ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 45 by nigel, Sat Feb 24 21:39:25 2007 UTC revision 1378 by zherczeg, Mon Oct 14 06:55:28 2013 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
4    Version 8.34 xx-xxxx-201x
5    --------------------------
7    1.  Add pcre[16|32]_jit_free_unused_memory to forcibly free unused JIT
8        executable memory. Patch inspired by Carsten Klein.
10    2.  ./configure --enable-coverage defined SUPPORT_GCOV in config.h, although
11        this macro is never tested and has no effect, because the work to support
12        coverage involves only compiling and linking options and special targets in
13        the Makefile. The comment in config.h implied that defining the macro would
14        enable coverage support, which is totally false. There was also support for
15        setting this macro in the CMake files (my fault, I just copied it from
16        configure). SUPPORT_GCOV has now been removed.
18    3.  Make a small performance improvement in strlen16() and strlen32() in
19        pcretest.
21    4.  Change 36 for 8.33 left some unreachable statements in pcre_exec.c,
22        detected by the Solaris compiler (gcc doesn't seem to be able to diagnose
23        these cases). There was also one in pcretest.c.
25    5.  Cleaned up a "may be uninitialized" compiler warning in pcre_exec.c.
27    6.  In UTF mode, the code for checking whether a group could match an empty
28        string (which is used for indefinitely repeated groups to allow for
29        breaking an infinite loop) was broken when the group contained a repeated
30        negated single-character class with a character that occupied more than one
31        data item and had a minimum repetition of zero (for example, [^\x{100}]* in
32        UTF-8 mode). The effect was undefined: the group might or might not be
33        deemed as matching an empty string, or the program might have crashed.
35    7.  The code for checking whether a group could match an empty string was not
36        recognizing that \h, \H, \v, \V, and \R must match a character.
38    8.  Implemented PCRE_INFO_MATCH_EMPTY, which yields 1 if the pattern can match
39        an empty string. If it can, pcretest shows this in its information output.
41    9.  Fixed two related bugs that applied to Unicode extended grapheme clusters
42        that were repeated with a maximizing qualifier (e.g. \X* or \X{2,5}) when
43        matched by pcre_exec() without using JIT:
45        (a) If the rest of the pattern did not match after a maximal run of
46            grapheme clusters, the code for backing up to try with fewer of them
47            did not always back up over a full grapheme when characters that do not
48            have the modifier quality were involved, e.g. Hangul syllables.
50        (b) If the match point in a subject started with modifier character, and
51            there was no match, the code could incorrectly back up beyond the match
52            point, and potentially beyond the first character in the subject,
53            leading to a segfault or an incorrect match result.
55    10. A conditional group with an assertion condition could lead to PCRE
56        recording an incorrect first data item for a match if no other first data
57        item was recorded. For example, the pattern (?(?=ab)ab) recorded "a" as a
58        first data item, and therefore matched "ca" after "c" instead of at the
59        start.
61    11. Change 40 for 8.33 (allowing pcregrep to find empty strings) showed up a
62        bug that caused the command "echo a | ./pcregrep -M '|a'" to loop.
64    12. The source of pcregrep now includes z/OS-specific code so that it can be
65        compiled for z/OS as part of the special z/OS distribution.
67    13. Added the -T and -TM options to pcretest.
69    14. The code in pcre_compile.c for creating the table of named capturing groups
70        has been refactored. Instead of creating the table dynamically during the
71        actual compiling pass, the information is remembered during the pre-compile
72        pass (on the stack unless there are more than 20 named groups, in which
73        case malloc() is used) and the whole table is created before the actual
74        compile happens. This has simplified the code (it is now nearly 150 lines
75        shorter) and prepared the way for better handling of references to groups
76        with duplicate names.
78    15. A back reference to a named subpattern when there is more than one of the
79        same name now checks them in the order in which they appear in the pattern.
80        The first one that is set is used for the reference. Previously only the
81        first one was inspected. This change makes PCRE more compatible with Perl.
83    16. Unicode character properties were updated from Unicode 6.3.0.
85    17. The compile-time code for auto-possessification has been refactored, based
86        on a patch by Zoltan Herczeg. It now happens after instead of during
87        compilation. The code is cleaner, and more cases are handled. The option
88        PCRE_NO_AUTO_POSSESSIFY is added for testing purposes, and the -O and /O
89        options in pcretest are provided to set it.
91    18. The character VT has been added to the set of characters that match \s and
92        are generally treated as white space, following this same change in Perl
93        5.18. There is now no difference between "Perl space" and "POSIX space".
95    19. The code for checking named groups as conditions, either for being set or
96        for being recursed, has been refactored (this is related to 14 and 15
97        above). Processing unduplicated named groups should now be as fast at
98        numerical groups, and processing duplicated groups should be faster than
99        before.
101    20. Two patches to the CMake build system, by Alexander Barkov:
103          (1) Replace the "source" command by "." in CMakeLists.txt because
104              "source" is a bash-ism.
106          (2) Add missing HAVE_STDINT_H and HAVE_INTTYPES_H to config-cmake.h.in;
107              without these the CMake build does not work on Solaris.
109    21. Perl has changed its handling of \8 and \9. If there is no previously
110        encountered capturing group of those numbers, they are treated as the
111        literal characters 8 and 9 instead of a binary zero followed by the
112        literals. PCRE now does the same.
114    22. Following Perl, added \o{} to specify codepoints in octal, making it
115        possible to specify values greater than 0777 and also making them
116        unambiguous.
118    23. Perl now gives an error for missing closing braces after \x{... instead of
119        treating the string as literal. PCRE now does the same.
121    24. RunTest used to grumble if an inappropriate test was selected explicitly,
122        but just skip it when running all tests. This make it awkward to run ranges
123        of tests when one of them was inappropriate. Now it just skips any
124        inappropriate tests, as it always did when running all tests.
126    25. If PCRE_AUTO_CALLOUT and PCRE_UCP were set for a pattern that contained
127        character types such as \d or \w, too many callouts were inserted, and the
128        data that they returned was rubbish.
130    26. In UCP mode, \s was not matching two of the characters that Perl matches,
131        namely NEL (U+0085) and MONGOLIAN VOWEL SEPARATOR (U+180E), though they
132        were matched by \h. The code has now been refactored so that the lists of
133        the horizontal and vertical whitespace characters used for \h and \v (which
134        are defined only in one place) are now also used for \s.
136    27. Add JIT support for the 64 bit TileGX architecture.
137        Patch by Jiong Wang (Tilera Corporation).
140    Version 8.33 28-May-2013
141    ------------------------
143    1.  Added 'U' to some constants that are compared to unsigned integers, to
144        avoid compiler signed/unsigned warnings. Added (int) casts to unsigned
145        variables that are added to signed variables, to ensure the result is
146        signed and can be negated.
148    2.  Applied patch by Daniel Richard G for quashing MSVC warnings to the
149        CMake config files.
151    3.  Revise the creation of config.h.generic so that all boolean macros are
152        #undefined, whereas non-boolean macros are #ifndef/#endif-ed. This makes
153        overriding via -D on the command line possible.
155    4.  Changing the definition of the variable "op" in pcre_exec.c from pcre_uchar
156        to unsigned int is reported to make a quite noticeable speed difference in
157        a specific Windows environment. Testing on Linux did also appear to show
158        some benefit (and it is clearly not harmful). Also fixed the definition of
159        Xop which should be unsigned.
161    5.  Related to (4), changing the definition of the intermediate variable cc
162        in repeated character loops from pcre_uchar to pcre_uint32 also gave speed
163        improvements.
165    6.  Fix forward search in JIT when link size is 3 or greater. Also removed some
166        unnecessary spaces.
168    7.  Adjust autogen.sh and configure.ac to lose warnings given by automake 1.12
169        and later.
171    8.  Fix two buffer over read issues in 16 and 32 bit modes. Affects JIT only.
173    9.  Optimizing fast_forward_start_bits in JIT.
175    10. Adding support for callouts in JIT, and fixing some issues revealed
176        during this work. Namely:
178        (a) Unoptimized capturing brackets incorrectly reset on backtrack.
180        (b) Minimum length was not checked before the matching is started.
182    11. The value of capture_last that is passed to callouts was incorrect in some
183        cases when there was a capture on one path that was subsequently abandoned
184        after a backtrack. Also, the capture_last value is now reset after a
185        recursion, since all captures are also reset in this case.
187    12. The interpreter no longer returns the "too many substrings" error in the
188        case when an overflowing capture is in a branch that is subsequently
189        abandoned after a backtrack.
191    13. In the pathological case when an offset vector of size 2 is used, pcretest
192        now prints out the matched string after a yield of 0 or 1.
194    14. Inlining subpatterns in recursions, when certain conditions are fulfilled.
195        Only supported by the JIT compiler at the moment.
197    15. JIT compiler now supports 32 bit Macs thanks to Lawrence Velazquez.
199    16. Partial matches now set offsets[2] to the "bumpalong" value, that is, the
200        offset of the starting point of the matching process, provided the offsets
201        vector is large enough.
203    17. The \A escape now records a lookbehind value of 1, though its execution
204        does not actually inspect the previous character. This is to ensure that,
205        in partial multi-segment matching, at least one character from the old
206        segment is retained when a new segment is processed. Otherwise, if there
207        are no lookbehinds in the pattern, \A might match incorrectly at the start
208        of a new segment.
