for JIT. Through this interface, the compiled
700        machine code can be directly executed. The purpose of this interface is to
701        provide fast pattern matching, so several sanity checks are not performed.
702        However, feature tests are still performed. The new interface provides
703        1.4x speedup compared to the old one.
705    29. If pcre_exec() or pcre_dfa_exec() was called with a negative value for
706        the subject string length, the error given was PCRE_ERROR_BADOFFSET, which
707        was confusing. There is now a new error PCRE_ERROR_BADLENGTH for this case.
709    30. In 8-bit UTF-8 mode, pcretest failed to give an error for data codepoints
710        greater than 0x7fffffff (which cannot be represented in UTF-8, even under
711        the "old" RFC 2279). Instead, it ended up passing a negative length to
712        pcre_exec().
714    31. Add support for GCC's visibility feature to hide internal functions.
716    32. Running "pcretest -C pcre8" or "pcretest -C pcre16" gave a spurious error
717        "unknown -C option" after outputting 0 or 1.
719    33. There is now support for generating a code coverage report for the test
720        suite in environments where gcc is the compiler and lcov is installed. This
721        is mainly for the benefit of the developers.
723    34. If PCRE is built with --enable-valgrind, certain memory regions are marked
724        unaddressable using valgrind annotations, allowing valgrind to detect
725        invalid memory accesses. This is mainly for the benefit of the developers.
727    25. (*UTF) can now be used to start a pattern in any of the three libraries.
729    26. Give configure error if --enable-cpp but no C++ compiler found.
732    Version 8.31 06-July-2012
733    -------------------------
735    1.  Fixing a wrong JIT test case and some compiler warnings.
737    2.  Removed a bashism from the RunTest script.
739    3.  Add a cast to pcre_exec.c to fix the warning "unary minus operator applied
740        to unsigned type, result still unsigned" that was given by an MS compiler
741        on encountering the code "-sizeof(xxx)".
743    4.  Partial matching support is added to the JIT compiler.
745    5.  Fixed several bugs concerned with partial matching of items that consist
746        of more than one character:
748        (a) /^(..)\1/ did not partially match "aba" because checking references was
749            done on an "all or nothing" basis. This also applied to repeated
750            references.
752        (b) \R did not give a hard partial match if \r was found at the end of the
753            subject.
755        (c) \X did not give a hard partial match after matching one or more
756            characters at the end of the subject.
758        (d) When newline was set to CRLF, a pattern such as /a$/ did not recognize
759            a partial match for the string "\r".
761        (e) When newline was set to CRLF, the metacharacter "." did not recognize
762            a partial match for a CR character at the end of the subject string.
764    6.  If JIT is requested using /S++ or -s++ (instead of just /S+ or -s+) when
765        running pcretest, the text "(JIT)" added to the output whenever JIT is
766        actually used to run the match.
768    7.  Individual JIT compile options can be set in pcretest by following -s+[+]
769        or /S+[+] with a digit between 1 and 7.
771    8.  OP_NOT now supports any UTF character not just single-byte ones.
773    9.  (*MARK) control verb is now supported by the JIT compiler.
775    10. The command "./RunTest list" lists the available tests without actually
776        running any of them. (Because I keep forgetting what they all are.)
780    12. Applied a (slightly modified) user-supplied patch that improves performance
781        when the heap is used for recursion (compiled with --disable-stack-for-
782        recursion). Instead of malloc and free for each heap frame each time a
783        logical recursion happens, frames are retained on a chain and re-used where
784        possible. This sometimes gives as much as 30% improvement.
786    13. As documented, (*COMMIT) is now confined to within a recursive subpattern
787        call.
789    14. As documented, (*COMMIT) is now confined to within a positive assertion.
791    15. It is now possible to link pcretest with libedit as an alternative to
792        libreadline.
794    16. (*COMMIT) control verb is now supported by the JIT compiler.
796    17. The Unicode data tables have been updated to Unicode 6.1.0.
798    18. Added --file-list option to pcregrep.
800    19. Added binary file support to pcregrep, including the -a, --binary-files,
801        -I, and --text options.
803    20. The madvise function is renamed for posix_madvise for QNX compatibility
804        reasons. Fixed by Giuseppe D'Angelo.
806    21. Fixed a bug for backward assertions with REVERSE 0 in the JIT compiler.
808    22. Changed the option for creating symbolic links for 16-bit man pages from
809        -s to -sf so that re-installing does not cause issues.
811    23. Support PCRE_NO_START_OPTIMIZE in JIT as (*MARK) support requires it.
813    24. Fixed a very old bug in pcretest that caused errors with restarted DFA
814        matches in certain environments (the workspace was not being correctly
815        retained). Also added to pcre_dfa_exec() a simple plausibility check on
816        some of the workspace data at the beginning of a restart.
818    25. \s*\R was auto-possessifying the \s* when it should not, whereas \S*\R
819        was not doing so when it should - probably a typo introduced by SVN 528
820        (change 8.10/14).
822    26. When PCRE_UCP was not set, \w+\x{c4} was incorrectly auto-possessifying the
823        \w+ when the character tables indicated that \x{c4} was a word character.
824        There were several related cases, all because the tests for doing a table
825        lookup were testing for characters less than 127 instead of 255.
827    27. If a pattern contains capturing parentheses that are not used in a match,
828        their slots in the ovector are set to -1. For those that are higher than
829        any matched groups, this happens at the end of processing. In the case when
830        there were back references that the ovector was too small to contain
831        (causing temporary malloc'd memory to be used during matching), and the
832        highest capturing number was not used, memory off the end of the ovector
833        was incorrectly being set to -1. (It was using the size of the temporary
834        memory instead of the true size.)
836    28. To catch bugs like 27 using valgrind, when pcretest is asked to specify an
837        ovector size, it uses memory at the end of the block that it has got.
839    29. Check for an overlong MARK name and give an error at compile time. The
840        limit is 255 for the 8-bit library and 65535 for the 16-bit library.
842    30. JIT compiler update.
844    31. JIT is now supported on jailbroken iOS devices. Thanks for Ruiger
845        Rill for the patch.
847    32. Put spaces around SLJIT_PRINT_D in the JIT compiler. Required by CXX11.
849    33. Variable renamings in the PCRE-JIT compiler. No functionality change.
851    34. Fixed typos in pcregrep: in two places there was SUPPORT_LIBZ2 instead of
852        SUPPORT_LIBBZ2. This caused a build problem when bzip2 but not gzip (zlib)
853        was enabled.
855    35. Improve JIT code generation for greedy plus quantifier.
857    36. When /((?:a?)*)*c/ or /((?>a?)*)*c/ was matched against "aac", it set group
858        1 to "aa" instead of to an empty string. The bug affected repeated groups
859        that could potentially match an empty string.
861    37. Optimizing single character iterators in JIT.
863    38. Wide characters specified with \uxxxx in JavaScript mode are now subject to
864        the same checks as \x{...} characters in non-JavaScript mode. Specifically,
865        codepoints that are too big for the mode are faulted, and in a UTF mode,
866        disallowed codepoints are also faulted.
868    39. If PCRE was compiled with UTF support, in three places in the DFA
869        matcher there was code that should only have been obeyed in UTF mode, but
870        was being obeyed unconditionally. In 8-bit mode this could cause incorrect
871        processing when bytes with values greater than 127 were present. In 16-bit
872        mode the bug would be provoked by values in the range 0xfc00 to 0xdc00. In
873        both cases the values are those that cannot be the first data item in a UTF
874        character. The three items that might have provoked this were recursions,
875        possessively repeated groups, and atomic groups.
877    40. Ensure that libpcre is explicitly listed in the link commands for pcretest
878        and pcregrep, because some OS require shared objects to be explicitly
879        passed to ld, causing the link step to fail if they are not.
881    41. There were two incorrect #ifdefs in pcre_study.c, meaning that, in 16-bit
882        mode, patterns that started with \h* or \R* might be incorrectly matched.
885    Version 8.30 04-February-2012
886    -----------------------------
888    1.  Renamed "isnumber" as "is_a_number" because in some Mac environments this
889        name is defined in ctype.h.
891    2.  Fixed a bug in fixed-length calculation for lookbehinds that would show up
892        only in quite long subpatterns.
894    3.  Removed the function pcre_info(), which has been obsolete and deprecated
895        since it was replaced by pcre_fullinfo() in February 2000.
897    4.  For a non-anchored pattern, if (*SKIP) was given with a name that did not
898        match a (*MARK), and the match failed at the start of the subject, a
899        reference to memory before the start of the subject could occur. This bug
900        was introduced by fix 17 of release 8.21.
902    5.  A reference to an unset group with zero minimum repetition was giving
903        totally wrong answers (in non-JavaScript-compatibility mode). For example,
904        /(another)?(\1?)test/ matched against "hello world test". This bug was
905        introduced in release 8.13.
907    6.  Add support for 16-bit character strings (a large amount of work involving
908        many changes and refactorings).
910    7.  RunGrepTest failed on msys because \r\n was replaced by whitespace when the
911        command "pattern=`printf 'xxx\r\njkl'`" was run. The pattern is now taken
912        from a file.
914    8.  Ovector size of 2 is also supported by JIT based pcre_exec (the ovector size
915        rounding is not applied in this particular case).
917    9.  The invalid Unicode surrogate codepoints U+D800 to U+DFFF are now rejected
918        if they appear, or are escaped, in patterns.
920    10. Get rid of a number of -Wunused-but-set-variable warnings.
922    11. The pattern /(?=(*:x))(q|)/ matches an empty string, and returns the mark
923        "x". The similar pattern /(?=(*:x))((*:y)q|)/ did not return a mark at all.
924        Oddly, Perl behaves the same way. PCRE has been fixed so that this pattern
925        also returns the mark "x". This bug applied to capturing parentheses,
926        non-capturing parentheses, and atomic parentheses. It also applied to some
927        assertions.
929    12. Stephen Kelly's patch to CMakeLists.txt allows it to parse the version
930        information out of configure.ac instead of relying on pcre.h.generic, which
931        is not stored in the repository.
