ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 267 by ph10, Wed Nov 14 11:48:49 2007 UTC revision 1496 by ph10, Fri Jul 18 08:24:35 2014 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
4  Version 7.5 12-Nov-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.
72    14. If a character class started [\Qx]... where x is any character, the class
73        was incorrectly terminated at the ].
75    15. If a pattern that started with a caseless match for a character with more
76        than one "other case" was studied, PCRE did not set up the starting code
77        unit bit map for the list of possible characters. Now it does. This is an
78        optimization improvement, not a bug fix.
80    16. The Unicode data tables have been updated to Unicode 7.0.0.
82    17. Fixed a number of memory leaks in pcregrep.
84    18. Avoid a compiler warning (from some compilers) for a function call with
85        a cast that removes "const" from an lvalue by using an intermediate
86        variable (to which the compiler does not object).
88    19. Incorrect code was compiled if a group that contained an internal recursive
89        back reference was optional (had quantifier with a minimum of zero). This
90        example compiled incorrect code: /(((a\2)|(a*)\g<-1>))*/ and other examples
91        caused segmentation faults because of stack overflows at compile time.
94    Version 8.35 04-April-2014
95    --------------------------
97    1.  A new flag is set, when property checks are present in an XCLASS.
98        When this flag is not set, PCRE can perform certain optimizations
99        such as studying these XCLASS-es.
101    2.  The auto-possessification of character sets were improved: a normal
102        and an extended character set can be compared now. Furthermore
103        the JIT compiler optimizes more character set checks.
105    3.  Got rid of some compiler warnings for potentially uninitialized variables
106        that show up only when compiled with -O2.
108    4.  A pattern such as (?=ab\K) that uses \K in an assertion can set the start
109        of a match later then the end of the match. The pcretest program was not
110        handling the case sensibly - it was outputting from the start to the next
111        binary zero. It now reports this situation in a message, and outputs the
112        text from the end to the start.
114    5.  Fast forward search is improved in JIT. Instead of the first three
115        characters, any three characters with fixed position can be searched.
116        Search order: first, last, middle.
118    6.  Improve character range checks in JIT. Characters are read by an inprecise
119        function now, which returns with an unknown value if the character code is
120        above a certain threshold (e.g: 256). The only limitation is that the value
121        must be bigger than the threshold as well. This function is useful when
122        the characters above the threshold are handled in the same way.
124    7.  The macros whose names start with RAWUCHAR are placeholders for a future
125        mode in which only the bottom 21 bits of 32-bit data items are used. To
126        make this more memorable for those maintaining the code, the names have
127        been changed to start with UCHAR21, and an extensive comment has been added
128        to their definition.
130    8.  Add missing (new) files sljitNativeTILEGX.c and sljitNativeTILEGX-encoder.c
131        to the export list in Makefile.am (they were accidentally omitted from the
132        8.34 tarball).
134    9.  The informational output from pcretest used the phrase "starting byte set"
135        which is inappropriate for the 16-bit and 32-bit libraries. As the output
136        for "first char" and "need char" really means "non-UTF-char", I've changed
137        "byte" to "char", and slightly reworded the output. The documentation about
138        these values has also been (I hope) clarified.
140    10. Another JIT related optimization: use table jumps for selecting the correct
141        backtracking path, when more than four alternatives are present inside a
142        bracket.
144    11. Empty match is not possible, when the minimum length is greater than zero,
145        and there is no \K in the pattern. JIT should avoid empty match checks in
146        such cases.
148    12. In a caseless character class with UCP support, when a character with more
149        than one alternative case was not the first character of a range, not all
150        the alternative cases were added to the class. For example, s and \x{17f}
151        are both alternative cases for S: the class [RST] was handled correctly,
152        but [R-T] was not.
154    13. The configure.ac file always checked for pthread support when JIT was
155        enabled. This is not used in Windows, so I have put this test inside a
156        check for the presence of windows.h (which was already tested for).
158    14. Improve pattern prefix search by a simplified Boyer-Moore algorithm in JIT.
159        The algorithm provides a way to skip certain starting offsets, and usually
160        faster than linear prefix searches.
162    15. Change 13 for 8.20 updated RunTest to check for the 'fr' locale as well
163        as for 'fr_FR' and 'french'. For some reason, however, it then used the
164        Windows-specific input and output files, which have 'french' screwed in.
165        So this could never have worked. One of the problems with locales is that
166        they aren't always the same. I have now updated RunTest so that it checks
167        the output of the locale test (test 3) against three different output
168        files, and it allows the test to pass if any one of them matches. With luck
169        this should make the test pass on some versions of Solaris where it was
170        failing. Because of the uncertainty, the script did not used to stop if
171        test 3 failed; it now does. If further versions of a French locale ever
172        come to light, they can now easily be added.
174    16. If --with-pcregrep-bufsize was given a non-integer value such as "50K",
175        there was a message during ./configure, but it did not stop. This now
176        provokes an error. The invalid example in README has been corrected.
177        If a value less than the minimum is given, the minimum value has always
178        been used, but now a warning is given.
180    17. If --enable-bsr-anycrlf was set, the special 16/32-bit test failed. This
181        was a bug in the test system, which is now fixed. Also, the list of various
182        configurations that are tested for each release did not have one with both
183        16/32 bits and --enable-bar-anycrlf. It now does.
185    18. pcretest was missing "-C bsr" for displaying the \R default setting.
187    19. Little endian PowerPC systems are supported now by the JIT compiler.
189    20. The fast forward newline mechanism could enter to an infinite loop on
190        certain invalid UTF-8 input. Although we don't support these cases
191        this issue can be fixed by a performance optimization.
193    21. Change 33 of 8.34 is not sufficient to ensure stack safety because it does
194        not take account if existing stack usage. There is now a new global
195        variable called pcre_stack_guard that can be set to point to an external
196        function to check stack availability. It is called at the start of
197        processing every parenthesized group.
199    22. A typo in the code meant that in ungreedy mode the max/min qualifier
200        behaved like a min-possessive qualifier, and, for example, /a{1,3}b/U did
201        not match "ab".
203    23. When UTF was disabled, the JIT program reported some incorrect compile
204        errors. These messages are silenced now.
206    24. Experimental support for ARM-64 and MIPS-64 has been added to the JIT
207        compiler.
209    25. Change all the temporary files used in RunGrepTest to be different to those
210        used by RunTest so that the tests can be run simultaneously, for example by
211        "make -j check".
214    Version 8.34 15-December-2013
215    -----------------------------
217    1.  Add pcre[16|32]_jit_free_unused_memory to forcibly free unused JIT
218        executable memory. Patch inspired by Carsten Klein.
220    2.  ./configure --enable-coverage defined SUPPORT_GCOV in config.h, although
221        this macro is never tested and has no effect, because the work to support
222        coverage involves only compiling and linking options and special targets in
223        the Makefile. The comment in config.h implied that defining the macro would
224        enable coverage support, which is totally false. There was also support for
225        setting this macro in the CMake files (my fault, I just copied it from
226        configure). SUPPORT_GCOV has now been removed.
228    3.  Make a small performance improvement in strlen16() and strlen32() in
229        pcretest.
231    4.  Change 36 for 8.33 left some unreachable statements in pcre_exec.c,
232        detected by the Solaris compiler (gcc doesn't seem to be able to diagnose
233        these cases). There was also one in pcretest.c.
235    5.  Cleaned up a "may be uninitialized" compiler warning in pcre_exec.c.
237    6.  In UTF mode, the code for checking whether a group could match an empty
238        string (which is used for indefinitely repeated groups to allow for
239        breaking an infinite loop) was broken when the group contained a repeated
240        negated single-character class with a character that occupied more than one
241        data item and had a minimum repetition of zero (for example, [^\x{100}]* in
242        UTF-8 mode). The effect was undefined: the group might or might not be
243        deemed as matching an empty string, or the program might have crashed.
245    7.  The code for checking whether a group could match an empty string was not
246        recognizing that \h, \H, \v, \V, and \R must match a character.
248    8.  Implemented PCRE_INFO_MATCH_EMPTY, which yields 1 if the pattern can match
249        an empty string. If it can, pcretest shows this in its information output.
251    9.  Fixed two related bugs that applied to Unicode extended grapheme clusters
252        that were repeated with a maximizing qualifier (e.g. \X* or \X{2,5}) when
253        matched by pcre_exec() without using JIT:
255        (a) If the rest of the pattern did not match after a maximal run of
256            grapheme clusters, the code for backing up to try with fewer of them
257            did not always back up over a full grapheme when characters that do not
258            have the modifier quality were involved, e.g. Hangul syllables.
260        (b) If the match point in a subject started with modifier character, and
261            there was no match, the code could incorrectly back up beyond the match
262            point, and potentially beyond the first character in the subject,
263            leading to a segfault or an incorrect match result.
265    10. A conditional group with an assertion condition could lead to PCRE
266        recording an incorrect first data item for a match if no other first data
267        item was recorded. For example, the pattern (?(?=ab)ab) recorded "a" as a
268        first data item, and therefore matched "ca" after "c" instead of at the
269        start.
271    11. Change 40 for 8.33 (allowing pcregrep to find empty strings) showed up a
272        bug that caused the command "echo a | ./pcregrep -M '|a'" to loop.
274    12. The source of pcregrep now includes z/OS-specific code so that it can be
275        compiled for z/OS as part of the special z/OS distribution.
277    13. Added the -T and -TM options to pcretest.
279    14. The code in pcre_compile.c for creating the table of named capturing groups
280        has been refactored. Instead of creating the table dynamically during the
281        actual compiling pass, the information is remembered during the pre-compile
282        pass (on the stack unless there are more than 20 named groups, in which
283        case malloc() is used) and the whole table is created before the actual
284        compile happens. This has simplified the code (it is now nearly 150 lines
285        shorter) and prepared the way for better handling of references to groups
286        with duplicate names.
288    15. A back reference to a named subpattern when there is more than one of the
289        same name now checks them in the order in which they appear in the pattern.
290        The first one that is set is used for the reference. Previously only the
291        first one was inspected. This change makes PCRE more compatible with Perl.
293    16. Unicode character properties were updated from Unicode 6.3.0.
295    17. The compile-time code for auto-possessification has been refactored, based
296        on a patch by Zoltan Herczeg. It now happens after instead of during
297        compilation. The code is cleaner, and more cases are handled. The option
298        PCRE_NO_AUTO_POSSESS is added for testing purposes, and the -O and /O
299        options in pcretest are provided to set it. It can also be set by
300        (*NO_AUTO_POSSESS) at the start of a pattern.
302    18. The character VT has been added to the default ("C" locale) set of
303        characters that match \s and are generally treated as white space,
304        following this same change in Perl 5.18. There is now no difference between
305        "Perl space" and "POSIX space". Whether VT is treated as white space in
306        other locales depends on the locale.
308    19. The code for checking named groups as conditions, either for being set or
309        for being recursed, has been refactored (this is related to 14 and 15
310        above). Processing unduplicated named groups should now be as fast at
311        numerical groups, and processing duplicated groups should be faster than
312        before.
314    20. Two patches to the CMake build system, by Alexander Barkov:
316          (1) Replace the "source" command by "." in CMakeLists.txt because
317              "source" is a bash-ism.
319          (2) Add missing HAVE_STDINT_H and HAVE_INTTYPES_H to config-cmake.h.in;
320              without these the CMake build does not work on Solaris.
322    21. Perl has changed its handling of \8 and \9. If there is no previously
323        encountered capturing group of those numbers, they are treated as the
324        literal characters 8 and 9 instead of a binary zero followed by the
325        literals. PCRE now does the same.
327    22. Following Perl, added \o{} to specify codepoints in octal, making it
328        possible to specify values greater than 0777 and also making them
329        unambiguous.
331    23. Perl now gives an error for missing closing braces after \x{... instead of
332        treating the string as literal. PCRE now does the same.
334    24. RunTest used to grumble if an inappropriate test was selected explicitly,
335        but just skip it when running all tests. This make it awkward to run ranges
336        of tests when one of them was inappropriate. Now it just skips any
337        inappropriate tests, as it always did when running all tests.