210    18. Added some #ifdef __VMS code into pcretest.c to help VMS implementations.
212    19. Redefined some pcre_uchar variables in pcre_exec.c as pcre_uint32; this
213        gives some modest performance improvement in 8-bit mode.
215    20. Added the PCRE-specific property \p{Xuc} for matching characters that can
216        be expressed in certain programming languages using Universal Character
217        Names.
219    21. Unicode validation has been updated in the light of Unicode Corrigendum #9,
220        which points out that "non characters" are not "characters that may not
221        appear in Unicode strings" but rather "characters that are reserved for
222        internal use and have only local meaning".
224    22. When a pattern was compiled with automatic callouts (PCRE_AUTO_CALLOUT) and
225        there was a conditional group that depended on an assertion, if the
226        assertion was false, the callout that immediately followed the alternation
227        in the condition was skipped when pcre_exec() was used for matching.
229    23. Allow an explicit callout to be inserted before an assertion that is the
230        condition for a conditional group, for compatibility with automatic
231        callouts, which always insert a callout at this point.
233    24. In 8.31, (*COMMIT) was confined to within a recursive subpattern. Perl also
234        confines (*SKIP) and (*PRUNE) in the same way, and this has now been done.
236    25. (*PRUNE) is now supported by the JIT compiler.
238    26. Fix infinite loop when /(?<=(*SKIP)ac)a/ is matched against aa.
240    27. Fix the case where there are two or more SKIPs with arguments that may be
241        ignored.
243    28. (*SKIP) is now supported by the JIT compiler.
245    29. (*THEN) is now supported by the JIT compiler.
247    30. Update RunTest with additional test selector options.
249    31. The way PCRE handles backtracking verbs has been changed in two ways.
251        (1) Previously, in something like (*COMMIT)(*SKIP), COMMIT would override
252        SKIP. Now, PCRE acts on whichever backtracking verb is reached first by
253        backtracking. In some cases this makes it more Perl-compatible, but Perl's
254        rather obscure rules do not always do the same thing.
256        (2) Previously, backtracking verbs were confined within assertions. This is
257        no longer the case for positive assertions, except for (*ACCEPT). Again,
258        this sometimes improves Perl compatibility, and sometimes does not.
260    32. A number of tests that were in test 2 because Perl did things differently
261        have been moved to test 1, because either Perl or PCRE has changed, and
262        these tests are now compatible.
264    32. Backtracking control verbs are now handled in the same way in JIT and
265        interpreter.
267    33. An opening parenthesis in a MARK/PRUNE/SKIP/THEN name in a pattern that
268        contained a forward subroutine reference caused a compile error.
270    34. Auto-detect and optimize limited repetitions in JIT.
272    35. Implement PCRE_NEVER_UTF to lock out the use of UTF, in particular,
273        blocking (*UTF) etc.
275    36. In the interpreter, maximizing pattern repetitions for characters and
276        character types now use tail recursion, which reduces stack usage.
278    37. The value of the max lookbehind was not correctly preserved if a compiled
279        and saved regex was reloaded on a host of different endianness.
281    38. Implemented (*LIMIT_MATCH) and (*LIMIT_RECURSION). As part of the extension
282        of the compiled pattern block, expand the flags field from 16 to 32 bits
283        because it was almost full.
285    39. Try madvise first before posix_madvise.
287    40. Change 7 for PCRE 7.9 made it impossible for pcregrep to find empty lines
288        with a pattern such as ^$. It has taken 4 years for anybody to notice! The
289        original change locked out all matches of empty strings. This has been
290        changed so that one match of an empty string per line is recognized.
291        Subsequent searches on the same line (for colouring or for --only-matching,
292        for example) do not recognize empty strings.
294    41. Applied a user patch to fix a number of spelling mistakes in comments.
296    42. Data lines longer than 65536 caused pcretest to crash.
298    43. Clarified the data type for length and startoffset arguments for pcre_exec
299        and pcre_dfa_exec in the function-specific man pages, where they were
300        explicitly stated to be in bytes, never having been updated. I also added
301        some clarification to the pcreapi man page.
303    44. A call to pcre_dfa_exec() with an output vector size less than 2 caused
304        a segmentation fault.
307    Version 8.32 30-November-2012
308    -----------------------------
310    1.  Improved JIT compiler optimizations for first character search and single
311        character iterators.
313    2.  Supporting IBM XL C compilers for PPC architectures in the JIT compiler.
314        Patch by Daniel Richard G.
316    3.  Single character iterator optimizations in the JIT compiler.
318    4.  Improved JIT compiler optimizations for character ranges.
320    5.  Rename the "leave" variable names to "quit" to improve WinCE compatibility.
321        Reported by Giuseppe D'Angelo.
323    6.  The PCRE_STARTLINE bit, indicating that a match can occur only at the start
324        of a line, was being set incorrectly in cases where .* appeared inside
325        atomic brackets at the start of a pattern, or where there was a subsequent
326        *PRUNE or *SKIP.
328    7.  Improved instruction cache flush for POWER/PowerPC.
329        Patch by Daniel Richard G.
331    8.  Fixed a number of issues in pcregrep, making it more compatible with GNU
332        grep:
334        (a) There is now no limit to the number of patterns to be matched.
336        (b) An error is given if a pattern is too long.
338        (c) Multiple uses of --exclude, --exclude-dir, --include, and --include-dir
339            are now supported.
341        (d) --exclude-from and --include-from (multiple use) have been added.
343        (e) Exclusions and inclusions now apply to all files and directories, not
344            just to those obtained from scanning a directory recursively.
346        (f) Multiple uses of -f and --file-list are now supported.
348        (g) In a Windows environment, the default for -d has been changed from
349            "read" (the GNU grep default) to "skip", because otherwise the presence
350            of a directory in the file list provokes an error.
352        (h) The documentation has been revised and clarified in places.
354    9.  Improve the matching speed of capturing brackets.
356    10. Changed the meaning of \X so that it now matches a Unicode extended
357        grapheme cluster.
359    11. Patch by Daniel Richard G to the autoconf files to add a macro for sorting
360        out POSIX threads when JIT support is configured.
362    12. Added support for PCRE_STUDY_EXTRA_NEEDED.
364    13. In the POSIX wrapper regcomp() function, setting re_nsub field in the preg
365        structure could go wrong in environments where size_t is not the same size
366        as int.
368    14. Applied user-supplied patch to pcrecpp.cc to allow PCRE_NO_UTF8_CHECK to be
369        set.
371    15. The EBCDIC support had decayed; later updates to the code had included
372        explicit references to (e.g.) \x0a instead of CHAR_LF. There has been a
373        general tidy up of EBCDIC-related issues, and the documentation was also
374        not quite right. There is now a test that can be run on ASCII systems to
375        check some of the EBCDIC-related things (but is it not a full test).
377    16. The new PCRE_STUDY_EXTRA_NEEDED option is now used by pcregrep, resulting
378        in a small tidy to the code.
380    17. Fix JIT tests when UTF is disabled and both 8 and 16 bit mode are enabled.
382    18. If the --only-matching (-o) option in pcregrep is specified multiple
383        times, each one causes appropriate output. For example, -o1 -o2 outputs the
384        substrings matched by the 1st and 2nd capturing parentheses. A separating
385        string can be specified by --om-separator (default empty).
387    19. Improving the first n character searches.
389    20. Turn case lists for horizontal and vertical white space into macros so that
390        they are defined only once.
392    21. This set of changes together give more compatible Unicode case-folding
393        behaviour for characters that have more than one other case when UCP
394        support is available.
396        (a) The Unicode property table now has offsets into a new table of sets of
397            three or more characters that are case-equivalent. The MultiStage2.py
398            script that generates these tables (the pcre_ucd.c file) now scans
399            CaseFolding.txt instead of UnicodeData.txt for character case
400            information.
402        (b) The code for adding characters or ranges of characters to a character
403            class has been abstracted into a generalized function that also handles
404            case-independence. In UTF-mode with UCP support, this uses the new data
405            to handle characters with more than one other case.
407        (c) A bug that is fixed as a result of (b) is that codepoints less than 256
408            whose other case is greater than 256 are now correctly matched
409            caselessly. Previously, the high codepoint matched the low one, but not
410            vice versa.
412        (d) The processing of \h, \H, \v, and \ in character classes now makes use
413            of the new class addition function, using character lists defined as
414            macros alongside the case definitions of 20 above.
416        (e) Caseless back references now work with characters that have more than
417            one other case.
419        (f) General caseless matching of characters with more than one other case
420            is supported.
422    22. Unicode character properties were updated from Unicode 6.2.0
424    23. Improved CMake support under Windows. Patch by Daniel Richard G.
426    24. Add support for 32-bit character strings, and UTF-32
428    25. Major JIT compiler update (code refactoring and bugfixing).
429        Experimental Sparc 32 support is added.
431    26. Applied a modified version of Daniel Richard G's patch to create
432        pcre.h.generic and config.h.generic by "make" instead of in the
433        PrepareRelease script.
435    27. Added a definition for CHAR_NULL (helpful for the z/OS port), and use it in
436        pcre_compile.c when checking for a zero character.
438    28. Introducing a native interface for JIT. Through this interface, the compiled
439        machine code can be directly executed. The purpose of this interface is to
440        provide fast pattern matching, so several sanity checks are not performed.
441        However, feature tests are still performed. The new interface provides
442        1.4x speedup compared to the old one.
