ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 166 by ph10, Wed May 9 14:48:28 2007 UTC revision 1462 by ph10, Mon Mar 10 16:28:54 2014 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
4  Version 7.2 01-May-07  Version 8.35-RC1 04-March-2014
5    ------------------------------
7    1.  A new flag is set, when property checks are present in an XCLASS.
8        When this flag is not set, PCRE can perform certain optimizations
9        such as studying these XCLASS-es.
11    2.  The auto-possessification of character sets were improved: a normal
12        and an extended character set can be compared now. Furthermore
13        the JIT compiler optimizes more character set checks.
15    3.  Got rid of some compiler warnings for potentially uninitialized variables
16        that show up only when compiled with -O2.
18    4.  A pattern such as (?=ab\K) that uses \K in an assertion can set the start
19        of a match later then the end of the match. The pcretest program was not
20        handling the case sensibly - it was outputting from the start to the next
21        binary zero. It now reports this situation in a message, and outputs the
22        text from the end to the start.
24    5.  Fast forward search is improved in JIT. Instead of the first three
25        characters, any three characters with fixed position can be searched.
26        Search order: first, last, middle.
28    6.  Improve character range checks in JIT. Characters are read by an inprecise
29        function now, which returns with an unknown value if the character code is
30        above a certain treshold (e.g: 256). The only limitation is that the value
31        must be bigger than the treshold as well. This function is useful, when
32        the characters above the treshold are handled in the same way.
34    7.  The macros whose names start with RAWUCHAR are placeholders for a future
35        mode in which only the bottom 21 bits of 32-bit data items are used. To
36        make this more memorable for those maintaining the code, the names have
37        been changed to start with UCHAR21, and an extensive comment has been added
38        to their definition.
40    8.  Add missing (new) files sljitNativeTILEGX.c and sljitNativeTILEGX-encoder.c
41        to the export list in Makefile.am (they were accidentally omitted from the
42        8.34 tarball).
44    9.  The informational output from pcretest used the phrase "starting byte set"
45        which is inappropriate for the 16-bit and 32-bit libraries. As the output
46        for "first char" and "need char" really means "non-UTF-char", I've changed
47        "byte" to "char", and slightly reworded the output. The documentation about
48        these values has also been (I hope) clarified.
50    10. Another JIT related optimization: use table jumps for selecting the correct
51        backtracking path, when more than four alternatives are present inside a
52        bracket.
54    11. Empty match is not possible, when the minimum length is greater than zero,
55        and there is no \K in the pattern. JIT should avoid empty match checks in
56        such cases.
58    12. In a caseless character class with UCP support, when a character with more
59        than one alternative case was not the first character of a range, not all
60        the alternative cases were added to the class. For example, s and \x{17f}
61        are both alternative cases for S: the class [RST] was handled correctly,
62        but [R-T] was not.
64    13. The configure.ac file always checked for pthread support when JIT was
65        enabled. This is not used in Windows, so I have put this test inside a
66        check for the presence of windows.h (which was already tested for).
68    14. Improve pattern prefix search by a simplified Boyer-Moore algorithm in JIT.
69        The algorithm provides a way to skip certain starting offsets, and usually
70        faster than linear prefix searches.
72    15. Change 13 for 8.20 updated RunTest to check for the 'fr' locale as well
73        as for 'fr_FR' and 'french'. For some reason, however, it then used the
74        Windows-specific input and output files, which have 'french' screwed in.
75        So this could never have worked. One of the problems with locales is that
76        they aren't always the same. I have now updated RunTest so that it checks
77        the output of the locale test (test 3) against three different output
78        files, and it allows the test to pass if any one of them matches. With luck
79        this should make the test pass on some versions of Solaris where it was
80        failing. Because of the uncertainty, the script did not used to stop if
81        test 3 failed; it now does. If further versions of a French locale ever
82        come to light, they can now easily be added.
84    16. If --with-pcregrep-bufsize was given a non-integer value such as "50K",
85        there was a message during ./configure, but it did not stop. This now
86        provokes an error. The invalid example in README has been corrected.
87        If a value less than the minimum is given, the minimum value has always
88        been used, but now a warning is given.
90    17. If --enable-bsr-anycrlf was set, the special 16/32-bit test failed. This
91        was a bug in the test system, which is now fixed. Also, the list of various
92        configurations that are tested for each release did not have one with both
93        16/32 bits and --enable-bar-anycrlf. It now does.
95    18. pcretest was missing "-C bsr" for displaying the \R default setting.
97    19. Little endian PowerPC systems are supported now by the JIT compiler.
99    20. The fast forward newline mechanism could enter to an infinite loop on
100        certain invalid UTF-8 input. Although we don't support these cases
101        this issue can be fixed by a performance optimization.
103    21. Change 33 of 8.34 is not sufficient to ensure stack safety because it does
104        not take account if existing stack usage. There is now a new global
105        variable called pcre_stack_guard that can be set to point to an external
106        function to check stack availability. It is called at the start of
107        processing every parenthesized group.
109    22. A typo in the code meant that in ungreedy mode the max/min qualifier
110        behaved like a min-possessive qualifier, and, for example, /a{1,3}b/U did
111        not match "ab".
114    Version 8.34 15-December-2013
115    -----------------------------
117    1.  Add pcre[16|32]_jit_free_unused_memory to forcibly free unused JIT
118        executable memory. Patch inspired by Carsten Klein.
120    2.  ./configure --enable-coverage defined SUPPORT_GCOV in config.h, although
121        this macro is never tested and has no effect, because the work to support
122        coverage involves only compiling and linking options and special targets in
123        the Makefile. The comment in config.h implied that defining the macro would
124        enable coverage support, which is totally false. There was also support for
125        setting this macro in the CMake files (my fault, I just copied it from
126        configure). SUPPORT_GCOV has now been removed.
128    3.  Make a small performance improvement in strlen16() and strlen32() in
129        pcretest.
131    4.  Change 36 for 8.33 left some unreachable statements in pcre_exec.c,
132        detected by the Solaris compiler (gcc doesn't seem to be able to diagnose
133        these cases). There was also one in pcretest.c.
135    5.  Cleaned up a "may be uninitialized" compiler warning in pcre_exec.c.
137    6.  In UTF mode, the code for checking whether a group could match an empty
138        string (which is used for indefinitely repeated groups to allow for
139        breaking an infinite loop) was broken when the group contained a repeated
140        negated single-character class with a character that occupied more than one
141        data item and had a minimum repetition of zero (for example, [^\x{100}]* in
142        UTF-8 mode). The effect was undefined: the group might or might not be
143        deemed as matching an empty string, or the program might have crashed.
145    7.  The code for checking whether a group could match an empty string was not
146        recognizing that \h, \H, \v, \V, and \R must match a character.
148    8.  Implemented PCRE_INFO_MATCH_EMPTY, which yields 1 if the pattern can match
149        an empty string. If it can, pcretest shows this in its information output.
151    9.  Fixed two related bugs that applied to Unicode extended grapheme clusters
152        that were repeated with a maximizing qualifier (e.g. \X* or \X{2,5}) when
153        matched by pcre_exec() without using JIT:
155        (a) If the rest of the pattern did not match after a maximal run of
156            grapheme clusters, the code for backing up to try with fewer of them
157            did not always back up over a full grapheme when characters that do not
158            have the modifier quality were involved, e.g. Hangul syllables.
160        (b) If the match point in a subject started with modifier character, and
161            there was no match, the code could incorrectly back up beyond the match
162            point, and potentially beyond the first character in the subject,
163            leading to a segfault or an incorrect match result.
165    10. A conditional group with an assertion condition could lead to PCRE
166        recording an incorrect first data item for a match if no other first data
167        item was recorded. For example, the pattern (?(?=ab)ab) recorded "a" as a
168        first data item, and therefore matched "ca" after "c" instead of at the
169        start.
171    11. Change 40 for 8.33 (allowing pcregrep to find empty strings) showed up a
172        bug that caused the command "echo a | ./pcregrep -M '|a'" to loop.
174    12. The source of pcregrep now includes z/OS-specific code so that it can be
175        compiled for z/OS as part of the special z/OS distribution.
177    13. Added the -T and -TM options to pcretest.
179    14. The code in pcre_compile.c for creating the table of named capturing groups
180        has been refactored. Instead of creating the table dynamically during the
181        actual compiling pass, the information is remembered during the pre-compile
182        pass (on the stack unless there are more than 20 named groups, in which
183        case malloc() is used) and the whole table is created before the actual
184        compile happens. This has simplified the code (it is now nearly 150 lines
185        shorter) and prepared the way for better handling of references to groups
186        with duplicate names.
188    15. A back reference to a named subpattern when there is more than one of the
189        same name now checks them in the order in which they appear in the pattern.
190        The first one that is set is used for the reference. Previously only the
191        first one was inspected. This change makes PCRE more compatible with Perl.
193    16. Unicode character properties were updated from Unicode 6.3.0.
195    17. The compile-time code for auto-possessification has been refactored, based
196        on a patch by Zoltan Herczeg. It now happens after instead of during
197        compilation. The code is cleaner, and more cases are handled. The option
198        PCRE_NO_AUTO_POSSESS is added for testing purposes, and the -O and /O
199        options in pcretest are provided to set it. It can also be set by
200        (*NO_AUTO_POSSESS) at the start of a pattern.
202    18. The character VT has been added to the default ("C" locale) set of
203        characters that match \s and are generally treated as white space,
204        following this same change in Perl 5.18. There is now no difference between
205        "Perl space" and "POSIX space". Whether VT is treated as white space in
206        other locales depends on the locale.
208    19. The code for checking named groups as conditions, either for being set or
209        for being recursed, has been refactored (this is related to 14 and 15
210        above). Processing unduplicated named groups should now be as fast at
211        numerical groups, and processing duplicated groups should be faster than
212        before.
214    20. Two patches to the CMake build system, by Alexander Barkov:
216          (1) Replace the "source" command by "." in CMakeLists.txt because
217              "source" is a bash-ism.
219          (2) Add missing HAVE_STDINT_H and HAVE_INTTYPES_H to config-cmake.h.in;
220              without these the CMake build does not work on Solaris.
222    21. Perl has changed its handling of \8 and \9. If there is no previously
223        encountered capturing group of those numbers, they are treated as the
224        literal characters 8 and 9 instead of a binary zero followed by the
225        literals. PCRE now does the same.
227    22. Following Perl, added \o{} to specify codepoints in octal, making it
228        possible to specify values greater than 0777 and also making them
229        unambiguous.
231    23. Perl now gives an error for missing closing braces after \x{... instead of
232        treating the string as literal. PCRE now does the same.
234    24. RunTest used to grumble if an inappropriate test was selected explicitly,
235        but just skip it when running all tests. This make it awkward to run ranges
236        of tests when one of them was inappropriate. Now it just skips any
237        inappropriate tests, as it always did when running all tests.
239    25. If PCRE_AUTO_CALLOUT and PCRE_UCP were set for a pattern that contained
240        character types such as \d or \w, too many callouts were inserted, and the
241        data that they returned was rubbish.
243    26. In UCP mode, \s was not matching two of the characters that Perl matches,
244        namely NEL (U+0085) and MONGOLIAN VOWEL SEPARATOR (U+180E), though they
245        were matched by \h. The code has now been refactored so that the lists of
246        the horizontal and vertical whitespace characters used for \h and \v (which
247        are defined only in one place) are now also used for \s.
249    27. Add JIT support for the 64 bit TileGX architecture.
250        Patch by Jiong Wang (Tilera Corporation).
252    28. Possessive quantifiers for classes (both explicit and automatically
253        generated) now use special opcodes instead of wrapping in ONCE brackets.
255    29. Whereas an item such as A{4}+ ignored the possessivenes of the quantifier
256        (because it's meaningless), this was not happening when PCRE_CASELESS was
257        set. Not wrong, but inefficient.
259    30. Updated perltest.pl to add /u (force Unicode mode) when /W (use Unicode
260        properties for \w, \d, etc) is present in a test regex. Otherwise if the
261        test contains no characters greater than 255, Perl doesn't realise it
262        should be using Unicode semantics.
264    31. Upgraded the handling of the POSIX classes [:graph:], [:print:], and
265        [:punct:] when PCRE_UCP is set so as to include the same characters as Perl
266        does in Unicode mode.
268    32. Added the "forbid" facility to pcretest so that putting tests into the
269        wrong test files can sometimes be quickly detected.
271    33. There is now a limit (default 250) on the depth of nesting of parentheses.
272        This limit is imposed to control the amount of system stack used at compile
273        time. It can be changed at build time by --with-parens-nest-limit=xxx or
274        the equivalent in CMake.
276    34. Character classes such as [A-\d] or [a-[:digit:]] now cause compile-time
277        errors. Perl warns for these when in warning mode, but PCRE has no facility
278        for giving warnings.
280    35. Change 34 for 8.13 allowed quantifiers on assertions, because Perl does.
281        However, this was not working for (?!) because it is optimized to (*FAIL),
282        for which PCRE does not allow quantifiers. The optimization is now disabled
283        when a quantifier follows (?!). I can't see any use for this, but it makes
284        things uniform.
286    36. Perl no longer allows group names to start with digits, so I have made this
287        change also in PCRE. It simplifies the code a bit.