339    25. If PCRE_AUTO_CALLOUT and PCRE_UCP were set for a pattern that contained
340        character types such as \d or \w, too many callouts were inserted, and the
341        data that they returned was rubbish.
343    26. In UCP mode, \s was not matching two of the characters that Perl matches,
344        namely NEL (U+0085) and MONGOLIAN VOWEL SEPARATOR (U+180E), though they
345        were matched by \h. The code has now been refactored so that the lists of
346        the horizontal and vertical whitespace characters used for \h and \v (which
347        are defined only in one place) are now also used for \s.
349    27. Add JIT support for the 64 bit TileGX architecture.
350        Patch by Jiong Wang (Tilera Corporation).
352    28. Possessive quantifiers for classes (both explicit and automatically
353        generated) now use special opcodes instead of wrapping in ONCE brackets.
355    29. Whereas an item such as A{4}+ ignored the possessivenes of the quantifier
356        (because it's meaningless), this was not happening when PCRE_CASELESS was
357        set. Not wrong, but inefficient.
359    30. Updated perltest.pl to add /u (force Unicode mode) when /W (use Unicode
360        properties for \w, \d, etc) is present in a test regex. Otherwise if the
361        test contains no characters greater than 255, Perl doesn't realise it
362        should be using Unicode semantics.
364    31. Upgraded the handling of the POSIX classes [:graph:], [:print:], and
365        [:punct:] when PCRE_UCP is set so as to include the same characters as Perl
366        does in Unicode mode.
368    32. Added the "forbid" facility to pcretest so that putting tests into the
369        wrong test files can sometimes be quickly detected.
371    33. There is now a limit (default 250) on the depth of nesting of parentheses.
372        This limit is imposed to control the amount of system stack used at compile
373        time. It can be changed at build time by --with-parens-nest-limit=xxx or
374        the equivalent in CMake.
376    34. Character classes such as [A-\d] or [a-[:digit:]] now cause compile-time
377        errors. Perl warns for these when in warning mode, but PCRE has no facility
378        for giving warnings.
380    35. Change 34 for 8.13 allowed quantifiers on assertions, because Perl does.
381        However, this was not working for (?!) because it is optimized to (*FAIL),
382        for which PCRE does not allow quantifiers. The optimization is now disabled
383        when a quantifier follows (?!). I can't see any use for this, but it makes
384        things uniform.
386    36. Perl no longer allows group names to start with digits, so I have made this
387        change also in PCRE. It simplifies the code a bit.
389    37. In extended mode, Perl ignores spaces before a + that indicates a
390        possessive quantifier. PCRE allowed a space before the quantifier, but not
391        before the possessive +. It now does.
393    38. The use of \K (reset reported match start) within a repeated possessive
394        group such as (a\Kb)*+ was not working.
396    40. Document that the same character tables must be used at compile time and
397        run time, and that the facility to pass tables to pcre_exec() and
398        pcre_dfa_exec() is for use only with saved/restored patterns.
400    41. Applied Jeff Trawick's patch CMakeLists.txt, which "provides two new
401        features for Builds with MSVC:
403        1. Support pcre.rc and/or pcreposix.rc (as is already done for MinGW
404           builds). The .rc files can be used to set FileDescription and many other
405           attributes.
407        2. Add an option (-DINSTALL_MSVC_PDB) to enable installation of .pdb files.
408           This allows higher-level build scripts which want .pdb files to avoid
409           hard-coding the exact files needed."
411    42. Added support for [[:<:]] and [[:>:]] as used in the BSD POSIX library to
412        mean "start of word" and "end of word", respectively, as a transition aid.
414    43. A minimizing repeat of a class containing codepoints greater than 255 in
415        non-UTF 16-bit or 32-bit modes caused an internal error when PCRE was
416        compiled to use the heap for recursion.
418    44. Got rid of some compiler warnings for unused variables when UTF but not UCP
419        is configured.
422    Version 8.33 28-May-2013
423    ------------------------
425    1.  Added 'U' to some constants that are compared to unsigned integers, to
426        avoid compiler signed/unsigned warnings. Added (int) casts to unsigned
427        variables that are added to signed variables, to ensure the result is
428        signed and can be negated.
430    2.  Applied patch by Daniel Richard G for quashing MSVC warnings to the
431        CMake config files.
433    3.  Revise the creation of config.h.generic so that all boolean macros are
434        #undefined, whereas non-boolean macros are #ifndef/#endif-ed. This makes
435        overriding via -D on the command line possible.
437    4.  Changing the definition of the variable "op" in pcre_exec.c from pcre_uchar
438        to unsigned int is reported to make a quite noticeable speed difference in
439        a specific Windows environment. Testing on Linux did also appear to show
440        some benefit (and it is clearly not harmful). Also fixed the definition of
441        Xop which should be unsigned.
443    5.  Related to (4), changing the definition of the intermediate variable cc
444        in repeated character loops from pcre_uchar to pcre_uint32 also gave speed
445        improvements.
447    6.  Fix forward search in JIT when link size is 3 or greater. Also removed some
448        unnecessary spaces.
450    7.  Adjust autogen.sh and configure.ac to lose warnings given by automake 1.12
451        and later.
453    8.  Fix two buffer over read issues in 16 and 32 bit modes. Affects JIT only.
455    9.  Optimizing fast_forward_start_bits in JIT.
457    10. Adding support for callouts in JIT, and fixing some issues revealed
458        during this work. Namely:
460        (a) Unoptimized capturing brackets incorrectly reset on backtrack.
462        (b) Minimum length was not checked before the matching is started.
464    11. The value of capture_last that is passed to callouts was incorrect in some
465        cases when there was a capture on one path that was subsequently abandoned
466        after a backtrack. Also, the capture_last value is now reset after a
467        recursion, since all captures are also reset in this case.
469    12. The interpreter no longer returns the "too many substrings" error in the
470        case when an overflowing capture is in a branch that is subsequently
471        abandoned after a backtrack.
473    13. In the pathological case when an offset vector of size 2 is used, pcretest
474        now prints out the matched string after a yield of 0 or 1.
476    14. Inlining subpatterns in recursions, when certain conditions are fulfilled.
477        Only supported by the JIT compiler at the moment.
479    15. JIT compiler now supports 32 bit Macs thanks to Lawrence Velazquez.
481    16. Partial matches now set offsets[2] to the "bumpalong" value, that is, the
482        offset of the starting point of the matching process, provided the offsets
483        vector is large enough.
485    17. The \A escape now records a lookbehind value of 1, though its execution
486        does not actually inspect the previous character. This is to ensure that,
487        in partial multi-segment matching, at least one character from the old
488        segment is retained when a new segment is processed. Otherwise, if there
489        are no lookbehinds in the pattern, \A might match incorrectly at the start
490        of a new segment.
492    18. Added some #ifdef __VMS code into pcretest.c to help VMS implementations.
494    19. Redefined some pcre_uchar variables in pcre_exec.c as pcre_uint32; this
495        gives some modest performance improvement in 8-bit mode.
497    20. Added the PCRE-specific property \p{Xuc} for matching characters that can
498        be expressed in certain programming languages using Universal Character
499        Names.
501    21. Unicode validation has been updated in the light of Unicode Corrigendum #9,
502        which points out that "non characters" are not "characters that may not
503        appear in Unicode strings" but rather "characters that are reserved for
504        internal use and have only local meaning".
506    22. When a pattern was compiled with automatic callouts (PCRE_AUTO_CALLOUT) and
507        there was a conditional group that depended on an assertion, if the
508        assertion was false, the callout that immediately followed the alternation
509        in the condition was skipped when pcre_exec() was used for matching.
511    23. Allow an explicit callout to be inserted before an assertion that is the
512        condition for a conditional group, for compatibility with automatic
513        callouts, which always insert a callout at this point.
515    24. In 8.31, (*COMMIT) was confined to within a recursive subpattern. Perl also
516        confines (*SKIP) and (*PRUNE) in the same way, and this has now been done.
518    25. (*PRUNE) is now supported by the JIT compiler.
520    26. Fix infinite loop when /(?<=(*SKIP)ac)a/ is matched against aa.
522    27. Fix the case where there are two or more SKIPs with arguments that may be
523        ignored.
525    28. (*SKIP) is now supported by the JIT compiler.
527    29. (*THEN) is now supported by the JIT compiler.
529    30. Update RunTest with additional test selector options.
531    31. The way PCRE handles backtracking verbs has been changed in two ways.
533        (1) Previously, in something like (*COMMIT)(*SKIP), COMMIT would override
534        SKIP. Now, PCRE acts on whichever backtracking verb is reached first by
535        backtracking. In some cases this makes it more Perl-compatible, but Perl's
536        rather obscure rules do not always do the same thing.
538        (2) Previously, backtracking verbs were confined within assertions. This is
539        no longer the case for positive assertions, except for (*ACCEPT). Again,
540        this sometimes improves Perl compatibility, and sometimes does not.
542    32. A number of tests that were in test 2 because Perl did things differently
543        have been moved to test 1, because either Perl or PCRE has changed, and
544        these tests are now compatible.
546    32. Backtracking control verbs are now handled in the same way in JIT and
547        interpreter.
549    33. An opening parenthesis in a MARK/PRUNE/SKIP/THEN name in a pattern that
550        contained a forward subroutine reference caused a compile error.
552    34. Auto-detect and optimize limited repetitions in JIT.
554    35. Implement PCRE_NEVER_UTF to lock out the use of UTF, in particular,
555        blocking (*UTF) etc.
557    36. In the interpreter, maximizing pattern repetitions for characters and
558        character types now use tail recursion, which reduces stack usage.
560    37. The value of the max lookbehind was not correctly preserved if a compiled
561        and saved regex was reloaded on a host of different endianness.
563    38. Implemented (*LIMIT_MATCH) and (*LIMIT_RECURSION). As part of the extension
564        of the compiled pattern block, expand the flags field from 16 to 32 bits
565        because it was almost full.
567    39. Try madvise first before posix_madvise.
569    40. Change 7 for PCRE 7.9 made it impossible for pcregrep to find empty lines
570        with a pattern such as ^$. It has taken 4 years for anybody to notice! The
571        original change locked out all matches of empty strings. This has been
572        changed so that one match of an empty string per line is recognized.
573        Subsequent searches on the same line (for colouring or for --only-matching,
574        for example) do not recognize empty strings.
576    41. Applied a user patch to fix a number of spelling mistakes in comments.
578    42. Data lines longer than 65536 caused pcretest to crash.
580    43. Clarified the data type for length and startoffset arguments for pcre_exec
581        and pcre_dfa_exec in the function-specific man pages, where they were
582        explicitly stated to be in bytes, never having been updated. I also added
583        some clarification to the pcreapi man page.
585    44. A call to pcre_dfa_exec() with an output vector size less than 2 caused
586        a segmentation fault.
589    Version 8.32 30-November-2012
590    -----------------------------
592    1.  Improved JIT compiler optimizations for first character search and single
593        character iterators.
595    2.  Supporting IBM XL C compilers for PPC architectures in the JIT compiler.
596        Patch by Daniel Richard G.
598    3.  Single character iterator optimizations in the JIT compiler.
600    4.  Improved JIT compiler optimizations for character ranges.
602    5.  Rename the "leave" variable names to "quit" to improve WinCE compatibility.
603        Reported by Giuseppe D'Angelo.
605    6.  The PCRE_STARTLINE bit, indicating that a match can occur only at the start
606        of a line, was being set incorrectly in cases where .* appeared inside
607        atomic brackets at the start of a pattern, or where there was a subsequent
608        *PRUNE or *SKIP.
610    7.  Improved instruction cache flush for POWER/PowerPC.
611        Patch by Daniel Richard G.
613    8.  Fixed a number of issues in pcregrep, making it more compatible with GNU
614        grep:
616        (a) There is now no limit to the number of patterns to be matched.
618        (b) An error is given if a pattern is too long.
620        (c) Multiple uses of --exclude, --exclude-dir, --include, and --include-dir
621            are now supported.
623        (d) --exclude-from and --include-from (multiple use) have been added.
625        (e) Exclusions and inclusions now apply to all files and directories, not
626            just to those obtained from scanning a directory recursively.
628        (f) Multiple uses of -f and --file-list are now supported.