444    29. If pcre_exec() or pcre_dfa_exec() was called with a negative value for
445        the subject string length, the error given was PCRE_ERROR_BADOFFSET, which
446        was confusing. There is now a new error PCRE_ERROR_BADLENGTH for this case.
448    30. In 8-bit UTF-8 mode, pcretest failed to give an error for data codepoints
449        greater than 0x7fffffff (which cannot be represented in UTF-8, even under
450        the "old" RFC 2279). Instead, it ended up passing a negative length to
451        pcre_exec().
453    31. Add support for GCC's visibility feature to hide internal functions.
455    32. Running "pcretest -C pcre8" or "pcretest -C pcre16" gave a spurious error
456        "unknown -C option" after outputting 0 or 1.
458    33. There is now support for generating a code coverage report for the test
459        suite in environments where gcc is the compiler and lcov is installed. This
460        is mainly for the benefit of the developers.
462    34. If PCRE is built with --enable-valgrind, certain memory regions are marked
463        unaddressable using valgrind annotations, allowing valgrind to detect
464        invalid memory accesses. This is mainly for the benefit of the developers.
466    25. (*UTF) can now be used to start a pattern in any of the three libraries.
468    26. Give configure error if --enable-cpp but no C++ compiler found.
471    Version 8.31 06-July-2012
472    -------------------------
474    1.  Fixing a wrong JIT test case and some compiler warnings.
476    2.  Removed a bashism from the RunTest script.
478    3.  Add a cast to pcre_exec.c to fix the warning "unary minus operator applied
479        to unsigned type, result still unsigned" that was given by an MS compiler
480        on encountering the code "-sizeof(xxx)".
482    4.  Partial matching support is added to the JIT compiler.
484    5.  Fixed several bugs concerned with partial matching of items that consist
485        of more than one character:
487        (a) /^(..)\1/ did not partially match "aba" because checking references was
488            done on an "all or nothing" basis. This also applied to repeated
489            references.
491        (b) \R did not give a hard partial match if \r was found at the end of the
492            subject.
494        (c) \X did not give a hard partial match after matching one or more
495            characters at the end of the subject.
497        (d) When newline was set to CRLF, a pattern such as /a$/ did not recognize
498            a partial match for the string "\r".
500        (e) When newline was set to CRLF, the metacharacter "." did not recognize
501            a partial match for a CR character at the end of the subject string.
503    6.  If JIT is requested using /S++ or -s++ (instead of just /S+ or -s+) when
504        running pcretest, the text "(JIT)" added to the output whenever JIT is
505        actually used to run the match.
507    7.  Individual JIT compile options can be set in pcretest by following -s+[+]
508        or /S+[+] with a digit between 1 and 7.
510    8.  OP_NOT now supports any UTF character not just single-byte ones.
512    9.  (*MARK) control verb is now supported by the JIT compiler.
514    10. The command "./RunTest list" lists the available tests without actually
515        running any of them. (Because I keep forgetting what they all are.)
519    12. Applied a (slightly modified) user-supplied patch that improves performance
520        when the heap is used for recursion (compiled with --disable-stack-for-
521        recursion). Instead of malloc and free for each heap frame each time a
522        logical recursion happens, frames are retained on a chain and re-used where
523        possible. This sometimes gives as much as 30% improvement.
525    13. As documented, (*COMMIT) is now confined to within a recursive subpattern
526        call.
528    14. As documented, (*COMMIT) is now confined to within a positive assertion.
530    15. It is now possible to link pcretest with libedit as an alternative to
531        libreadline.
533    16. (*COMMIT) control verb is now supported by the JIT compiler.
535    17. The Unicode data tables have been updated to Unicode 6.1.0.
537    18. Added --file-list option to pcregrep.
539    19. Added binary file support to pcregrep, including the -a, --binary-files,
540        -I, and --text options.
542    20. The madvise function is renamed for posix_madvise for QNX compatibility
543        reasons. Fixed by Giuseppe D'Angelo.
545    21. Fixed a bug for backward assertions with REVERSE 0 in the JIT compiler.
547    22. Changed the option for creating symbolic links for 16-bit man pages from
548        -s to -sf so that re-installing does not cause issues.
550    23. Support PCRE_NO_START_OPTIMIZE in JIT as (*MARK) support requires it.
552    24. Fixed a very old bug in pcretest that caused errors with restarted DFA
553        matches in certain environments (the workspace was not being correctly
554        retained). Also added to pcre_dfa_exec() a simple plausibility check on
555        some of the workspace data at the beginning of a restart.
557    25. \s*\R was auto-possessifying the \s* when it should not, whereas \S*\R
558        was not doing so when it should - probably a typo introduced by SVN 528
559        (change 8.10/14).
561    26. When PCRE_UCP was not set, \w+\x{c4} was incorrectly auto-possessifying the
562        \w+ when the character tables indicated that \x{c4} was a word character.
563        There were several related cases, all because the tests for doing a table
564        lookup were testing for characters less than 127 instead of 255.
566    27. If a pattern contains capturing parentheses that are not used in a match,
567        their slots in the ovector are set to -1. For those that are higher than
568        any matched groups, this happens at the end of processing. In the case when
569        there were back references that the ovector was too small to contain
570        (causing temporary malloc'd memory to be used during matching), and the
571        highest capturing number was not used, memory off the end of the ovector
572        was incorrectly being set to -1. (It was using the size of the temporary
573        memory instead of the true size.)
575    28. To catch bugs like 27 using valgrind, when pcretest is asked to specify an
576        ovector size, it uses memory at the end of the block that it has got.
578    29. Check for an overlong MARK name and give an error at compile time. The
579        limit is 255 for the 8-bit library and 65535 for the 16-bit library.
581    30. JIT compiler update.
583    31. JIT is now supported on jailbroken iOS devices. Thanks for Ruiger
584        Rill for the patch.
586    32. Put spaces around SLJIT_PRINT_D in the JIT compiler. Required by CXX11.
588    33. Variable renamings in the PCRE-JIT compiler. No functionality change.
590    34. Fixed typos in pcregrep: in two places there was SUPPORT_LIBZ2 instead of
591        SUPPORT_LIBBZ2. This caused a build problem when bzip2 but not gzip (zlib)
592        was enabled.
594    35. Improve JIT code generation for greedy plus quantifier.
596    36. When /((?:a?)*)*c/ or /((?>a?)*)*c/ was matched against "aac", it set group
597        1 to "aa" instead of to an empty string. The bug affected repeated groups
598        that could potentially match an empty string.
600    37. Optimizing single character iterators in JIT.
602    38. Wide characters specified with \uxxxx in JavaScript mode are now subject to
603        the same checks as \x{...} characters in non-JavaScript mode. Specifically,
604        codepoints that are too big for the mode are faulted, and in a UTF mode,
605        disallowed codepoints are also faulted.
607    39. If PCRE was compiled with UTF support, in three places in the DFA
608        matcher there was code that should only have been obeyed in UTF mode, but
609        was being obeyed unconditionally. In 8-bit mode this could cause incorrect
610        processing when bytes with values greater than 127 were present. In 16-bit
611        mode the bug would be provoked by values in the range 0xfc00 to 0xdc00. In
612        both cases the values are those that cannot be the first data item in a UTF
613        character. The three items that might have provoked this were recursions,
614        possessively repeated groups, and atomic groups.
616    40. Ensure that libpcre is explicitly listed in the link commands for pcretest
617        and pcregrep, because some OS require shared objects to be explicitly
618        passed to ld, causing the link step to fail if they are not.
620    41. There were two incorrect #ifdefs in pcre_study.c, meaning that, in 16-bit
621        mode, patterns that started with \h* or \R* might be incorrectly matched.
624    Version 8.30 04-February-2012
625    -----------------------------
627    1.  Renamed "isnumber" as "is_a_number" because in some Mac environments this
628        name is defined in ctype.h.
630    2.  Fixed a bug in fixed-length calculation for lookbehinds that would show up
631        only in quite long subpatterns.
633    3.  Removed the function pcre_info(), which has been obsolete and deprecated
634        since it was replaced by pcre_fullinfo() in February 2000.
636    4.  For a non-anchored pattern, if (*SKIP) was given with a name that did not
637        match a (*MARK), and the match failed at the start of the subject, a
638        reference to memory before the start of the subject could occur. This bug
639        was introduced by fix 17 of release 8.21.
641    5.  A reference to an unset group with zero minimum repetition was giving
642        totally wrong answers (in non-JavaScript-compatibility mode). For example,
643        /(another)?(\1?)test/ matched against "hello world test". This bug was
644        introduced in release 8.13.
646    6.  Add support for 16-bit character strings (a large amount of work involving
647        many changes and refactorings).
649    7.  RunGrepTest failed on msys because \r\n was replaced by whitespace when the
650        command "pattern=`printf 'xxx\r\njkl'`" was run. The pattern is now taken
651        from a file.
653    8.  Ovector size of 2 is also supported by JIT based pcre_exec (the ovector size
654        rounding is not applied in this particular case).
656    9.  The invalid Unicode surrogate codepoints U+D800 to U+DFFF are now rejected
657        if they appear, or are escaped, in patterns.
659    10. Get rid of a number of -Wunused-but-set-variable warnings.
661    11. The pattern /(?=(*:x))(q|)/ matches an empty string, and returns the mark
662        "x". The similar pattern /(?=(*:x))((*:y)q|)/ did not return a mark at all.