289    37. In extended mode, Perl ignores spaces before a + that indicates a
290        possessive quantifier. PCRE allowed a space before the quantifier, but not
291        before the possessive +. It now does.
293    38. The use of \K (reset reported match start) within a repeated possessive
294        group such as (a\Kb)*+ was not working.
296    40. Document that the same character tables must be used at compile time and
297        run time, and that the facility to pass tables to pcre_exec() and
298        pcre_dfa_exec() is for use only with saved/restored patterns.
300    41. Applied Jeff Trawick's patch CMakeLists.txt, which "provides two new
301        features for Builds with MSVC:
303        1. Support pcre.rc and/or pcreposix.rc (as is already done for MinGW
304           builds). The .rc files can be used to set FileDescription and many other
305           attributes.
307        2. Add an option (-DINSTALL_MSVC_PDB) to enable installation of .pdb files.
308           This allows higher-level build scripts which want .pdb files to avoid
309           hard-coding the exact files needed."
311    42. Added support for [[:<:]] and [[:>:]] as used in the BSD POSIX library to
312        mean "start of word" and "end of word", respectively, as a transition aid.
314    43. A minimizing repeat of a class containing codepoints greater than 255 in
315        non-UTF 16-bit or 32-bit modes caused an internal error when PCRE was
316        compiled to use the heap for recursion.
318    44. Got rid of some compiler warnings for unused variables when UTF but not UCP
319        is configured.
322    Version 8.33 28-May-2013
323    ------------------------
325    1.  Added 'U' to some constants that are compared to unsigned integers, to
326        avoid compiler signed/unsigned warnings. Added (int) casts to unsigned
327        variables that are added to signed variables, to ensure the result is
328        signed and can be negated.
330    2.  Applied patch by Daniel Richard G for quashing MSVC warnings to the
331        CMake config files.
333    3.  Revise the creation of config.h.generic so that all boolean macros are
334        #undefined, whereas non-boolean macros are #ifndef/#endif-ed. This makes
335        overriding via -D on the command line possible.
337    4.  Changing the definition of the variable "op" in pcre_exec.c from pcre_uchar
338        to unsigned int is reported to make a quite noticeable speed difference in
339        a specific Windows environment. Testing on Linux did also appear to show
340        some benefit (and it is clearly not harmful). Also fixed the definition of
341        Xop which should be unsigned.
343    5.  Related to (4), changing the definition of the intermediate variable cc
344        in repeated character loops from pcre_uchar to pcre_uint32 also gave speed
345        improvements.
347    6.  Fix forward search in JIT when link size is 3 or greater. Also removed some
348        unnecessary spaces.
350    7.  Adjust autogen.sh and configure.ac to lose warnings given by automake 1.12
351        and later.
353    8.  Fix two buffer over read issues in 16 and 32 bit modes. Affects JIT only.
355    9.  Optimizing fast_forward_start_bits in JIT.
357    10. Adding support for callouts in JIT, and fixing some issues revealed
358        during this work. Namely:
360        (a) Unoptimized capturing brackets incorrectly reset on backtrack.
362        (b) Minimum length was not checked before the matching is started.
364    11. The value of capture_last that is passed to callouts was incorrect in some
365        cases when there was a capture on one path that was subsequently abandoned
366        after a backtrack. Also, the capture_last value is now reset after a
367        recursion, since all captures are also reset in this case.
369    12. The interpreter no longer returns the "too many substrings" error in the
370        case when an overflowing capture is in a branch that is subsequently
371        abandoned after a backtrack.
373    13. In the pathological case when an offset vector of size 2 is used, pcretest
374        now prints out the matched string after a yield of 0 or 1.
376    14. Inlining subpatterns in recursions, when certain conditions are fulfilled.
377        Only supported by the JIT compiler at the moment.
379    15. JIT compiler now supports 32 bit Macs thanks to Lawrence Velazquez.
381    16. Partial matches now set offsets[2] to the "bumpalong" value, that is, the
382        offset of the starting point of the matching process, provided the offsets
383        vector is large enough.
385    17. The \A escape now records a lookbehind value of 1, though its execution
386        does not actually inspect the previous character. This is to ensure that,
387        in partial multi-segment matching, at least one character from the old
388        segment is retained when a new segment is processed. Otherwise, if there
389        are no lookbehinds in the pattern, \A might match incorrectly at the start
390        of a new segment.
392    18. Added some #ifdef __VMS code into pcretest.c to help VMS implementations.
394    19. Redefined some pcre_uchar variables in pcre_exec.c as pcre_uint32; this
395        gives some modest performance improvement in 8-bit mode.
397    20. Added the PCRE-specific property \p{Xuc} for matching characters that can
398        be expressed in certain programming languages using Universal Character
399        Names.
401    21. Unicode validation has been updated in the light of Unicode Corrigendum #9,
402        which points out that "non characters" are not "characters that may not
403        appear in Unicode strings" but rather "characters that are reserved for
404        internal use and have only local meaning".
406    22. When a pattern was compiled with automatic callouts (PCRE_AUTO_CALLOUT) and
407        there was a conditional group that depended on an assertion, if the
408        assertion was false, the callout that immediately followed the alternation
409        in the condition was skipped when pcre_exec() was used for matching.
411    23. Allow an explicit callout to be inserted before an assertion that is the
412        condition for a conditional group, for compatibility with automatic
413        callouts, which always insert a callout at this point.
415    24. In 8.31, (*COMMIT) was confined to within a recursive subpattern. Perl also
416        confines (*SKIP) and (*PRUNE) in the same way, and this has now been done.
418    25. (*PRUNE) is now supported by the JIT compiler.
420    26. Fix infinite loop when /(?<=(*SKIP)ac)a/ is matched against aa.
422    27. Fix the case where there are two or more SKIPs with arguments that may be
423        ignored.
425    28. (*SKIP) is now supported by the JIT compiler.
427    29. (*THEN) is now supported by the JIT compiler.
429    30. Update RunTest with additional test selector options.
431    31. The way PCRE handles backtracking verbs has been changed in two ways.
433        (1) Previously, in something like (*COMMIT)(*SKIP), COMMIT would override
434        SKIP. Now, PCRE acts on whichever backtracking verb is reached first by
435        backtracking. In some cases this makes it more Perl-compatible, but Perl's
436        rather obscure rules do not always do the same thing.
438        (2) Previously, backtracking verbs were confined within assertions. This is
439        no longer the case for positive assertions, except for (*ACCEPT). Again,
440        this sometimes improves Perl compatibility, and sometimes does not.
442    32. A number of tests that were in test 2 because Perl did things differently
443        have been moved to test 1, because either Perl or PCRE has changed, and
444        these tests are now compatible.
446    32. Backtracking control verbs are now handled in the same way in JIT and
447        interpreter.
449    33. An opening parenthesis in a MARK/PRUNE/SKIP/THEN name in a pattern that
450        contained a forward subroutine reference caused a compile error.
452    34. Auto-detect and optimize limited repetitions in JIT.
454    35. Implement PCRE_NEVER_UTF to lock out the use of UTF, in particular,
455        blocking (*UTF) etc.
457    36. In the interpreter, maximizing pattern repetitions for characters and
458        character types now use tail recursion, which reduces stack usage.
460    37. The value of the max lookbehind was not correctly preserved if a compiled
461        and saved regex was reloaded on a host of different endianness.
463    38. Implemented (*LIMIT_MATCH) and (*LIMIT_RECURSION). As part of the extension
464        of the compiled pattern block, expand the flags field from 16 to 32 bits
465        because it was almost full.
467    39. Try madvise first before posix_madvise.
469    40. Change 7 for PCRE 7.9 made it impossible for pcregrep to find empty lines
470        with a pattern such as ^$. It has taken 4 years for anybody to notice! The
471        original change locked out all matches of empty strings. This has been
472        changed so that one match of an empty string per line is recognized.
473        Subsequent searches on the same line (for colouring or for --only-matching,
474        for example) do not recognize empty strings.
476    41. Applied a user patch to fix a number of spelling mistakes in comments.
478    42. Data lines longer than 65536 caused pcretest to crash.
480    43. Clarified the data type for length and startoffset arguments for pcre_exec
481        and pcre_dfa_exec in the function-specific man pages, where they were
482        explicitly stated to be in bytes, never having been updated. I also added
483        some clarification to the pcreapi man page.
485    44. A call to pcre_dfa_exec() with an output vector size less than 2 caused
486        a segmentation fault.
489    Version 8.32 30-November-2012
490    -----------------------------
492    1.  Improved JIT compiler optimizations for first character search and single
493        character iterators.
495    2.  Supporting IBM XL C compilers for PPC architectures in the JIT compiler.
496        Patch by Daniel Richard G.
498    3.  Single character iterator optimizations in the JIT compiler.
500    4.  Improved JIT compiler optimizations for character ranges.
502    5.  Rename the "leave" variable names to "quit" to improve WinCE compatibility.
503        Reported by Giuseppe D'Angelo.
505    6.  The PCRE_STARTLINE bit, indicating that a match can occur only at the start
506        of a line, was being set incorrectly in cases where .* appeared inside
507        atomic brackets at the start of a pattern, or where there was a subsequent
508        *PRUNE or *SKIP.
510    7.  Improved instruction cache flush for POWER/PowerPC.
511        Patch by Daniel Richard G.
513    8.  Fixed a number of issues in pcregrep, making it more compatible with GNU
514        grep:
516        (a) There is now no limit to the number of patterns to be matched.
518        (b) An error is given if a pattern is too long.
520        (c) Multiple uses of --exclude, --exclude-dir, --include, and --include-dir
521            are now supported.
523        (d) --exclude-from and --include-from (multiple use) have been added.
525        (e) Exclusions and inclusions now apply to all files and directories, not
526            just to those obtained from scanning a directory recursively.
528        (f) Multiple uses of -f and --file-list are now supported.
530        (g) In a Windows environment, the default for -d has been changed from
531            "read" (the GNU grep default) to "skip", because otherwise the presence
532            of a directory in the file list provokes an error.
534        (h) The documentation has been revised and clarified in places.
536    9.  Improve the matching speed of capturing brackets.
538    10. Changed the meaning of \X so that it now matches a Unicode extended
539        grapheme cluster.
541    11. Patch by Daniel Richard G to the autoconf files to add a macro for sorting
542        out POSIX threads when JIT support is configured.
544    12. Added support for PCRE_STUDY_EXTRA_NEEDED.
546    13. In the POSIX wrapper regcomp() function, setting re_nsub field in the preg
547        structure could go wrong in environments where size_t is not the same size
548        as int.
550    14. Applied user-supplied patch to pcrecpp.cc to allow PCRE_NO_UTF8_CHECK to be
551        set.
553    15. The EBCDIC support had decayed; later updates to the code had included
554        explicit references to (e.g.) \x0a instead of CHAR_LF. There has been a
555        general tidy up of EBCDIC-related issues, and the documentation was also
556        not quite right. There is now a test that can be run on ASCII systems to
557        check some of the EBCDIC-related things (but is it not a full test).
559    16. The new PCRE_STUDY_EXTRA_NEEDED option is now used by pcregrep, resulting
560        in a small tidy to the code.
562    17. Fix JIT tests when UTF is disabled and both 8 and 16 bit mode are enabled.
564    18. If the --only-matching (-o) option in pcregrep is specified multiple
565        times, each one causes appropriate output. For example, -o1 -o2 outputs the
566        substrings matched by the 1st and 2nd capturing parentheses. A separating
567        string can be specified by --om-separator (default empty).
569    19. Improving the first n character searches.
571    20. Turn case lists for horizontal and vertical white space into macros so that
572        they are defined only once.
574    21. This set of changes together give more compatible Unicode case-folding
575        behaviour for characters that have more than one other case when UCP
576        support is available.
578        (a) The Unicode property table now has offsets into a new table of sets of
579            three or more characters that are case-equivalent. The MultiStage2.py
580            script that generates these tables (the pcre_ucd.c file) now scans
581            CaseFolding.txt instead of UnicodeData.txt for character case
582            information.
584        (b) The code for adding characters or ranges of characters to a character
585            class has been abstracted into a generalized function that also handles
586            case-independence. In UTF-mode with UCP support, this uses the new data
587            to handle characters with more than one other case.
589        (c) A bug that is fixed as a result of (b) is that codepoints less than 256
590            whose other case is greater than 256 are now correctly matched
591            caselessly. Previously, the high codepoint matched the low one, but not
592            vice versa.
594        (d) The processing of \h, \H, \v, and \ in character classes now makes use
595            of the new class addition function, using character lists defined as
596            macros alongside the case definitions of 20 above.
598        (e) Caseless back references now work with characters that have more than
599            one other case.
601        (f) General caseless matching of characters with more than one other case
602            is supported.
604    22. Unicode character properties were updated from Unicode 6.2.0
606    23. Improved CMake support under Windows. Patch by Daniel Richard G.
608    24. Add support for 32-bit character strings, and UTF-32
610    25. Major JIT compiler update (code refactoring and bugfixing).
611        Experimental Sparc 32 support is added.
613    26. Applied a modified version of Daniel Richard G's patch to create
614        pcre.h.generic and config.h.generic by "make" instead of in the
615        PrepareRelease script.
617    27. Added a definition for CHAR_NULL (helpful for the z/OS port), and use it in
618        pcre_compile.c when checking for a zero character.