630        (g) In a Windows environment, the default for -d has been changed from
631            "read" (the GNU grep default) to "skip", because otherwise the presence
632            of a directory in the file list provokes an error.
634        (h) The documentation has been revised and clarified in places.
636    9.  Improve the matching speed of capturing brackets.
638    10. Changed the meaning of \X so that it now matches a Unicode extended
639        grapheme cluster.
641    11. Patch by Daniel Richard G to the autoconf files to add a macro for sorting
642        out POSIX threads when JIT support is configured.
644    12. Added support for PCRE_STUDY_EXTRA_NEEDED.
646    13. In the POSIX wrapper regcomp() function, setting re_nsub field in the preg
647        structure could go wrong in environments where size_t is not the same size
648        as int.
650    14. Applied user-supplied patch to pcrecpp.cc to allow PCRE_NO_UTF8_CHECK to be
651        set.
653    15. The EBCDIC support had decayed; later updates to the code had included
654        explicit references to (e.g.) \x0a instead of CHAR_LF. There has been a
655        general tidy up of EBCDIC-related issues, and the documentation was also
656        not quite right. There is now a test that can be run on ASCII systems to
657        check some of the EBCDIC-related things (but is it not a full test).
659    16. The new PCRE_STUDY_EXTRA_NEEDED option is now used by pcregrep, resulting
660        in a small tidy to the code.
662    17. Fix JIT tests when UTF is disabled and both 8 and 16 bit mode are enabled.
664    18. If the --only-matching (-o) option in pcregrep is specified multiple
665        times, each one causes appropriate output. For example, -o1 -o2 outputs the
666        substrings matched by the 1st and 2nd capturing parentheses. A separating
667        string can be specified by --om-separator (default empty).
669    19. Improving the first n character searches.
671    20. Turn case lists for horizontal and vertical white space into macros so that
672        they are defined only once.
674    21. This set of changes together give more compatible Unicode case-folding
675        behaviour for characters that have more than one other case when UCP
676        support is available.
678        (a) The Unicode property table now has offsets into a new table of sets of
679            three or more characters that are case-equivalent. The MultiStage2.py
680            script that generates these tables (the pcre_ucd.c file) now scans
681            CaseFolding.txt instead of UnicodeData.txt for character case
682            information.
684        (b) The code for adding characters or ranges of characters to a character
685            class has been abstracted into a generalized function that also handles
686            case-independence. In UTF-mode with UCP support, this uses the new data
687            to handle characters with more than one other case.
689        (c) A bug that is fixed as a result of (b) is that codepoints less than 256
690            whose other case is greater than 256 are now correctly matched
691            caselessly. Previously, the high codepoint matched the low one, but not
692            vice versa.
694        (d) The processing of \h, \H, \v, and \ in character classes now makes use
695            of the new class addition function, using character lists defined as
696            macros alongside the case definitions of 20 above.
698        (e) Caseless back references now work with characters that have more than
699            one other case.
701        (f) General caseless matching of characters with more than one other case
702            is supported.
704    22. Unicode character properties were updated from Unicode 6.2.0
706    23. Improved CMake support under Windows. Patch by Daniel Richard G.
708    24. Add support for 32-bit character strings, and UTF-32
710    25. Major JIT compiler update (code refactoring and bugfixing).
711        Experimental Sparc 32 support is added.
713    26. Applied a modified version of Daniel Richard G's patch to create
714        pcre.h.generic and config.h.generic by "make" instead of in the
715        PrepareRelease script.
717    27. Added a definition for CHAR_NULL (helpful for the z/OS port), and use it in
718        pcre_compile.c when checking for a zero character.
720    28. Introducing a native interface for JIT. Through this interface, the compiled
721        machine code can be directly executed. The purpose of this interface is to
722        provide fast pattern matching, so several sanity checks are not performed.
723        However, feature tests are still performed. The new interface provides
724        1.4x speedup compared to the old one.
726    29. If pcre_exec() or pcre_dfa_exec() was called with a negative value for
727        the subject string length, the error given was PCRE_ERROR_BADOFFSET, which
728        was confusing. There is now a new error PCRE_ERROR_BADLENGTH for this case.
730    30. In 8-bit UTF-8 mode, pcretest failed to give an error for data codepoints
731        greater than 0x7fffffff (which cannot be represented in UTF-8, even under
732        the "old" RFC 2279). Instead, it ended up passing a negative length to
733        pcre_exec().
735    31. Add support for GCC's visibility feature to hide internal functions.
737    32. Running "pcretest -C pcre8" or "pcretest -C pcre16" gave a spurious error
738        "unknown -C option" after outputting 0 or 1.
740    33. There is now support for generating a code coverage report for the test
741        suite in environments where gcc is the compiler and lcov is installed. This
742        is mainly for the benefit of the developers.
744    34. If PCRE is built with --enable-valgrind, certain memory regions are marked
745        unaddressable using valgrind annotations, allowing valgrind to detect
746        invalid memory accesses. This is mainly for the benefit of the developers.
748    25. (*UTF) can now be used to start a pattern in any of the three libraries.
750    26. Give configure error if --enable-cpp but no C++ compiler found.
753    Version 8.31 06-July-2012
754    -------------------------
756    1.  Fixing a wrong JIT test case and some compiler warnings.
758    2.  Removed a bashism from the RunTest script.
760    3.  Add a cast to pcre_exec.c to fix the warning "unary minus operator applied
761        to unsigned type, result still unsigned" that was given by an MS compiler
762        on encountering the code "-sizeof(xxx)".
764    4.  Partial matching support is added to the JIT compiler.
766    5.  Fixed several bugs concerned with partial matching of items that consist
767        of more than one character:
769        (a) /^(..)\1/ did not partially match "aba" because checking references was
770            done on an "all or nothing" basis. This also applied to repeated
771            references.
773        (b) \R did not give a hard partial match if \r was found at the end of the
774            subject.
776        (c) \X did not give a hard partial match after matching one or more
777            characters at the end of the subject.
779        (d) When newline was set to CRLF, a pattern such as /a$/ did not recognize
780            a partial match for the string "\r".
782        (e) When newline was set to CRLF, the metacharacter "." did not recognize
783            a partial match for a CR character at the end of the subject string.
785    6.  If JIT is requested using /S++ or -s++ (instead of just /S+ or -s+) when
786        running pcretest, the text "(JIT)" added to the output whenever JIT is
787        actually used to run the match.
789    7.  Individual JIT compile options can be set in pcretest by following -s+[+]
790        or /S+[+] with a digit between 1 and 7.
792    8.  OP_NOT now supports any UTF character not just single-byte ones.
794    9.  (*MARK) control verb is now supported by the JIT compiler.
796    10. The command "./RunTest list" lists the available tests without actually
797        running any of them. (Because I keep forgetting what they all are.)
801    12. Applied a (slightly modified) user-supplied patch that improves performance
802        when the heap is used for recursion (compiled with --disable-stack-for-
803        recursion). Instead of malloc and free for each heap frame each time a
804        logical recursion happens, frames are retained on a chain and re-used where
805        possible. This sometimes gives as much as 30% improvement.
807    13. As documented, (*COMMIT) is now confined to within a recursive subpattern
808        call.
810    14. As documented, (*COMMIT) is now confined to within a positive assertion.
812    15. It is now possible to link pcretest with libedit as an alternative to
813        libreadline.
815    16. (*COMMIT) control verb is now supported by the JIT compiler.
817    17. The Unicode data tables have been updated to Unicode 6.1.0.
819    18. Added --file-list option to pcregrep.
821    19. Added binary file support to pcregrep, including the -a, --binary-files,
822        -I, and --text options.
824    20. The madvise function is renamed for posix_madvise for QNX compatibility
825        reasons. Fixed by Giuseppe D'Angelo.
827    21. Fixed a bug for backward assertions with REVERSE 0 in the JIT compiler.
829    22. Changed the option for creating symbolic links for 16-bit man pages from
830        -s to -sf so that re-installing does not cause issues.
832    23. Support PCRE_NO_START_OPTIMIZE in JIT as (*MARK) support requires it.
834    24. Fixed a very old bug in pcretest that caused errors with restarted DFA
835        matches in certain environments (the workspace was not being correctly
836        retained). Also added to pcre_dfa_exec() a simple plausibility check on
837        some of the workspace data at the beginning of a restart.
839    25. \s*\R was auto-possessifying the \s* when it should not, whereas \S*\R
840        was not doing so when it should - probably a typo introduced by SVN 528
841        (change 8.10/14).
843    26. When PCRE_UCP was not set, \w+\x{c4} was incorrectly auto-possessifying the
844        \w+ when the character tables indicated that \x{c4} was a word character.
845        There were several related cases, all because the tests for doing a table
846        lookup were testing for characters less than 127 instead of 255.
848    27. If a pattern contains capturing parentheses that are not used in a match,
849        their slots in the ovector are set to -1. For those that are higher than
850        any matched groups, this happens at the end of processing. In the case when
851        there were back references that the ovector was too small to contain
852        (causing temporary malloc'd memory to be used during matching), and the
853        highest capturing number was not used, memory off the end of the ovector
854        was incorrectly being set to -1. (It was using the size of the temporary
855        memory instead of the true size.)
857    28. To catch bugs like 27 using valgrind, when pcretest is asked to specify an
858        ovector size, it uses memory at the end of the block that it has got.
860    29. Check for an overlong MARK name and give an error at compile time. The
861        limit is 255 for the 8-bit library and 65535 for the 16-bit library.
863    30. JIT compiler update.
865    31. JIT is now supported on jailbroken iOS devices. Thanks for Ruiger
866        Rill for the patch.
868    32. Put spaces around SLJIT_PRINT_D in the JIT compiler. Required by CXX11.
870    33. Variable renamings in the PCRE-JIT compiler. No functionality change.
872    34. Fixed typos in pcregrep: in two places there was SUPPORT_LIBZ2 instead of
873        SUPPORT_LIBBZ2. This caused a build problem when bzip2 but not gzip (zlib)
874        was enabled.
876    35. Improve JIT code generation for greedy plus quantifier.
878    36. When /((?:a?)*)*c/ or /((?>a?)*)*c/ was matched against "aac", it set group
879        1 to "aa" instead of to an empty string. The bug affected repeated groups
880        that could potentially match an empty string.
882    37. Optimizing single character iterators in JIT.
884    38. Wide characters specified with \uxxxx in JavaScript mode are now subject to
885        the same checks as \x{...} characters in non-JavaScript mode. Specifically,
886        codepoints that are too big for the mode are faulted, and in a UTF mode,
887        disallowed codepoints are also faulted.
889    39. If PCRE was compiled with UTF support, in three places in the DFA
890        matcher there was code that should only have been obeyed in UTF mode, but
891        was being obeyed unconditionally. In 8-bit mode this could cause incorrect
892        processing when bytes with values greater than 127 were present. In 16-bit
893        mode the bug would be provoked by values in the range 0xfc00 to 0xdc00. In
894        both cases the values are those that cannot be the first data item in a UTF
895        character. The three items that might have provoked this were recursions,
896        possessively repeated groups, and atomic groups.
898    40. Ensure that libpcre is explicitly listed in the link commands for pcretest
899        and pcregrep, because some OS require shared objects to be explicitly
900        passed to ld, causing the link step to fail if they are not.
902    41. There were two incorrect #ifdefs in pcre_study.c, meaning that, in 16-bit
903        mode, patterns that started with \h* or \R* might be incorrectly matched.
906    Version 8.30 04-February-2012
907    -----------------------------
909    1.  Renamed "isnumber" as "is_a_number" because in some Mac environments this
910        name is defined in ctype.h.
912    2.  Fixed a bug in fixed-length calculation for lookbehinds that would show up
913        only in quite long subpatterns.
915    3.  Removed the function pcre_info(), which has been obsolete and deprecated
916        since it was replaced by pcre_fullinfo() in February 2000.
918    4.  For a non-anchored pattern, if (*SKIP) was given with a name that did not
919        match a (*MARK), and the match failed at the start of the subject, a
920        reference to memory before the start of the subject could occur. This bug
921        was introduced by fix 17 of release 8.21.
923    5.  A reference to an unset group with zero minimum repetition was giving
924        totally wrong answers (in non-JavaScript-compatibility mode). For example,
925        /(another)?(\1?)test/ matched against "hello world test". This bug was
926        introduced in release 8.13.