663        Oddly, Perl behaves the same way. PCRE has been fixed so that this pattern
664        also returns the mark "x". This bug applied to capturing parentheses,
665        non-capturing parentheses, and atomic parentheses. It also applied to some
666        assertions.
668    12. Stephen Kelly's patch to CMakeLists.txt allows it to parse the version
669        information out of configure.ac instead of relying on pcre.h.generic, which
670        is not stored in the repository.
672    13. Applied Dmitry V. Levin's patch for a more portable method for linking with
673        -lreadline.
675    14. ZH added PCRE_CONFIG_JITTARGET; added its output to pcretest -C.
677    15. Applied Graycode's patch to put the top-level frame on the stack rather
678        than the heap when not using the stack for recursion. This gives a
679        performance improvement in many cases when recursion is not deep.
681    16. Experimental code added to "pcretest -C" to output the stack frame size.
684    Version 8.21 12-Dec-2011
685    ------------------------
687    1.  Updating the JIT compiler.
689    2.  JIT compiler now supports OP_NCREF, OP_RREF and OP_NRREF. New test cases
690        are added as well.
692    3.  Fix cache-flush issue on PowerPC (It is still an experimental JIT port).
693        PCRE_EXTRA_TABLES is not suported by JIT, and should be checked before
694        calling _pcre_jit_exec. Some extra comments are added.
696    4.  (*MARK) settings inside atomic groups that do not contain any capturing
697        parentheses, for example, (?>a(*:m)), were not being passed out. This bug
698        was introduced by change 18 for 8.20.
700    5.  Supporting of \x, \U and \u in JavaScript compatibility mode based on the
701        ECMA-262 standard.
703    6.  Lookbehinds such as (?<=a{2}b) that contained a fixed repetition were
704        erroneously being rejected as "not fixed length" if PCRE_CASELESS was set.
705        This bug was probably introduced by change 9 of 8.13.
707    7.  While fixing 6 above, I noticed that a number of other items were being
708        incorrectly rejected as "not fixed length". This arose partly because newer
709        opcodes had not been added to the fixed-length checking code. I have (a)
710        corrected the bug and added tests for these items, and (b) arranged for an
711        error to occur if an unknown opcode is encountered while checking for fixed
712        length instead of just assuming "not fixed length". The items that were
713        rejected were: (*ACCEPT), (*COMMIT), (*FAIL), (*MARK), (*PRUNE), (*SKIP),
714        (*THEN), \h, \H, \v, \V, and single character negative classes with fixed
715        repetitions, e.g. [^a]{3}, with and without PCRE_CASELESS.
717    8.  A possessively repeated conditional subpattern such as (?(?=c)c|d)++ was
718        being incorrectly compiled and would have given unpredicatble results.
720    9.  A possessively repeated subpattern with minimum repeat count greater than
721        one behaved incorrectly. For example, (A){2,}+ behaved as if it was
722        (A)(A)++ which meant that, after a subsequent mismatch, backtracking into
723        the first (A) could occur when it should not.
725    10. Add a cast and remove a redundant test from the code.
727    11. JIT should use pcre_malloc/pcre_free for allocation.
729    12. Updated pcre-config so that it no longer shows -L/usr/lib, which seems
730        best practice nowadays, and helps with cross-compiling. (If the exec_prefix
731        is anything other than /usr, -L is still shown).
733    13. In non-UTF-8 mode, \C is now supported in lookbehinds and DFA matching.
735    14. Perl does not support \N without a following name in a [] class; PCRE now
736        also gives an error.
738    15. If a forward reference was repeated with an upper limit of around 2000,
739        it caused the error "internal error: overran compiling workspace". The
740        maximum number of forward references (including repeats) was limited by the
741        internal workspace, and dependent on the LINK_SIZE. The code has been
742        rewritten so that the workspace expands (via pcre_malloc) if necessary, and
743        the default depends on LINK_SIZE. There is a new upper limit (for safety)
744        of around 200,000 forward references. While doing this, I also speeded up
745        the filling in of repeated forward references.
747    16. A repeated forward reference in a pattern such as (a)(?2){2}(.) was
748        incorrectly expecting the subject to contain another "a" after the start.
750    17. When (*SKIP:name) is activated without a corresponding (*MARK:name) earlier
751        in the match, the SKIP should be ignored. This was not happening; instead
752        the SKIP was being treated as NOMATCH. For patterns such as
753        /A(*MARK:A)A+(*SKIP:B)Z|AAC/ this meant that the AAC branch was never
754        tested.
756    18. The behaviour of (*MARK), (*PRUNE), and (*THEN) has been reworked and is
757        now much more compatible with Perl, in particular in cases where the result
758        is a non-match for a non-anchored pattern. For example, if
759        /b(*:m)f|a(*:n)w/ is matched against "abc", the non-match returns the name
760        "m", where previously it did not return a name. A side effect of this
761        change is that for partial matches, the last encountered mark name is
762        returned, as for non matches. A number of tests that were previously not
763        Perl-compatible have been moved into the Perl-compatible test files. The
764        refactoring has had the pleasing side effect of removing one argument from
765        the match() function, thus reducing its stack requirements.
767    19. If the /S+ option was used in pcretest to study a pattern using JIT,
768        subsequent uses of /S (without +) incorrectly behaved like /S+.
770    21. Retrieve executable code size support for the JIT compiler and fixing
771        some warnings.
773    22. A caseless match of a UTF-8 character whose other case uses fewer bytes did
774        not work when the shorter character appeared right at the end of the
775        subject string.
777    23. Added some (int) casts to non-JIT modules to reduce warnings on 64-bit
778        systems.
780    24. Added PCRE_INFO_JITSIZE to pass on the value from (21) above, and also
781        output it when the /M option is used in pcretest.
783    25. The CheckMan script was not being included in the distribution. Also, added
784        an explicit "perl" to run Perl scripts from the PrepareRelease script
785        because this is reportedly needed in Windows.
787    26. If study data was being save in a file and studying had not found a set of
788        "starts with" bytes for the pattern, the data written to the file (though
789        never used) was taken from uninitialized memory and so caused valgrind to
790        complain.
792    27. Updated RunTest.bat as provided by Sheri Pierce.
794    28. Fixed a possible uninitialized memory bug in pcre_jit_compile.c.
796    29. Computation of memory usage for the table of capturing group names was
797        giving an unnecessarily large value.
800    Version 8.20 21-Oct-2011
801    ------------------------
803    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
804        a POSIX class. After further experiments with Perl, which convinced me that
805        Perl has bugs and confusions, a closing square bracket is no longer allowed
806        in a POSIX name. This bug also affected patterns with classes that started
807        with full stops.
809    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
810        captured substring, but while checking the failing first alternative,
811        substring 1 is temporarily captured. If the output vector supplied to
812        pcre_exec() was not big enough for this capture, the yield of the function
813        was still zero ("insufficient space for captured substrings"). This cannot
814        be totally fixed without adding another stack variable, which seems a lot
815        of expense for a edge case. However, I have improved the situation in cases
816        such as /(a)(b)x|abc/ matched against "abc", where the return code
817        indicates that fewer than the maximum number of slots in the ovector have
818        been set.
820    3.  Related to (2) above: when there are more back references in a pattern than
821        slots in the output vector, pcre_exec() uses temporary memory during
822        matching, and copies in the captures as far as possible afterwards. It was
823        using the entire output vector, but this conflicts with the specification
824        that only 2/3 is used for passing back captured substrings. Now it uses
825        only the first 2/3, for compatibility. This is, of course, another edge
826        case.
828    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
829        main code base, and can be used by building with --enable-jit. When this is
830        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
831        runtime --no-jit option is given.
833    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
834        ovector, the return from the function was zero, implying that there were
835        other matches that did not fit. The correct "exactly full" value is now
836        returned.
838    6.  If a subpattern that was called recursively or as a subroutine contained
839        (*PRUNE) or any other control that caused it to give a non-standard return,
840        invalid errors such as "Error -26 (nested recursion at the same subject
841        position)" or even infinite loops could occur.
843    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
844        computing the minimum length on reaching *ACCEPT, and so ended up with the
845        wrong value of 1 rather than 0. Further investigation indicates that
846        computing a minimum subject length in the presence of *ACCEPT is difficult
847        (think back references, subroutine calls), and so I have changed the code
848        so that no minimum is registered for a pattern that contains *ACCEPT.
850    8.  If (*THEN) was present in the first (true) branch of a conditional group,
851        it was not handled as intended. [But see 16 below.]
853    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
854        Sheri Pierce.
856    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
857        the first byte in a match must be "a".
859    11. Change 17 for 8.13 increased the recursion depth for patterns like
860        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
861        pattern contains any instances of (*THEN). If it does not, the old
862        optimizations are restored. It would be nice to do this on a per-group
863        basis, but at the moment that is not feasible.
865    12. In some environments, the output of pcretest -C is CRLF terminated. This
866        broke RunTest's code that checks for the link size. A single white space
867        character after the value is now allowed for.
869    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
870        For "fr", it uses the Windows-specific input and output files.
872    14. If (*THEN) appeared in a group that was called recursively or as a
873        subroutine, it did not work as intended. [But see next item.]