620    28. Introducing a native interface for JIT. Through this interface, the compiled
621        machine code can be directly executed. The purpose of this interface is to
622        provide fast pattern matching, so several sanity checks are not performed.
623        However, feature tests are still performed. The new interface provides
624        1.4x speedup compared to the old one.
626    29. If pcre_exec() or pcre_dfa_exec() was called with a negative value for
627        the subject string length, the error given was PCRE_ERROR_BADOFFSET, which
628        was confusing. There is now a new error PCRE_ERROR_BADLENGTH for this case.
630    30. In 8-bit UTF-8 mode, pcretest failed to give an error for data codepoints
631        greater than 0x7fffffff (which cannot be represented in UTF-8, even under
632        the "old" RFC 2279). Instead, it ended up passing a negative length to
633        pcre_exec().
635    31. Add support for GCC's visibility feature to hide internal functions.
637    32. Running "pcretest -C pcre8" or "pcretest -C pcre16" gave a spurious error
638        "unknown -C option" after outputting 0 or 1.
640    33. There is now support for generating a code coverage report for the test
641        suite in environments where gcc is the compiler and lcov is installed. This
642        is mainly for the benefit of the developers.
644    34. If PCRE is built with --enable-valgrind, certain memory regions are marked
645        unaddressable using valgrind annotations, allowing valgrind to detect
646        invalid memory accesses. This is mainly for the benefit of the developers.
648    25. (*UTF) can now be used to start a pattern in any of the three libraries.
650    26. Give configure error if --enable-cpp but no C++ compiler found.
653    Version 8.31 06-July-2012
654    -------------------------
656    1.  Fixing a wrong JIT test case and some compiler warnings.
658    2.  Removed a bashism from the RunTest script.
660    3.  Add a cast to pcre_exec.c to fix the warning "unary minus operator applied
661        to unsigned type, result still unsigned" that was given by an MS compiler
662        on encountering the code "-sizeof(xxx)".
664    4.  Partial matching support is added to the JIT compiler.
666    5.  Fixed several bugs concerned with partial matching of items that consist
667        of more than one character:
669        (a) /^(..)\1/ did not partially match "aba" because checking references was
670            done on an "all or nothing" basis. This also applied to repeated
671            references.
673        (b) \R did not give a hard partial match if \r was found at the end of the
674            subject.
676        (c) \X did not give a hard partial match after matching one or more
677            characters at the end of the subject.
679        (d) When newline was set to CRLF, a pattern such as /a$/ did not recognize
680            a partial match for the string "\r".
682        (e) When newline was set to CRLF, the metacharacter "." did not recognize
683            a partial match for a CR character at the end of the subject string.
685    6.  If JIT is requested using /S++ or -s++ (instead of just /S+ or -s+) when
686        running pcretest, the text "(JIT)" added to the output whenever JIT is
687        actually used to run the match.
689    7.  Individual JIT compile options can be set in pcretest by following -s+[+]
690        or /S+[+] with a digit between 1 and 7.
692    8.  OP_NOT now supports any UTF character not just single-byte ones.
694    9.  (*MARK) control verb is now supported by the JIT compiler.
696    10. The command "./RunTest list" lists the available tests without actually
697        running any of them. (Because I keep forgetting what they all are.)
701    12. Applied a (slightly modified) user-supplied patch that improves performance
702        when the heap is used for recursion (compiled with --disable-stack-for-
703        recursion). Instead of malloc and free for each heap frame each time a
704        logical recursion happens, frames are retained on a chain and re-used where
705        possible. This sometimes gives as much as 30% improvement.
707    13. As documented, (*COMMIT) is now confined to within a recursive subpattern
708        call.
710    14. As documented, (*COMMIT) is now confined to within a positive assertion.
712    15. It is now possible to link pcretest with libedit as an alternative to
713        libreadline.
715    16. (*COMMIT) control verb is now supported by the JIT compiler.
717    17. The Unicode data tables have been updated to Unicode 6.1.0.
719    18. Added --file-list option to pcregrep.
721    19. Added binary file support to pcregrep, including the -a, --binary-files,
722        -I, and --text options.
724    20. The madvise function is renamed for posix_madvise for QNX compatibility
725        reasons. Fixed by Giuseppe D'Angelo.
727    21. Fixed a bug for backward assertions with REVERSE 0 in the JIT compiler.
729    22. Changed the option for creating symbolic links for 16-bit man pages from
730        -s to -sf so that re-installing does not cause issues.
732    23. Support PCRE_NO_START_OPTIMIZE in JIT as (*MARK) support requires it.
734    24. Fixed a very old bug in pcretest that caused errors with restarted DFA
735        matches in certain environments (the workspace was not being correctly
736        retained). Also added to pcre_dfa_exec() a simple plausibility check on
737        some of the workspace data at the beginning of a restart.
739    25. \s*\R was auto-possessifying the \s* when it should not, whereas \S*\R
740        was not doing so when it should - probably a typo introduced by SVN 528
741        (change 8.10/14).
743    26. When PCRE_UCP was not set, \w+\x{c4} was incorrectly auto-possessifying the
744        \w+ when the character tables indicated that \x{c4} was a word character.
745        There were several related cases, all because the tests for doing a table
746        lookup were testing for characters less than 127 instead of 255.
748    27. If a pattern contains capturing parentheses that are not used in a match,
749        their slots in the ovector are set to -1. For those that are higher than
750        any matched groups, this happens at the end of processing. In the case when
751        there were back references that the ovector was too small to contain
752        (causing temporary malloc'd memory to be used during matching), and the
753        highest capturing number was not used, memory off the end of the ovector
754        was incorrectly being set to -1. (It was using the size of the temporary
755        memory instead of the true size.)
757    28. To catch bugs like 27 using valgrind, when pcretest is asked to specify an
758        ovector size, it uses memory at the end of the block that it has got.
760    29. Check for an overlong MARK name and give an error at compile time. The
761        limit is 255 for the 8-bit library and 65535 for the 16-bit library.
763    30. JIT compiler update.
765    31. JIT is now supported on jailbroken iOS devices. Thanks for Ruiger
766        Rill for the patch.
768    32. Put spaces around SLJIT_PRINT_D in the JIT compiler. Required by CXX11.
770    33. Variable renamings in the PCRE-JIT compiler. No functionality change.
772    34. Fixed typos in pcregrep: in two places there was SUPPORT_LIBZ2 instead of
773        SUPPORT_LIBBZ2. This caused a build problem when bzip2 but not gzip (zlib)
774        was enabled.
776    35. Improve JIT code generation for greedy plus quantifier.
778    36. When /((?:a?)*)*c/ or /((?>a?)*)*c/ was matched against "aac", it set group
779        1 to "aa" instead of to an empty string. The bug affected repeated groups
780        that could potentially match an empty string.
782    37. Optimizing single character iterators in JIT.
784    38. Wide characters specified with \uxxxx in JavaScript mode are now subject to
785        the same checks as \x{...} characters in non-JavaScript mode. Specifically,
786        codepoints that are too big for the mode are faulted, and in a UTF mode,
787        disallowed codepoints are also faulted.
789    39. If PCRE was compiled with UTF support, in three places in the DFA
790        matcher there was code that should only have been obeyed in UTF mode, but
791        was being obeyed unconditionally. In 8-bit mode this could cause incorrect
792        processing when bytes with values greater than 127 were present. In 16-bit
793        mode the bug would be provoked by values in the range 0xfc00 to 0xdc00. In
794        both cases the values are those that cannot be the first data item in a UTF
795        character. The three items that might have provoked this were recursions,
796        possessively repeated groups, and atomic groups.
798    40. Ensure that libpcre is explicitly listed in the link commands for pcretest
799        and pcregrep, because some OS require shared objects to be explicitly
800        passed to ld, causing the link step to fail if they are not.
802    41. There were two incorrect #ifdefs in pcre_study.c, meaning that, in 16-bit
803        mode, patterns that started with \h* or \R* might be incorrectly matched.
806    Version 8.30 04-February-2012
807    -----------------------------
809    1.  Renamed "isnumber" as "is_a_number" because in some Mac environments this
810        name is defined in ctype.h.
812    2.  Fixed a bug in fixed-length calculation for lookbehinds that would show up
813        only in quite long subpatterns.
815    3.  Removed the function pcre_info(), which has been obsolete and deprecated
816        since it was replaced by pcre_fullinfo() in February 2000.
818    4.  For a non-anchored pattern, if (*SKIP) was given with a name that did not
819        match a (*MARK), and the match failed at the start of the subject, a
820        reference to memory before the start of the subject could occur. This bug
821        was introduced by fix 17 of release 8.21.
823    5.  A reference to an unset group with zero minimum repetition was giving
824        totally wrong answers (in non-JavaScript-compatibility mode). For example,
825        /(another)?(\1?)test/ matched against "hello world test". This bug was
826        introduced in release 8.13.
828    6.  Add support for 16-bit character strings (a large amount of work involving
829        many changes and refactorings).
831    7.  RunGrepTest failed on msys because \r\n was replaced by whitespace when the
832        command "pattern=`printf 'xxx\r\njkl'`" was run. The pattern is now taken
833        from a file.
835    8.  Ovector size of 2 is also supported by JIT based pcre_exec (the ovector size
836        rounding is not applied in this particular case).
838    9.  The invalid Unicode surrogate codepoints U+D800 to U+DFFF are now rejected
839        if they appear, or are escaped, in patterns.
841    10. Get rid of a number of -Wunused-but-set-variable warnings.
843    11. The pattern /(?=(*:x))(q|)/ matches an empty string, and returns the mark
844        "x". The similar pattern /(?=(*:x))((*:y)q|)/ did not return a mark at all.
845        Oddly, Perl behaves the same way. PCRE has been fixed so that this pattern
846        also returns the mark "x". This bug applied to capturing parentheses,
847        non-capturing parentheses, and atomic parentheses. It also applied to some
848        assertions.
850    12. Stephen Kelly's patch to CMakeLists.txt allows it to parse the version
851        information out of configure.ac instead of relying on pcre.h.generic, which
852        is not stored in the repository.
854    13. Applied Dmitry V. Levin's patch for a more portable method for linking with
855        -lreadline.
857    14. ZH added PCRE_CONFIG_JITTARGET; added its output to pcretest -C.
859    15. Applied Graycode's patch to put the top-level frame on the stack rather
860        than the heap when not using the stack for recursion. This gives a
861        performance improvement in many cases when recursion is not deep.
863    16. Experimental code added to "pcretest -C" to output the stack frame size.
866    Version 8.21 12-Dec-2011
867    ------------------------
869    1.  Updating the JIT compiler.
871    2.  JIT compiler now supports OP_NCREF, OP_RREF and OP_NRREF. New test cases
872        are added as well.
874    3.  Fix cache-flush issue on PowerPC (It is still an experimental JIT port).
875        PCRE_EXTRA_TABLES is not suported by JIT, and should be checked before
876        calling _pcre_jit_exec. Some extra comments are added.
878    4.  (*MARK) settings inside atomic groups that do not contain any capturing
879        parentheses, for example, (?>a(*:m)), were not being passed out. This bug
880        was introduced by change 18 for 8.20.
882    5.  Supporting of \x, \U and \u in JavaScript compatibility mode based on the
883        ECMA-262 standard.
885    6.  Lookbehinds such as (?<=a{2}b) that contained a fixed repetition were
886        erroneously being rejected as "not fixed length" if PCRE_CASELESS was set.
887        This bug was probably introduced by change 9 of 8.13.
889    7.  While fixing 6 above, I noticed that a number of other items were being
890        incorrectly rejected as "not fixed length". This arose partly because newer
891        opcodes had not been added to the fixed-length checking code. I have (a)
892        corrected the bug and added tests for these items, and (b) arranged for an
893        error to occur if an unknown opcode is encountered while checking for fixed
894        length instead of just assuming "not fixed length". The items that were
895        rejected were: (*ACCEPT), (*COMMIT), (*FAIL), (*MARK), (*PRUNE), (*SKIP),
896        (*THEN), \h, \H, \v, \V, and single character negative classes with fixed
897        repetitions, e.g. [^a]{3}, with and without PCRE_CASELESS.
899    8.  A possessively repeated conditional subpattern such as (?(?=c)c|d)++ was
900        being incorrectly compiled and would have given unpredicatble results.
902    9.  A possessively repeated subpattern with minimum repeat count greater than
903        one behaved incorrectly. For example, (A){2,}+ behaved as if it was
904        (A)(A)++ which meant that, after a subsequent mismatch, backtracking into
905        the first (A) could occur when it should not.
907    10. Add a cast and remove a redundant test from the code.
909    11. JIT should use pcre_malloc/pcre_free for allocation.
911    12. Updated pcre-config so that it no longer shows -L/usr/lib, which seems
912        best practice nowadays, and helps with cross-compiling. (If the exec_prefix
913        is anything other than /usr, -L is still shown).
915    13. In non-UTF-8 mode, \C is now supported in lookbehinds and DFA matching.
917    14. Perl does not support \N without a following name in a [] class; PCRE now
918        also gives an error.