933    13. Applied Dmitry V. Levin's patch for a more portable method for linking with
934        -lreadline.
936    14. ZH added PCRE_CONFIG_JITTARGET; added its output to pcretest -C.
938    15. Applied Graycode's patch to put the top-level frame on the stack rather
939        than the heap when not using the stack for recursion. This gives a
940        performance improvement in many cases when recursion is not deep.
942    16. Experimental code added to "pcretest -C" to output the stack frame size.
945    Version 8.21 12-Dec-2011
946    ------------------------
948    1.  Updating the JIT compiler.
950    2.  JIT compiler now supports OP_NCREF, OP_RREF and OP_NRREF. New test cases
951        are added as well.
953    3.  Fix cache-flush issue on PowerPC (It is still an experimental JIT port).
954        PCRE_EXTRA_TABLES is not suported by JIT, and should be checked before
955        calling _pcre_jit_exec. Some extra comments are added.
957    4.  (*MARK) settings inside atomic groups that do not contain any capturing
958        parentheses, for example, (?>a(*:m)), were not being passed out. This bug
959        was introduced by change 18 for 8.20.
961    5.  Supporting of \x, \U and \u in JavaScript compatibility mode based on the
962        ECMA-262 standard.
964    6.  Lookbehinds such as (?<=a{2}b) that contained a fixed repetition were
965        erroneously being rejected as "not fixed length" if PCRE_CASELESS was set.
966        This bug was probably introduced by change 9 of 8.13.
968    7.  While fixing 6 above, I noticed that a number of other items were being
969        incorrectly rejected as "not fixed length". This arose partly because newer
970        opcodes had not been added to the fixed-length checking code. I have (a)
971        corrected the bug and added tests for these items, and (b) arranged for an
972        error to occur if an unknown opcode is encountered while checking for fixed
973        length instead of just assuming "not fixed length". The items that were
974        rejected were: (*ACCEPT), (*COMMIT), (*FAIL), (*MARK), (*PRUNE), (*SKIP),
975        (*THEN), \h, \H, \v, \V, and single character negative classes with fixed
976        repetitions, e.g. [^a]{3}, with and without PCRE_CASELESS.
978    8.  A possessively repeated conditional subpattern such as (?(?=c)c|d)++ was
979        being incorrectly compiled and would have given unpredicatble results.
981    9.  A possessively repeated subpattern with minimum repeat count greater than
982        one behaved incorrectly. For example, (A){2,}+ behaved as if it was
983        (A)(A)++ which meant that, after a subsequent mismatch, backtracking into
984        the first (A) could occur when it should not.
986    10. Add a cast and remove a redundant test from the code.
988    11. JIT should use pcre_malloc/pcre_free for allocation.
990    12. Updated pcre-config so that it no longer shows -L/usr/lib, which seems
991        best practice nowadays, and helps with cross-compiling. (If the exec_prefix
992        is anything other than /usr, -L is still shown).
994    13. In non-UTF-8 mode, \C is now supported in lookbehinds and DFA matching.
996    14. Perl does not support \N without a following name in a [] class; PCRE now
997        also gives an error.
999    15. If a forward reference was repeated with an upper limit of around 2000,
1000        it caused the error "internal error: overran compiling workspace". The
1001        maximum number of forward references (including repeats) was limited by the
1002        internal workspace, and dependent on the LINK_SIZE. The code has been
1003        rewritten so that the workspace expands (via pcre_malloc) if necessary, and
1004        the default depends on LINK_SIZE. There is a new upper limit (for safety)
1005        of around 200,000 forward references. While doing this, I also speeded up
1006        the filling in of repeated forward references.
1008    16. A repeated forward reference in a pattern such as (a)(?2){2}(.) was
1009        incorrectly expecting the subject to contain another "a" after the start.
1011    17. When (*SKIP:name) is activated without a corresponding (*MARK:name) earlier
1012        in the match, the SKIP should be ignored. This was not happening; instead
1013        the SKIP was being treated as NOMATCH. For patterns such as
1014        /A(*MARK:A)A+(*SKIP:B)Z|AAC/ this meant that the AAC branch was never
1015        tested.
1017    18. The behaviour of (*MARK), (*PRUNE), and (*THEN) has been reworked and is
1018        now much more compatible with Perl, in particular in cases where the result
1019        is a non-match for a non-anchored pattern. For example, if
1020        /b(*:m)f|a(*:n)w/ is matched against "abc", the non-match returns the name
1021        "m", where previously it did not return a name. A side effect of this
1022        change is that for partial matches, the last encountered mark name is
1023        returned, as for non matches. A number of tests that were previously not
1024        Perl-compatible have been moved into the Perl-compatible test files. The
1025        refactoring has had the pleasing side effect of removing one argument from
1026        the match() function, thus reducing its stack requirements.
1028    19. If the /S+ option was used in pcretest to study a pattern using JIT,
1029        subsequent uses of /S (without +) incorrectly behaved like /S+.
1031    21. Retrieve executable code size support for the JIT compiler and fixing
1032        some warnings.
1034    22. A caseless match of a UTF-8 character whose other case uses fewer bytes did
1035        not work when the shorter character appeared right at the end of the
1036        subject string.
1038    23. Added some (int) casts to non-JIT modules to reduce warnings on 64-bit
1039        systems.
1041    24. Added PCRE_INFO_JITSIZE to pass on the value from (21) above, and also
1042        output it when the /M option is used in pcretest.
1044    25. The CheckMan script was not being included in the distribution. Also, added
1045        an explicit "perl" to run Perl scripts from the PrepareRelease script
1046        because this is reportedly needed in Windows.
1048    26. If study data was being save in a file and studying had not found a set of
1049        "starts with" bytes for the pattern, the data written to the file (though
1050        never used) was taken from uninitialized memory and so caused valgrind to
1051        complain.
1053    27. Updated RunTest.bat as provided by Sheri Pierce.
1055    28. Fixed a possible uninitialized memory bug in pcre_jit_compile.c.
1057    29. Computation of memory usage for the table of capturing group names was
1058        giving an unnecessarily large value.
1061    Version 8.20 21-Oct-2011
1062    ------------------------
1064    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
1065        a POSIX class. After further experiments with Perl, which convinced me that
1066        Perl has bugs and confusions, a closing square bracket is no longer allowed
1067        in a POSIX name. This bug also affected patterns with classes that started
1068        with full stops.
1070    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
1071        captured substring, but while checking the failing first alternative,
1072        substring 1 is temporarily captured. If the output vector supplied to
1073        pcre_exec() was not big enough for this capture, the yield of the function
1074        was still zero ("insufficient space for captured substrings"). This cannot
1075        be totally fixed without adding another stack variable, which seems a lot
1076        of expense for a edge case. However, I have improved the situation in cases
1077        such as /(a)(b)x|abc/ matched against "abc", where the return code
1078        indicates that fewer than the maximum number of slots in the ovector have
1079        been set.
1081    3.  Related to (2) above: when there are more back references in a pattern than
1082        slots in the output vector, pcre_exec() uses temporary memory during
1083        matching, and copies in the captures as far as possible afterwards. It was
1084        using the entire output vector, but this conflicts with the specification
1085        that only 2/3 is used for passing back captured substrings. Now it uses
1086        only the first 2/3, for compatibility. This is, of course, another edge
1087        case.
1089    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
1090        main code base, and can be used by building with --enable-jit. When this is
1091        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
1092        runtime --no-jit option is given.
1094    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
1095        ovector, the return from the function was zero, implying that there were
1096        other matches that did not fit. The correct "exactly full" value is now
1097        returned.
1099    6.  If a subpattern that was called recursively or as a subroutine contained
1100        (*PRUNE) or any other control that caused it to give a non-standard return,
1101        invalid errors such as "Error -26 (nested recursion at the same subject
1102        position)" or even infinite loops could occur.
1104    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
1105        computing the minimum length on reaching *ACCEPT, and so ended up with the
1106        wrong value of 1 rather than 0. Further investigation indicates that
1107        computing a minimum subject length in the presence of *ACCEPT is difficult
1108        (think back references, subroutine calls), and so I have changed the code
1109        so that no minimum is registered for a pattern that contains *ACCEPT.
1111    8.  If (*THEN) was present in the first (true) branch of a conditional group,
1112        it was not handled as intended. [But see 16 below.]
1114    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
1115        Sheri Pierce.
1117    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
1118        the first byte in a match must be "a".
1120    11. Change 17 for 8.13 increased the recursion depth for patterns like
1121        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
1122        pattern contains any instances of (*THEN). If it does not, the old
1123        optimizations are restored. It would be nice to do this on a per-group
1124        basis, but at the moment that is not feasible.
1126    12. In some environments, the output of pcretest -C is CRLF terminated. This
1127        broke RunTest's code that checks for the link size. A single white space
1128        character after the value is now allowed for.
1130    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
1131        For "fr", it uses the Windows-specific input and output files.
1133    14. If (*THEN) appeared in a group that was called recursively or as a
1134        subroutine, it did not work as intended. [But see next item.]
1136    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
1137        pattern fragments (but not containing any | characters). If A and B are
1138        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
1139        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
1140        D. In other words, Perl considers parentheses that do not contain any |
1141        characters to be part of a surrounding alternative, whereas PCRE was
1142        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
1143        differently. PCRE now behaves in the same way as Perl, except in the case
1144        of subroutine/recursion calls such as (?1) which have in any case always
1145        been different (but PCRE had them first :-).
1147    16. Related to 15 above: Perl does not treat the | in a conditional group as
1148        creating alternatives. Such a group is treated in the same way as an
1149        ordinary group without any | characters when processing (*THEN). PCRE has
1150        been changed to match Perl's behaviour.
1152    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
1153        RunGrepTest script failed.
1155    18. Change 22 for version 13 caused atomic groups to use more stack. This is
1156        inevitable for groups that contain captures, but it can lead to a lot of
1157        stack use in large patterns. The old behaviour has been restored for atomic
1158        groups that do not contain any capturing parentheses.