875    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
876        pattern fragments (but not containing any | characters). If A and B are
877        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
878        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
879        D. In other words, Perl considers parentheses that do not contain any |
880        characters to be part of a surrounding alternative, whereas PCRE was
881        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
882        differently. PCRE now behaves in the same way as Perl, except in the case
883        of subroutine/recursion calls such as (?1) which have in any case always
884        been different (but PCRE had them first :-).
886    16. Related to 15 above: Perl does not treat the | in a conditional group as
887        creating alternatives. Such a group is treated in the same way as an
888        ordinary group without any | characters when processing (*THEN). PCRE has
889        been changed to match Perl's behaviour.
891    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
892        RunGrepTest script failed.
894    18. Change 22 for version 13 caused atomic groups to use more stack. This is
895        inevitable for groups that contain captures, but it can lead to a lot of
896        stack use in large patterns. The old behaviour has been restored for atomic
897        groups that do not contain any capturing parentheses.
899    19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not
900        suppress the check for a minimum subject length at run time. (If it was
901        given to pcre_exec() or pcre_dfa_exec() it did work.)
903    20. Fixed an ASCII-dependent infelicity in pcretest that would have made it
904        fail to work when decoding hex characters in data strings in EBCDIC
905        environments.
907    21. It appears that in at least one Mac OS environment, the isxdigit() function
908        is implemented as a macro that evaluates to its argument more than once,
909        contravening the C 90 Standard (I haven't checked a later standard). There
910        was an instance in pcretest which caused it to go wrong when processing
911        \x{...} escapes in subject strings. The has been rewritten to avoid using
912        things like p++ in the argument of isxdigit().
915    Version 8.13 16-Aug-2011
916    ------------------------
918    1.  The Unicode data tables have been updated to Unicode 6.0.0.
920    2.  Two minor typos in pcre_internal.h have been fixed.
922    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
923        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
924        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
926    4.  There were a number of related bugs in the code for matching backrefences
927        caselessly in UTF-8 mode when codes for the characters concerned were
928        different numbers of bytes. For example, U+023A and U+2C65 are an upper
929        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
930        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
931        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
932        2-byte code at the end of the subject (it thought there wasn't enough data
933        left).
935    5.  Comprehensive information about what went wrong is now returned by
936        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
937        as the output vector has at least 2 elements. The offset of the start of
938        the failing character and a reason code are placed in the vector.
940    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
941        now returned is for the first byte of the failing character, instead of the
942        last byte inspected. This is an incompatible change, but I hope it is small
943        enough not to be a problem. It makes the returned offset consistent with
944        pcre_exec() and pcre_dfa_exec().
946    7.  pcretest now gives a text phrase as well as the error number when
947        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
948        failure, the offset and reason code are output.
950    8.  When \R was used with a maximizing quantifier it failed to skip backwards
951        over a \r\n pair if the subsequent match failed. Instead, it just skipped
952        back over a single character (\n). This seems wrong (because it treated the
953        two characters as a single entity when going forwards), conflicts with the
954        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
955        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
956        has been changed.
958    9.  Some internal refactoring has changed the processing so that the handling
959        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
960        time (the PCRE_DOTALL option was changed this way some time ago: version
961        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
962        which was always a bit of a fudge. It also means that there is one less
963        argument for the match() function, which reduces its stack requirements
964        slightly. This change also fixes an incompatibility with Perl: the pattern
965        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
967    10. More internal refactoring has drastically reduced the number of recursive
968        calls to match() for possessively repeated groups such as (abc)++ when
969        using pcre_exec().
971    11. While implementing 10, a number of bugs in the handling of groups were
972        discovered and fixed:
974        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
975        (a|)*(?1) gave a compile-time internal error.
976        ((a|)+)+  did not notice that the outer group could match an empty string.
977        (^a|^)+   was not marked as anchored.
978        (.*a|.*)+ was not marked as matching at start or after a newline.
980    12. Yet more internal refactoring has removed another argument from the match()
981        function. Special calls to this function are now indicated by setting a
982        value in a variable in the "match data" data block.
984    13. Be more explicit in pcre_study() instead of relying on "default" for
985        opcodes that mean there is no starting character; this means that when new
986        ones are added and accidentally left out of pcre_study(), testing should
987        pick them up.
989    14. The -s option of pcretest has been documented for ages as being an old
990        synonym of -m (show memory usage). I have changed it to mean "force study
991        for every regex", that is, assume /S for every regex. This is similar to -i
992        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
993        using it. It makes it easier to run collections of tests with and without
994        study enabled, and thereby test pcre_study() more easily. All the standard
995        tests are now run with and without -s (but some patterns can be marked as
996        "never study" - see 20 below).
998    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
999        restoration of the capturing data to the outer values was not happening
1000        correctly.
1002    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
1003        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
1004        pattern had matched an empty string, and so incorrectly returned a no
1005        match.
1007    17. There was optimizing code for the last branch of non-capturing parentheses,
1008        and also for the obeyed branch of a conditional subexpression, which used
1009        tail recursion to cut down on stack usage. Unfortunately, now that there is
1010        the possibility of (*THEN) occurring in these branches, tail recursion is
1011        no longer possible because the return has to be checked for (*THEN). These
1012        two optimizations have therefore been removed. [But see 8.20/11 above.]
1014    18. If a pattern containing \R was studied, it was assumed that \R always
1015        matched two bytes, thus causing the minimum subject length to be
1016        incorrectly computed because \R can also match just one byte.
1018    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
1019        was incorrectly computed.
1021    20. If /S is present twice on a test pattern in pcretest input, it now
1022        *disables* studying, thereby overriding the use of -s on the command line
1023        (see 14 above). This is necessary for one or two tests to keep the output
1024        identical in both cases.
1026    21. When (*ACCEPT) was used in an assertion that matched an empty string and
1027        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
1029    22. When an atomic group that contained a capturing parenthesis was
1030        successfully matched, but the branch in which it appeared failed, the
1031        capturing was not being forgotten if a higher numbered group was later
1032        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
1033        group 1 to "a", when in fact it should be unset. This applied to multi-
1034        branched capturing and non-capturing groups, repeated or not, and also to
1035        positive assertions (capturing in negative assertions does not happen
1036        in PCRE) and also to nested atomic groups.
1038    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
1039        subject after a captured substring, to make it easier to tell which of a
1040        number of identical substrings has been captured.
1042    24. The way atomic groups are processed by pcre_exec() has been changed so that
1043        if they are repeated, backtracking one repetition now resets captured
1044        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
1045        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
1046        "aaa". Previously, it would have been "a". As part of this code
1047        refactoring, the way recursive calls are handled has also been changed.
1049    25. If an assertion condition captured any substrings, they were not passed
1050        back unless some other capturing happened later. For example, if
1051        (?(?=(a))a) was matched against "a", no capturing was returned.
1053    26. When studying a pattern that contained subroutine calls or assertions,
1054        the code for finding the minimum length of a possible match was handling
1055        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
1056        group 1 called group 2 while simultaneously a separate group 2 called group
1057        1). A stack overflow occurred in this case. I have fixed this by limiting
1058        the recursion depth to 10.
1060    27. Updated RunTest.bat in the distribution to the version supplied by Tom
1061        Fortmann. This supports explicit test numbers on the command line, and has
1062        argument validation and error reporting.
1064    28. An instance of \X with an unlimited repeat could fail if at any point the
1065        first character it looked at was a mark character.
1067    29. Some minor code refactoring concerning Unicode properties and scripts
1068        should reduce the stack requirement of match() slightly.
1070    30. Added the '=' option to pcretest to check the setting of unused capturing
1071        slots at the end of the pattern, which are documented as being -1, but are
1072        not included in the return count.
1074    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
1075        compiled something random. Now it gives a compile-time error (as does
1076        Perl).
1078    32. A *MARK encountered during the processing of a positive assertion is now
1079        recorded and passed back (compatible with Perl).
1081    33. If --only-matching or --colour was set on a pcregrep call whose pattern
1082        had alternative anchored branches, the search for a second match in a line
1083        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
1084        matched the line "0102" twice. The same bug affected patterns that started
1085        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
1086        twice.
1088    34. Previously, PCRE did not allow quantification of assertions. However, Perl
1089        does, and because of capturing effects, quantifying parenthesized
1090        assertions may at times be useful. Quantifiers are now allowed for
1091        parenthesized assertions.
1093    35. A minor code tidy in pcre_compile() when checking options for \R usage.
1095    36. \g was being checked for fancy things in a character class, when it should
1096        just be a literal "g".
1098    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
1099        appearance of a nested POSIX class supersedes an apparent external class.
1100        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
1101        unescaped square brackets may also appear as part of class names. For
1102        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
1103        more like Perl. (But see 8.20/1 above.)
1105    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
1106        was because it thought it was \N{name}, which is not supported).
1108    39. Add minix to OS list not supporting the -S option in pcretest.
1110    40. PCRE tries to detect cases of infinite recursion at compile time, but it
1111        cannot analyze patterns in sufficient detail to catch mutual recursions
1112        such as ((?1))((?2)). There is now a runtime test that gives an error if a
1113        subgroup is called recursively as a subpattern for a second time at the
1114        same position in the subject string. In previous releases this might have
1115        been caught by the recursion limit, or it might have run out of stack.
1117    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
1118        happen only once. PCRE was, however incorrectly giving a compile time error
1119        "recursive call could loop indefinitely" because it cannot analyze the
1120        pattern in sufficient detail. The compile time test no longer happens when
1121        PCRE is compiling a conditional subpattern, but actual runaway loops are
1122        now caught at runtime (see 40 above).