920    15. If a forward reference was repeated with an upper limit of around 2000,
921        it caused the error "internal error: overran compiling workspace". The
922        maximum number of forward references (including repeats) was limited by the
923        internal workspace, and dependent on the LINK_SIZE. The code has been
924        rewritten so that the workspace expands (via pcre_malloc) if necessary, and
925        the default depends on LINK_SIZE. There is a new upper limit (for safety)
926        of around 200,000 forward references. While doing this, I also speeded up
927        the filling in of repeated forward references.
929    16. A repeated forward reference in a pattern such as (a)(?2){2}(.) was
930        incorrectly expecting the subject to contain another "a" after the start.
932    17. When (*SKIP:name) is activated without a corresponding (*MARK:name) earlier
933        in the match, the SKIP should be ignored. This was not happening; instead
934        the SKIP was being treated as NOMATCH. For patterns such as
935        /A(*MARK:A)A+(*SKIP:B)Z|AAC/ this meant that the AAC branch was never
936        tested.
938    18. The behaviour of (*MARK), (*PRUNE), and (*THEN) has been reworked and is
939        now much more compatible with Perl, in particular in cases where the result
940        is a non-match for a non-anchored pattern. For example, if
941        /b(*:m)f|a(*:n)w/ is matched against "abc", the non-match returns the name
942        "m", where previously it did not return a name. A side effect of this
943        change is that for partial matches, the last encountered mark name is
944        returned, as for non matches. A number of tests that were previously not
945        Perl-compatible have been moved into the Perl-compatible test files. The
946        refactoring has had the pleasing side effect of removing one argument from
947        the match() function, thus reducing its stack requirements.
949    19. If the /S+ option was used in pcretest to study a pattern using JIT,
950        subsequent uses of /S (without +) incorrectly behaved like /S+.
952    21. Retrieve executable code size support for the JIT compiler and fixing
953        some warnings.
955    22. A caseless match of a UTF-8 character whose other case uses fewer bytes did
956        not work when the shorter character appeared right at the end of the
957        subject string.
959    23. Added some (int) casts to non-JIT modules to reduce warnings on 64-bit
960        systems.
962    24. Added PCRE_INFO_JITSIZE to pass on the value from (21) above, and also
963        output it when the /M option is used in pcretest.
965    25. The CheckMan script was not being included in the distribution. Also, added
966        an explicit "perl" to run Perl scripts from the PrepareRelease script
967        because this is reportedly needed in Windows.
969    26. If study data was being save in a file and studying had not found a set of
970        "starts with" bytes for the pattern, the data written to the file (though
971        never used) was taken from uninitialized memory and so caused valgrind to
972        complain.
974    27. Updated RunTest.bat as provided by Sheri Pierce.
976    28. Fixed a possible uninitialized memory bug in pcre_jit_compile.c.
978    29. Computation of memory usage for the table of capturing group names was
979        giving an unnecessarily large value.
982    Version 8.20 21-Oct-2011
983    ------------------------
985    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
986        a POSIX class. After further experiments with Perl, which convinced me that
987        Perl has bugs and confusions, a closing square bracket is no longer allowed
988        in a POSIX name. This bug also affected patterns with classes that started
989        with full stops.
991    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
992        captured substring, but while checking the failing first alternative,
993        substring 1 is temporarily captured. If the output vector supplied to
994        pcre_exec() was not big enough for this capture, the yield of the function
995        was still zero ("insufficient space for captured substrings"). This cannot
996        be totally fixed without adding another stack variable, which seems a lot
997        of expense for a edge case. However, I have improved the situation in cases
998        such as /(a)(b)x|abc/ matched against "abc", where the return code
999        indicates that fewer than the maximum number of slots in the ovector have
1000        been set.
1002    3.  Related to (2) above: when there are more back references in a pattern than
1003        slots in the output vector, pcre_exec() uses temporary memory during
1004        matching, and copies in the captures as far as possible afterwards. It was
1005        using the entire output vector, but this conflicts with the specification
1006        that only 2/3 is used for passing back captured substrings. Now it uses
1007        only the first 2/3, for compatibility. This is, of course, another edge
1008        case.
1010    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
1011        main code base, and can be used by building with --enable-jit. When this is
1012        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
1013        runtime --no-jit option is given.
1015    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
1016        ovector, the return from the function was zero, implying that there were
1017        other matches that did not fit. The correct "exactly full" value is now
1018        returned.
1020    6.  If a subpattern that was called recursively or as a subroutine contained
1021        (*PRUNE) or any other control that caused it to give a non-standard return,
1022        invalid errors such as "Error -26 (nested recursion at the same subject
1023        position)" or even infinite loops could occur.
1025    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
1026        computing the minimum length on reaching *ACCEPT, and so ended up with the
1027        wrong value of 1 rather than 0. Further investigation indicates that
1028        computing a minimum subject length in the presence of *ACCEPT is difficult
1029        (think back references, subroutine calls), and so I have changed the code
1030        so that no minimum is registered for a pattern that contains *ACCEPT.
1032    8.  If (*THEN) was present in the first (true) branch of a conditional group,
1033        it was not handled as intended. [But see 16 below.]
1035    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
1036        Sheri Pierce.
1038    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
1039        the first byte in a match must be "a".
1041    11. Change 17 for 8.13 increased the recursion depth for patterns like
1042        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
1043        pattern contains any instances of (*THEN). If it does not, the old
1044        optimizations are restored. It would be nice to do this on a per-group
1045        basis, but at the moment that is not feasible.
1047    12. In some environments, the output of pcretest -C is CRLF terminated. This
1048        broke RunTest's code that checks for the link size. A single white space
1049        character after the value is now allowed for.
1051    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
1052        For "fr", it uses the Windows-specific input and output files.
1054    14. If (*THEN) appeared in a group that was called recursively or as a
1055        subroutine, it did not work as intended. [But see next item.]
1057    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
1058        pattern fragments (but not containing any | characters). If A and B are
1059        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
1060        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
1061        D. In other words, Perl considers parentheses that do not contain any |
1062        characters to be part of a surrounding alternative, whereas PCRE was
1063        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
1064        differently. PCRE now behaves in the same way as Perl, except in the case
1065        of subroutine/recursion calls such as (?1) which have in any case always
1066        been different (but PCRE had them first :-).
1068    16. Related to 15 above: Perl does not treat the | in a conditional group as
1069        creating alternatives. Such a group is treated in the same way as an
1070        ordinary group without any | characters when processing (*THEN). PCRE has
1071        been changed to match Perl's behaviour.
1073    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
1074        RunGrepTest script failed.
1076    18. Change 22 for version 13 caused atomic groups to use more stack. This is
1077        inevitable for groups that contain captures, but it can lead to a lot of
1078        stack use in large patterns. The old behaviour has been restored for atomic
1079        groups that do not contain any capturing parentheses.
1081    19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not
1082        suppress the check for a minimum subject length at run time. (If it was
1083        given to pcre_exec() or pcre_dfa_exec() it did work.)
1085    20. Fixed an ASCII-dependent infelicity in pcretest that would have made it
1086        fail to work when decoding hex characters in data strings in EBCDIC
1087        environments.
1089    21. It appears that in at least one Mac OS environment, the isxdigit() function
1090        is implemented as a macro that evaluates to its argument more than once,
1091        contravening the C 90 Standard (I haven't checked a later standard). There
1092        was an instance in pcretest which caused it to go wrong when processing
1093        \x{...} escapes in subject strings. The has been rewritten to avoid using
1094        things like p++ in the argument of isxdigit().
1097    Version 8.13 16-Aug-2011
1098    ------------------------
1100    1.  The Unicode data tables have been updated to Unicode 6.0.0.
1102    2.  Two minor typos in pcre_internal.h have been fixed.
1104    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
1105        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
1106        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
1108    4.  There were a number of related bugs in the code for matching backrefences
1109        caselessly in UTF-8 mode when codes for the characters concerned were
1110        different numbers of bytes. For example, U+023A and U+2C65 are an upper
1111        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
1112        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
1113        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
1114        2-byte code at the end of the subject (it thought there wasn't enough data
1115        left).
1117    5.  Comprehensive information about what went wrong is now returned by
1118        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
1119        as the output vector has at least 2 elements. The offset of the start of
1120        the failing character and a reason code are placed in the vector.
1122    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
1123        now returned is for the first byte of the failing character, instead of the
1124        last byte inspected. This is an incompatible change, but I hope it is small
1125        enough not to be a problem. It makes the returned offset consistent with
1126        pcre_exec() and pcre_dfa_exec().
1128    7.  pcretest now gives a text phrase as well as the error number when
1129        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
1130        failure, the offset and reason code are output.
1132    8.  When \R was used with a maximizing quantifier it failed to skip backwards
1133        over a \r\n pair if the subsequent match failed. Instead, it just skipped
1134        back over a single character (\n). This seems wrong (because it treated the
1135        two characters as a single entity when going forwards), conflicts with the
1136        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
1137        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
1138        has been changed.
1140    9.  Some internal refactoring has changed the processing so that the handling
1141        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
1142        time (the PCRE_DOTALL option was changed this way some time ago: version
1143        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
1144        which was always a bit of a fudge. It also means that there is one less
1145        argument for the match() function, which reduces its stack requirements
1146        slightly. This change also fixes an incompatibility with Perl: the pattern
1147        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
1149    10. More internal refactoring has drastically reduced the number of recursive
1150        calls to match() for possessively repeated groups such as (abc)++ when
1151        using pcre_exec().
1153    11. While implementing 10, a number of bugs in the handling of groups were
1154        discovered and fixed:
1156        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
1157        (a|)*(?1) gave a compile-time internal error.
1158        ((a|)+)+  did not notice that the outer group could match an empty string.
1159        (^a|^)+   was not marked as anchored.
1160        (.*a|.*)+ was not marked as matching at start or after a newline.
1162    12. Yet more internal refactoring has removed another argument from the match()
1163        function. Special calls to this function are now indicated by setting a
1164        value in a variable in the "match data" data block.
1166    13. Be more explicit in pcre_study() instead of relying on "default" for
1167        opcodes that mean there is no starting character; this means that when new
1168        ones are added and accidentally left out of pcre_study(), testing should
1169        pick them up.
1171    14. The -s option of pcretest has been documented for ages as being an old
1172        synonym of -m (show memory usage). I have changed it to mean "force study
1173        for every regex", that is, assume /S for every regex. This is similar to -i
1174        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
1175        using it. It makes it easier to run collections of tests with and without
1176        study enabled, and thereby test pcre_study() more easily. All the standard
1177        tests are now run with and without -s (but some patterns can be marked as
1178        "never study" - see 20 below).
1180    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
1181        restoration of the capturing data to the outer values was not happening
1182        correctly.
1184    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
1185        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
1186        pattern had matched an empty string, and so incorrectly returned a no
1187        match.
1189    17. There was optimizing code for the last branch of non-capturing parentheses,
1190        and also for the obeyed branch of a conditional subexpression, which used
1191        tail recursion to cut down on stack usage. Unfortunately, now that there is
1192        the possibility of (*THEN) occurring in these branches, tail recursion is
1193        no longer possible because the return has to be checked for (*THEN). These
1194        two optimizations have therefore been removed. [But see 8.20/11 above.]
1196    18. If a pattern containing \R was studied, it was assumed that \R always
1197        matched two bytes, thus causing the minimum subject length to be
1198        incorrectly computed because \R can also match just one byte.
1200    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
1201        was incorrectly computed.
1203    20. If /S is present twice on a test pattern in pcretest input, it now
1204        *disables* studying, thereby overriding the use of -s on the command line
1205        (see 14 above). This is necessary for one or two tests to keep the output
1206        identical in both cases.
1208    21. When (*ACCEPT) was used in an assertion that matched an empty string and
1209        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
1211    22. When an atomic group that contained a capturing parenthesis was
1212        successfully matched, but the branch in which it appeared failed, the
1213        capturing was not being forgotten if a higher numbered group was later
1214        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
1215        group 1 to "a", when in fact it should be unset. This applied to multi-
1216        branched capturing and non-capturing groups, repeated or not, and also to
1217        positive assertions (capturing in negative assertions does not happen
1218        in PCRE) and also to nested atomic groups.
1220    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
1221        subject after a captured substring, to make it easier to tell which of a
1222        number of identical substrings has been captured.
1224    24. The way atomic groups are processed by pcre_exec() has been changed so that
1225        if they are repeated, backtracking one repetition now resets captured
1226        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
1227        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
1228        "aaa". Previously, it would have been "a". As part of this code
1229        refactoring, the way recursive calls are handled has also been changed.
1231    25. If an assertion condition captured any substrings, they were not passed
1232        back unless some other capturing happened later. For example, if
1233        (?(?=(a))a) was matched against "a", no capturing was returned.
1235    26. When studying a pattern that contained subroutine calls or assertions,
1236        the code for finding the minimum length of a possible match was handling
1237        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
1238        group 1 called group 2 while simultaneously a separate group 2 called group
1239        1). A stack overflow occurred in this case. I have fixed this by limiting
1240        the recursion depth to 10.
1242    27. Updated RunTest.bat in the distribution to the version supplied by Tom
1243        Fortmann. This supports explicit test numbers on the command line, and has
1244        argument validation and error reporting.