928    6.  Add support for 16-bit character strings (a large amount of work involving
929        many changes and refactorings).
931    7.  RunGrepTest failed on msys because \r\n was replaced by whitespace when the
932        command "pattern=`printf 'xxx\r\njkl'`" was run. The pattern is now taken
933        from a file.
935    8.  Ovector size of 2 is also supported by JIT based pcre_exec (the ovector size
936        rounding is not applied in this particular case).
938    9.  The invalid Unicode surrogate codepoints U+D800 to U+DFFF are now rejected
939        if they appear, or are escaped, in patterns.
941    10. Get rid of a number of -Wunused-but-set-variable warnings.
943    11. The pattern /(?=(*:x))(q|)/ matches an empty string, and returns the mark
944        "x". The similar pattern /(?=(*:x))((*:y)q|)/ did not return a mark at all.
945        Oddly, Perl behaves the same way. PCRE has been fixed so that this pattern
946        also returns the mark "x". This bug applied to capturing parentheses,
947        non-capturing parentheses, and atomic parentheses. It also applied to some
948        assertions.
950    12. Stephen Kelly's patch to CMakeLists.txt allows it to parse the version
951        information out of configure.ac instead of relying on pcre.h.generic, which
952        is not stored in the repository.
954    13. Applied Dmitry V. Levin's patch for a more portable method for linking with
955        -lreadline.
957    14. ZH added PCRE_CONFIG_JITTARGET; added its output to pcretest -C.
959    15. Applied Graycode's patch to put the top-level frame on the stack rather
960        than the heap when not using the stack for recursion. This gives a
961        performance improvement in many cases when recursion is not deep.
963    16. Experimental code added to "pcretest -C" to output the stack frame size.
966    Version 8.21 12-Dec-2011
967    ------------------------
969    1.  Updating the JIT compiler.
971    2.  JIT compiler now supports OP_NCREF, OP_RREF and OP_NRREF. New test cases
972        are added as well.
974    3.  Fix cache-flush issue on PowerPC (It is still an experimental JIT port).
975        PCRE_EXTRA_TABLES is not suported by JIT, and should be checked before
976        calling _pcre_jit_exec. Some extra comments are added.
978    4.  (*MARK) settings inside atomic groups that do not contain any capturing
979        parentheses, for example, (?>a(*:m)), were not being passed out. This bug
980        was introduced by change 18 for 8.20.
982    5.  Supporting of \x, \U and \u in JavaScript compatibility mode based on the
983        ECMA-262 standard.
985    6.  Lookbehinds such as (?<=a{2}b) that contained a fixed repetition were
986        erroneously being rejected as "not fixed length" if PCRE_CASELESS was set.
987        This bug was probably introduced by change 9 of 8.13.
989    7.  While fixing 6 above, I noticed that a number of other items were being
990        incorrectly rejected as "not fixed length". This arose partly because newer
991        opcodes had not been added to the fixed-length checking code. I have (a)
992        corrected the bug and added tests for these items, and (b) arranged for an
993        error to occur if an unknown opcode is encountered while checking for fixed
994        length instead of just assuming "not fixed length". The items that were
995        rejected were: (*ACCEPT), (*COMMIT), (*FAIL), (*MARK), (*PRUNE), (*SKIP),
996        (*THEN), \h, \H, \v, \V, and single character negative classes with fixed
997        repetitions, e.g. [^a]{3}, with and without PCRE_CASELESS.
999    8.  A possessively repeated conditional subpattern such as (?(?=c)c|d)++ was
1000        being incorrectly compiled and would have given unpredicatble results.
1002    9.  A possessively repeated subpattern with minimum repeat count greater than
1003        one behaved incorrectly. For example, (A){2,}+ behaved as if it was
1004        (A)(A)++ which meant that, after a subsequent mismatch, backtracking into
1005        the first (A) could occur when it should not.
1007    10. Add a cast and remove a redundant test from the code.
1009    11. JIT should use pcre_malloc/pcre_free for allocation.
1011    12. Updated pcre-config so that it no longer shows -L/usr/lib, which seems
1012        best practice nowadays, and helps with cross-compiling. (If the exec_prefix
1013        is anything other than /usr, -L is still shown).
1015    13. In non-UTF-8 mode, \C is now supported in lookbehinds and DFA matching.
1017    14. Perl does not support \N without a following name in a [] class; PCRE now
1018        also gives an error.
1020    15. If a forward reference was repeated with an upper limit of around 2000,
1021        it caused the error "internal error: overran compiling workspace". The
1022        maximum number of forward references (including repeats) was limited by the
1023        internal workspace, and dependent on the LINK_SIZE. The code has been
1024        rewritten so that the workspace expands (via pcre_malloc) if necessary, and
1025        the default depends on LINK_SIZE. There is a new upper limit (for safety)
1026        of around 200,000 forward references. While doing this, I also speeded up
1027        the filling in of repeated forward references.
1029    16. A repeated forward reference in a pattern such as (a)(?2){2}(.) was
1030        incorrectly expecting the subject to contain another "a" after the start.
1032    17. When (*SKIP:name) is activated without a corresponding (*MARK:name) earlier
1033        in the match, the SKIP should be ignored. This was not happening; instead
1034        the SKIP was being treated as NOMATCH. For patterns such as
1035        /A(*MARK:A)A+(*SKIP:B)Z|AAC/ this meant that the AAC branch was never
1036        tested.
1038    18. The behaviour of (*MARK), (*PRUNE), and (*THEN) has been reworked and is
1039        now much more compatible with Perl, in particular in cases where the result
1040        is a non-match for a non-anchored pattern. For example, if
1041        /b(*:m)f|a(*:n)w/ is matched against "abc", the non-match returns the name
1042        "m", where previously it did not return a name. A side effect of this
1043        change is that for partial matches, the last encountered mark name is
1044        returned, as for non matches. A number of tests that were previously not
1045        Perl-compatible have been moved into the Perl-compatible test files. The
1046        refactoring has had the pleasing side effect of removing one argument from
1047        the match() function, thus reducing its stack requirements.
1049    19. If the /S+ option was used in pcretest to study a pattern using JIT,
1050        subsequent uses of /S (without +) incorrectly behaved like /S+.
1052    21. Retrieve executable code size support for the JIT compiler and fixing
1053        some warnings.
1055    22. A caseless match of a UTF-8 character whose other case uses fewer bytes did
1056        not work when the shorter character appeared right at the end of the
1057        subject string.
1059    23. Added some (int) casts to non-JIT modules to reduce warnings on 64-bit
1060        systems.
1062    24. Added PCRE_INFO_JITSIZE to pass on the value from (21) above, and also
1063        output it when the /M option is used in pcretest.
1065    25. The CheckMan script was not being included in the distribution. Also, added
1066        an explicit "perl" to run Perl scripts from the PrepareRelease script
1067        because this is reportedly needed in Windows.
1069    26. If study data was being save in a file and studying had not found a set of
1070        "starts with" bytes for the pattern, the data written to the file (though
1071        never used) was taken from uninitialized memory and so caused valgrind to
1072        complain.
1074    27. Updated RunTest.bat as provided by Sheri Pierce.
1076    28. Fixed a possible uninitialized memory bug in pcre_jit_compile.c.
1078    29. Computation of memory usage for the table of capturing group names was
1079        giving an unnecessarily large value.
1082    Version 8.20 21-Oct-2011
1083    ------------------------
1085    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
1086        a POSIX class. After further experiments with Perl, which convinced me that
1087        Perl has bugs and confusions, a closing square bracket is no longer allowed
1088        in a POSIX name. This bug also affected patterns with classes that started
1089        with full stops.
1091    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
1092        captured substring, but while checking the failing first alternative,
1093        substring 1 is temporarily captured. If the output vector supplied to
1094        pcre_exec() was not big enough for this capture, the yield of the function
1095        was still zero ("insufficient space for captured substrings"). This cannot
1096        be totally fixed without adding another stack variable, which seems a lot
1097        of expense for a edge case. However, I have improved the situation in cases
1098        such as /(a)(b)x|abc/ matched against "abc", where the return code
1099        indicates that fewer than the maximum number of slots in the ovector have
1100        been set.
1102    3.  Related to (2) above: when there are more back references in a pattern than
1103        slots in the output vector, pcre_exec() uses temporary memory during
1104        matching, and copies in the captures as far as possible afterwards. It was
1105        using the entire output vector, but this conflicts with the specification
1106        that only 2/3 is used for passing back captured substrings. Now it uses
1107        only the first 2/3, for compatibility. This is, of course, another edge
1108        case.
1110    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
1111        main code base, and can be used by building with --enable-jit. When this is
1112        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
1113        runtime --no-jit option is given.
1115    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
1116        ovector, the return from the function was zero, implying that there were
1117        other matches that did not fit. The correct "exactly full" value is now
1118        returned.
1120    6.  If a subpattern that was called recursively or as a subroutine contained
1121        (*PRUNE) or any other control that caused it to give a non-standard return,
1122        invalid errors such as "Error -26 (nested recursion at the same subject
1123        position)" or even infinite loops could occur.
1125    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
1126        computing the minimum length on reaching *ACCEPT, and so ended up with the
1127        wrong value of 1 rather than 0. Further investigation indicates that
1128        computing a minimum subject length in the presence of *ACCEPT is difficult
1129        (think back references, subroutine calls), and so I have changed the code
1130        so that no minimum is registered for a pattern that contains *ACCEPT.
1132    8.  If (*THEN) was present in the first (true) branch of a conditional group,
1133        it was not handled as intended. [But see 16 below.]
1135    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
1136        Sheri Pierce.
1138    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
1139        the first byte in a match must be "a".
1141    11. Change 17 for 8.13 increased the recursion depth for patterns like
1142        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
1143        pattern contains any instances of (*THEN). If it does not, the old
1144        optimizations are restored. It would be nice to do this on a per-group
1145        basis, but at the moment that is not feasible.
1147    12. In some environments, the output of pcretest -C is CRLF terminated. This
1148        broke RunTest's code that checks for the link size. A single white space
1149        character after the value is now allowed for.
1151    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
1152        For "fr", it uses the Windows-specific input and output files.
1154    14. If (*THEN) appeared in a group that was called recursively or as a
1155        subroutine, it did not work as intended. [But see next item.]
1157    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
1158        pattern fragments (but not containing any | characters). If A and B are
1159        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
1160        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
1161        D. In other words, Perl considers parentheses that do not contain any |
1162        characters to be part of a surrounding alternative, whereas PCRE was
1163        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
1164        differently. PCRE now behaves in the same way as Perl, except in the case
1165        of subroutine/recursion calls such as (?1) which have in any case always
1166        been different (but PCRE had them first :-).
1168    16. Related to 15 above: Perl does not treat the | in a conditional group as
1169        creating alternatives. Such a group is treated in the same way as an
1170        ordinary group without any | characters when processing (*THEN). PCRE has
1171        been changed to match Perl's behaviour.
1173    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
1174        RunGrepTest script failed.
1176    18. Change 22 for version 13 caused atomic groups to use more stack. This is
1177        inevitable for groups that contain captures, but it can lead to a lot of
1178        stack use in large patterns. The old behaviour has been restored for atomic
1179        groups that do not contain any capturing parentheses.
1181    19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not
1182        suppress the check for a minimum subject length at run time. (If it was
1183        given to pcre_exec() or pcre_dfa_exec() it did work.)
1185    20. Fixed an ASCII-dependent infelicity in pcretest that would have made it
1186        fail to work when decoding hex characters in data strings in EBCDIC
1187        environments.
1189    21. It appears that in at least one Mac OS environment, the isxdigit() function
1190        is implemented as a macro that evaluates to its argument more than once,
1191        contravening the C 90 Standard (I haven't checked a later standard). There
1192        was an instance in pcretest which caused it to go wrong when processing
1193        \x{...} escapes in subject strings. The has been rewritten to avoid using
1194        things like p++ in the argument of isxdigit().
1197    Version 8.13 16-Aug-2011
1198    ------------------------
1200    1.  The Unicode data tables have been updated to Unicode 6.0.0.
1202    2.  Two minor typos in pcre_internal.h have been fixed.