1124    42. It seems that Perl allows any characters other than a closing parenthesis
1125        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
1126        has been changed to be the same.
1128    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
1129        as not to get warnings when autogen.sh is called. Also changed
1130        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
1132    44. To help people who use pcregrep to scan files containing exceedingly long
1133        lines, the following changes have been made:
1135        (a) The default value of the buffer size parameter has been increased from
1136            8K to 20K. (The actual buffer used is three times this size.)
1138        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
1139            PCRE is built.
1141        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
1142            to be set at run time.
1144        (d) Numerical values in pcregrep options can be followed by K or M, for
1145            example --buffer-size=50K.
1147        (e) If a line being scanned overflows pcregrep's buffer, an error is now
1148            given and the return code is set to 2.
1150    45. Add a pointer to the latest mark to the callout data block.
1152    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
1153        partial match of an empty string instead of no match. This was specific to
1154        the use of ".".
1156    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
1157        complete match instead of a partial match. This bug was dependent on both
1158        the PCRE_UTF8 and PCRE_DOTALL options being set.
1160    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
1161        starting byte set, because \b was not being ignored.
1164    Version 8.12 15-Jan-2011
1165    ------------------------
1167    1.  Fixed some typos in the markup of the man pages, and wrote a script that
1168        checks for such things as part of the documentation building process.
1170    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
1171        --match-limit and --recursion-limit options (added for 8.11). In
1172        particular, this made one of the standard tests fail. (The integer value
1173        went into the wrong half of a long int.)
1175    3.  If the --colour option was given to pcregrep with -v (invert match), it
1176        did strange things, either producing crazy output, or crashing. It should,
1177        of course, ignore a request for colour when reporting lines that do not
1178        match.
1180    4.  Another pcregrep bug caused similar problems if --colour was specified with
1181        -M (multiline) and the pattern match finished with a line ending.
1183    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
1184        matched in multiline mode, the following line was shown as part of the
1185        match. This seems wrong, so I have changed it.
1187    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
1188        the check for further matches in the same line (so they could be coloured)
1189        to overrun the end of the current line. If another match was found, it was
1190        incorrectly shown (and then shown again when found in the next line).
1192    7.  If pcregrep was compiled under Windows, there was a reference to the
1193        function pcregrep_exit() before it was defined. I am assuming this was
1194        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
1195        reported by a user. I've moved the definition above the reference.
1198    Version 8.11 10-Dec-2010
1199    ------------------------
1201    1.  (*THEN) was not working properly if there were untried alternatives prior
1202        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
1203        backtracked to try for "b" instead of moving to the next alternative branch
1204        at the same level (in this case, to look for "c"). The Perl documentation
1205        is clear that when (*THEN) is backtracked onto, it goes to the "next
1206        alternative in the innermost enclosing group".
1208    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
1209        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
1210        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
1211        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
1212        (*THEN).
1214    3.  If \s appeared in a character class, it removed the VT character from
1215        the class, even if it had been included by some previous item, for example
1216        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
1217        of \s, but is part of the POSIX "space" class.)
1219    4.  A partial match never returns an empty string (because you can always
1220        match an empty string at the end of the subject); however the checking for
1221        an empty string was starting at the "start of match" point. This has been
1222        changed to the "earliest inspected character" point, because the returned
1223        data for a partial match starts at this character. This means that, for
1224        example, /(?<=abc)def/ gives a partial match for the subject "abc"
1225        (previously it gave "no match").
1227    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
1228        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
1229        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
1230        has an implication that the given string is incomplete (because a partial
1231        match is preferred over a full match). For this reason, these items now
1232        give a partial match in this situation. [Aside: previously, the one case
1233        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
1234        match rather than a full match, which was wrong by the old rules, but is
1235        now correct.]
1237    6.  There was a bug in the handling of #-introduced comments, recognized when
1238        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
1239        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
1240        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
1241        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
1242        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
1243        places in pcre_compile().
1245    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
1246        comments when looking ahead for named forward references to subpatterns,
1247        the only newline sequence it recognized was NL. It now handles newlines
1248        according to the set newline convention.
1250    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
1251        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
1252        cater for a lack of strerror(). These oversights have been fixed.
1254    9.  Added --match-limit and --recursion-limit to pcregrep.
1256    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
1258    11. When the -o option was used, pcregrep was setting a return code of 1, even
1259        when matches were found, and --line-buffered was not being honoured.
1261    12. Added an optional parentheses number to the -o and --only-matching options
1262        of pcregrep.
1264    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
1265        can match an empty string. The code to do it in pcretest and pcredemo
1266        needed fixing:
1268        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
1269            only one byte after an empty string match just before CRLF (this case
1270            just got forgotten; "any" and "anycrlf" were OK).
1272        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
1273            mode when an empty string match preceded an ASCII character followed by
1274            a non-ASCII character. (The code for advancing by one character rather
1275            than one byte was nonsense.)
1277        (c) The pcredemo.c sample program did not have any code at all to handle
1278            the cases when CRLF is a valid newline sequence.
1280    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
1281        as a starting offset was within the subject string. There is now a new
1282        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
1283        negative or greater than the length of the string. In order to test this,
1284        pcretest is extended to allow the setting of negative starting offsets.
1286    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
1287        starting offset points to the beginning of a UTF-8 character was
1288        unnecessarily clumsy. I tidied it up.
1290    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
1291        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
1293    17. Nobody had reported that the --include_dir option, which was added in
1294        release 7.7 should have been called --include-dir (hyphen, not underscore)
1295        for compatibility with GNU grep. I have changed it to --include-dir, but
1296        left --include_dir as an undocumented synonym, and the same for
1297        --exclude-dir, though that is not available in GNU grep, at least as of
1298        release 2.5.4.
1300    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
1301        characters from a string of bytes) have been redefined so as not to use
1302        loops, in order to improve performance in some environments. At the same
1303        time, I abstracted some of the common code into auxiliary macros to save
1304        repetition (this should not affect the compiled code).
1306    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
1307        compile-time error is now given if \c is not followed by an ASCII
1308        character, that is, a byte less than 128. (In EBCDIC mode, the code is
1309        different, and any byte value is allowed.)
1311    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
1312        START_OPTIMIZE option, which is now allowed at compile time - but just
1313        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
1314        to pcregrep and other applications that have no direct access to PCRE
1315        options. The new /Y option in pcretest sets this option when calling
1316        pcre_compile().
1318    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
1319        back references. Groups containing recursive back references were forced to
1320        be atomic by that change, but in the case of named groups, the amount of
1321        memory required was incorrectly computed, leading to "Failed: internal
1322        error: code overflow". This has been fixed.
1324    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
1325        pcretest.c, to avoid build problems in some Borland environments.
1328    Version 8.10 25-Jun-2010
1329    ------------------------
1331    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
1332        THEN.
1334    2.  (*ACCEPT) was not working when inside an atomic group.
1336    3.  Inside a character class, \B is treated as a literal by default, but
1337        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
1338        causes the error). The code is unchanged, but I tidied the documentation.
1340    4.  Inside a character class, PCRE always treated \R and \X as literals,
1341        whereas Perl faults them if its -w option is set. I have changed PCRE so
1342        that it faults them when PCRE_EXTRA is set.
1344    5.  Added support for \N, which always matches any character other than
1345        newline. (It is the same as "." when PCRE_DOTALL is not set.)
1347    6.  When compiling pcregrep with newer versions of gcc which may have
1348        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
1349        declared with attribute warn_unused_result" were given. Just casting the
1350        result to (void) does not stop the warnings; a more elaborate fudge is
1351        needed. I've used a macro to implement this.
1353    7.  Minor change to pcretest.c to avoid a compiler warning.
1355    8.  Added four artifical Unicode properties to help with an option to make
1356        \s etc use properties (see next item). The new properties are: Xan
1357        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
1359    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
1360        use Unicode properties. (*UCP) at the start of a pattern can be used to set
1361        this option. Modified pcretest to add /W to test this facility. Added
1362        REG_UCP to make it available via the POSIX interface.
1364    10. Added --line-buffered to pcregrep.
1366    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
1367        studied, and the match started with a letter with a code point greater than
1368        127 whose first byte was different to the first byte of the other case of
1369        the letter, the other case of this starting letter was not recognized
1370        (#976).
1372    12. If a pattern that was studied started with a repeated Unicode property
1373        test, for example, \p{Nd}+, there was the theoretical possibility of
1374        setting up an incorrect bitmap of starting bytes, but fortunately it could
1375        not have actually happened in practice until change 8 above was made (it
1376        added property types that matched character-matching opcodes).
1378    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
1379        possible starting bytes for non-anchored patterns.
1381    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
1382        \R, and also a number of cases that involve Unicode properties, both
1383        explicit and implicit when PCRE_UCP is set.
1385    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
1386        input, it could crash or give wrong results if characters with values
1387        greater than 0xc0 were present in the subject string. (Detail: it assumed
1388        UTF-8 input when processing these items.)
1390    16. Added a lot of (int) casts to avoid compiler warnings in systems where
1391        size_t is 64-bit (#991).
1393    17. Added a check for running out of memory when PCRE is compiled with
1394        --disable-stack-for-recursion (#990).
1396    18. If the last data line in a file for pcretest does not have a newline on
1397        the end, a newline was missing in the output.