1246    28. An instance of \X with an unlimited repeat could fail if at any point the
1247        first character it looked at was a mark character.
1249    29. Some minor code refactoring concerning Unicode properties and scripts
1250        should reduce the stack requirement of match() slightly.
1252    30. Added the '=' option to pcretest to check the setting of unused capturing
1253        slots at the end of the pattern, which are documented as being -1, but are
1254        not included in the return count.
1256    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
1257        compiled something random. Now it gives a compile-time error (as does
1258        Perl).
1260    32. A *MARK encountered during the processing of a positive assertion is now
1261        recorded and passed back (compatible with Perl).
1263    33. If --only-matching or --colour was set on a pcregrep call whose pattern
1264        had alternative anchored branches, the search for a second match in a line
1265        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
1266        matched the line "0102" twice. The same bug affected patterns that started
1267        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
1268        twice.
1270    34. Previously, PCRE did not allow quantification of assertions. However, Perl
1271        does, and because of capturing effects, quantifying parenthesized
1272        assertions may at times be useful. Quantifiers are now allowed for
1273        parenthesized assertions.
1275    35. A minor code tidy in pcre_compile() when checking options for \R usage.
1277    36. \g was being checked for fancy things in a character class, when it should
1278        just be a literal "g".
1280    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
1281        appearance of a nested POSIX class supersedes an apparent external class.
1282        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
1283        unescaped square brackets may also appear as part of class names. For
1284        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
1285        more like Perl. (But see 8.20/1 above.)
1287    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
1288        was because it thought it was \N{name}, which is not supported).
1290    39. Add minix to OS list not supporting the -S option in pcretest.
1292    40. PCRE tries to detect cases of infinite recursion at compile time, but it
1293        cannot analyze patterns in sufficient detail to catch mutual recursions
1294        such as ((?1))((?2)). There is now a runtime test that gives an error if a
1295        subgroup is called recursively as a subpattern for a second time at the
1296        same position in the subject string. In previous releases this might have
1297        been caught by the recursion limit, or it might have run out of stack.
1299    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
1300        happen only once. PCRE was, however incorrectly giving a compile time error
1301        "recursive call could loop indefinitely" because it cannot analyze the
1302        pattern in sufficient detail. The compile time test no longer happens when
1303        PCRE is compiling a conditional subpattern, but actual runaway loops are
1304        now caught at runtime (see 40 above).
1306    42. It seems that Perl allows any characters other than a closing parenthesis
1307        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
1308        has been changed to be the same.
1310    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
1311        as not to get warnings when autogen.sh is called. Also changed
1312        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
1314    44. To help people who use pcregrep to scan files containing exceedingly long
1315        lines, the following changes have been made:
1317        (a) The default value of the buffer size parameter has been increased from
1318            8K to 20K. (The actual buffer used is three times this size.)
1320        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
1321            PCRE is built.
1323        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
1324            to be set at run time.
1326        (d) Numerical values in pcregrep options can be followed by K or M, for
1327            example --buffer-size=50K.
1329        (e) If a line being scanned overflows pcregrep's buffer, an error is now
1330            given and the return code is set to 2.
1332    45. Add a pointer to the latest mark to the callout data block.
1334    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
1335        partial match of an empty string instead of no match. This was specific to
1336        the use of ".".
1338    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
1339        complete match instead of a partial match. This bug was dependent on both
1340        the PCRE_UTF8 and PCRE_DOTALL options being set.
1342    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
1343        starting byte set, because \b was not being ignored.
1346    Version 8.12 15-Jan-2011
1347    ------------------------
1349    1.  Fixed some typos in the markup of the man pages, and wrote a script that
1350        checks for such things as part of the documentation building process.
1352    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
1353        --match-limit and --recursion-limit options (added for 8.11). In
1354        particular, this made one of the standard tests fail. (The integer value
1355        went into the wrong half of a long int.)
1357    3.  If the --colour option was given to pcregrep with -v (invert match), it
1358        did strange things, either producing crazy output, or crashing. It should,
1359        of course, ignore a request for colour when reporting lines that do not
1360        match.
1362    4.  Another pcregrep bug caused similar problems if --colour was specified with
1363        -M (multiline) and the pattern match finished with a line ending.
1365    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
1366        matched in multiline mode, the following line was shown as part of the
1367        match. This seems wrong, so I have changed it.
1369    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
1370        the check for further matches in the same line (so they could be coloured)
1371        to overrun the end of the current line. If another match was found, it was
1372        incorrectly shown (and then shown again when found in the next line).
1374    7.  If pcregrep was compiled under Windows, there was a reference to the
1375        function pcregrep_exit() before it was defined. I am assuming this was
1376        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
1377        reported by a user. I've moved the definition above the reference.
1380    Version 8.11 10-Dec-2010
1381    ------------------------
1383    1.  (*THEN) was not working properly if there were untried alternatives prior
1384        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
1385        backtracked to try for "b" instead of moving to the next alternative branch
1386        at the same level (in this case, to look for "c"). The Perl documentation
1387        is clear that when (*THEN) is backtracked onto, it goes to the "next
1388        alternative in the innermost enclosing group".
1390    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
1391        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
1392        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
1393        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
1394        (*THEN).
1396    3.  If \s appeared in a character class, it removed the VT character from
1397        the class, even if it had been included by some previous item, for example
1398        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
1399        of \s, but is part of the POSIX "space" class.)
1401    4.  A partial match never returns an empty string (because you can always
1402        match an empty string at the end of the subject); however the checking for
1403        an empty string was starting at the "start of match" point. This has been
1404        changed to the "earliest inspected character" point, because the returned
1405        data for a partial match starts at this character. This means that, for
1406        example, /(?<=abc)def/ gives a partial match for the subject "abc"
1407        (previously it gave "no match").
1409    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
1410        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
1411        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
1412        has an implication that the given string is incomplete (because a partial
1413        match is preferred over a full match). For this reason, these items now
1414        give a partial match in this situation. [Aside: previously, the one case
1415        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
1416        match rather than a full match, which was wrong by the old rules, but is
1417        now correct.]
1419    6.  There was a bug in the handling of #-introduced comments, recognized when
1420        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
1421        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
1422        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
1423        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
1424        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
1425        places in pcre_compile().
1427    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
1428        comments when looking ahead for named forward references to subpatterns,
1429        the only newline sequence it recognized was NL. It now handles newlines
1430        according to the set newline convention.
1432    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
1433        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
1434        cater for a lack of strerror(). These oversights have been fixed.
1436    9.  Added --match-limit and --recursion-limit to pcregrep.
1438    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
1440    11. When the -o option was used, pcregrep was setting a return code of 1, even
1441        when matches were found, and --line-buffered was not being honoured.
1443    12. Added an optional parentheses number to the -o and --only-matching options
1444        of pcregrep.
1446    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
1447        can match an empty string. The code to do it in pcretest and pcredemo
1448        needed fixing:
1450        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
1451            only one byte after an empty string match just before CRLF (this case
1452            just got forgotten; "any" and "anycrlf" were OK).
1454        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
1455            mode when an empty string match preceded an ASCII character followed by
1456            a non-ASCII character. (The code for advancing by one character rather
1457            than one byte was nonsense.)
1459        (c) The pcredemo.c sample program did not have any code at all to handle
1460            the cases when CRLF is a valid newline sequence.
1462    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
1463        as a starting offset was within the subject string. There is now a new
1464        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
1465        negative or greater than the length of the string. In order to test this,
1466        pcretest is extended to allow the setting of negative starting offsets.
1468    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
1469        starting offset points to the beginning of a UTF-8 character was
1470        unnecessarily clumsy. I tidied it up.
1472    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
1473        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
1475    17. Nobody had reported that the --include_dir option, which was added in
1476        release 7.7 should have been called --include-dir (hyphen, not underscore)
1477        for compatibility with GNU grep. I have changed it to --include-dir, but
1478        left --include_dir as an undocumented synonym, and the same for
1479        --exclude-dir, though that is not available in GNU grep, at least as of
1480        release 2.5.4.
1482    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
1483        characters from a string of bytes) have been redefined so as not to use
1484        loops, in order to improve performance in some environments. At the same
1485        time, I abstracted some of the common code into auxiliary macros to save
1486        repetition (this should not affect the compiled code).
1488    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
1489        compile-time error is now given if \c is not followed by an ASCII
1490        character, that is, a byte less than 128. (In EBCDIC mode, the code is
1491        different, and any byte value is allowed.)
1493    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
1494        START_OPTIMIZE option, which is now allowed at compile time - but just
1495        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
1496        to pcregrep and other applications that have no direct access to PCRE
1497        options. The new /Y option in pcretest sets this option when calling
1498        pcre_compile().
1500    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
1501        back references. Groups containing recursive back references were forced to
1502        be atomic by that change, but in the case of named groups, the amount of
1503        memory required was incorrectly computed, leading to "Failed: internal
1504        error: code overflow". This has been fixed.
1506    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
1507        pcretest.c, to avoid build problems in some Borland environments.
1510    Version 8.10 25-Jun-2010
1511    ------------------------
1513    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
1514        THEN.
1516    2.  (*ACCEPT) was not working when inside an atomic group.
1518    3.  Inside a character class, \B is treated as a literal by default, but
1519        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
1520        causes the error). The code is unchanged, but I tidied the documentation.
1522    4.  Inside a character class, PCRE always treated \R and \X as literals,
1523        whereas Perl faults them if its -w option is set. I have changed PCRE so
1524        that it faults them when PCRE_EXTRA is set.
1526    5.  Added support for \N, which always matches any character other than
1527        newline. (It is the same as "." when PCRE_DOTALL is not set.)
1529    6.  When compiling pcregrep with newer versions of gcc which may have
1530        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
1531        declared with attribute warn_unused_result" were given. Just casting the
1532        result to (void) does not stop the warnings; a more elaborate fudge is
1533        needed. I've used a macro to implement this.
1535    7.  Minor change to pcretest.c to avoid a compiler warning.
1537    8.  Added four artifical Unicode properties to help with an option to make
1538        \s etc use properties (see next item). The new properties are: Xan
1539        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
1541    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
1542        use Unicode properties. (*UCP) at the start of a pattern can be used to set
1543        this option. Modified pcretest to add /W to test this facility. Added
1544        REG_UCP to make it available via the POSIX interface.
1546    10. Added --line-buffered to pcregrep.
1548    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
1549        studied, and the match started with a letter with a code point greater than
1550        127 whose first byte was different to the first byte of the other case of
1551        the letter, the other case of this starting letter was not recognized
1552        (#976).
1554    12. If a pattern that was studied started with a repeated Unicode property
1555        test, for example, \p{Nd}+, there was the theoretical possibility of
1556        setting up an incorrect bitmap of starting bytes, but fortunately it could
1557        not have actually happened in practice until change 8 above was made (it
1558        added property types that matched character-matching opcodes).
1560    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
1561        possible starting bytes for non-anchored patterns.
1563    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
1564        \R, and also a number of cases that involve Unicode properties, both
1565        explicit and implicit when PCRE_UCP is set.
1567    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
1568        input, it could crash or give wrong results if characters with values
1569        greater than 0xc0 were present in the subject string. (Detail: it assumed
1570        UTF-8 input when processing these items.)
1572    16. Added a lot of (int) casts to avoid compiler warnings in systems where
1573        size_t is 64-bit (#991).
1575    17. Added a check for running out of memory when PCRE is compiled with
1576        --disable-stack-for-recursion (#990).
1578    18. If the last data line in a file for pcretest does not have a newline on
1579        the end, a newline was missing in the output.
1581    19. The default pcre_chartables.c file recognizes only ASCII characters (values
1582        less than 128) in its various bitmaps. However, there is a facility for
1583        generating tables according to the current locale when PCRE is compiled. It
1584        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
1585        characters, are recognized by isspace() and therefore were getting set in
1586        these tables, and indeed these tables seem to approximate to ISO 8859. This
1587        caused a problem in UTF-8 mode when pcre_study() was used to create a list
1588        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
1589        which of course cannot start UTF-8 characters. I have changed the code so
1590        that only real ASCII characters (less than 128) and the correct starting
1591        bytes for UTF-8 encodings are set for characters greater than 127 when in
1592        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
1593        altogether.)
1595    20. Added the /T option to pcretest so as to be able to run tests with non-
1596        standard character tables, thus making it possible to include the tests
1597        used for 19 above in the standard set of tests.
1599    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
1600        reference to a subpattern the other side of a comment that contains an
1601        opening parenthesis caused either an internal compiling error, or a
1602        reference to the wrong subpattern.
1605    Version 8.02 19-Mar-2010
1606    ------------------------
1608    1.  The Unicode data tables have been updated to Unicode 5.2.0.
1610    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
1611        configured.
1613    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
1614        original author of that file, following a query about its status.
1616    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
1617        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
1619    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
1620        quantifier applied to a forward-referencing subroutine call, could compile
1621        incorrect code or give the error "internal error: previously-checked
1622        referenced subpattern not found".
1624    6.  Both MS Visual Studio and Symbian OS have problems with initializing
1625        variables to point to external functions. For these systems, therefore,
1626        pcre_malloc etc. are now initialized to local functions that call the
1627        relevant global functions.