1160    19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not
1161        suppress the check for a minimum subject length at run time. (If it was
1162        given to pcre_exec() or pcre_dfa_exec() it did work.)
1164    20. Fixed an ASCII-dependent infelicity in pcretest that would have made it
1165        fail to work when decoding hex characters in data strings in EBCDIC
1166        environments.
1168    21. It appears that in at least one Mac OS environment, the isxdigit() function
1169        is implemented as a macro that evaluates to its argument more than once,
1170        contravening the C 90 Standard (I haven't checked a later standard). There
1171        was an instance in pcretest which caused it to go wrong when processing
1172        \x{...} escapes in subject strings. The has been rewritten to avoid using
1173        things like p++ in the argument of isxdigit().
1176    Version 8.13 16-Aug-2011
1177    ------------------------
1179    1.  The Unicode data tables have been updated to Unicode 6.0.0.
1181    2.  Two minor typos in pcre_internal.h have been fixed.
1183    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
1184        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
1185        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
1187    4.  There were a number of related bugs in the code for matching backrefences
1188        caselessly in UTF-8 mode when codes for the characters concerned were
1189        different numbers of bytes. For example, U+023A and U+2C65 are an upper
1190        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
1191        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
1192        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
1193        2-byte code at the end of the subject (it thought there wasn't enough data
1194        left).
1196    5.  Comprehensive information about what went wrong is now returned by
1197        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
1198        as the output vector has at least 2 elements. The offset of the start of
1199        the failing character and a reason code are placed in the vector.
1201    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
1202        now returned is for the first byte of the failing character, instead of the
1203        last byte inspected. This is an incompatible change, but I hope it is small
1204        enough not to be a problem. It makes the returned offset consistent with
1205        pcre_exec() and pcre_dfa_exec().
1207    7.  pcretest now gives a text phrase as well as the error number when
1208        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
1209        failure, the offset and reason code are output.
1211    8.  When \R was used with a maximizing quantifier it failed to skip backwards
1212        over a \r\n pair if the subsequent match failed. Instead, it just skipped
1213        back over a single character (\n). This seems wrong (because it treated the
1214        two characters as a single entity when going forwards), conflicts with the
1215        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
1216        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
1217        has been changed.
1219    9.  Some internal refactoring has changed the processing so that the handling
1220        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
1221        time (the PCRE_DOTALL option was changed this way some time ago: version
1222        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
1223        which was always a bit of a fudge. It also means that there is one less
1224        argument for the match() function, which reduces its stack requirements
1225        slightly. This change also fixes an incompatibility with Perl: the pattern
1226        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
1228    10. More internal refactoring has drastically reduced the number of recursive
1229        calls to match() for possessively repeated groups such as (abc)++ when
1230        using pcre_exec().
1232    11. While implementing 10, a number of bugs in the handling of groups were
1233        discovered and fixed:
1235        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
1236        (a|)*(?1) gave a compile-time internal error.
1237        ((a|)+)+  did not notice that the outer group could match an empty string.
1238        (^a|^)+   was not marked as anchored.
1239        (.*a|.*)+ was not marked as matching at start or after a newline.
1241    12. Yet more internal refactoring has removed another argument from the match()
1242        function. Special calls to this function are now indicated by setting a
1243        value in a variable in the "match data" data block.
1245    13. Be more explicit in pcre_study() instead of relying on "default" for
1246        opcodes that mean there is no starting character; this means that when new
1247        ones are added and accidentally left out of pcre_study(), testing should
1248        pick them up.
1250    14. The -s option of pcretest has been documented for ages as being an old
1251        synonym of -m (show memory usage). I have changed it to mean "force study
1252        for every regex", that is, assume /S for every regex. This is similar to -i
1253        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
1254        using it. It makes it easier to run collections of tests with and without
1255        study enabled, and thereby test pcre_study() more easily. All the standard
1256        tests are now run with and without -s (but some patterns can be marked as
1257        "never study" - see 20 below).
1259    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
1260        restoration of the capturing data to the outer values was not happening
1261        correctly.
1263    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
1264        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
1265        pattern had matched an empty string, and so incorrectly returned a no
1266        match.
1268    17. There was optimizing code for the last branch of non-capturing parentheses,
1269        and also for the obeyed branch of a conditional subexpression, which used
1270        tail recursion to cut down on stack usage. Unfortunately, now that there is
1271        the possibility of (*THEN) occurring in these branches, tail recursion is
1272        no longer possible because the return has to be checked for (*THEN). These
1273        two optimizations have therefore been removed. [But see 8.20/11 above.]
1275    18. If a pattern containing \R was studied, it was assumed that \R always
1276        matched two bytes, thus causing the minimum subject length to be
1277        incorrectly computed because \R can also match just one byte.
1279    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
1280        was incorrectly computed.
1282    20. If /S is present twice on a test pattern in pcretest input, it now
1283        *disables* studying, thereby overriding the use of -s on the command line
1284        (see 14 above). This is necessary for one or two tests to keep the output
1285        identical in both cases.
1287    21. When (*ACCEPT) was used in an assertion that matched an empty string and
1288        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
1290    22. When an atomic group that contained a capturing parenthesis was
1291        successfully matched, but the branch in which it appeared failed, the
1292        capturing was not being forgotten if a higher numbered group was later
1293        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
1294        group 1 to "a", when in fact it should be unset. This applied to multi-
1295        branched capturing and non-capturing groups, repeated or not, and also to
1296        positive assertions (capturing in negative assertions does not happen
1297        in PCRE) and also to nested atomic groups.
1299    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
1300        subject after a captured substring, to make it easier to tell which of a
1301        number of identical substrings has been captured.
1303    24. The way atomic groups are processed by pcre_exec() has been changed so that
1304        if they are repeated, backtracking one repetition now resets captured
1305        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
1306        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
1307        "aaa". Previously, it would have been "a". As part of this code
1308        refactoring, the way recursive calls are handled has also been changed.
1310    25. If an assertion condition captured any substrings, they were not passed
1311        back unless some other capturing happened later. For example, if
1312        (?(?=(a))a) was matched against "a", no capturing was returned.
1314    26. When studying a pattern that contained subroutine calls or assertions,
1315        the code for finding the minimum length of a possible match was handling
1316        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
1317        group 1 called group 2 while simultaneously a separate group 2 called group
1318        1). A stack overflow occurred in this case. I have fixed this by limiting
1319        the recursion depth to 10.
1321    27. Updated RunTest.bat in the distribution to the version supplied by Tom
1322        Fortmann. This supports explicit test numbers on the command line, and has
1323        argument validation and error reporting.
1325    28. An instance of \X with an unlimited repeat could fail if at any point the
1326        first character it looked at was a mark character.
1328    29. Some minor code refactoring concerning Unicode properties and scripts
1329        should reduce the stack requirement of match() slightly.
1331    30. Added the '=' option to pcretest to check the setting of unused capturing
1332        slots at the end of the pattern, which are documented as being -1, but are
1333        not included in the return count.
1335    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
1336        compiled something random. Now it gives a compile-time error (as does
1337        Perl).
1339    32. A *MARK encountered during the processing of a positive assertion is now
1340        recorded and passed back (compatible with Perl).
1342    33. If --only-matching or --colour was set on a pcregrep call whose pattern
1343        had alternative anchored branches, the search for a second match in a line
1344        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
1345        matched the line "0102" twice. The same bug affected patterns that started
1346        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
1347        twice.
1349    34. Previously, PCRE did not allow quantification of assertions. However, Perl
1350        does, and because of capturing effects, quantifying parenthesized
1351        assertions may at times be useful. Quantifiers are now allowed for
1352        parenthesized assertions.
1354    35. A minor code tidy in pcre_compile() when checking options for \R usage.
1356    36. \g was being checked for fancy things in a character class, when it should
1357        just be a literal "g".
1359    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
1360        appearance of a nested POSIX class supersedes an apparent external class.
1361        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
1362        unescaped square brackets may also appear as part of class names. For
1363        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
1364        more like Perl. (But see 8.20/1 above.)
1366    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
1367        was because it thought it was \N{name}, which is not supported).
1369    39. Add minix to OS list not supporting the -S option in pcretest.
1371    40. PCRE tries to detect cases of infinite recursion at compile time, but it
1372        cannot analyze patterns in sufficient detail to catch mutual recursions
1373        such as ((?1))((?2)). There is now a runtime test that gives an error if a
1374        subgroup is called recursively as a subpattern for a second time at the
1375        same position in the subject string. In previous releases this might have
1376        been caught by the recursion limit, or it might have run out of stack.
1378    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
1379        happen only once. PCRE was, however incorrectly giving a compile time error
1380        "recursive call could loop indefinitely" because it cannot analyze the
1381        pattern in sufficient detail. The compile time test no longer happens when
1382        PCRE is compiling a conditional subpattern, but actual runaway loops are
1383        now caught at runtime (see 40 above).
1385    42. It seems that Perl allows any characters other than a closing parenthesis
1386        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
1387        has been changed to be the same.
1389    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
1390        as not to get warnings when autogen.sh is called. Also changed
1391        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
1393    44. To help people who use pcregrep to scan files containing exceedingly long
1394        lines, the following changes have been made:
1396        (a) The default value of the buffer size parameter has been increased from
1397            8K to 20K. (The actual buffer used is three times this size.)
1399        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
1400            PCRE is built.
1402        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
1403            to be set at run time.
1405        (d) Numerical values in pcregrep options can be followed by K or M, for
1406            example --buffer-size=50K.
1408        (e) If a line being scanned overflows pcregrep's buffer, an error is now
1409            given and the return code is set to 2.
1411    45. Add a pointer to the latest mark to the callout data block.
1413    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
1414        partial match of an empty string instead of no match. This was specific to
1415        the use of ".".
1417    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
1418        complete match instead of a partial match. This bug was dependent on both
1419        the PCRE_UTF8 and PCRE_DOTALL options being set.