1399    19. The default pcre_chartables.c file recognizes only ASCII characters (values
1400        less than 128) in its various bitmaps. However, there is a facility for
1401        generating tables according to the current locale when PCRE is compiled. It
1402        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
1403        characters, are recognized by isspace() and therefore were getting set in
1404        these tables, and indeed these tables seem to approximate to ISO 8859. This
1405        caused a problem in UTF-8 mode when pcre_study() was used to create a list
1406        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
1407        which of course cannot start UTF-8 characters. I have changed the code so
1408        that only real ASCII characters (less than 128) and the correct starting
1409        bytes for UTF-8 encodings are set for characters greater than 127 when in
1410        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
1411        altogether.)
1413    20. Added the /T option to pcretest so as to be able to run tests with non-
1414        standard character tables, thus making it possible to include the tests
1415        used for 19 above in the standard set of tests.
1417    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
1418        reference to a subpattern the other side of a comment that contains an
1419        opening parenthesis caused either an internal compiling error, or a
1420        reference to the wrong subpattern.
1423    Version 8.02 19-Mar-2010
1424    ------------------------
1426    1.  The Unicode data tables have been updated to Unicode 5.2.0.
1428    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
1429        configured.
1431    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
1432        original author of that file, following a query about its status.
1434    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
1435        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
1437    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
1438        quantifier applied to a forward-referencing subroutine call, could compile
1439        incorrect code or give the error "internal error: previously-checked
1440        referenced subpattern not found".
1442    6.  Both MS Visual Studio and Symbian OS have problems with initializing
1443        variables to point to external functions. For these systems, therefore,
1444        pcre_malloc etc. are now initialized to local functions that call the
1445        relevant global functions.
1447    7.  There were two entries missing in the vectors called coptable and poptable
1448        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
1449        I've fixed the data, and added a kludgy way of testing at compile time that
1450        the lengths are correct (equal to the number of opcodes).
1452    8.  Following on from 7, I added a similar kludge to check the length of the
1453        eint vector in pcreposix.c.
1455    9.  Error texts for pcre_compile() are held as one long string to avoid too
1456        much relocation at load time. To find a text, the string is searched,
1457        counting zeros. There was no check for running off the end of the string,
1458        which could happen if a new error number was added without updating the
1459        string.
1461    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
1463    11. \K was not working if it appeared in an atomic group or in a group that
1464        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
1465        \K is "not well defined" if used in an assertion. PCRE now accepts it if
1466        the assertion is positive, but not if it is negative.
1468    12. Change 11 fortuitously reduced the size of the stack frame used in the
1469        "match()" function of pcre_exec.c by one pointer. Forthcoming
1470        implementation of support for (*MARK) will need an extra pointer on the
1471        stack; I have reserved it now, so that the stack frame size does not
1472        decrease.
1474    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
1475        item in branch that calls a recursion is a subroutine call - as in the
1476        second branch in the above example - was incorrectly given the compile-
1477        time error "recursive call could loop indefinitely" because pcre_compile()
1478        was not correctly checking the subroutine for matching a non-empty string.
1480    14. The checks for overrunning compiling workspace could trigger after an
1481        overrun had occurred. This is a "should never occur" error, but it can be
1482        triggered by pathological patterns such as hundreds of nested parentheses.
1483        The checks now trigger 100 bytes before the end of the workspace.
1485    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
1488    Version 8.01 19-Jan-2010
1489    ------------------------
1491    1.  If a pattern contained a conditional subpattern with only one branch (in
1492        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
1493        computed the wrong minimum data length (which is of course zero for such
1494        subpatterns). This could cause incorrect "no match" results.
1496    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
1497        the pattern is reset in the first branch, pcre_compile() failed with
1498        "internal error: code overflow at offset...". This happened only when
1499        the reset was to the original external option setting. (An optimization
1500        abstracts leading options settings into an external setting, which was the
1501        cause of this.)
1503    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
1504        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
1505        assertion pattern did not match (meaning that the assertion was true), it
1506        was incorrectly treated as false if the SKIP had been reached during the
1507        matching. This also applied to assertions used as conditions.
1509    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
1510        assertion subpattern, including such a pattern used as a condition,
1511        unpredictable results occurred, instead of the error return
1514    5.  The C++ GlobalReplace function was not working like Perl for the special
1515        situation when an empty string is matched. It now does the fancy magic
1516        stuff that is necessary.
1518    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
1519        removed. (These were left over from very, very early versions of PCRE.)
1521    7.  Some cosmetic changes to the code to make life easier when compiling it
1522        as part of something else:
1524        (a) Change DEBUG to PCRE_DEBUG.
1526        (b) In pcre_compile(), rename the member of the "branch_chain" structure
1527            called "current" as "current_branch", to prevent a collision with the
1528            Linux macro when compiled as a kernel module.
1530        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
1531            prevent a collision with the Linux macro when compiled as a kernel
1532            module.
1534    8.  In pcre_compile() there are some checks for integer overflows that used to
1535        cast potentially large values to (double). This has been changed to that
1536        when building, a check for int64_t is made, and if it is found, it is used
1537        instead, thus avoiding the use of floating point arithmetic. (There is no
1538        other use of FP in PCRE.) If int64_t is not found, the fallback is to
1539        double.
1541    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
1542        2005 (difference between two addresses compared to an unsigned value).
1544    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
1545        custom one, because of the following reported problem in Windows:
1547          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
1548              under Win32.
1549          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
1550              therefore missing the function definition.
1551          - The compiler thus generates a "C" signature for the test function.
1552          - The linker fails to find the "C" function.
1553          - PCRE fails to configure if asked to do so against libbz2.
1555    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
1556        messages were output:
1558          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
1559          rerunning libtoolize, to keep the correct libtool macros in-tree.
1560          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
1562        I have done both of these things.
1564    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
1565        most of the time, it *can* run out if it is given a pattern that contains a
1566        runaway infinite recursion. I updated the discussion in the pcrestack man
1567        page.
1569    13. Now that we have gone to the x.xx style of version numbers, the minor
1570        version may start with zero. Using 08 or 09 is a bad idea because users
1571        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
1572        interpreted as invalid octal numbers. I've updated the previous comment in
1573        configure.ac, and also added a check that gives an error if 08 or 09 are
1574        used.
1576    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
1577        causing partial matching to fail when the end of the subject matched \W
1578        in a UTF-8 pattern where \W was quantified with a minimum of 3.
1580    15. There were some discrepancies between the declarations in pcre_internal.h
1581        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
1582        their definitions. The declarations used "const uschar *" and the
1583        definitions used USPTR. Even though USPTR is normally defined as "const
1584        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
1585        reported that: "This difference in casting confuses some C++ compilers, for
1586        example, SunCC recognizes above declarations as different functions and
1587        generates broken code for hbpcre." I have changed the declarations to use
1588        USPTR.
1590    16. GNU libtool is named differently on some systems. The autogen.sh script now
1591        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
1592        (FreeBSD).
1594    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
1595        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
1596        comment: "Figure out how to create a longlong from a string: strtoll and
1597        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
1598        instance, but it only takes 2 args instead of 3!"
1600    18. A subtle bug concerned with back references has been fixed by a change of
1601        specification, with a corresponding code fix. A pattern such as
1602        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
1603        refers, was giving matches when it shouldn't. For example, xa=xaaa would
1604        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
1605        same bug. Such groups have to be quantified to be useful, or contained
1606        inside another quantified group. (If there's no repetition, the reference
1607        can never match.) The problem arises because, having left the group and
1608        moved on to the rest of the pattern, a later failure that backtracks into
1609        the group uses the captured value from the final iteration of the group
1610        rather than the correct earlier one. I have fixed this in PCRE by forcing
1611        any group that contains a reference to itself to be an atomic group; that
1612        is, there cannot be any backtracking into it once it has completed. This is
1613        similar to recursive and subroutine calls.
1616    Version 8.00 19-Oct-09
1617    ----------------------
1619    1.  The table for translating pcre_compile() error codes into POSIX error codes
1620        was out-of-date, and there was no check on the pcre_compile() error code
1621        being within the table. This could lead to an OK return being given in
1622        error.
1624    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
1625        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
1626        in a Windows environment.
1628    3.  The pcregrep --count option prints the count for each file even when it is
1629        zero, as does GNU grep. However, pcregrep was also printing all files when
1630        --files-with-matches was added. Now, when both options are given, it prints
1631        counts only for those files that have at least one match. (GNU grep just
1632        prints the file name in this circumstance, but including the count seems
1633        more useful - otherwise, why use --count?) Also ensured that the
1634        combination -clh just lists non-zero counts, with no names.
1636    4.  The long form of the pcregrep -F option was incorrectly implemented as
1637        --fixed_strings instead of --fixed-strings. This is an incompatible change,
1638        but it seems right to fix it, and I didn't think it was worth preserving
1639        the old behaviour.
1641    5.  The command line items --regex=pattern and --regexp=pattern were not
1642        recognized by pcregrep, which required --regex pattern or --regexp pattern
1643        (with a space rather than an '='). The man page documented the '=' forms,
1644        which are compatible with GNU grep; these now work.
1646    6.  No libpcreposix.pc file was created for pkg-config; there was just
1647        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
1649    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
1650        when UCP support is not needed, by modifying the Python script that
1651        generates it from Unicode data files. This should not matter if the module
1652        is correctly used as a library, but I received one complaint about 50K of
1653        unwanted data. My guess is that the person linked everything into his
1654        program rather than using a library. Anyway, it does no harm.