1629    7.  There were two entries missing in the vectors called coptable and poptable
1630        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
1631        I've fixed the data, and added a kludgy way of testing at compile time that
1632        the lengths are correct (equal to the number of opcodes).
1634    8.  Following on from 7, I added a similar kludge to check the length of the
1635        eint vector in pcreposix.c.
1637    9.  Error texts for pcre_compile() are held as one long string to avoid too
1638        much relocation at load time. To find a text, the string is searched,
1639        counting zeros. There was no check for running off the end of the string,
1640        which could happen if a new error number was added without updating the
1641        string.
1643    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
1645    11. \K was not working if it appeared in an atomic group or in a group that
1646        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
1647        \K is "not well defined" if used in an assertion. PCRE now accepts it if
1648        the assertion is positive, but not if it is negative.
1650    12. Change 11 fortuitously reduced the size of the stack frame used in the
1651        "match()" function of pcre_exec.c by one pointer. Forthcoming
1652        implementation of support for (*MARK) will need an extra pointer on the
1653        stack; I have reserved it now, so that the stack frame size does not
1654        decrease.
1656    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
1657        item in branch that calls a recursion is a subroutine call - as in the
1658        second branch in the above example - was incorrectly given the compile-
1659        time error "recursive call could loop indefinitely" because pcre_compile()
1660        was not correctly checking the subroutine for matching a non-empty string.
1662    14. The checks for overrunning compiling workspace could trigger after an
1663        overrun had occurred. This is a "should never occur" error, but it can be
1664        triggered by pathological patterns such as hundreds of nested parentheses.
1665        The checks now trigger 100 bytes before the end of the workspace.
1667    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
1670    Version 8.01 19-Jan-2010
1671    ------------------------
1673    1.  If a pattern contained a conditional subpattern with only one branch (in
1674        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
1675        computed the wrong minimum data length (which is of course zero for such
1676        subpatterns). This could cause incorrect "no match" results.
1678    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
1679        the pattern is reset in the first branch, pcre_compile() failed with
1680        "internal error: code overflow at offset...". This happened only when
1681        the reset was to the original external option setting. (An optimization
1682        abstracts leading options settings into an external setting, which was the
1683        cause of this.)
1685    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
1686        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
1687        assertion pattern did not match (meaning that the assertion was true), it
1688        was incorrectly treated as false if the SKIP had been reached during the
1689        matching. This also applied to assertions used as conditions.
1691    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
1692        assertion subpattern, including such a pattern used as a condition,
1693        unpredictable results occurred, instead of the error return
1696    5.  The C++ GlobalReplace function was not working like Perl for the special
1697        situation when an empty string is matched. It now does the fancy magic
1698        stuff that is necessary.
1700    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
1701        removed. (These were left over from very, very early versions of PCRE.)
1703    7.  Some cosmetic changes to the code to make life easier when compiling it
1704        as part of something else:
1706        (a) Change DEBUG to PCRE_DEBUG.
1708        (b) In pcre_compile(), rename the member of the "branch_chain" structure
1709            called "current" as "current_branch", to prevent a collision with the
1710            Linux macro when compiled as a kernel module.
1712        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
1713            prevent a collision with the Linux macro when compiled as a kernel
1714            module.
1716    8.  In pcre_compile() there are some checks for integer overflows that used to
1717        cast potentially large values to (double). This has been changed to that
1718        when building, a check for int64_t is made, and if it is found, it is used
1719        instead, thus avoiding the use of floating point arithmetic. (There is no
1720        other use of FP in PCRE.) If int64_t is not found, the fallback is to
1721        double.
1723    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
1724        2005 (difference between two addresses compared to an unsigned value).
1726    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
1727        custom one, because of the following reported problem in Windows:
1729          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
1730              under Win32.
1731          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
1732              therefore missing the function definition.
1733          - The compiler thus generates a "C" signature for the test function.
1734          - The linker fails to find the "C" function.
1735          - PCRE fails to configure if asked to do so against libbz2.
1737    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
1738        messages were output:
1740          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
1741          rerunning libtoolize, to keep the correct libtool macros in-tree.
1742          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
1744        I have done both of these things.
1746    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
1747        most of the time, it *can* run out if it is given a pattern that contains a
1748        runaway infinite recursion. I updated the discussion in the pcrestack man
1749        page.
1751    13. Now that we have gone to the x.xx style of version numbers, the minor
1752        version may start with zero. Using 08 or 09 is a bad idea because users
1753        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
1754        interpreted as invalid octal numbers. I've updated the previous comment in
1755        configure.ac, and also added a check that gives an error if 08 or 09 are
1756        used.
1758    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
1759        causing partial matching to fail when the end of the subject matched \W
1760        in a UTF-8 pattern where \W was quantified with a minimum of 3.
1762    15. There were some discrepancies between the declarations in pcre_internal.h
1763        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
1764        their definitions. The declarations used "const uschar *" and the
1765        definitions used USPTR. Even though USPTR is normally defined as "const
1766        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
1767        reported that: "This difference in casting confuses some C++ compilers, for
1768        example, SunCC recognizes above declarations as different functions and
1769        generates broken code for hbpcre." I have changed the declarations to use
1770        USPTR.
1772    16. GNU libtool is named differently on some systems. The autogen.sh script now
1773        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
1774        (FreeBSD).
1776    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
1777        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
1778        comment: "Figure out how to create a longlong from a string: strtoll and
1779        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
1780        instance, but it only takes 2 args instead of 3!"
1782    18. A subtle bug concerned with back references has been fixed by a change of
1783        specification, with a corresponding code fix. A pattern such as
1784        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
1785        refers, was giving matches when it shouldn't. For example, xa=xaaa would
1786        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
1787        same bug. Such groups have to be quantified to be useful, or contained
1788        inside another quantified group. (If there's no repetition, the reference
1789        can never match.) The problem arises because, having left the group and
1790        moved on to the rest of the pattern, a later failure that backtracks into
1791        the group uses the captured value from the final iteration of the group
1792        rather than the correct earlier one. I have fixed this in PCRE by forcing
1793        any group that contains a reference to itself to be an atomic group; that
1794        is, there cannot be any backtracking into it once it has completed. This is
1795        similar to recursive and subroutine calls.
1798    Version 8.00 19-Oct-09
1799    ----------------------
1801    1.  The table for translating pcre_compile() error codes into POSIX error codes
1802        was out-of-date, and there was no check on the pcre_compile() error code
1803        being within the table. This could lead to an OK return being given in
1804        error.
1806    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
1807        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
1808        in a Windows environment.
1810    3.  The pcregrep --count option prints the count for each file even when it is
1811        zero, as does GNU grep. However, pcregrep was also printing all files when
1812        --files-with-matches was added. Now, when both options are given, it prints
1813        counts only for those files that have at least one match. (GNU grep just
1814        prints the file name in this circumstance, but including the count seems
1815        more useful - otherwise, why use --count?) Also ensured that the
1816        combination -clh just lists non-zero counts, with no names.
1818    4.  The long form of the pcregrep -F option was incorrectly implemented as
1819        --fixed_strings instead of --fixed-strings. This is an incompatible change,
1820        but it seems right to fix it, and I didn't think it was worth preserving
1821        the old behaviour.
1823    5.  The command line items --regex=pattern and --regexp=pattern were not
1824        recognized by pcregrep, which required --regex pattern or --regexp pattern
1825        (with a space rather than an '='). The man page documented the '=' forms,
1826        which are compatible with GNU grep; these now work.
1828    6.  No libpcreposix.pc file was created for pkg-config; there was just
1829        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
1831    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
1832        when UCP support is not needed, by modifying the Python script that
1833        generates it from Unicode data files. This should not matter if the module
1834        is correctly used as a library, but I received one complaint about 50K of
1835        unwanted data. My guess is that the person linked everything into his
1836        program rather than using a library. Anyway, it does no harm.
1838    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
1839        was a minimum greater than 1 for a wide character in a possessive
1840        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
1841        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
1842        character. Chaos in the form of incorrect output or a compiling loop could
1843        result.
1845    9.  The restrictions on what a pattern can contain when partial matching is
1846        requested for pcre_exec() have been removed. All patterns can now be
1847        partially matched by this function. In addition, if there are at least two
1848        slots in the offset vector, the offset of the earliest inspected character
1849        for the match and the offset of the end of the subject are set in them when
1850        PCRE_ERROR_PARTIAL is returned.
1852    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
1853        synonymous with PCRE_PARTIAL, for backwards compatibility, and
1854        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
1855        and may be more useful for multi-segment matching.
1857    11. Partial matching with pcre_exec() is now more intuitive. A partial match
1858        used to be given if ever the end of the subject was reached; now it is
1859        given only if matching could not proceed because another character was
1860        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
1861        string "Z", which now yields "no match" instead of "partial match". In the
1862        case of pcre_dfa_exec(), "no match" is given if every matching path for the
1863        final character ended with (*FAIL).
1865    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
1866        if the pattern had a "must contain" character that was already found in the
1867        earlier partial match, unless partial matching was again requested. For
1868        example, with the pattern /dog.(body)?/, the "must contain" character is
1869        "g". If the first part-match was for the string "dog", restarting with
1870        "sbody" failed. This bug has been fixed.
1872    13. The string returned by pcre_dfa_exec() after a partial match has been
1873        changed so that it starts at the first inspected character rather than the
1874        first character of the match. This makes a difference only if the pattern
1875        starts with a lookbehind assertion or \b or \B (\K is not supported by
1876        pcre_dfa_exec()). It's an incompatible change, but it makes the two
1877        matching functions compatible, and I think it's the right thing to do.
1879    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
1880        so that the demonstration program is easily available in environments where
1881        PCRE has not been installed from source.
1883    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
1884        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
1885        library.
1887    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
1888        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
1889        is not the first non-POSIX option to be added. Clearly some people find
1890        these options useful.
1892    17. If a caller to the POSIX matching function regexec() passes a non-zero
1893        value for nmatch with a NULL value for pmatch, the value of
1894        nmatch is forced to zero.
1896    18. RunGrepTest did not have a test for the availability of the -u option of
1897        the diff command, as RunTest does. It now checks in the same way as
1898        RunTest, and also checks for the -b option.
1900    19. If an odd number of negated classes containing just a single character
1901        interposed, within parentheses, between a forward reference to a named
1902        subpattern and the definition of the subpattern, compilation crashed with
1903        an internal error, complaining that it could not find the referenced
1904        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
1905        [The bug was that it was starting one character too far in when skipping
1906        over the character class, thus treating the ] as data rather than
1907        terminating the class. This meant it could skip too much.]
1909    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
1910        /g option in pcretest when the pattern contains \K, which makes it possible
1911        to have an empty string match not at the start, even when the pattern is
1912        anchored. Updated pcretest and pcredemo to use this option.
1914    21. If the maximum number of capturing subpatterns in a recursion was greater
1915        than the maximum at the outer level, the higher number was returned, but
1916        with unset values at the outer level. The correct (outer level) value is
1917        now given.
1919    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
1920        PCRE did not set those parentheses (unlike Perl). I have now found a way to
1921        make it do so. The string so far is captured, making this feature
1922        compatible with Perl.
1924    23. The tests have been re-organized, adding tests 11 and 12, to make it
1925        possible to check the Perl 5.10 features against Perl 5.10.
1927    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
1928        pattern matches a fixed length string. PCRE did not allow this; now it
1929        does. Neither allows recursion.
1931    25. I finally figured out how to implement a request to provide the minimum
1932        length of subject string that was needed in order to match a given pattern.
1933        (It was back references and recursion that I had previously got hung up
1934        on.) This code has now been added to pcre_study(); it finds a lower bound
1935        to the length of subject needed. It is not necessarily the greatest lower
1936        bound, but using it to avoid searching strings that are too short does give
1937        some useful speed-ups. The value is available to calling programs via
1938        pcre_fullinfo().
1940    26. While implementing 25, I discovered to my embarrassment that pcretest had
1941        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
1942        study optimizations had never been tested with that matching function.
1943        Oops. What is worse, even when it was passed study data, there was a bug in
1944        pcre_dfa_exec() that meant it never actually used it. Double oops. There
1945        were also very few tests of studied patterns with pcre_dfa_exec().
1947    27. If (?| is used to create subpatterns with duplicate numbers, they are now
1948        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
1949        on the other side of the coin, they are no longer allowed to have different
1950        names, because these cannot be distinguished in PCRE, and this has caused
1951        confusion. (This is a difference from Perl.)
1953    28. When duplicate subpattern names are present (necessarily with different
1954        numbers, as required by 27 above), and a test is made by name in a
1955        conditional pattern, either for a subpattern having been matched, or for
1956        recursion in such a pattern, all the associated numbered subpatterns are
1957        tested, and the overall condition is true if the condition is true for any
1958        one of them. This is the way Perl works, and is also more like the way
1959        testing by number works.
1962    Version 7.9 11-Apr-09
1963    ---------------------
1965    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
1966        (pcretest), all targets were linked against these libraries. This included
1967        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
1968        libraries. This caused unwanted dependencies to be created. This problem
1969        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
1970        pcretest is linked with readline.
1972    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
1973        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
1974        moved outside it again, because FALSE and TRUE are already defined in AIX,
1975        but BOOL is not.