1204    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
1205        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
1206        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
1208    4.  There were a number of related bugs in the code for matching backrefences
1209        caselessly in UTF-8 mode when codes for the characters concerned were
1210        different numbers of bytes. For example, U+023A and U+2C65 are an upper
1211        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
1212        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
1213        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
1214        2-byte code at the end of the subject (it thought there wasn't enough data
1215        left).
1217    5.  Comprehensive information about what went wrong is now returned by
1218        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
1219        as the output vector has at least 2 elements. The offset of the start of
1220        the failing character and a reason code are placed in the vector.
1222    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
1223        now returned is for the first byte of the failing character, instead of the
1224        last byte inspected. This is an incompatible change, but I hope it is small
1225        enough not to be a problem. It makes the returned offset consistent with
1226        pcre_exec() and pcre_dfa_exec().
1228    7.  pcretest now gives a text phrase as well as the error number when
1229        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
1230        failure, the offset and reason code are output.
1232    8.  When \R was used with a maximizing quantifier it failed to skip backwards
1233        over a \r\n pair if the subsequent match failed. Instead, it just skipped
1234        back over a single character (\n). This seems wrong (because it treated the
1235        two characters as a single entity when going forwards), conflicts with the
1236        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
1237        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
1238        has been changed.
1240    9.  Some internal refactoring has changed the processing so that the handling
1241        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
1242        time (the PCRE_DOTALL option was changed this way some time ago: version
1243        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
1244        which was always a bit of a fudge. It also means that there is one less
1245        argument for the match() function, which reduces its stack requirements
1246        slightly. This change also fixes an incompatibility with Perl: the pattern
1247        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
1249    10. More internal refactoring has drastically reduced the number of recursive
1250        calls to match() for possessively repeated groups such as (abc)++ when
1251        using pcre_exec().
1253    11. While implementing 10, a number of bugs in the handling of groups were
1254        discovered and fixed:
1256        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
1257        (a|)*(?1) gave a compile-time internal error.
1258        ((a|)+)+  did not notice that the outer group could match an empty string.
1259        (^a|^)+   was not marked as anchored.
1260        (.*a|.*)+ was not marked as matching at start or after a newline.
1262    12. Yet more internal refactoring has removed another argument from the match()
1263        function. Special calls to this function are now indicated by setting a
1264        value in a variable in the "match data" data block.
1266    13. Be more explicit in pcre_study() instead of relying on "default" for
1267        opcodes that mean there is no starting character; this means that when new
1268        ones are added and accidentally left out of pcre_study(), testing should
1269        pick them up.
1271    14. The -s option of pcretest has been documented for ages as being an old
1272        synonym of -m (show memory usage). I have changed it to mean "force study
1273        for every regex", that is, assume /S for every regex. This is similar to -i
1274        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
1275        using it. It makes it easier to run collections of tests with and without
1276        study enabled, and thereby test pcre_study() more easily. All the standard
1277        tests are now run with and without -s (but some patterns can be marked as
1278        "never study" - see 20 below).
1280    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
1281        restoration of the capturing data to the outer values was not happening
1282        correctly.
1284    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
1285        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
1286        pattern had matched an empty string, and so incorrectly returned a no
1287        match.
1289    17. There was optimizing code for the last branch of non-capturing parentheses,
1290        and also for the obeyed branch of a conditional subexpression, which used
1291        tail recursion to cut down on stack usage. Unfortunately, now that there is
1292        the possibility of (*THEN) occurring in these branches, tail recursion is
1293        no longer possible because the return has to be checked for (*THEN). These
1294        two optimizations have therefore been removed. [But see 8.20/11 above.]
1296    18. If a pattern containing \R was studied, it was assumed that \R always
1297        matched two bytes, thus causing the minimum subject length to be
1298        incorrectly computed because \R can also match just one byte.
1300    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
1301        was incorrectly computed.
1303    20. If /S is present twice on a test pattern in pcretest input, it now
1304        *disables* studying, thereby overriding the use of -s on the command line
1305        (see 14 above). This is necessary for one or two tests to keep the output
1306        identical in both cases.
1308    21. When (*ACCEPT) was used in an assertion that matched an empty string and
1309        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
1311    22. When an atomic group that contained a capturing parenthesis was
1312        successfully matched, but the branch in which it appeared failed, the
1313        capturing was not being forgotten if a higher numbered group was later
1314        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
1315        group 1 to "a", when in fact it should be unset. This applied to multi-
1316        branched capturing and non-capturing groups, repeated or not, and also to
1317        positive assertions (capturing in negative assertions does not happen
1318        in PCRE) and also to nested atomic groups.
1320    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
1321        subject after a captured substring, to make it easier to tell which of a
1322        number of identical substrings has been captured.
1324    24. The way atomic groups are processed by pcre_exec() has been changed so that
1325        if they are repeated, backtracking one repetition now resets captured
1326        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
1327        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
1328        "aaa". Previously, it would have been "a". As part of this code
1329        refactoring, the way recursive calls are handled has also been changed.
1331    25. If an assertion condition captured any substrings, they were not passed
1332        back unless some other capturing happened later. For example, if
1333        (?(?=(a))a) was matched against "a", no capturing was returned.
1335    26. When studying a pattern that contained subroutine calls or assertions,
1336        the code for finding the minimum length of a possible match was handling
1337        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
1338        group 1 called group 2 while simultaneously a separate group 2 called group
1339        1). A stack overflow occurred in this case. I have fixed this by limiting
1340        the recursion depth to 10.
1342    27. Updated RunTest.bat in the distribution to the version supplied by Tom
1343        Fortmann. This supports explicit test numbers on the command line, and has
1344        argument validation and error reporting.
1346    28. An instance of \X with an unlimited repeat could fail if at any point the
1347        first character it looked at was a mark character.
1349    29. Some minor code refactoring concerning Unicode properties and scripts
1350        should reduce the stack requirement of match() slightly.
1352    30. Added the '=' option to pcretest to check the setting of unused capturing
1353        slots at the end of the pattern, which are documented as being -1, but are
1354        not included in the return count.
1356    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
1357        compiled something random. Now it gives a compile-time error (as does
1358        Perl).
1360    32. A *MARK encountered during the processing of a positive assertion is now
1361        recorded and passed back (compatible with Perl).
1363    33. If --only-matching or --colour was set on a pcregrep call whose pattern
1364        had alternative anchored branches, the search for a second match in a line
1365        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
1366        matched the line "0102" twice. The same bug affected patterns that started
1367        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
1368        twice.
1370    34. Previously, PCRE did not allow quantification of assertions. However, Perl
1371        does, and because of capturing effects, quantifying parenthesized
1372        assertions may at times be useful. Quantifiers are now allowed for
1373        parenthesized assertions.
1375    35. A minor code tidy in pcre_compile() when checking options for \R usage.
1377    36. \g was being checked for fancy things in a character class, when it should
1378        just be a literal "g".
1380    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
1381        appearance of a nested POSIX class supersedes an apparent external class.
1382        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
1383        unescaped square brackets may also appear as part of class names. For
1384        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
1385        more like Perl. (But see 8.20/1 above.)
1387    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
1388        was because it thought it was \N{name}, which is not supported).
1390    39. Add minix to OS list not supporting the -S option in pcretest.
1392    40. PCRE tries to detect cases of infinite recursion at compile time, but it
1393        cannot analyze patterns in sufficient detail to catch mutual recursions
1394        such as ((?1))((?2)). There is now a runtime test that gives an error if a
1395        subgroup is called recursively as a subpattern for a second time at the
1396        same position in the subject string. In previous releases this might have
1397        been caught by the recursion limit, or it might have run out of stack.
1399    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
1400        happen only once. PCRE was, however incorrectly giving a compile time error
1401        "recursive call could loop indefinitely" because it cannot analyze the
1402        pattern in sufficient detail. The compile time test no longer happens when
1403        PCRE is compiling a conditional subpattern, but actual runaway loops are
1404        now caught at runtime (see 40 above).
1406    42. It seems that Perl allows any characters other than a closing parenthesis
1407        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
1408        has been changed to be the same.
1410    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
1411        as not to get warnings when autogen.sh is called. Also changed
1412        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
1414    44. To help people who use pcregrep to scan files containing exceedingly long
1415        lines, the following changes have been made:
1417        (a) The default value of the buffer size parameter has been increased from
1418            8K to 20K. (The actual buffer used is three times this size.)
1420        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
1421            PCRE is built.
1423        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
1424            to be set at run time.
1426        (d) Numerical values in pcregrep options can be followed by K or M, for
1427            example --buffer-size=50K.
1429        (e) If a line being scanned overflows pcregrep's buffer, an error is now
1430            given and the return code is set to 2.
1432    45. Add a pointer to the latest mark to the callout data block.
1434    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
1435        partial match of an empty string instead of no match. This was specific to
1436        the use of ".".
1438    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
1439        complete match instead of a partial match. This bug was dependent on both
1440        the PCRE_UTF8 and PCRE_DOTALL options being set.
1442    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
1443        starting byte set, because \b was not being ignored.
1446    Version 8.12 15-Jan-2011
1447    ------------------------
1449    1.  Fixed some typos in the markup of the man pages, and wrote a script that
1450        checks for such things as part of the documentation building process.
1452    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
1453        --match-limit and --recursion-limit options (added for 8.11). In
1454        particular, this made one of the standard tests fail. (The integer value
1455        went into the wrong half of a long int.)
1457    3.  If the --colour option was given to pcregrep with -v (invert match), it
1458        did strange things, either producing crazy output, or crashing. It should,
1459        of course, ignore a request for colour when reporting lines that do not
1460        match.
1462    4.  Another pcregrep bug caused similar problems if --colour was specified with
1463        -M (multiline) and the pattern match finished with a line ending.
1465    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
1466        matched in multiline mode, the following line was shown as part of the
1467        match. This seems wrong, so I have changed it.
1469    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
1470        the check for further matches in the same line (so they could be coloured)
1471        to overrun the end of the current line. If another match was found, it was
1472        incorrectly shown (and then shown again when found in the next line).
1474    7.  If pcregrep was compiled under Windows, there was a reference to the
1475        function pcregrep_exit() before it was defined. I am assuming this was
1476        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
1477        reported by a user. I've moved the definition above the reference.
1480    Version 8.11 10-Dec-2010
1481    ------------------------
1483    1.  (*THEN) was not working properly if there were untried alternatives prior
1484        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
1485        backtracked to try for "b" instead of moving to the next alternative branch
1486        at the same level (in this case, to look for "c"). The Perl documentation
1487        is clear that when (*THEN) is backtracked onto, it goes to the "next
1488        alternative in the innermost enclosing group".
1490    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
1491        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
1492        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
1493        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
1494        (*THEN).
1496    3.  If \s appeared in a character class, it removed the VT character from
1497        the class, even if it had been included by some previous item, for example
1498        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
1499        of \s, but is part of the POSIX "space" class.)
1501    4.  A partial match never returns an empty string (because you can always
1502        match an empty string at the end of the subject); however the checking for
1503        an empty string was starting at the "start of match" point. This has been
1504        changed to the "earliest inspected character" point, because the returned
1505        data for a partial match starts at this character. This means that, for
1506        example, /(?<=abc)def/ gives a partial match for the subject "abc"
1507        (previously it gave "no match").
1509    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
1510        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
1511        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
1512        has an implication that the given string is incomplete (because a partial
1513        match is preferred over a full match). For this reason, these items now
1514        give a partial match in this situation. [Aside: previously, the one case
1515        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
1516        match rather than a full match, which was wrong by the old rules, but is
1517        now correct.]
1519    6.  There was a bug in the handling of #-introduced comments, recognized when
1520        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
1521        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
1522        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
1523        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
1524        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
1525        places in pcre_compile().
1527    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
1528        comments when looking ahead for named forward references to subpatterns,
1529        the only newline sequence it recognized was NL. It now handles newlines
1530        according to the set newline convention.
1532    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
1533        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
1534        cater for a lack of strerror(). These oversights have been fixed.
1536    9.  Added --match-limit and --recursion-limit to pcregrep.
1538    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
1540    11. When the -o option was used, pcregrep was setting a return code of 1, even
1541        when matches were found, and --line-buffered was not being honoured.