1421    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
1422        starting byte set, because \b was not being ignored.
1425    Version 8.12 15-Jan-2011
1426    ------------------------
1428    1.  Fixed some typos in the markup of the man pages, and wrote a script that
1429        checks for such things as part of the documentation building process.
1431    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
1432        --match-limit and --recursion-limit options (added for 8.11). In
1433        particular, this made one of the standard tests fail. (The integer value
1434        went into the wrong half of a long int.)
1436    3.  If the --colour option was given to pcregrep with -v (invert match), it
1437        did strange things, either producing crazy output, or crashing. It should,
1438        of course, ignore a request for colour when reporting lines that do not
1439        match.
1441    4.  Another pcregrep bug caused similar problems if --colour was specified with
1442        -M (multiline) and the pattern match finished with a line ending.
1444    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
1445        matched in multiline mode, the following line was shown as part of the
1446        match. This seems wrong, so I have changed it.
1448    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
1449        the check for further matches in the same line (so they could be coloured)
1450        to overrun the end of the current line. If another match was found, it was
1451        incorrectly shown (and then shown again when found in the next line).
1453    7.  If pcregrep was compiled under Windows, there was a reference to the
1454        function pcregrep_exit() before it was defined. I am assuming this was
1455        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
1456        reported by a user. I've moved the definition above the reference.
1459    Version 8.11 10-Dec-2010
1460    ------------------------
1462    1.  (*THEN) was not working properly if there were untried alternatives prior
1463        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
1464        backtracked to try for "b" instead of moving to the next alternative branch
1465        at the same level (in this case, to look for "c"). The Perl documentation
1466        is clear that when (*THEN) is backtracked onto, it goes to the "next
1467        alternative in the innermost enclosing group".
1469    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
1470        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
1471        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
1472        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
1473        (*THEN).
1475    3.  If \s appeared in a character class, it removed the VT character from
1476        the class, even if it had been included by some previous item, for example
1477        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
1478        of \s, but is part of the POSIX "space" class.)
1480    4.  A partial match never returns an empty string (because you can always
1481        match an empty string at the end of the subject); however the checking for
1482        an empty string was starting at the "start of match" point. This has been
1483        changed to the "earliest inspected character" point, because the returned
1484        data for a partial match starts at this character. This means that, for
1485        example, /(?<=abc)def/ gives a partial match for the subject "abc"
1486        (previously it gave "no match").
1488    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
1489        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
1490        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
1491        has an implication that the given string is incomplete (because a partial
1492        match is preferred over a full match). For this reason, these items now
1493        give a partial match in this situation. [Aside: previously, the one case
1494        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
1495        match rather than a full match, which was wrong by the old rules, but is
1496        now correct.]
1498    6.  There was a bug in the handling of #-introduced comments, recognized when
1499        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
1500        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
1501        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
1502        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
1503        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
1504        places in pcre_compile().
1506    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
1507        comments when looking ahead for named forward references to subpatterns,
1508        the only newline sequence it recognized was NL. It now handles newlines
1509        according to the set newline convention.
1511    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
1512        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
1513        cater for a lack of strerror(). These oversights have been fixed.
1515    9.  Added --match-limit and --recursion-limit to pcregrep.
1517    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
1519    11. When the -o option was used, pcregrep was setting a return code of 1, even
1520        when matches were found, and --line-buffered was not being honoured.
1522    12. Added an optional parentheses number to the -o and --only-matching options
1523        of pcregrep.
1525    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
1526        can match an empty string. The code to do it in pcretest and pcredemo
1527        needed fixing:
1529        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
1530            only one byte after an empty string match just before CRLF (this case
1531            just got forgotten; "any" and "anycrlf" were OK).
1533        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
1534            mode when an empty string match preceded an ASCII character followed by
1535            a non-ASCII character. (The code for advancing by one character rather
1536            than one byte was nonsense.)
1538        (c) The pcredemo.c sample program did not have any code at all to handle
1539            the cases when CRLF is a valid newline sequence.
1541    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
1542        as a starting offset was within the subject string. There is now a new
1543        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
1544        negative or greater than the length of the string. In order to test this,
1545        pcretest is extended to allow the setting of negative starting offsets.
1547    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
1548        starting offset points to the beginning of a UTF-8 character was
1549        unnecessarily clumsy. I tidied it up.
1551    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
1552        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
1554    17. Nobody had reported that the --include_dir option, which was added in
1555        release 7.7 should have been called --include-dir (hyphen, not underscore)
1556        for compatibility with GNU grep. I have changed it to --include-dir, but
1557        left --include_dir as an undocumented synonym, and the same for
1558        --exclude-dir, though that is not available in GNU grep, at least as of
1559        release 2.5.4.
1561    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
1562        characters from a string of bytes) have been redefined so as not to use
1563        loops, in order to improve performance in some environments. At the same
1564        time, I abstracted some of the common code into auxiliary macros to save
1565        repetition (this should not affect the compiled code).
1567    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
1568        compile-time error is now given if \c is not followed by an ASCII
1569        character, that is, a byte less than 128. (In EBCDIC mode, the code is
1570        different, and any byte value is allowed.)
1572    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
1573        START_OPTIMIZE option, which is now allowed at compile time - but just
1574        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
1575        to pcregrep and other applications that have no direct access to PCRE
1576        options. The new /Y option in pcretest sets this option when calling
1577        pcre_compile().
1579    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
1580        back references. Groups containing recursive back references were forced to
1581        be atomic by that change, but in the case of named groups, the amount of
1582        memory required was incorrectly computed, leading to "Failed: internal
1583        error: code overflow". This has been fixed.
1585    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
1586        pcretest.c, to avoid build problems in some Borland environments.
1589    Version 8.10 25-Jun-2010
1590    ------------------------
1592    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
1593        THEN.
1595    2.  (*ACCEPT) was not working when inside an atomic group.
1597    3.  Inside a character class, \B is treated as a literal by default, but
1598        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
1599        causes the error). The code is unchanged, but I tidied the documentation.
1601    4.  Inside a character class, PCRE always treated \R and \X as literals,
1602        whereas Perl faults them if its -w option is set. I have changed PCRE so
1603        that it faults them when PCRE_EXTRA is set.
1605    5.  Added support for \N, which always matches any character other than
1606        newline. (It is the same as "." when PCRE_DOTALL is not set.)
1608    6.  When compiling pcregrep with newer versions of gcc which may have
1609        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
1610        declared with attribute warn_unused_result" were given. Just casting the
1611        result to (void) does not stop the warnings; a more elaborate fudge is
1612        needed. I've used a macro to implement this.
1614    7.  Minor change to pcretest.c to avoid a compiler warning.
1616    8.  Added four artifical Unicode properties to help with an option to make
1617        \s etc use properties (see next item). The new properties are: Xan
1618        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
1620    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
1621        use Unicode properties. (*UCP) at the start of a pattern can be used to set
1622        this option. Modified pcretest to add /W to test this facility. Added
1623        REG_UCP to make it available via the POSIX interface.
1625    10. Added --line-buffered to pcregrep.
1627    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
1628        studied, and the match started with a letter with a code point greater than
1629        127 whose first byte was different to the first byte of the other case of
1630        the letter, the other case of this starting letter was not recognized
1631        (#976).
1633    12. If a pattern that was studied started with a repeated Unicode property
1634        test, for example, \p{Nd}+, there was the theoretical possibility of
1635        setting up an incorrect bitmap of starting bytes, but fortunately it could
1636        not have actually happened in practice until change 8 above was made (it
1637        added property types that matched character-matching opcodes).
1639    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
1640        possible starting bytes for non-anchored patterns.
1642    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
1643        \R, and also a number of cases that involve Unicode properties, both
1644        explicit and implicit when PCRE_UCP is set.
1646    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
1647        input, it could crash or give wrong results if characters with values
1648        greater than 0xc0 were present in the subject string. (Detail: it assumed
1649        UTF-8 input when processing these items.)
1651    16. Added a lot of (int) casts to avoid compiler warnings in systems where
1652        size_t is 64-bit (#991).
1654    17. Added a check for running out of memory when PCRE is compiled with
1655        --disable-stack-for-recursion (#990).
1657    18. If the last data line in a file for pcretest does not have a newline on
1658        the end, a newline was missing in the output.
1660    19. The default pcre_chartables.c file recognizes only ASCII characters (values
1661        less than 128) in its various bitmaps. However, there is a facility for
1662        generating tables according to the current locale when PCRE is compiled. It
1663        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
1664        characters, are recognized by isspace() and therefore were getting set in
1665        these tables, and indeed these tables seem to approximate to ISO 8859. This
1666        caused a problem in UTF-8 mode when pcre_study() was used to create a list
1667        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
1668        which of course cannot start UTF-8 characters. I have changed the code so
1669        that only real ASCII characters (less than 128) and the correct starting
1670        bytes for UTF-8 encodings are set for characters greater than 127 when in
1671        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
1672        altogether.)
1674    20. Added the /T option to pcretest so as to be able to run tests with non-
1675        standard character tables, thus making it possible to include the tests
1676        used for 19 above in the standard set of tests.
1678    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
1679        reference to a subpattern the other side of a comment that contains an
1680        opening parenthesis caused either an internal compiling error, or a
1681        reference to the wrong subpattern.
1684    Version 8.02 19-Mar-2010
1685    ------------------------
1687    1.  The Unicode data tables have been updated to Unicode 5.2.0.
1689    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
1690        configured.
1692    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
1693        original author of that file, following a query about its status.
1695    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
1696        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
1698    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
1699        quantifier applied to a forward-referencing subroutine call, could compile
1700        incorrect code or give the error "internal error: previously-checked
1701        referenced subpattern not found".
1703    6.  Both MS Visual Studio and Symbian OS have problems with initializing
1704        variables to point to external functions. For these systems, therefore,
1705        pcre_malloc etc. are now initialized to local functions that call the
1706        relevant global functions.