1977    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
1978        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
1980    4.  The pcregrep documentation said spaces were inserted as well as colons (or
1981        hyphens) following file names and line numbers when outputting matching
1982        lines. This is not true; no spaces are inserted. I have also clarified the
1983        wording for the --colour (or --color) option.
1985    5.  In pcregrep, when --colour was used with -o, the list of matching strings
1986        was not coloured; this is different to GNU grep, so I have changed it to be
1987        the same.
1989    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
1990        each matching line was coloured. Now it goes on to look for further matches
1991        of any of the test patterns, which is the same behaviour as GNU grep.
1993    7.  A pattern that could match an empty string could cause pcregrep to loop; it
1994        doesn't make sense to accept an empty string match in pcregrep, so I have
1995        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
1996        seems to be how GNU grep behaves. [But see later change 40 for release
1997        8.33.]
1999    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
2000        start or after a newline", because the conditional assertion was not being
2001        correctly handled. The rule now is that both the assertion and what follows
2002        in the first alternative must satisfy the test.
2004    9.  If auto-callout was enabled in a pattern with a conditional group whose
2005        condition was an assertion, PCRE could crash during matching, both with
2006        pcre_exec() and pcre_dfa_exec().
2008    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
2009        used for matching.
2011    11. Unicode property support in character classes was not working for
2012        characters (bytes) greater than 127 when not in UTF-8 mode.
2014    12. Added the -M command line option to pcretest.
2016    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
2018    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
2020    16. Added comments and documentation about mis-use of no_arg in the C++
2021        wrapper.
2023    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
2024        from Martin Jerabek that uses macro names for all relevant character and
2025        string constants.
2027    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
2028        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
2029        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
2030        these, but not everybody uses configure.
2032    19. A conditional group that had only one branch was not being correctly
2033        recognized as an item that could match an empty string. This meant that an
2034        enclosing group might also not be so recognized, causing infinite looping
2035        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
2036        with the subject "ab", where knowledge that the repeated group can match
2037        nothing is needed in order to break the loop.
2039    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
2040        exec(), but without supplying a callout function, matching went wrong.
2042    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
2043        leak if the size of the offset vector was greater than 30. When the vector
2044        is smaller, the saved offsets during recursion go onto a local stack
2045        vector, but for larger vectors malloc() is used. It was failing to free
2046        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
2047        error, in fact).
2049    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
2050        heapframe that is used only when UTF-8 support is enabled. This caused no
2051        problem, but was untidy.
2053    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
2054        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
2055        included within another project.
2057    24. Steven Van Ingelgem's patches to add more options to the CMake support,
2058        slightly modified by me:
2060          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
2061              not building pcregrep.
2063          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
2064              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
2066    25. Forward references, both numeric and by name, in patterns that made use of
2067        duplicate group numbers, could behave incorrectly or give incorrect errors,
2068        because when scanning forward to find the reference group, PCRE was not
2069        taking into account the duplicate group numbers. A pattern such as
2070        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
2072    26. Changed a few more instances of "const unsigned char *" to USPTR, making
2073        the feature of a custom pointer more persuasive (as requested by a user).
2075    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
2076        pcretest.c, inside #ifndefs, because it seems they are sometimes already
2077        pre-defined.
2079    28. Added support for (*UTF8) at the start of a pattern.
2081    29. Arrange for flags added by the "release type" setting in CMake to be shown
2082        in the configuration summary.
2085    Version 7.8 05-Sep-08
2086    ---------------------
2088    1.  Replaced UCP searching code with optimized version as implemented for Ad
2089        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
2090        stage table and inline lookup instead of a function, giving speed ups of 2
2091        to 5 times on some simple patterns that I tested. Permission was given to
2092        distribute the MultiStage2.py script that generates the tables (it's not in
2093        the tarball, but is in the Subversion repository).
2095    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
2096        scripts.
2098    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
2099        a group with a zero qualifier. The result of the study could be incorrect,
2100        or the function might crash, depending on the pattern.
2102    4.  Caseless matching was not working for non-ASCII characters in back
2103        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
2104        It now works when Unicode Property Support is available.
2106    5.  In pcretest, an escape such as \x{de} in the data was always generating
2107        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
2108        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
2109        truncation.
2111    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
2113    7.  Added two (int) casts to pcregrep when printing the difference of two
2114        pointers, in case they are 64-bit values.
2116    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
2117        test 2 if it fails.
2119    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
2120        and a #define of that name to empty if it is not externally set. This is to
2121        allow users of MSVC to set it if necessary.
2123    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
2124        the convenience functions in the pcre_get.c source file.
2126    11. An option change at the start of a pattern that had top-level alternatives
2127        could cause overwriting and/or a crash. This command provoked a crash in
2128        some environments:
2130          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
2132        This potential security problem was recorded as CVE-2008-2371.
2134    12. For a pattern where the match had to start at the beginning or immediately
2135        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
2136        pcre_dfa_exec() could read past the end of the passed subject if there was
2137        no match. To help with detecting such bugs (e.g. with valgrind), I modified
2138        pcretest so that it places the subject at the end of its malloc-ed buffer.
2140    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
2141        exec() might read past the end of the data buffer in UTF-8 mode.
2143    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
2144        the data contained the byte 0x85 as part of a UTF-8 character within its
2145        first line. This applied both to normal and DFA matching.
2147    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
2148        /^[^d]*?$/8 failed to match "abc".
2150    16. Added a missing copyright notice to pcrecpp_internal.h.
2152    17. Make it more clear in the documentation that values returned from
2153        pcre_exec() in ovector are byte offsets, not character counts.
2155    18. Tidied a few places to stop certain compilers from issuing warnings.
2157    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
2158        supplied by Stefan Weber. I made a further small update for 7.8 because
2159        there is a change of source arrangements: the pcre_searchfuncs.c module is
2160        replaced by pcre_ucd.c.
2163    Version 7.7 07-May-08
2164    ---------------------
2166    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
2167        a string to a long long, pretend we don't even have a long long." This is
2168        done by checking for the strtoq, strtoll, and _strtoi64 functions.
2170    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
2171        pre-7.6 versions, which defined a global no_arg variable instead of putting
2172        it in the RE class. (See also #8 below.)
2174    3.  Remove a line of dead code, identified by coverity and reported by Nuno
2175        Lopes.
2177    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
2179        (1) The include/exclude patterns were being applied to the whole pathnames
2180            of files, instead of just to the final components.
2182        (2) If there was more than one level of directory, the subdirectories were
2183            skipped unless they satisfied the include/exclude conditions. This is
2184            inconsistent with GNU grep (and could even be seen as contrary to the
2185            pcregrep specification - which I improved to make it absolutely clear).
2186            The action now is always to scan all levels of directory, and just
2187            apply the include/exclude patterns to regular files.
2189    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
2190        --exclude_dir in the tests to avoid scanning .svn directories.
2192    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
2193        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
2194        doesn't support NULs in patterns.
2196    7.  Added some missing "const"s to declarations of static tables in
2197        pcre_compile.c and pcre_dfa_exec.c.
2199    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
2200        caused by fix #2  above. (Subsequently also a second patch to fix the
2201        first patch. And a third patch - this was a messy problem.)
2203    9.  Applied Craig's patch to remove the use of push_back().
2205    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
2206        matching function regexec().
2208    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
2209        which, however, unlike Perl's \g{...}, are subroutine calls, not back
2210        references. PCRE supports relative numbers with this syntax (I don't think
2211        Oniguruma does).
2213    12. Previously, a group with a zero repeat such as (...){0} was completely
2214        omitted from the compiled regex. However, this means that if the group
2215        was called as a subroutine from elsewhere in the pattern, things went wrong
2216        (an internal error was given). Such groups are now left in the compiled
2217        pattern, with a new opcode that causes them to be skipped at execution
2218        time.
2220    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
2221        to the way PCRE behaves:
2223        (a) A lone ] character is dis-allowed (Perl treats it as data).
2225        (b) A back reference to an unmatched subpattern matches an empty string
2226            (Perl fails the current match path).
2228        (c) A data ] in a character class must be notated as \] because if the
2229            first data character in a class is ], it defines an empty class. (In
2230            Perl it is not possible to have an empty class.) The empty class []
2231            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
2232            The negative empty class [^] matches any one character, independently
2233            of the DOTALL setting.
2235    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
2236        non-existent subpattern following a character class starting with ']' and
2237        containing () gave an internal compiling error instead of "reference to
2238        non-existent subpattern". Fortunately, when the pattern did exist, the
2239        compiled code was correct. (When scanning forwards to check for the
2240        existence of the subpattern, it was treating the data ']' as terminating
2241        the class, so got the count wrong. When actually compiling, the reference
2242        was subsequently set up correctly.)
2244    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
2245        it was being rejected as not supported by pcre_dfa_exec(), even though
2246        other assertions are supported. I have made pcre_dfa_exec() support
2247        (*FAIL).
2249    16. The implementation of 13c above involved the invention of a new opcode,
2250        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
2251        cannot be changed at match time, I realized I could make a small
2252        improvement to matching performance by compiling OP_ALLANY instead of
2253        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
2254        on the OP_ANY path.
2256    17. Compiling pcretest on Windows with readline support failed without the
2257        following two fixes: (1) Make the unistd.h include conditional on
2258        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
2260    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
2261        ncurses library to be included for pcretest when ReadLine support is
2262        requested, but also to allow for it to be overridden. This patch came from
2263        Daniel Bergström.
2265    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
2266        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
2267        any errors with the current Unicode tables. Thanks to Peter Kankowski for
2268        spotting this.
2271    Version 7.6 28-Jan-08
2272    ---------------------
2274    1.  A character class containing a very large number of characters with
2275        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
2276        overflow.
2278    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
2279        HAVE_LONG_LONG is not defined.
2281    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
2282        bring them up to date and include new features. This patch includes:
2284        - Fixed PH's badly added libz and libbz2 support.
2285        - Fixed a problem with static linking.
2286        - Added pcredemo. [But later removed - see 7 below.]
2287        - Fixed dftables problem and added an option.
2288        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
2289            HAVE_LONG_LONG.
2290        - Added readline support for pcretest.
2291        - Added an listing of the option settings after cmake has run.
2293    4.  A user submitted a patch to Makefile that makes it easy to create
2294        "pcre.dll" under mingw when using Configure/Make. I added stuff to
2295        Makefile.am that cause it to include this special target, without
2296        affecting anything else. Note that the same mingw target plus all
2297        the other distribution libraries and programs are now supported
2298        when configuring with CMake (see 6 below) instead of with
2299        Configure/Make.
2301    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
2302        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
2303        exported in the Windows port". It has not yet been confirmed that the patch
2304        solves the problem, but it does no harm.
2306    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
2307        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
2308        with CMake, and also correct the comment about stack recursion.
2310    7.  Remove the automatic building of pcredemo from the ./configure system and
2311        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
2312        of a program that users should build themselves after PCRE is installed, so
2313        building it automatically is not really right. What is more, it gave
2314        trouble in some build environments.
2316    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
2319    Version 7.5 10-Jan-08
2320    ---------------------
2322    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
2323        values in parens when parsing an RE using the C++ wrapper."
2325    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
2326        Characters greater than 255 were excluded from the class instead of being
2327        included.
2329    3.  The same bug as (2) above applied to negated POSIX classes such as
2330        [:^space:].
2332    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
2333        defined or documented. It seems to have been a typo for PCRE_STATIC, so
2334        I have changed it.
2336    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
2337        first named subpattern) and a construct such as (?&a) would reference the
2338        first named subpattern whose name started with "a" (in other words, the
2339        length check was missing). Both these problems are fixed. "Subpattern name
2340        expected" is now given for (?&) (a zero-length name), and this patch also
2341        makes it give the same error for \k'' (previously it complained that that
2342        was a reference to a non-existent subpattern).
2344    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
2345        this is right because (?- can be followed by option settings as well as by
2346        digits. I have, however, made the messages clearer.
2348    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
2349        than the number used in the conditional) now cause a compile-time error.
2350        This is actually not compatible with Perl, which accepts such patterns, but
2351        treats the conditional as always being FALSE (as PCRE used to), but it
2352        seems to me that giving a diagnostic is better.
2354    8.  Change "alphameric" to the more common word "alphanumeric" in comments
2355        and messages.
2357    9.  Fix two occurrences of "backslash" in comments that should have been
2358        "backspace".
2360    10. Remove two redundant lines of code that can never be obeyed (their function
2361        was moved elsewhere).
2363    11. The program that makes PCRE's Unicode character property table had a bug
2364        which caused it to generate incorrect table entries for sequences of
2365        characters that have the same character type, but are in different scripts.
2366        It amalgamated them into a single range, with the script of the first of
2367        them. In other words, some characters were in the wrong script. There were
2368        thirteen such cases, affecting characters in the following ranges:
2370          U+002b0 - U+002c1
2371          U+0060c - U+0060d
2372          U+0061e - U+00612
2373          U+0064b - U+0065e
2374          U+0074d - U+0076d
2375          U+01800 - U+01805
2376          U+01d00 - U+01d77
2377          U+01d9b - U+01dbf
2378          U+0200b - U+0200f
2379          U+030fc - U+030fe
2380          U+03260 - U+0327f
2381          U+0fb46 - U+0fbb1
2382          U+10450 - U+1049d
2384    12. The -o option (show only the matching part of a line) for pcregrep was not
2385        compatible with GNU grep in that, if there was more than one match in a
2386        line, it showed only the first of them. It now behaves in the same way as
2387        GNU grep.
2389    13. If the -o and -v options were combined for pcregrep, it printed a blank
2390        line for every non-matching line. GNU grep prints nothing, and pcregrep now
2391        does the same. The return code can be used to tell if there were any
2392        non-matching lines.