1543    12. Added an optional parentheses number to the -o and --only-matching options
1544        of pcregrep.
1546    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
1547        can match an empty string. The code to do it in pcretest and pcredemo
1548        needed fixing:
1550        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
1551            only one byte after an empty string match just before CRLF (this case
1552            just got forgotten; "any" and "anycrlf" were OK).
1554        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
1555            mode when an empty string match preceded an ASCII character followed by
1556            a non-ASCII character. (The code for advancing by one character rather
1557            than one byte was nonsense.)
1559        (c) The pcredemo.c sample program did not have any code at all to handle
1560            the cases when CRLF is a valid newline sequence.
1562    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
1563        as a starting offset was within the subject string. There is now a new
1564        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
1565        negative or greater than the length of the string. In order to test this,
1566        pcretest is extended to allow the setting of negative starting offsets.
1568    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
1569        starting offset points to the beginning of a UTF-8 character was
1570        unnecessarily clumsy. I tidied it up.
1572    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
1573        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
1575    17. Nobody had reported that the --include_dir option, which was added in
1576        release 7.7 should have been called --include-dir (hyphen, not underscore)
1577        for compatibility with GNU grep. I have changed it to --include-dir, but
1578        left --include_dir as an undocumented synonym, and the same for
1579        --exclude-dir, though that is not available in GNU grep, at least as of
1580        release 2.5.4.
1582    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
1583        characters from a string of bytes) have been redefined so as not to use
1584        loops, in order to improve performance in some environments. At the same
1585        time, I abstracted some of the common code into auxiliary macros to save
1586        repetition (this should not affect the compiled code).
1588    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
1589        compile-time error is now given if \c is not followed by an ASCII
1590        character, that is, a byte less than 128. (In EBCDIC mode, the code is
1591        different, and any byte value is allowed.)
1593    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
1594        START_OPTIMIZE option, which is now allowed at compile time - but just
1595        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
1596        to pcregrep and other applications that have no direct access to PCRE
1597        options. The new /Y option in pcretest sets this option when calling
1598        pcre_compile().
1600    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
1601        back references. Groups containing recursive back references were forced to
1602        be atomic by that change, but in the case of named groups, the amount of
1603        memory required was incorrectly computed, leading to "Failed: internal
1604        error: code overflow". This has been fixed.
1606    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
1607        pcretest.c, to avoid build problems in some Borland environments.
1610    Version 8.10 25-Jun-2010
1611    ------------------------
1613    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
1614        THEN.
1616    2.  (*ACCEPT) was not working when inside an atomic group.
1618    3.  Inside a character class, \B is treated as a literal by default, but
1619        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
1620        causes the error). The code is unchanged, but I tidied the documentation.
1622    4.  Inside a character class, PCRE always treated \R and \X as literals,
1623        whereas Perl faults them if its -w option is set. I have changed PCRE so
1624        that it faults them when PCRE_EXTRA is set.
1626    5.  Added support for \N, which always matches any character other than
1627        newline. (It is the same as "." when PCRE_DOTALL is not set.)
1629    6.  When compiling pcregrep with newer versions of gcc which may have
1630        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
1631        declared with attribute warn_unused_result" were given. Just casting the
1632        result to (void) does not stop the warnings; a more elaborate fudge is
1633        needed. I've used a macro to implement this.
1635    7.  Minor change to pcretest.c to avoid a compiler warning.
1637    8.  Added four artifical Unicode properties to help with an option to make
1638        \s etc use properties (see next item). The new properties are: Xan
1639        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
1641    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
1642        use Unicode properties. (*UCP) at the start of a pattern can be used to set
1643        this option. Modified pcretest to add /W to test this facility. Added
1644        REG_UCP to make it available via the POSIX interface.
1646    10. Added --line-buffered to pcregrep.
1648    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
1649        studied, and the match started with a letter with a code point greater than
1650        127 whose first byte was different to the first byte of the other case of
1651        the letter, the other case of this starting letter was not recognized
1652        (#976).
1654    12. If a pattern that was studied started with a repeated Unicode property
1655        test, for example, \p{Nd}+, there was the theoretical possibility of
1656        setting up an incorrect bitmap of starting bytes, but fortunately it could
1657        not have actually happened in practice until change 8 above was made (it
1658        added property types that matched character-matching opcodes).
1660    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
1661        possible starting bytes for non-anchored patterns.
1663    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
1664        \R, and also a number of cases that involve Unicode properties, both
1665        explicit and implicit when PCRE_UCP is set.
1667    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
1668        input, it could crash or give wrong results if characters with values
1669        greater than 0xc0 were present in the subject string. (Detail: it assumed
1670        UTF-8 input when processing these items.)
1672    16. Added a lot of (int) casts to avoid compiler warnings in systems where
1673        size_t is 64-bit (#991).
1675    17. Added a check for running out of memory when PCRE is compiled with
1676        --disable-stack-for-recursion (#990).
1678    18. If the last data line in a file for pcretest does not have a newline on
1679        the end, a newline was missing in the output.
1681    19. The default pcre_chartables.c file recognizes only ASCII characters (values
1682        less than 128) in its various bitmaps. However, there is a facility for
1683        generating tables according to the current locale when PCRE is compiled. It
1684        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
1685        characters, are recognized by isspace() and therefore were getting set in
1686        these tables, and indeed these tables seem to approximate to ISO 8859. This
1687        caused a problem in UTF-8 mode when pcre_study() was used to create a list
1688        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
1689        which of course cannot start UTF-8 characters. I have changed the code so
1690        that only real ASCII characters (less than 128) and the correct starting
1691        bytes for UTF-8 encodings are set for characters greater than 127 when in
1692        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
1693        altogether.)
1695    20. Added the /T option to pcretest so as to be able to run tests with non-
1696        standard character tables, thus making it possible to include the tests
1697        used for 19 above in the standard set of tests.
1699    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
1700        reference to a subpattern the other side of a comment that contains an
1701        opening parenthesis caused either an internal compiling error, or a
1702        reference to the wrong subpattern.
1705    Version 8.02 19-Mar-2010
1706    ------------------------
1708    1.  The Unicode data tables have been updated to Unicode 5.2.0.
1710    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
1711        configured.
1713    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
1714        original author of that file, following a query about its status.
1716    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
1717        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
1719    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
1720        quantifier applied to a forward-referencing subroutine call, could compile
1721        incorrect code or give the error "internal error: previously-checked
1722        referenced subpattern not found".
1724    6.  Both MS Visual Studio and Symbian OS have problems with initializing
1725        variables to point to external functions. For these systems, therefore,
1726        pcre_malloc etc. are now initialized to local functions that call the
1727        relevant global functions.
1729    7.  There were two entries missing in the vectors called coptable and poptable
1730        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
1731        I've fixed the data, and added a kludgy way of testing at compile time that
1732        the lengths are correct (equal to the number of opcodes).
1734    8.  Following on from 7, I added a similar kludge to check the length of the
1735        eint vector in pcreposix.c.
1737    9.  Error texts for pcre_compile() are held as one long string to avoid too
1738        much relocation at load time. To find a text, the string is searched,
1739        counting zeros. There was no check for running off the end of the string,
1740        which could happen if a new error number was added without updating the
1741        string.
1743    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
1745    11. \K was not working if it appeared in an atomic group or in a group that
1746        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
1747        \K is "not well defined" if used in an assertion. PCRE now accepts it if
1748        the assertion is positive, but not if it is negative.
1750    12. Change 11 fortuitously reduced the size of the stack frame used in the
1751        "match()" function of pcre_exec.c by one pointer. Forthcoming
1752        implementation of support for (*MARK) will need an extra pointer on the
1753        stack; I have reserved it now, so that the stack frame size does not
1754        decrease.
1756    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
1757        item in branch that calls a recursion is a subroutine call - as in the
1758        second branch in the above example - was incorrectly given the compile-
1759        time error "recursive call could loop indefinitely" because pcre_compile()
1760        was not correctly checking the subroutine for matching a non-empty string.
1762    14. The checks for overrunning compiling workspace could trigger after an
1763        overrun had occurred. This is a "should never occur" error, but it can be
1764        triggered by pathological patterns such as hundreds of nested parentheses.
1765        The checks now trigger 100 bytes before the end of the workspace.
1767    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
1770    Version 8.01 19-Jan-2010
1771    ------------------------
1773    1.  If a pattern contained a conditional subpattern with only one branch (in
1774        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
1775        computed the wrong minimum data length (which is of course zero for such
1776        subpatterns). This could cause incorrect "no match" results.
1778    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
1779        the pattern is reset in the first branch, pcre_compile() failed with
1780        "internal error: code overflow at offset...". This happened only when
1781        the reset was to the original external option setting. (An optimization
1782        abstracts leading options settings into an external setting, which was the
1783        cause of this.)
1785    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
1786        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
1787        assertion pattern did not match (meaning that the assertion was true), it
1788        was incorrectly treated as false if the SKIP had been reached during the
1789        matching. This also applied to assertions used as conditions.
1791    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
1792        assertion subpattern, including such a pattern used as a condition,
1793        unpredictable results occurred, instead of the error return
1796    5.  The C++ GlobalReplace function was not working like Perl for the special
1797        situation when an empty string is matched. It now does the fancy magic
1798        stuff that is necessary.
1800    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
1801        removed. (These were left over from very, very early versions of PCRE.)
1803    7.  Some cosmetic changes to the code to make life easier when compiling it
1804        as part of something else:
1806        (a) Change DEBUG to PCRE_DEBUG.
1808        (b) In pcre_compile(), rename the member of the "branch_chain" structure
1809            called "current" as "current_branch", to prevent a collision with the
1810            Linux macro when compiled as a kernel module.
1812        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
1813            prevent a collision with the Linux macro when compiled as a kernel
1814            module.
1816    8.  In pcre_compile() there are some checks for integer overflows that used to
1817        cast potentially large values to (double). This has been changed to that
1818        when building, a check for int64_t is made, and if it is found, it is used
1819        instead, thus avoiding the use of floating point arithmetic. (There is no
1820        other use of FP in PCRE.) If int64_t is not found, the fallback is to
1821        double.
1823    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
1824        2005 (difference between two addresses compared to an unsigned value).
1826    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
1827        custom one, because of the following reported problem in Windows:
1829          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
1830              under Win32.
1831          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
1832              therefore missing the function definition.
1833          - The compiler thus generates a "C" signature for the test function.
1834          - The linker fails to find the "C" function.
1835          - PCRE fails to configure if asked to do so against libbz2.
1837    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
1838        messages were output:
1840          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
1841          rerunning libtoolize, to keep the correct libtool macros in-tree.
1842          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
1844        I have done both of these things.
1846    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
1847        most of the time, it *can* run out if it is given a pattern that contains a
1848        runaway infinite recursion. I updated the discussion in the pcrestack man
1849        page.
1851    13. Now that we have gone to the x.xx style of version numbers, the minor
1852        version may start with zero. Using 08 or 09 is a bad idea because users
1853        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
1854        interpreted as invalid octal numbers. I've updated the previous comment in
1855        configure.ac, and also added a check that gives an error if 08 or 09 are
1856        used.
1858    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
1859        causing partial matching to fail when the end of the subject matched \W
1860        in a UTF-8 pattern where \W was quantified with a minimum of 3.
1862    15. There were some discrepancies between the declarations in pcre_internal.h
1863        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
1864        their definitions. The declarations used "const uschar *" and the
1865        definitions used USPTR. Even though USPTR is normally defined as "const
1866        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
1867        reported that: "This difference in casting confuses some C++ compilers, for
1868        example, SunCC recognizes above declarations as different functions and
1869        generates broken code for hbpcre." I have changed the declarations to use
1870        USPTR.
1872    16. GNU libtool is named differently on some systems. The autogen.sh script now
1873        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
1874        (FreeBSD).
1876    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
1877        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
1878        comment: "Figure out how to create a longlong from a string: strtoll and
1879        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
1880        instance, but it only takes 2 args instead of 3!"