1708    7.  There were two entries missing in the vectors called coptable and poptable
1709        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
1710        I've fixed the data, and added a kludgy way of testing at compile time that
1711        the lengths are correct (equal to the number of opcodes).
1713    8.  Following on from 7, I added a similar kludge to check the length of the
1714        eint vector in pcreposix.c.
1716    9.  Error texts for pcre_compile() are held as one long string to avoid too
1717        much relocation at load time. To find a text, the string is searched,
1718        counting zeros. There was no check for running off the end of the string,
1719        which could happen if a new error number was added without updating the
1720        string.
1722    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
1724    11. \K was not working if it appeared in an atomic group or in a group that
1725        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
1726        \K is "not well defined" if used in an assertion. PCRE now accepts it if
1727        the assertion is positive, but not if it is negative.
1729    12. Change 11 fortuitously reduced the size of the stack frame used in the
1730        "match()" function of pcre_exec.c by one pointer. Forthcoming
1731        implementation of support for (*MARK) will need an extra pointer on the
1732        stack; I have reserved it now, so that the stack frame size does not
1733        decrease.
1735    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
1736        item in branch that calls a recursion is a subroutine call - as in the
1737        second branch in the above example - was incorrectly given the compile-
1738        time error "recursive call could loop indefinitely" because pcre_compile()
1739        was not correctly checking the subroutine for matching a non-empty string.
1741    14. The checks for overrunning compiling workspace could trigger after an
1742        overrun had occurred. This is a "should never occur" error, but it can be
1743        triggered by pathological patterns such as hundreds of nested parentheses.
1744        The checks now trigger 100 bytes before the end of the workspace.
1746    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
1749    Version 8.01 19-Jan-2010
1750    ------------------------
1752    1.  If a pattern contained a conditional subpattern with only one branch (in
1753        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
1754        computed the wrong minimum data length (which is of course zero for such
1755        subpatterns). This could cause incorrect "no match" results.
1757    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
1758        the pattern is reset in the first branch, pcre_compile() failed with
1759        "internal error: code overflow at offset...". This happened only when
1760        the reset was to the original external option setting. (An optimization
1761        abstracts leading options settings into an external setting, which was the
1762        cause of this.)
1764    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
1765        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
1766        assertion pattern did not match (meaning that the assertion was true), it
1767        was incorrectly treated as false if the SKIP had been reached during the
1768        matching. This also applied to assertions used as conditions.
1770    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
1771        assertion subpattern, including such a pattern used as a condition,
1772        unpredictable results occurred, instead of the error return
1775    5.  The C++ GlobalReplace function was not working like Perl for the special
1776        situation when an empty string is matched. It now does the fancy magic
1777        stuff that is necessary.
1779    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
1780        removed. (These were left over from very, very early versions of PCRE.)
1782    7.  Some cosmetic changes to the code to make life easier when compiling it
1783        as part of something else:
1785        (a) Change DEBUG to PCRE_DEBUG.
1787        (b) In pcre_compile(), rename the member of the "branch_chain" structure
1788            called "current" as "current_branch", to prevent a collision with the
1789            Linux macro when compiled as a kernel module.
1791        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
1792            prevent a collision with the Linux macro when compiled as a kernel
1793            module.
1795    8.  In pcre_compile() there are some checks for integer overflows that used to
1796        cast potentially large values to (double). This has been changed to that
1797        when building, a check for int64_t is made, and if it is found, it is used
1798        instead, thus avoiding the use of floating point arithmetic. (There is no
1799        other use of FP in PCRE.) If int64_t is not found, the fallback is to
1800        double.
1802    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
1803        2005 (difference between two addresses compared to an unsigned value).
1805    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
1806        custom one, because of the following reported problem in Windows:
1808          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
1809              under Win32.
1810          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
1811              therefore missing the function definition.
1812          - The compiler thus generates a "C" signature for the test function.
1813          - The linker fails to find the "C" function.
1814          - PCRE fails to configure if asked to do so against libbz2.
1816    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
1817        messages were output:
1819          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
1820          rerunning libtoolize, to keep the correct libtool macros in-tree.
1821          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
1823        I have done both of these things.
1825    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
1826        most of the time, it *can* run out if it is given a pattern that contains a
1827        runaway infinite recursion. I updated the discussion in the pcrestack man
1828        page.
1830    13. Now that we have gone to the x.xx style of version numbers, the minor
1831        version may start with zero. Using 08 or 09 is a bad idea because users
1832        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
1833        interpreted as invalid octal numbers. I've updated the previous comment in
1834        configure.ac, and also added a check that gives an error if 08 or 09 are
1835        used.
1837    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
1838        causing partial matching to fail when the end of the subject matched \W
1839        in a UTF-8 pattern where \W was quantified with a minimum of 3.
1841    15. There were some discrepancies between the declarations in pcre_internal.h
1842        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
1843        their definitions. The declarations used "const uschar *" and the
1844        definitions used USPTR. Even though USPTR is normally defined as "const
1845        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
1846        reported that: "This difference in casting confuses some C++ compilers, for
1847        example, SunCC recognizes above declarations as different functions and
1848        generates broken code for hbpcre." I have changed the declarations to use
1849        USPTR.
1851    16. GNU libtool is named differently on some systems. The autogen.sh script now
1852        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
1853        (FreeBSD).
1855    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
1856        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
1857        comment: "Figure out how to create a longlong from a string: strtoll and
1858        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
1859        instance, but it only takes 2 args instead of 3!"
1861    18. A subtle bug concerned with back references has been fixed by a change of
1862        specification, with a corresponding code fix. A pattern such as
1863        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
1864        refers, was giving matches when it shouldn't. For example, xa=xaaa would
1865        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
1866        same bug. Such groups have to be quantified to be useful, or contained
1867        inside another quantified group. (If there's no repetition, the reference
1868        can never match.) The problem arises because, having left the group and
1869        moved on to the rest of the pattern, a later failure that backtracks into
1870        the group uses the captured value from the final iteration of the group
1871        rather than the correct earlier one. I have fixed this in PCRE by forcing
1872        any group that contains a reference to itself to be an atomic group; that
1873        is, there cannot be any backtracking into it once it has completed. This is
1874        similar to recursive and subroutine calls.
1877    Version 8.00 19-Oct-09
1878    ----------------------
1880    1.  The table for translating pcre_compile() error codes into POSIX error codes
1881        was out-of-date, and there was no check on the pcre_compile() error code
1882        being within the table. This could lead to an OK return being given in
1883        error.
1885    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
1886        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
1887        in a Windows environment.
1889    3.  The pcregrep --count option prints the count for each file even when it is
1890        zero, as does GNU grep. However, pcregrep was also printing all files when
1891        --files-with-matches was added. Now, when both options are given, it prints
1892        counts only for those files that have at least one match. (GNU grep just
1893        prints the file name in this circumstance, but including the count seems
1894        more useful - otherwise, why use --count?) Also ensured that the
1895        combination -clh just lists non-zero counts, with no names.
1897    4.  The long form of the pcregrep -F option was incorrectly implemented as
1898        --fixed_strings instead of --fixed-strings. This is an incompatible change,
1899        but it seems right to fix it, and I didn't think it was worth preserving
1900        the old behaviour.
1902    5.  The command line items --regex=pattern and --regexp=pattern were not
1903        recognized by pcregrep, which required --regex pattern or --regexp pattern
1904        (with a space rather than an '='). The man page documented the '=' forms,
1905        which are compatible with GNU grep; these now work.
1907    6.  No libpcreposix.pc file was created for pkg-config; there was just
1908        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
1910    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
1911        when UCP support is not needed, by modifying the Python script that
1912        generates it from Unicode data files. This should not matter if the module
1913        is correctly used as a library, but I received one complaint about 50K of
1914        unwanted data. My guess is that the person linked everything into his
1915        program rather than using a library. Anyway, it does no harm.
1917    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
1918        was a minimum greater than 1 for a wide character in a possessive
1919        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
1920        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
1921        character. Chaos in the form of incorrect output or a compiling loop could
1922        result.
1924    9.  The restrictions on what a pattern can contain when partial matching is
1925        requested for pcre_exec() have been removed. All patterns can now be
1926        partially matched by this function. In addition, if there are at least two
1927        slots in the offset vector, the offset of the earliest inspected character
1928        for the match and the offset of the end of the subject are set in them when
1929        PCRE_ERROR_PARTIAL is returned.
1931    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
1932        synonymous with PCRE_PARTIAL, for backwards compatibility, and
1933        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
1934        and may be more useful for multi-segment matching.
1936    11. Partial matching with pcre_exec() is now more intuitive. A partial match
1937        used to be given if ever the end of the subject was reached; now it is
1938        given only if matching could not proceed because another character was
1939        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
1940        string "Z", which now yields "no match" instead of "partial match". In the
1941        case of pcre_dfa_exec(), "no match" is given if every matching path for the
1942        final character ended with (*FAIL).
1944    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
1945        if the pattern had a "must contain" character that was already found in the
1946        earlier partial match, unless partial matching was again requested. For
1947        example, with the pattern /dog.(body)?/, the "must contain" character is
1948        "g". If the first part-match was for the string "dog", restarting with
1949        "sbody" failed. This bug has been fixed.
1951    13. The string returned by pcre_dfa_exec() after a partial match has been
1952        changed so that it starts at the first inspected character rather than the
1953        first character of the match. This makes a difference only if the pattern
1954        starts with a lookbehind assertion or \b or \B (\K is not supported by
1955        pcre_dfa_exec()). It's an incompatible change, but it makes the two
1956        matching functions compatible, and I think it's the right thing to do.
1958    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
1959        so that the demonstration program is easily available in environments where
1960        PCRE has not been installed from source.
1962    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
1963        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
1964        library.
1966    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
1967        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
1968        is not the first non-POSIX option to be added. Clearly some people find
1969        these options useful.
1971    17. If a caller to the POSIX matching function regexec() passes a non-zero
1972        value for nmatch with a NULL value for pmatch, the value of
1973        nmatch is forced to zero.