2394    14. Added --file-offsets and --line-offsets to pcregrep.
2396    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
2397        infinitely looping recursion. The bug was that positive lookaheads were not
2398        being skipped when checking for a possible empty match (negative lookaheads
2399        and both kinds of lookbehind were skipped).
2401    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
2402        inclusion of <windows.h> to before rather than after the definition of
2403        INVALID_FILE_ATTRIBUTES (patch from David Byron).
2405    17. Specifying a possessive quantifier with a specific limit for a Unicode
2406        character property caused pcre_compile() to compile bad code, which led at
2407        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
2408        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
2409        caused the error; without that there was no problem.
2411    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
2413    19. Added --enable-pcretest-libreadline.
2415    20. In pcrecpp.cc, the variable 'count' was incremented twice in
2416        RE::GlobalReplace(). As a result, the number of replacements returned was
2417        double what it should be. I removed one of the increments, but Craig sent a
2418        later patch that removed the other one (the right fix) and added unit tests
2419        that check the return values (which was not done before).
2421    21. Several CMake things:
2423        (1) Arranged that, when cmake is used on Unix, the libraries end up with
2424            the names libpcre and libpcreposix, not just pcre and pcreposix.
2426        (2) The above change means that pcretest and pcregrep are now correctly
2427            linked with the newly-built libraries, not previously installed ones.
2431    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
2432        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
2433        UTF-8 newline character). The key issue is that the pattern starts .*;
2434        this means that the match must be either at the beginning, or after a
2435        newline. The bug was in the code for advancing after a failed match and
2436        checking that the new position followed a newline. It was not taking
2437        account of UTF-8 characters correctly.
2439    23. PCRE was behaving differently from Perl in the way it recognized POSIX
2440        character classes. PCRE was not treating the sequence [:...:] as a
2441        character class unless the ... were all letters. Perl, however, seems to
2442        allow any characters between [: and :], though of course it rejects as
2443        unknown any "names" that contain non-letters, because all the known class
2444        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
2445        for example, whereas PCRE did not - it did not recognize a POSIX character
2446        class. This seemed a bit dangerous, so the code has been changed to be
2447        closer to Perl. The behaviour is not identical to Perl, because PCRE will
2448        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
2449        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
2450        Perl does, and where it didn't before.
2452    24. Rewrite so as to remove the single use of %n from pcregrep because in some
2453        Windows environments %n is disabled by default.
2456    Version 7.4 21-Sep-07
2457    ---------------------
2459    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
2460        means that a class such as [\s] counted as "explicit reference to CR or
2461        LF". That isn't really right - the whole point of the change was to try to
2462        help when there was an actual mention of one of the two characters. So now
2463        the change happens only if \r or \n (or a literal CR or LF) character is
2464        encountered.
2466    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
2467        of both had grown to the point where there were only 3 bits left.
2468        Fortunately, there was spare space in the data structure, and so I have
2469        moved the internal flags into a new 16-bit field to free up more option
2470        bits.
2472    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
2473        but did not set the internal JCHANGED flag - either of these is enough to
2474        control the way the "get" function works - but the PCRE_INFO_JCHANGED
2475        facility is supposed to tell if (?J) was ever used, so now (?J) at the
2476        start sets both bits.
2478    4.  Added options (at build time, compile time, exec time) to change \R from
2479        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
2481    5.  doc/pcresyntax.html was missing from the distribution.
2483    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
2484        compatibility, even though it is no longer used.
2486    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
2487        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
2488        windows.h file is present (where different names are used). [This was
2489        reversed later after testing - see 16 below.]
2491    8.  Changed all #include <config.h> to #include "config.h". There were also
2492        some further <pcre.h> cases that I changed to "pcre.h".
2494    9.  When pcregrep was used with the --colour option, it missed the line ending
2495        sequence off the lines that it output.
2497    10. It was pointed out to me that arrays of string pointers cause lots of
2498        relocations when a shared library is dynamically loaded. A technique of
2499        using a single long string with a table of offsets can drastically reduce
2500        these. I have refactored PCRE in four places to do this. The result is
2501        dramatic:
2503          Originally:                          290
2504          After changing UCP table:            187
2505          After changing error message table:   43
2506          After changing table of "verbs"       36
2507          After changing table of Posix names   22
2509        Thanks to the folks working on Gregex for glib for this insight.
2511    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
2512        unicode-properties was also set.
2514    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
2516    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
2517        checked only for CRLF.
2519    14. Added casts to pcretest.c to avoid compiler warnings.
2521    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
2523    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
2524        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
2525        entirely. This removes changes made in 7 above.
2527    17. The CMake files have been updated, and there is now more information about
2528        building with CMake in the NON-UNIX-USE document.
2531    Version 7.3 28-Aug-07
2532    ---------------------
2534     1. In the rejigging of the build system that eventually resulted in 7.1, the
2535        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
2536        brackets there is not right, since it causes compilers to look for an
2537        installed pcre.h, not the version that is in the source that is being
2538        compiled (which of course may be different). I have changed it back to:
2540          #include "pcre.h"
2542        I have a vague recollection that the change was concerned with compiling in
2543        different directories, but in the new build system, that is taken care of
2544        by the VPATH setting the Makefile.
2546     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
2547        when the subject happened to end in the byte 0x85 (e.g. if the last
2548        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
2549        characters but of course it shouldn't be taken as a newline when it is part
2550        of another character. The bug was that, for an unlimited repeat of . in
2551        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
2552        characters when looking for a newline.
2554     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
2556     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
2557        in debug output.
2559     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
2560        long printing in the pcrecpp unittest when running under MinGW.
2562     6. ESC_K was left out of the EBCDIC table.
2564     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
2565        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
2566        limit also applies to "virtual nesting" when a pattern is recursive, and in
2567        this case 1000 isn't so big. I have been able to remove this limit at the
2568        expense of backing off one optimization in certain circumstances. Normally,
2569        when pcre_exec() would call its internal match() function recursively and
2570        immediately return the result unconditionally, it uses a "tail recursion"
2571        feature to save stack. However, when a subpattern that can match an empty
2572        string has an unlimited repetition quantifier, it no longer makes this
2573        optimization. That gives it a stack frame in which to save the data for
2574        checking that an empty string has been matched. Previously this was taken
2575        from the 1000-entry workspace that had been reserved. So now there is no
2576        explicit limit, but more stack is used.
2578     8. Applied Daniel's patches to solve problems with the import/export magic
2579        syntax that is required for Windows, and which was going wrong for the
2580        pcreposix and pcrecpp parts of the library. These were overlooked when this
2581        problem was solved for the main library.
2583     9. There were some crude static tests to avoid integer overflow when computing
2584        the size of patterns that contain repeated groups with explicit upper
2585        limits. As the maximum quantifier is 65535, the maximum group length was
2586        set at 30,000 so that the product of these two numbers did not overflow a
2587        32-bit integer. However, it turns out that people want to use groups that
2588        are longer than 30,000 bytes (though not repeat them that many times).
2589        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
2590        made it possible to implement the integer overflow checks in a much more
2591        dynamic way, which I have now done. The artificial limitation on group
2592        length has been removed - we now have only the limit on the total length of
2593        the compiled pattern, which depends on the LINK_SIZE setting.
2595    10. Fixed a bug in the documentation for get/copy named substring when
2596        duplicate names are permitted. If none of the named substrings are set, the
2597        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
2598        empty string.
2600    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
2601        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
2602        because the ] is interpreted as the first data character and the
2603        terminating ] is not found. PCRE has been made compatible with Perl in this
2604        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
2605        cause memory overwriting.
2607    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
2608        string has been matched (to stop an infinite loop). It was not recognizing
2609        a conditional subpattern that could match an empty string if that
2610        subpattern was within another subpattern. For example, it looped when
2611        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
2612        condition was not nested. This bug has been fixed.
2614    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
2615        past the start of the subject in the presence of bytes with the top bit
2616        set, for example "\x8aBCD".
2618    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
2619        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
2621    14. Optimized (?!) to (*FAIL).
2623    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
2624        This restricts code points to be within the range 0 to 0x10FFFF, excluding
2625        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
2626        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
2627        does: it's just the validity check that is more restrictive.
2629    16. Inserted checks for integer overflows during escape sequence (backslash)
2630        processing, and also fixed erroneous offset values for syntax errors during
2631        backslash processing.
2633    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
2634        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
2636    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
2637        caused an overrun.
2639    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
2640        something other than just ASCII characters) inside a group that had an
2641        unlimited repeat caused a loop at compile time (while checking to see
2642        whether the group could match an empty string).
2644    20. Debugging a pattern containing \p or \P could cause a crash. For example,
2645        [\P{Any}] did so. (Error in the code for printing property names.)
2647    21. An orphan \E inside a character class could cause a crash.
2649    22. A repeated capturing bracket such as (A)? could cause a wild memory
2650        reference during compilation.
2652    23. There are several functions in pcre_compile() that scan along a compiled
2653        expression for various reasons (e.g. to see if it's fixed length for look
2654        behind). There were bugs in these functions when a repeated \p or \P was
2655        present in the pattern. These operators have additional parameters compared
2656        with \d, etc, and these were not being taken into account when moving along
2657        the compiled data. Specifically:
2659        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
2660            length.
2662        (b) An item such as \pL+ within a repeated group could cause crashes or
2663            loops.
2665        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
2666            "reference to non-existent subpattern" error.
2668        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
2670    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
2671        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
2673    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
2675    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
2676        character were causing crashes (broken optimization).
2678    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
2679        \p or \P) caused a compile-time loop.
2681    28. More problems have arisen in unanchored patterns when CRLF is a valid line
2682        break. For example, the unstudied pattern [\r\n]A does not match the string
2683        "\r\nA" because change 7.0/46 below moves the current point on by two
2684        characters after failing to match at the start. However, the pattern \nA
2685        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
2686        the same is true. There doesn't seem any very clean way out of this, but
2687        what I have chosen to do makes the common cases work: PCRE now takes note
2688        of whether there can be an explicit match for \r or \n anywhere in the
2689        pattern, and if so, 7.0/46 no longer applies. As part of this change,
2690        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
2691        pattern has explicit CR or LF references.
2693    29. Added (*CR) etc for changing newline setting at start of pattern.
2696    Version 7.2 19-Jun-07
2697  ---------------------  ---------------------
2699   1. If the fr_FR locale cannot be found for test 3, try the "french" locale,   1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
# Line 21  Version 7.2 01-May-07 Line 2713  Version 7.2 01-May-07
2713      stack recursion. This gives a massive performance boost under BSD, but just      stack recursion. This gives a massive performance boost under BSD, but just
2714      a small improvement under Linux. However, it saves one field in the frame      a small improvement under Linux. However, it saves one field in the frame
2715      in all cases.      in all cases.
2717   6. Added more features from the forthcoming Perl 5.10:   6. Added more features from the forthcoming Perl 5.10:
2719      (a) (?-n) (where n is a string of digits) is a relative subroutine or      (a) (?-n) (where n is a string of digits) is a relative subroutine or
2720          recursion call. It refers to the nth most recently opened parentheses.          recursion call. It refers to the nth most recently opened parentheses.
2722      (b) (?+n) is also a relative subroutine call; it refers to the nth next      (b) (?+n) is also a relative subroutine call; it refers to the nth next
2723          to be opened parentheses.          to be opened parentheses.
2725        (c) Conditions that refer to capturing parentheses can be specified
2726            relatively, for example, (?(-2)... or (?(+3)...
2728        (d) \K resets the start of the current match so that everything before
2729            is not part of it.
2731        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
2733        (f) \g{name} is another synonym - part of Perl 5.10's unification of
2734            reference syntax.
2736        (g) (?| introduces a group in which the numbering of parentheses in each
2737            alternative starts with the same number.
2739        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
2741     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
2744     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
2745        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
2746        for detecting groups that can match an empty string.
2748     9. A pattern with a very large number of alternatives (more than several
2749        hundred) was running out of internal workspace during the pre-compile
2750        phase, where pcre_compile() figures out how much memory will be needed. A
2751        bit of new cunning has reduced the workspace needed for groups with
2752        alternatives. The 1000-alternative test pattern now uses 12 bytes of
2753        workspace instead of running out of the 4096 that are available.
2755    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
2757    11. Applied patch from Google to remove an optimization that didn't quite work.
2758        The report of the bug said:
2760          pcrecpp::RE("a*").FullMatch("aaa") matches, while
2761          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
2762          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
2764    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
2765        it matched the wrong number of bytes.
2768  Version 7.1 24-Apr-07  Version 7.1 24-Apr-07

Removed from v.166  
changed lines
  Added in v.1462

  ViewVC Help
Powered by ViewVC 1.1.5