1882    18. A subtle bug concerned with back references has been fixed by a change of
1883        specification, with a corresponding code fix. A pattern such as
1884        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
1885        refers, was giving matches when it shouldn't. For example, xa=xaaa would
1886        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
1887        same bug. Such groups have to be quantified to be useful, or contained
1888        inside another quantified group. (If there's no repetition, the reference
1889        can never match.) The problem arises because, having left the group and
1890        moved on to the rest of the pattern, a later failure that backtracks into
1891        the group uses the captured value from the final iteration of the group
1892        rather than the correct earlier one. I have fixed this in PCRE by forcing
1893        any group that contains a reference to itself to be an atomic group; that
1894        is, there cannot be any backtracking into it once it has completed. This is
1895        similar to recursive and subroutine calls.
1898    Version 8.00 19-Oct-09
1899    ----------------------
1901    1.  The table for translating pcre_compile() error codes into POSIX error codes
1902        was out-of-date, and there was no check on the pcre_compile() error code
1903        being within the table. This could lead to an OK return being given in
1904        error.
1906    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
1907        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
1908        in a Windows environment.
1910    3.  The pcregrep --count option prints the count for each file even when it is
1911        zero, as does GNU grep. However, pcregrep was also printing all files when
1912        --files-with-matches was added. Now, when both options are given, it prints
1913        counts only for those files that have at least one match. (GNU grep just
1914        prints the file name in this circumstance, but including the count seems
1915        more useful - otherwise, why use --count?) Also ensured that the
1916        combination -clh just lists non-zero counts, with no names.
1918    4.  The long form of the pcregrep -F option was incorrectly implemented as
1919        --fixed_strings instead of --fixed-strings. This is an incompatible change,
1920        but it seems right to fix it, and I didn't think it was worth preserving
1921        the old behaviour.
1923    5.  The command line items --regex=pattern and --regexp=pattern were not
1924        recognized by pcregrep, which required --regex pattern or --regexp pattern
1925        (with a space rather than an '='). The man page documented the '=' forms,
1926        which are compatible with GNU grep; these now work.
1928    6.  No libpcreposix.pc file was created for pkg-config; there was just
1929        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
1931    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
1932        when UCP support is not needed, by modifying the Python script that
1933        generates it from Unicode data files. This should not matter if the module
1934        is correctly used as a library, but I received one complaint about 50K of
1935        unwanted data. My guess is that the person linked everything into his
1936        program rather than using a library. Anyway, it does no harm.
1938    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
1939        was a minimum greater than 1 for a wide character in a possessive
1940        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
1941        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
1942        character. Chaos in the form of incorrect output or a compiling loop could
1943        result.
1945    9.  The restrictions on what a pattern can contain when partial matching is
1946        requested for pcre_exec() have been removed. All patterns can now be
1947        partially matched by this function. In addition, if there are at least two
1948        slots in the offset vector, the offset of the earliest inspected character
1949        for the match and the offset of the end of the subject are set in them when
1950        PCRE_ERROR_PARTIAL is returned.
1952    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
1953        synonymous with PCRE_PARTIAL, for backwards compatibility, and
1954        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
1955        and may be more useful for multi-segment matching.
1957    11. Partial matching with pcre_exec() is now more intuitive. A partial match
1958        used to be given if ever the end of the subject was reached; now it is
1959        given only if matching could not proceed because another character was
1960        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
1961        string "Z", which now yields "no match" instead of "partial match". In the
1962        case of pcre_dfa_exec(), "no match" is given if every matching path for the
1963        final character ended with (*FAIL).
1965    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
1966        if the pattern had a "must contain" character that was already found in the
1967        earlier partial match, unless partial matching was again requested. For
1968        example, with the pattern /dog.(body)?/, the "must contain" character is
1969        "g". If the first part-match was for the string "dog", restarting with
1970        "sbody" failed. This bug has been fixed.
1972    13. The string returned by pcre_dfa_exec() after a partial match has been
1973        changed so that it starts at the first inspected character rather than the
1974        first character of the match. This makes a difference only if the pattern
1975        starts with a lookbehind assertion or \b or \B (\K is not supported by
1976        pcre_dfa_exec()). It's an incompatible change, but it makes the two
1977        matching functions compatible, and I think it's the right thing to do.
1979    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
1980        so that the demonstration program is easily available in environments where
1981        PCRE has not been installed from source.
1983    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
1984        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
1985        library.
1987    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
1988        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
1989        is not the first non-POSIX option to be added. Clearly some people find
1990        these options useful.
1992    17. If a caller to the POSIX matching function regexec() passes a non-zero
1993        value for nmatch with a NULL value for pmatch, the value of
1994        nmatch is forced to zero.
1996    18. RunGrepTest did not have a test for the availability of the -u option of
1997        the diff command, as RunTest does. It now checks in the same way as
1998        RunTest, and also checks for the -b option.
2000    19. If an odd number of negated classes containing just a single character
2001        interposed, within parentheses, between a forward reference to a named
2002        subpattern and the definition of the subpattern, compilation crashed with
2003        an internal error, complaining that it could not find the referenced
2004        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
2005        [The bug was that it was starting one character too far in when skipping
2006        over the character class, thus treating the ] as data rather than
2007        terminating the class. This meant it could skip too much.]
2009    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
2010        /g option in pcretest when the pattern contains \K, which makes it possible
2011        to have an empty string match not at the start, even when the pattern is
2012        anchored. Updated pcretest and pcredemo to use this option.
2014    21. If the maximum number of capturing subpatterns in a recursion was greater
2015        than the maximum at the outer level, the higher number was returned, but
2016        with unset values at the outer level. The correct (outer level) value is
2017        now given.
2019    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
2020        PCRE did not set those parentheses (unlike Perl). I have now found a way to
2021        make it do so. The string so far is captured, making this feature
2022        compatible with Perl.
2024    23. The tests have been re-organized, adding tests 11 and 12, to make it
2025        possible to check the Perl 5.10 features against Perl 5.10.
2027    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
2028        pattern matches a fixed length string. PCRE did not allow this; now it
2029        does. Neither allows recursion.
2031    25. I finally figured out how to implement a request to provide the minimum
2032        length of subject string that was needed in order to match a given pattern.
2033        (It was back references and recursion that I had previously got hung up
2034        on.) This code has now been added to pcre_study(); it finds a lower bound
2035        to the length of subject needed. It is not necessarily the greatest lower
2036        bound, but using it to avoid searching strings that are too short does give
2037        some useful speed-ups. The value is available to calling programs via
2038        pcre_fullinfo().
2040    26. While implementing 25, I discovered to my embarrassment that pcretest had
2041        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
2042        study optimizations had never been tested with that matching function.
2043        Oops. What is worse, even when it was passed study data, there was a bug in
2044        pcre_dfa_exec() that meant it never actually used it. Double oops. There
2045        were also very few tests of studied patterns with pcre_dfa_exec().
2047    27. If (?| is used to create subpatterns with duplicate numbers, they are now
2048        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
2049        on the other side of the coin, they are no longer allowed to have different
2050        names, because these cannot be distinguished in PCRE, and this has caused
2051        confusion. (This is a difference from Perl.)
2053    28. When duplicate subpattern names are present (necessarily with different
2054        numbers, as required by 27 above), and a test is made by name in a
2055        conditional pattern, either for a subpattern having been matched, or for
2056        recursion in such a pattern, all the associated numbered subpatterns are
2057        tested, and the overall condition is true if the condition is true for any
2058        one of them. This is the way Perl works, and is also more like the way
2059        testing by number works.
2062    Version 7.9 11-Apr-09
2063    ---------------------
2065    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
2066        (pcretest), all targets were linked against these libraries. This included
2067        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
2068        libraries. This caused unwanted dependencies to be created. This problem
2069        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
2070        pcretest is linked with readline.
2072    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
2073        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
2074        moved outside it again, because FALSE and TRUE are already defined in AIX,
2075        but BOOL is not.
2077    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
2078        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
2080    4.  The pcregrep documentation said spaces were inserted as well as colons (or
2081        hyphens) following file names and line numbers when outputting matching
2082        lines. This is not true; no spaces are inserted. I have also clarified the
2083        wording for the --colour (or --color) option.
2085    5.  In pcregrep, when --colour was used with -o, the list of matching strings
2086        was not coloured; this is different to GNU grep, so I have changed it to be
2087        the same.
2089    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
2090        each matching line was coloured. Now it goes on to look for further matches
2091        of any of the test patterns, which is the same behaviour as GNU grep.
2093    7.  A pattern that could match an empty string could cause pcregrep to loop; it
2094        doesn't make sense to accept an empty string match in pcregrep, so I have
2095        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
2096        seems to be how GNU grep behaves. [But see later change 40 for release
2097        8.33.]
2099    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
2100        start or after a newline", because the conditional assertion was not being
2101        correctly handled. The rule now is that both the assertion and what follows
2102        in the first alternative must satisfy the test.
2104    9.  If auto-callout was enabled in a pattern with a conditional group whose
2105        condition was an assertion, PCRE could crash during matching, both with
2106        pcre_exec() and pcre_dfa_exec().
2108    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
2109        used for matching.
2111    11. Unicode property support in character classes was not working for
2112        characters (bytes) greater than 127 when not in UTF-8 mode.
2114    12. Added the -M command line option to pcretest.
2116    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
2118    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
2120    16. Added comments and documentation about mis-use of no_arg in the C++
2121        wrapper.
2123    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
2124        from Martin Jerabek that uses macro names for all relevant character and
2125        string constants.
2127    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
2128        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
2129        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
2130        these, but not everybody uses configure.
2132    19. A conditional group that had only one branch was not being correctly
2133        recognized as an item that could match an empty string. This meant that an
2134        enclosing group might also not be so recognized, causing infinite looping
2135        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
2136        with the subject "ab", where knowledge that the repeated group can match
2137        nothing is needed in order to break the loop.
2139    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
2140        exec(), but without supplying a callout function, matching went wrong.
2142    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
2143        leak if the size of the offset vector was greater than 30. When the vector
2144        is smaller, the saved offsets during recursion go onto a local stack
2145        vector, but for larger vectors malloc() is used. It was failing to free
2146        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
2147        error, in fact).
2149    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
2150        heapframe that is used only when UTF-8 support is enabled. This caused no
2151        problem, but was untidy.
2153    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
2154        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
2155        included within another project.
2157    24. Steven Van Ingelgem's patches to add more options to the CMake support,
2158        slightly modified by me:
2160          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
2161              not building pcregrep.
2163          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
2164              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
2166    25. Forward references, both numeric and by name, in patterns that made use of
2167        duplicate group numbers, could behave incorrectly or give incorrect errors,
2168        because when scanning forward to find the reference group, PCRE was not
2169        taking into account the duplicate group numbers. A pattern such as
2170        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
2172    26. Changed a few more instances of "const unsigned char *" to USPTR, making
2173        the feature of a custom pointer more persuasive (as requested by a user).
2175    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
2176        pcretest.c, inside #ifndefs, because it seems they are sometimes already
2177        pre-defined.
2179    28. Added support for (*UTF8) at the start of a pattern.
2181    29. Arrange for flags added by the "release type" setting in CMake to be shown
2182        in the configuration summary.
2185    Version 7.8 05-Sep-08
2186    ---------------------
2188    1.  Replaced UCP searching code with optimized version as implemented for Ad
2189        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
2190        stage table and inline lookup instead of a function, giving speed ups of 2
2191        to 5 times on some simple patterns that I tested. Permission was given to
2192        distribute the MultiStage2.py script that generates the tables (it's not in
2193        the tarball, but is in the Subversion repository).
2195    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
2196        scripts.
2198    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
2199        a group with a zero qualifier. The result of the study could be incorrect,
2200        or the function might crash, depending on the pattern.
2202    4.  Caseless matching was not working for non-ASCII characters in back
2203        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
2204        It now works when Unicode Property Support is available.
2206    5.  In pcretest, an escape such as \x{de} in the data was always generating
2207        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
2208        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
2209        truncation.
2211    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
2213    7.  Added two (int) casts to pcregrep when printing the difference of two
2214        pointers, in case they are 64-bit values.
2216    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
2217        test 2 if it fails.
2219    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
2220        and a #define of that name to empty if it is not externally set. This is to
2221        allow users of MSVC to set it if necessary.
2223    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
2224        the convenience functions in the pcre_get.c source file.