1975    18. RunGrepTest did not have a test for the availability of the -u option of
1976        the diff command, as RunTest does. It now checks in the same way as
1977        RunTest, and also checks for the -b option.
1979    19. If an odd number of negated classes containing just a single character
1980        interposed, within parentheses, between a forward reference to a named
1981        subpattern and the definition of the subpattern, compilation crashed with
1982        an internal error, complaining that it could not find the referenced
1983        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
1984        [The bug was that it was starting one character too far in when skipping
1985        over the character class, thus treating the ] as data rather than
1986        terminating the class. This meant it could skip too much.]
1988    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
1989        /g option in pcretest when the pattern contains \K, which makes it possible
1990        to have an empty string match not at the start, even when the pattern is
1991        anchored. Updated pcretest and pcredemo to use this option.
1993    21. If the maximum number of capturing subpatterns in a recursion was greater
1994        than the maximum at the outer level, the higher number was returned, but
1995        with unset values at the outer level. The correct (outer level) value is
1996        now given.
1998    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
1999        PCRE did not set those parentheses (unlike Perl). I have now found a way to
2000        make it do so. The string so far is captured, making this feature
2001        compatible with Perl.
2003    23. The tests have been re-organized, adding tests 11 and 12, to make it
2004        possible to check the Perl 5.10 features against Perl 5.10.
2006    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
2007        pattern matches a fixed length string. PCRE did not allow this; now it
2008        does. Neither allows recursion.
2010    25. I finally figured out how to implement a request to provide the minimum
2011        length of subject string that was needed in order to match a given pattern.
2012        (It was back references and recursion that I had previously got hung up
2013        on.) This code has now been added to pcre_study(); it finds a lower bound
2014        to the length of subject needed. It is not necessarily the greatest lower
2015        bound, but using it to avoid searching strings that are too short does give
2016        some useful speed-ups. The value is available to calling programs via
2017        pcre_fullinfo().
2019    26. While implementing 25, I discovered to my embarrassment that pcretest had
2020        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
2021        study optimizations had never been tested with that matching function.
2022        Oops. What is worse, even when it was passed study data, there was a bug in
2023        pcre_dfa_exec() that meant it never actually used it. Double oops. There
2024        were also very few tests of studied patterns with pcre_dfa_exec().
2026    27. If (?| is used to create subpatterns with duplicate numbers, they are now
2027        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
2028        on the other side of the coin, they are no longer allowed to have different
2029        names, because these cannot be distinguished in PCRE, and this has caused
2030        confusion. (This is a difference from Perl.)
2032    28. When duplicate subpattern names are present (necessarily with different
2033        numbers, as required by 27 above), and a test is made by name in a
2034        conditional pattern, either for a subpattern having been matched, or for
2035        recursion in such a pattern, all the associated numbered subpatterns are
2036        tested, and the overall condition is true if the condition is true for any
2037        one of them. This is the way Perl works, and is also more like the way
2038        testing by number works.
2041    Version 7.9 11-Apr-09
2042    ---------------------
2044    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
2045        (pcretest), all targets were linked against these libraries. This included
2046        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
2047        libraries. This caused unwanted dependencies to be created. This problem
2048        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
2049        pcretest is linked with readline.
2051    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
2052        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
2053        moved outside it again, because FALSE and TRUE are already defined in AIX,
2054        but BOOL is not.
2056    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
2057        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
2059    4.  The pcregrep documentation said spaces were inserted as well as colons (or
2060        hyphens) following file names and line numbers when outputting matching
2061        lines. This is not true; no spaces are inserted. I have also clarified the
2062        wording for the --colour (or --color) option.
2064    5.  In pcregrep, when --colour was used with -o, the list of matching strings
2065        was not coloured; this is different to GNU grep, so I have changed it to be
2066        the same.
2068    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
2069        each matching line was coloured. Now it goes on to look for further matches
2070        of any of the test patterns, which is the same behaviour as GNU grep.
2072    7.  A pattern that could match an empty string could cause pcregrep to loop; it
2073        doesn't make sense to accept an empty string match in pcregrep, so I have
2074        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
2075        seems to be how GNU grep behaves. [But see later change 40 for release
2076        8.33.]
2078    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
2079        start or after a newline", because the conditional assertion was not being
2080        correctly handled. The rule now is that both the assertion and what follows
2081        in the first alternative must satisfy the test.
2083    9.  If auto-callout was enabled in a pattern with a conditional group whose
2084        condition was an assertion, PCRE could crash during matching, both with
2085        pcre_exec() and pcre_dfa_exec().
2087    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
2088        used for matching.
2090    11. Unicode property support in character classes was not working for
2091        characters (bytes) greater than 127 when not in UTF-8 mode.
2093    12. Added the -M command line option to pcretest.
2095    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
2097    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
2099    16. Added comments and documentation about mis-use of no_arg in the C++
2100        wrapper.
2102    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
2103        from Martin Jerabek that uses macro names for all relevant character and
2104        string constants.
2106    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
2107        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
2108        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
2109        these, but not everybody uses configure.
2111    19. A conditional group that had only one branch was not being correctly
2112        recognized as an item that could match an empty string. This meant that an
2113        enclosing group might also not be so recognized, causing infinite looping
2114        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
2115        with the subject "ab", where knowledge that the repeated group can match
2116        nothing is needed in order to break the loop.
2118    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
2119        exec(), but without supplying a callout function, matching went wrong.
2121    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
2122        leak if the size of the offset vector was greater than 30. When the vector
2123        is smaller, the saved offsets during recursion go onto a local stack
2124        vector, but for larger vectors malloc() is used. It was failing to free
2125        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
2126        error, in fact).
2128    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
2129        heapframe that is used only when UTF-8 support is enabled. This caused no
2130        problem, but was untidy.
2132    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
2133        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
2134        included within another project.
2136    24. Steven Van Ingelgem's patches to add more options to the CMake support,
2137        slightly modified by me:
2139          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
2140              not building pcregrep.
2142          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
2143              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
2145    25. Forward references, both numeric and by name, in patterns that made use of
2146        duplicate group numbers, could behave incorrectly or give incorrect errors,
2147        because when scanning forward to find the reference group, PCRE was not
2148        taking into account the duplicate group numbers. A pattern such as
2149        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
2151    26. Changed a few more instances of "const unsigned char *" to USPTR, making
2152        the feature of a custom pointer more persuasive (as requested by a user).
2154    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
2155        pcretest.c, inside #ifndefs, because it seems they are sometimes already
2156        pre-defined.
2158    28. Added support for (*UTF8) at the start of a pattern.
2160    29. Arrange for flags added by the "release type" setting in CMake to be shown
2161        in the configuration summary.
2164    Version 7.8 05-Sep-08
2165    ---------------------
2167    1.  Replaced UCP searching code with optimized version as implemented for Ad
2168        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
2169        stage table and inline lookup instead of a function, giving speed ups of 2
2170        to 5 times on some simple patterns that I tested. Permission was given to
2171        distribute the MultiStage2.py script that generates the tables (it's not in
2172        the tarball, but is in the Subversion repository).
2174    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
2175        scripts.
2177    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
2178        a group with a zero qualifier. The result of the study could be incorrect,
2179        or the function might crash, depending on the pattern.
2181    4.  Caseless matching was not working for non-ASCII characters in back
2182        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
2183        It now works when Unicode Property Support is available.
2185    5.  In pcretest, an escape such as \x{de} in the data was always generating
2186        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
2187        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
2188        truncation.
2190    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
2192    7.  Added two (int) casts to pcregrep when printing the difference of two
2193        pointers, in case they are 64-bit values.
2195    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
2196        test 2 if it fails.
2198    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
2199        and a #define of that name to empty if it is not externally set. This is to
2200        allow users of MSVC to set it if necessary.
2202    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
2203        the convenience functions in the pcre_get.c source file.
2205    11. An option change at the start of a pattern that had top-level alternatives
2206        could cause overwriting and/or a crash. This command provoked a crash in
2207        some environments:
2209          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
2211        This potential security problem was recorded as CVE-2008-2371.
2213    12. For a pattern where the match had to start at the beginning or immediately
2214        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
2215        pcre_dfa_exec() could read past the end of the passed subject if there was
2216        no match. To help with detecting such bugs (e.g. with valgrind), I modified
2217        pcretest so that it places the subject at the end of its malloc-ed buffer.
2219    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
2220        exec() might read past the end of the data buffer in UTF-8 mode.
2222    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
2223        the data contained the byte 0x85 as part of a UTF-8 character within its
2224        first line. This applied both to normal and DFA matching.
2226    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
2227        /^[^d]*?$/8 failed to match "abc".
2229    16. Added a missing copyright notice to pcrecpp_internal.h.
2231    17. Make it more clear in the documentation that values returned from
2232        pcre_exec() in ovector are byte offsets, not character counts.
2234    18. Tidied a few places to stop certain compilers from issuing warnings.
2236    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
2237        supplied by Stefan Weber. I made a further small update for 7.8 because
2238        there is a change of source arrangements: the pcre_searchfuncs.c module is
2239        replaced by pcre_ucd.c.
2242    Version 7.7 07-May-08
2243    ---------------------
2245    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
2246        a string to a long long, pretend we don't even have a long long." This is
2247        done by checking for the strtoq, strtoll, and _strtoi64 functions.
2249    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
2250        pre-7.6 versions, which defined a global no_arg variable instead of putting
2251        it in the RE class. (See also #8 below.)
2253    3.  Remove a line of dead code, identified by coverity and reported by Nuno
2254        Lopes.
2256    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
2258        (1) The include/exclude patterns were being applied to the whole pathnames
2259            of files, instead of just to the final components.
2261        (2) If there was more than one level of directory, the subdirectories were
2262            skipped unless they satisfied the include/exclude conditions. This is
2263            inconsistent with GNU grep (and could even be seen as contrary to the
2264            pcregrep specification - which I improved to make it absolutely clear).
2265            The action now is always to scan all levels of directory, and just
2266            apply the include/exclude patterns to regular files.