2226    11. An option change at the start of a pattern that had top-level alternatives
2227        could cause overwriting and/or a crash. This command provoked a crash in
2228        some environments:
2230          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
2232        This potential security problem was recorded as CVE-2008-2371.
2234    12. For a pattern where the match had to start at the beginning or immediately
2235        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
2236        pcre_dfa_exec() could read past the end of the passed subject if there was
2237        no match. To help with detecting such bugs (e.g. with valgrind), I modified
2238        pcretest so that it places the subject at the end of its malloc-ed buffer.
2240    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
2241        exec() might read past the end of the data buffer in UTF-8 mode.
2243    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
2244        the data contained the byte 0x85 as part of a UTF-8 character within its
2245        first line. This applied both to normal and DFA matching.
2247    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
2248        /^[^d]*?$/8 failed to match "abc".
2250    16. Added a missing copyright notice to pcrecpp_internal.h.
2252    17. Make it more clear in the documentation that values returned from
2253        pcre_exec() in ovector are byte offsets, not character counts.
2255    18. Tidied a few places to stop certain compilers from issuing warnings.
2257    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
2258        supplied by Stefan Weber. I made a further small update for 7.8 because
2259        there is a change of source arrangements: the pcre_searchfuncs.c module is
2260        replaced by pcre_ucd.c.
2263    Version 7.7 07-May-08
2264    ---------------------
2266    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
2267        a string to a long long, pretend we don't even have a long long." This is
2268        done by checking for the strtoq, strtoll, and _strtoi64 functions.
2270    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
2271        pre-7.6 versions, which defined a global no_arg variable instead of putting
2272        it in the RE class. (See also #8 below.)
2274    3.  Remove a line of dead code, identified by coverity and reported by Nuno
2275        Lopes.
2277    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
2279        (1) The include/exclude patterns were being applied to the whole pathnames
2280            of files, instead of just to the final components.
2282        (2) If there was more than one level of directory, the subdirectories were
2283            skipped unless they satisfied the include/exclude conditions. This is
2284            inconsistent with GNU grep (and could even be seen as contrary to the
2285            pcregrep specification - which I improved to make it absolutely clear).
2286            The action now is always to scan all levels of directory, and just
2287            apply the include/exclude patterns to regular files.
2289    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
2290        --exclude_dir in the tests to avoid scanning .svn directories.
2292    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
2293        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
2294        doesn't support NULs in patterns.
2296    7.  Added some missing "const"s to declarations of static tables in
2297        pcre_compile.c and pcre_dfa_exec.c.
2299    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
2300        caused by fix #2  above. (Subsequently also a second patch to fix the
2301        first patch. And a third patch - this was a messy problem.)
2303    9.  Applied Craig's patch to remove the use of push_back().
2305    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
2306        matching function regexec().
2308    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
2309        which, however, unlike Perl's \g{...}, are subroutine calls, not back
2310        references. PCRE supports relative numbers with this syntax (I don't think
2311        Oniguruma does).
2313    12. Previously, a group with a zero repeat such as (...){0} was completely
2314        omitted from the compiled regex. However, this means that if the group
2315        was called as a subroutine from elsewhere in the pattern, things went wrong
2316        (an internal error was given). Such groups are now left in the compiled
2317        pattern, with a new opcode that causes them to be skipped at execution
2318        time.
2320    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
2321        to the way PCRE behaves:
2323        (a) A lone ] character is dis-allowed (Perl treats it as data).
2325        (b) A back reference to an unmatched subpattern matches an empty string
2326            (Perl fails the current match path).
2328        (c) A data ] in a character class must be notated as \] because if the
2329            first data character in a class is ], it defines an empty class. (In
2330            Perl it is not possible to have an empty class.) The empty class []
2331            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
2332            The negative empty class [^] matches any one character, independently
2333            of the DOTALL setting.
2335    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
2336        non-existent subpattern following a character class starting with ']' and
2337        containing () gave an internal compiling error instead of "reference to
2338        non-existent subpattern". Fortunately, when the pattern did exist, the
2339        compiled code was correct. (When scanning forwards to check for the
2340        existence of the subpattern, it was treating the data ']' as terminating
2341        the class, so got the count wrong. When actually compiling, the reference
2342        was subsequently set up correctly.)
2344    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
2345        it was being rejected as not supported by pcre_dfa_exec(), even though
2346        other assertions are supported. I have made pcre_dfa_exec() support
2347        (*FAIL).
2349    16. The implementation of 13c above involved the invention of a new opcode,
2350        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
2351        cannot be changed at match time, I realized I could make a small
2352        improvement to matching performance by compiling OP_ALLANY instead of
2353        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
2354        on the OP_ANY path.
2356    17. Compiling pcretest on Windows with readline support failed without the
2357        following two fixes: (1) Make the unistd.h include conditional on
2358        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
2360    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
2361        ncurses library to be included for pcretest when ReadLine support is
2362        requested, but also to allow for it to be overridden. This patch came from
2363        Daniel Bergström.
2365    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
2366        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
2367        any errors with the current Unicode tables. Thanks to Peter Kankowski for
2368        spotting this.
2371    Version 7.6 28-Jan-08
2372    ---------------------
2374    1.  A character class containing a very large number of characters with
2375        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
2376        overflow.
2378    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
2379        HAVE_LONG_LONG is not defined.
2381    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
2382        bring them up to date and include new features. This patch includes:
2384        - Fixed PH's badly added libz and libbz2 support.
2385        - Fixed a problem with static linking.
2386        - Added pcredemo. [But later removed - see 7 below.]
2387        - Fixed dftables problem and added an option.
2388        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
2389            HAVE_LONG_LONG.
2390        - Added readline support for pcretest.
2391        - Added an listing of the option settings after cmake has run.
2393    4.  A user submitted a patch to Makefile that makes it easy to create
2394        "pcre.dll" under mingw when using Configure/Make. I added stuff to
2395        Makefile.am that cause it to include this special target, without
2396        affecting anything else. Note that the same mingw target plus all
2397        the other distribution libraries and programs are now supported
2398        when configuring with CMake (see 6 below) instead of with
2399        Configure/Make.
2401    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
2402        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
2403        exported in the Windows port". It has not yet been confirmed that the patch
2404        solves the problem, but it does no harm.
2406    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
2407        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
2408        with CMake, and also correct the comment about stack recursion.
2410    7.  Remove the automatic building of pcredemo from the ./configure system and
2411        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
2412        of a program that users should build themselves after PCRE is installed, so
2413        building it automatically is not really right. What is more, it gave
2414        trouble in some build environments.
2416    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
2419    Version 7.5 10-Jan-08
2420  ---------------------  ---------------------
2422  1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'  1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
2423      values in parens when parsing an RE using the C++ wrapper."      values in parens when parsing an RE using the C++ wrapper."
2425  2.  Negative specials like \S did not work in character classes in UTF-8 mode.  2.  Negative specials like \S did not work in character classes in UTF-8 mode.
2426      Characters greater than 255 were excluded from the class instead of being      Characters greater than 255 were excluded from the class instead of being
2427      included.      included.
2429  3.  The same bug as (2) above applied to negated POSIX classes such as  3.  The same bug as (2) above applied to negated POSIX classes such as
2430      [:^space:].      [:^space:].
2432  4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it  4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
2433      defined or documented. It seems to have been a typo for PCRE_STATIC, so      defined or documented. It seems to have been a typo for PCRE_STATIC, so
2434      I have changed it.      I have changed it.
2436    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
2437        first named subpattern) and a construct such as (?&a) would reference the
2438        first named subpattern whose name started with "a" (in other words, the
2439        length check was missing). Both these problems are fixed. "Subpattern name
2440        expected" is now given for (?&) (a zero-length name), and this patch also
2441        makes it give the same error for \k'' (previously it complained that that
2442        was a reference to a non-existent subpattern).
2444    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
2445        this is right because (?- can be followed by option settings as well as by
2446        digits. I have, however, made the messages clearer.
2448    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
2449        than the number used in the conditional) now cause a compile-time error.
2450        This is actually not compatible with Perl, which accepts such patterns, but
2451        treats the conditional as always being FALSE (as PCRE used to), but it
2452        seems to me that giving a diagnostic is better.
2454    8.  Change "alphameric" to the more common word "alphanumeric" in comments
2455        and messages.
2457    9.  Fix two occurrences of "backslash" in comments that should have been
2458        "backspace".
2460    10. Remove two redundant lines of code that can never be obeyed (their function
2461        was moved elsewhere).
2463    11. The program that makes PCRE's Unicode character property table had a bug
2464        which caused it to generate incorrect table entries for sequences of
2465        characters that have the same character type, but are in different scripts.
2466        It amalgamated them into a single range, with the script of the first of
2467        them. In other words, some characters were in the wrong script. There were
2468        thirteen such cases, affecting characters in the following ranges:
2470          U+002b0 - U+002c1
2471          U+0060c - U+0060d
2472          U+0061e - U+00612
2473          U+0064b - U+0065e
2474          U+0074d - U+0076d
2475          U+01800 - U+01805
2476          U+01d00 - U+01d77
2477          U+01d9b - U+01dbf
2478          U+0200b - U+0200f
2479          U+030fc - U+030fe
2480          U+03260 - U+0327f
2481          U+0fb46 - U+0fbb1
2482          U+10450 - U+1049d
2484    12. The -o option (show only the matching part of a line) for pcregrep was not
2485        compatible with GNU grep in that, if there was more than one match in a
2486        line, it showed only the first of them. It now behaves in the same way as
2487        GNU grep.
2489    13. If the -o and -v options were combined for pcregrep, it printed a blank
2490        line for every non-matching line. GNU grep prints nothing, and pcregrep now
2491        does the same. The return code can be used to tell if there were any
2492        non-matching lines.
2494    14. Added --file-offsets and --line-offsets to pcregrep.
2496    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
2497        infinitely looping recursion. The bug was that positive lookaheads were not
2498        being skipped when checking for a possible empty match (negative lookaheads
2499        and both kinds of lookbehind were skipped).
2501    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
2502        inclusion of <windows.h> to before rather than after the definition of
2503        INVALID_FILE_ATTRIBUTES (patch from David Byron).
2505    17. Specifying a possessive quantifier with a specific limit for a Unicode
2506        character property caused pcre_compile() to compile bad code, which led at
2507        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
2508        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
2509        caused the error; without that there was no problem.
2511    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
2513    19. Added --enable-pcretest-libreadline.
2515    20. In pcrecpp.cc, the variable 'count' was incremented twice in
2516        RE::GlobalReplace(). As a result, the number of replacements returned was
2517        double what it should be. I removed one of the increments, but Craig sent a
2518        later patch that removed the other one (the right fix) and added unit tests
2519        that check the return values (which was not done before).
2521    21. Several CMake things:
2523        (1) Arranged that, when cmake is used on Unix, the libraries end up with
2524            the names libpcre and libpcreposix, not just pcre and pcreposix.
2526        (2) The above change means that pcretest and pcregrep are now correctly
2527            linked with the newly-built libraries, not previously installed ones.
2531    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
2532        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
2533        UTF-8 newline character). The key issue is that the pattern starts .*;
2534        this means that the match must be either at the beginning, or after a
2535        newline. The bug was in the code for advancing after a failed match and
2536        checking that the new position followed a newline. It was not taking
2537        account of UTF-8 characters correctly.
2539    23. PCRE was behaving differently from Perl in the way it recognized POSIX
2540        character classes. PCRE was not treating the sequence [:...:] as a
2541        character class unless the ... were all letters. Perl, however, seems to
2542        allow any characters between [: and :], though of course it rejects as
2543        unknown any "names" that contain non-letters, because all the known class
2544        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
2545        for example, whereas PCRE did not - it did not recognize a POSIX character
2546        class. This seemed a bit dangerous, so the code has been changed to be
2547        closer to Perl. The behaviour is not identical to Perl, because PCRE will
2548        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
2549        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
2550        Perl does, and where it didn't before.
2552    24. Rewrite so as to remove the single use of %n from pcregrep because in some
2553        Windows environments %n is disabled by default.
2556  Version 7.4 21-Sep-07  Version 7.4 21-Sep-07

Removed from v.267  
changed lines
  Added in v.1496

  ViewVC Help
Powered by ViewVC 1.1.5