2268    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
2269        --exclude_dir in the tests to avoid scanning .svn directories.
2271    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
2272        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
2273        doesn't support NULs in patterns.
2275    7.  Added some missing "const"s to declarations of static tables in
2276        pcre_compile.c and pcre_dfa_exec.c.
2278    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
2279        caused by fix #2  above. (Subsequently also a second patch to fix the
2280        first patch. And a third patch - this was a messy problem.)
2282    9.  Applied Craig's patch to remove the use of push_back().
2284    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
2285        matching function regexec().
2287    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
2288        which, however, unlike Perl's \g{...}, are subroutine calls, not back
2289        references. PCRE supports relative numbers with this syntax (I don't think
2290        Oniguruma does).
2292    12. Previously, a group with a zero repeat such as (...){0} was completely
2293        omitted from the compiled regex. However, this means that if the group
2294        was called as a subroutine from elsewhere in the pattern, things went wrong
2295        (an internal error was given). Such groups are now left in the compiled
2296        pattern, with a new opcode that causes them to be skipped at execution
2297        time.
2299    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
2300        to the way PCRE behaves:
2302        (a) A lone ] character is dis-allowed (Perl treats it as data).
2304        (b) A back reference to an unmatched subpattern matches an empty string
2305            (Perl fails the current match path).
2307        (c) A data ] in a character class must be notated as \] because if the
2308            first data character in a class is ], it defines an empty class. (In
2309            Perl it is not possible to have an empty class.) The empty class []
2310            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
2311            The negative empty class [^] matches any one character, independently
2312            of the DOTALL setting.
2314    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
2315        non-existent subpattern following a character class starting with ']' and
2316        containing () gave an internal compiling error instead of "reference to
2317        non-existent subpattern". Fortunately, when the pattern did exist, the
2318        compiled code was correct. (When scanning forwards to check for the
2319        existence of the subpattern, it was treating the data ']' as terminating
2320        the class, so got the count wrong. When actually compiling, the reference
2321        was subsequently set up correctly.)
2323    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
2324        it was being rejected as not supported by pcre_dfa_exec(), even though
2325        other assertions are supported. I have made pcre_dfa_exec() support
2326        (*FAIL).
2328    16. The implementation of 13c above involved the invention of a new opcode,
2329        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
2330        cannot be changed at match time, I realized I could make a small
2331        improvement to matching performance by compiling OP_ALLANY instead of
2332        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
2333        on the OP_ANY path.
2335    17. Compiling pcretest on Windows with readline support failed without the
2336        following two fixes: (1) Make the unistd.h include conditional on
2337        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
2339    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
2340        ncurses library to be included for pcretest when ReadLine support is
2341        requested, but also to allow for it to be overridden. This patch came from
2342        Daniel Bergström.
2344    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
2345        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
2346        any errors with the current Unicode tables. Thanks to Peter Kankowski for
2347        spotting this.
2350    Version 7.6 28-Jan-08
2351    ---------------------
2353    1.  A character class containing a very large number of characters with
2354        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
2355        overflow.
2357    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
2358        HAVE_LONG_LONG is not defined.
2360    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
2361        bring them up to date and include new features. This patch includes:
2363        - Fixed PH's badly added libz and libbz2 support.
2364        - Fixed a problem with static linking.
2365        - Added pcredemo. [But later removed - see 7 below.]
2366        - Fixed dftables problem and added an option.
2367        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
2368            HAVE_LONG_LONG.
2369        - Added readline support for pcretest.
2370        - Added an listing of the option settings after cmake has run.
2372    4.  A user submitted a patch to Makefile that makes it easy to create
2373        "pcre.dll" under mingw when using Configure/Make. I added stuff to
2374        Makefile.am that cause it to include this special target, without
2375        affecting anything else. Note that the same mingw target plus all
2376        the other distribution libraries and programs are now supported
2377        when configuring with CMake (see 6 below) instead of with
2378        Configure/Make.
2380    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
2381        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
2382        exported in the Windows port". It has not yet been confirmed that the patch
2383        solves the problem, but it does no harm.
2385    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
2386        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
2387        with CMake, and also correct the comment about stack recursion.
2389    7.  Remove the automatic building of pcredemo from the ./configure system and
2390        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
2391        of a program that users should build themselves after PCRE is installed, so
2392        building it automatically is not really right. What is more, it gave
2393        trouble in some build environments.
2395    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
2398    Version 7.5 10-Jan-08
2399    ---------------------
2401    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
2402        values in parens when parsing an RE using the C++ wrapper."
2404    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
2405        Characters greater than 255 were excluded from the class instead of being
2406        included.
2408    3.  The same bug as (2) above applied to negated POSIX classes such as
2409        [:^space:].
2411    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
2412        defined or documented. It seems to have been a typo for PCRE_STATIC, so
2413        I have changed it.
2415    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
2416        first named subpattern) and a construct such as (?&a) would reference the
2417        first named subpattern whose name started with "a" (in other words, the
2418        length check was missing). Both these problems are fixed. "Subpattern name
2419        expected" is now given for (?&) (a zero-length name), and this patch also
2420        makes it give the same error for \k'' (previously it complained that that
2421        was a reference to a non-existent subpattern).
2423    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
2424        this is right because (?- can be followed by option settings as well as by
2425        digits. I have, however, made the messages clearer.
2427    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
2428        than the number used in the conditional) now cause a compile-time error.
2429        This is actually not compatible with Perl, which accepts such patterns, but
2430        treats the conditional as always being FALSE (as PCRE used to), but it
2431        seems to me that giving a diagnostic is better.
2433    8.  Change "alphameric" to the more common word "alphanumeric" in comments
2434        and messages.
2436    9.  Fix two occurrences of "backslash" in comments that should have been
2437        "backspace".
2439    10. Remove two redundant lines of code that can never be obeyed (their function
2440        was moved elsewhere).
2442    11. The program that makes PCRE's Unicode character property table had a bug
2443        which caused it to generate incorrect table entries for sequences of
2444        characters that have the same character type, but are in different scripts.
2445        It amalgamated them into a single range, with the script of the first of
2446        them. In other words, some characters were in the wrong script. There were
2447        thirteen such cases, affecting characters in the following ranges:
2449          U+002b0 - U+002c1
2450          U+0060c - U+0060d
2451          U+0061e - U+00612
2452          U+0064b - U+0065e
2453          U+0074d - U+0076d
2454          U+01800 - U+01805
2455          U+01d00 - U+01d77
2456          U+01d9b - U+01dbf
2457          U+0200b - U+0200f
2458          U+030fc - U+030fe
2459          U+03260 - U+0327f
2460          U+0fb46 - U+0fbb1
2461          U+10450 - U+1049d
2463    12. The -o option (show only the matching part of a line) for pcregrep was not
2464        compatible with GNU grep in that, if there was more than one match in a
2465        line, it showed only the first of them. It now behaves in the same way as
2466        GNU grep.
2468    13. If the -o and -v options were combined for pcregrep, it printed a blank
2469        line for every non-matching line. GNU grep prints nothing, and pcregrep now
2470        does the same. The return code can be used to tell if there were any
2471        non-matching lines.
2473    14. Added --file-offsets and --line-offsets to pcregrep.
2475    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
2476        infinitely looping recursion. The bug was that positive lookaheads were not
2477        being skipped when checking for a possible empty match (negative lookaheads
2478        and both kinds of lookbehind were skipped).
2480    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
2481        inclusion of <windows.h> to before rather than after the definition of
2482        INVALID_FILE_ATTRIBUTES (patch from David Byron).
2484    17. Specifying a possessive quantifier with a specific limit for a Unicode
2485        character property caused pcre_compile() to compile bad code, which led at
2486        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
2487        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
2488        caused the error; without that there was no problem.
2490    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
2492    19. Added --enable-pcretest-libreadline.
2494    20. In pcrecpp.cc, the variable 'count' was incremented twice in
2495        RE::GlobalReplace(). As a result, the number of replacements returned was
2496        double what it should be. I removed one of the increments, but Craig sent a
2497        later patch that removed the other one (the right fix) and added unit tests
2498        that check the return values (which was not done before).
2500    21. Several CMake things:
2502        (1) Arranged that, when cmake is used on Unix, the libraries end up with
2503            the names libpcre and libpcreposix, not just pcre and pcreposix.
2505        (2) The above change means that pcretest and pcregrep are now correctly
2506            linked with the newly-built libraries, not previously installed ones.
2510    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
2511        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
2512        UTF-8 newline character). The key issue is that the pattern starts .*;
2513        this means that the match must be either at the beginning, or after a
2514        newline. The bug was in the code for advancing after a failed match and
2515        checking that the new position followed a newline. It was not taking
2516        account of UTF-8 characters correctly.
2518    23. PCRE was behaving differently from Perl in the way it recognized POSIX
2519        character classes. PCRE was not treating the sequence [:...:] as a
2520        character class unless the ... were all letters. Perl, however, seems to
2521        allow any characters between [: and :], though of course it rejects as
2522        unknown any "names" that contain non-letters, because all the known class
2523        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
2524        for example, whereas PCRE did not - it did not recognize a POSIX character
2525        class. This seemed a bit dangerous, so the code has been changed to be
2526        closer to Perl. The behaviour is not identical to Perl, because PCRE will
2527        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
2528        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
2529        Perl does, and where it didn't before.
2531    24. Rewrite so as to remove the single use of %n from pcregrep because in some
2532        Windows environments %n is disabled by default.
2535    Version 7.4 21-Sep-07
2536    ---------------------
2538    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
2539        means that a class such as [\s] counted as "explicit reference to CR or
2540        LF". That isn't really right - the whole point of the change was to try to
2541        help when there was an actual mention of one of the two characters. So now
2542        the change happens only if \r or \n (or a literal CR or LF) character is
2543        encountered.
2545    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
2546        of both had grown to the point where there were only 3 bits left.
2547        Fortunately, there was spare space in the data structure, and so I have
2548        moved the internal flags into a new 16-bit field to free up more option
2549        bits.
2551    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
2552        but did not set the internal JCHANGED flag - either of these is enough to
2553        control the way the "get" function works - but the PCRE_INFO_JCHANGED
2554        facility is supposed to tell if (?J) was ever used, so now (?J) at the
2555        start sets both bits.
2557    4.  Added options (at build time, compile time, exec time) to change \R from
2558        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
2560    5.  doc/pcresyntax.html was missing from the distribution.
2562    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
2563        compatibility, even though it is no longer used.
2565    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
2566        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
2567        windows.h file is present (where different names are used). [This was
2568        reversed later after testing - see 16 below.]
2570    8.  Changed all #include <config.h> to #include "config.h". There were also
2571        some further <pcre.h> cases that I changed to "pcre.h".
2573    9.  When pcregrep was used with the --colour option, it missed the line ending
2574        sequence off the lines that it output.
2576    10. It was pointed out to me that arrays of string pointers cause lots of
2577        relocations when a shared library is dynamically loaded. A technique of
2578        using a single long string with a table of offsets can drastically reduce
2579        these. I have refactored PCRE in four places to do this. The result is
2580        dramatic:
2582          Originally:                          290
2583          After changing UCP table:            187
2584          After changing error message table:   43
2585          After changing table of "verbs"       36
2586          After changing table of Posix names   22
2588        Thanks to the folks working on Gregex for glib for this insight.
2590    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
2591        unicode-properties was also set.
2593    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
2595    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
2596        checked only for CRLF.
2598    14. Added casts to pcretest.c to avoid compiler warnings.
2600    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
2602    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
2603        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
2604        entirely. This removes changes made in 7 above.
2606    17. The CMake files have been updated, and there is now more information about
2607        building with CMake in the NON-UNIX-USE document.
2610    Version 7.3 28-Aug-07
2611    ---------------------
2613     1. In the rejigging of the build system that eventually resulted in 7.1, the
2614        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
2615        brackets there is not right, since it causes compilers to look for an
2616        installed pcre.h, not the version that is in the source that is being
2617        compiled (which of course may be different). I have changed it back to:
2619          #include "pcre.h"
2621        I have a vague recollection that the change was concerned with compiling in
2622        different directories, but in the new build system, that is taken care of
2623        by the VPATH setting the Makefile.
2625     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
2626        when the subject happened to end in the byte 0x85 (e.g. if the last
2627        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
2628        characters but of course it shouldn't be taken as a newline when it is part
2629        of another character. The bug was that, for an unlimited repeat of . in
2630        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
2631        characters when looking for a newline.
2633     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
2635     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
2636        in debug output.
2638     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
2639        long printing in the pcrecpp unittest when running under MinGW.
2641     6. ESC_K was left out of the EBCDIC table.
2643     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
2644        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
2645        limit also applies to "virtual nesting" when a pattern is recursive, and in
2646        this case 1000 isn't so big. I have been able to remove this limit at the
2647        expense of backing off one optimization in certain circumstances. Normally,
2648        when pcre_exec() would call its internal match() function recursively and
2649        immediately return the result unconditionally, it uses a "tail recursion"
2650        feature to save stack. However, when a subpattern that can match an empty
2651        string has an unlimited repetition quantifier, it no longer makes this
2652        optimization. That gives it a stack frame in which to save the data for
2653        checking that an empty string has been matched. Previously this was taken
2654        from the 1000-entry workspace that had been reserved. So now there is no
2655        explicit limit, but more stack is used.
2657     8. Applied Daniel's patches to solve problems with the import/export magic
2658        syntax that is required for Windows, and which was going wrong for the
2659        pcreposix and pcrecpp parts of the library. These were overlooked when this
2660        problem was solved for the main library.
2662     9. There were some crude static tests to avoid integer overflow when computing
2663        the size of patterns that contain repeated groups with explicit upper
2664        limits. As the maximum quantifier is 65535, the maximum group length was
2665        set at 30,000 so that the product of these two numbers did not overflow a
2666        32-bit integer. However, it turns out that people want to use groups that
2667        are longer than 30,000 bytes (though not repeat them that many times).
2668        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
2669        made it possible to implement the integer overflow checks in a much more
2670        dynamic way, which I have now done. The artificial limitation on group
2671        length has been removed - we now have only the limit on the total length of
2672        the compiled pattern, which depends on the LINK_SIZE setting.
2674    10. Fixed a bug in the documentation for get/copy named substring when
2675        duplicate names are permitted. If none of the named substrings are set, the
2676        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
2677        empty string.
2679    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
2680        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
2681        because the ] is interpreted as the first data character and the
2682        terminating ] is not found. PCRE has been made compatible with Perl in this
2683        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
2684        cause memory overwriting.
2686    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
2687        string has been matched (to stop an infinite loop). It was not recognizing
2688        a conditional subpattern that could match an empty string if that
2689        subpattern was within another subpattern. For example, it looped when
2690        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
2691        condition was not nested. This bug has been fixed.
2693    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
2694        past the start of the subject in the presence of bytes with the top bit
2695        set, for example "\x8aBCD".
2697    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
2698        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
2700    14. Optimized (?!) to (*FAIL).
2702    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
2703        This restricts code points to be within the range 0 to 0x10FFFF, excluding
2704        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
2705        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
2706        does: it's just the validity check that is more restrictive.
2708    16. Inserted checks for integer overflows during escape sequence (backslash)
2709        processing, and also fixed erroneous offset values for syntax errors during
2710        backslash processing.
2712    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
2713        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
2715    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
2716        caused an overrun.
2718    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
2719        something other than just ASCII characters) inside a group that had an
2720        unlimited repeat caused a loop at compile time (while checking to see
2721        whether the group could match an empty string).
2723    20. Debugging a pattern containing \p or \P could cause a crash. For example,
2724        [\P{Any}] did so. (Error in the code for printing property names.)
2726    21. An orphan \E inside a character class could cause a crash.
2728    22. A repeated capturing bracket such as (A)? could cause a wild memory
2729        reference during compilation.
2731    23. There are several functions in pcre_compile() that scan along a compiled
2732        expression for various reasons (e.g. to see if it's fixed length for look
2733        behind). There were bugs in these functions when a repeated \p or \P was
2734        present in the pattern. These operators have additional parameters compared
2735        with \d, etc, and these were not being taken into account when moving along
2736        the compiled data. Specifically:
2738        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
2739            length.
2741        (b) An item such as \pL+ within a repeated group could cause crashes or
2742            loops.
2744        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
2745            "reference to non-existent subpattern" error.
2747        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
2749    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
2750        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
2752    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
2754    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
2755        character were causing crashes (broken optimization).
2757    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
2758        \p or \P) caused a compile-time loop.
2760    28. More problems have arisen in unanchored patterns when CRLF is a valid line
2761        break. For example, the unstudied pattern [\r\n]A does not match the string
2762        "\r\nA" because change 7.0/46 below moves the current point on by two
2763        characters after failing to match at the start. However, the pattern \nA
2764        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
2765        the same is true. There doesn't seem any very clean way out of this, but
2766        what I have chosen to do makes the common cases work: PCRE now takes note
2767        of whether there can be an explicit match for \r or \n anywhere in the
2768        pattern, and if so, 7.0/46 no longer applies. As part of this change,
2769        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
2770        pattern has explicit CR or LF references.
2772    29. Added (*CR) etc for changing newline setting at start of pattern.
2775    Version 7.2 19-Jun-07
2776  ---------------------  ---------------------
2778   1. If the fr_FR locale cannot be found for test 3, try the "french" locale,   1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
# Line 21  Version 7.2 01-May-07 Line 2792  Version 7.2 01-May-07
2792      stack recursion. This gives a massive performance boost under BSD, but just      stack recursion. This gives a massive performance boost under BSD, but just
2793      a small improvement under Linux. However, it saves one field in the frame      a small improvement under Linux. However, it saves one field in the frame
2794      in all cases.      in all cases.
2796   6. Added more features from the forthcoming Perl 5.10:   6. Added more features from the forthcoming Perl 5.10:
2798      (a) (?-n) (where n is a string of digits) is a relative subroutine or      (a) (?-n) (where n is a string of digits) is a relative subroutine or
2799          recursion call. It refers to the nth most recently opened parentheses.          recursion call. It refers to the nth most recently opened parentheses.
2801      (b) (?+n) is also a relative subroutine call; it refers to the nth next      (b) (?+n) is also a relative subroutine call; it refers to the nth next
2802          to be opened parentheses.          to be opened parentheses.
2804      (c) Conditions that refer to capturing parentheses can be specified      (c) Conditions that refer to capturing parentheses can be specified
2805          relatively, for example, (?(-2)... or (?(+3)...          relatively, for example, (?(-2)... or (?(+3)...
2807        (d) \K resets the start of the current match so that everything before
2808            is not part of it.
2810        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
2812        (f) \g{name} is another synonym - part of Perl 5.10's unification of
2813            reference syntax.
2815        (g) (?| introduces a group in which the numbering of parentheses in each
2816            alternative starts with the same number.
2818        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
2820     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
2823     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
2824        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
2825        for detecting groups that can match an empty string.
2827     9. A pattern with a very large number of alternatives (more than several
2828        hundred) was running out of internal workspace during the pre-compile
2829        phase, where pcre_compile() figures out how much memory will be needed. A
2830        bit of new cunning has reduced the workspace needed for groups with
2831        alternatives. The 1000-alternative test pattern now uses 12 bytes of
2832        workspace instead of running out of the 4096 that are available.
2834    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
2836    11. Applied patch from Google to remove an optimization that didn't quite work.
2837        The report of the bug said:
2839          pcrecpp::RE("a*").FullMatch("aaa") matches, while
2840          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
2841          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
2843    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
2844        it matched the wrong number of bytes.
2847  Version 7.1 24-Apr-07  Version 7.1 24-Apr-07
2848  ---------------------  ---------------------

Removed from v.167  
changed lines
  Added in v.1486

  ViewVC Help
Powered by ViewVC 1.1.5