1656    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
1657        was a minimum greater than 1 for a wide character in a possessive
1658        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
1659        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
1660        character. Chaos in the form of incorrect output or a compiling loop could
1661        result.
1663    9.  The restrictions on what a pattern can contain when partial matching is
1664        requested for pcre_exec() have been removed. All patterns can now be
1665        partially matched by this function. In addition, if there are at least two
1666        slots in the offset vector, the offset of the earliest inspected character
1667        for the match and the offset of the end of the subject are set in them when
1668        PCRE_ERROR_PARTIAL is returned.
1670    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
1671        synonymous with PCRE_PARTIAL, for backwards compatibility, and
1672        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
1673        and may be more useful for multi-segment matching.
1675    11. Partial matching with pcre_exec() is now more intuitive. A partial match
1676        used to be given if ever the end of the subject was reached; now it is
1677        given only if matching could not proceed because another character was
1678        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
1679        string "Z", which now yields "no match" instead of "partial match". In the
1680        case of pcre_dfa_exec(), "no match" is given if every matching path for the
1681        final character ended with (*FAIL).
1683    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
1684        if the pattern had a "must contain" character that was already found in the
1685        earlier partial match, unless partial matching was again requested. For
1686        example, with the pattern /dog.(body)?/, the "must contain" character is
1687        "g". If the first part-match was for the string "dog", restarting with
1688        "sbody" failed. This bug has been fixed.
1690    13. The string returned by pcre_dfa_exec() after a partial match has been
1691        changed so that it starts at the first inspected character rather than the
1692        first character of the match. This makes a difference only if the pattern
1693        starts with a lookbehind assertion or \b or \B (\K is not supported by
1694        pcre_dfa_exec()). It's an incompatible change, but it makes the two
1695        matching functions compatible, and I think it's the right thing to do.
1697    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
1698        so that the demonstration program is easily available in environments where
1699        PCRE has not been installed from source.
1701    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
1702        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
1703        library.
1705    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
1706        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
1707        is not the first non-POSIX option to be added. Clearly some people find
1708        these options useful.
1710    17. If a caller to the POSIX matching function regexec() passes a non-zero
1711        value for nmatch with a NULL value for pmatch, the value of
1712        nmatch is forced to zero.
1714    18. RunGrepTest did not have a test for the availability of the -u option of
1715        the diff command, as RunTest does. It now checks in the same way as
1716        RunTest, and also checks for the -b option.
1718    19. If an odd number of negated classes containing just a single character
1719        interposed, within parentheses, between a forward reference to a named
1720        subpattern and the definition of the subpattern, compilation crashed with
1721        an internal error, complaining that it could not find the referenced
1722        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
1723        [The bug was that it was starting one character too far in when skipping
1724        over the character class, thus treating the ] as data rather than
1725        terminating the class. This meant it could skip too much.]
1727    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
1728        /g option in pcretest when the pattern contains \K, which makes it possible
1729        to have an empty string match not at the start, even when the pattern is
1730        anchored. Updated pcretest and pcredemo to use this option.
1732    21. If the maximum number of capturing subpatterns in a recursion was greater
1733        than the maximum at the outer level, the higher number was returned, but
1734        with unset values at the outer level. The correct (outer level) value is
1735        now given.
1737    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
1738        PCRE did not set those parentheses (unlike Perl). I have now found a way to
1739        make it do so. The string so far is captured, making this feature
1740        compatible with Perl.
1742    23. The tests have been re-organized, adding tests 11 and 12, to make it
1743        possible to check the Perl 5.10 features against Perl 5.10.
1745    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
1746        pattern matches a fixed length string. PCRE did not allow this; now it
1747        does. Neither allows recursion.
1749    25. I finally figured out how to implement a request to provide the minimum
1750        length of subject string that was needed in order to match a given pattern.
1751        (It was back references and recursion that I had previously got hung up
1752        on.) This code has now been added to pcre_study(); it finds a lower bound
1753        to the length of subject needed. It is not necessarily the greatest lower
1754        bound, but using it to avoid searching strings that are too short does give
1755        some useful speed-ups. The value is available to calling programs via
1756        pcre_fullinfo().
1758    26. While implementing 25, I discovered to my embarrassment that pcretest had
1759        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
1760        study optimizations had never been tested with that matching function.
1761        Oops. What is worse, even when it was passed study data, there was a bug in
1762        pcre_dfa_exec() that meant it never actually used it. Double oops. There
1763        were also very few tests of studied patterns with pcre_dfa_exec().
1765    27. If (?| is used to create subpatterns with duplicate numbers, they are now
1766        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
1767        on the other side of the coin, they are no longer allowed to have different
1768        names, because these cannot be distinguished in PCRE, and this has caused
1769        confusion. (This is a difference from Perl.)
1771    28. When duplicate subpattern names are present (necessarily with different
1772        numbers, as required by 27 above), and a test is made by name in a
1773        conditional pattern, either for a subpattern having been matched, or for
1774        recursion in such a pattern, all the associated numbered subpatterns are
1775        tested, and the overall condition is true if the condition is true for any
1776        one of them. This is the way Perl works, and is also more like the way
1777        testing by number works.
1780    Version 7.9 11-Apr-09
1781    ---------------------
1783    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
1784        (pcretest), all targets were linked against these libraries. This included
1785        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
1786        libraries. This caused unwanted dependencies to be created. This problem
1787        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
1788        pcretest is linked with readline.
1790    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
1791        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
1792        moved outside it again, because FALSE and TRUE are already defined in AIX,
1793        but BOOL is not.
1795    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
1796        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
1798    4.  The pcregrep documentation said spaces were inserted as well as colons (or
1799        hyphens) following file names and line numbers when outputting matching
1800        lines. This is not true; no spaces are inserted. I have also clarified the
1801        wording for the --colour (or --color) option.
1803    5.  In pcregrep, when --colour was used with -o, the list of matching strings
1804        was not coloured; this is different to GNU grep, so I have changed it to be
1805        the same.
1807    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
1808        each matching line was coloured. Now it goes on to look for further matches
1809        of any of the test patterns, which is the same behaviour as GNU grep.
1811    7.  A pattern that could match an empty string could cause pcregrep to loop; it
1812        doesn't make sense to accept an empty string match in pcregrep, so I have
1813        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
1814        seems to be how GNU grep behaves. [But see later change 40 for release
1815        8.33.]
1817    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
1818        start or after a newline", because the conditional assertion was not being
1819        correctly handled. The rule now is that both the assertion and what follows
1820        in the first alternative must satisfy the test.
1822    9.  If auto-callout was enabled in a pattern with a conditional group whose
1823        condition was an assertion, PCRE could crash during matching, both with
1824        pcre_exec() and pcre_dfa_exec().
1826    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
1827        used for matching.
1829    11. Unicode property support in character classes was not working for
1830        characters (bytes) greater than 127 when not in UTF-8 mode.
1832    12. Added the -M command line option to pcretest.
1834    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
1836    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
1838    16. Added comments and documentation about mis-use of no_arg in the C++
1839        wrapper.
1841    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
1842        from Martin Jerabek that uses macro names for all relevant character and
1843        string constants.
1845    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
1846        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
1847        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
1848        these, but not everybody uses configure.
1850    19. A conditional group that had only one branch was not being correctly
1851        recognized as an item that could match an empty string. This meant that an
1852        enclosing group might also not be so recognized, causing infinite looping
1853        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
1854        with the subject "ab", where knowledge that the repeated group can match
1855        nothing is needed in order to break the loop.
1857    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
1858        exec(), but without supplying a callout function, matching went wrong.
1860    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
1861        leak if the size of the offset vector was greater than 30. When the vector
1862        is smaller, the saved offsets during recursion go onto a local stack
1863        vector, but for larger vectors malloc() is used. It was failing to free
1864        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
1865        error, in fact).
1867    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
1868        heapframe that is used only when UTF-8 support is enabled. This caused no
1869        problem, but was untidy.
1871    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
1872        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
1873        included within another project.
1875    24. Steven Van Ingelgem's patches to add more options to the CMake support,
1876        slightly modified by me:
1878          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
1879              not building pcregrep.
1881          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
1882              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
1884    25. Forward references, both numeric and by name, in patterns that made use of
1885        duplicate group numbers, could behave incorrectly or give incorrect errors,
1886        because when scanning forward to find the reference group, PCRE was not
1887        taking into account the duplicate group numbers. A pattern such as
1888        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
1890    26. Changed a few more instances of "const unsigned char *" to USPTR, making
1891        the feature of a custom pointer more persuasive (as requested by a user).
1893    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
1894        pcretest.c, inside #ifndefs, because it seems they are sometimes already
1895        pre-defined.
1897    28. Added support for (*UTF8) at the start of a pattern.
1899    29. Arrange for flags added by the "release type" setting in CMake to be shown
1900        in the configuration summary.
1903    Version 7.8 05-Sep-08
1904    ---------------------
1906    1.  Replaced UCP searching code with optimized version as implemented for Ad
1907        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
1908        stage table and inline lookup instead of a function, giving speed ups of 2
1909        to 5 times on some simple patterns that I tested. Permission was given to
1910        distribute the MultiStage2.py script that generates the tables (it's not in
1911        the tarball, but is in the Subversion repository).
1913    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
1914        scripts.
1916    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
1917        a group with a zero qualifier. The result of the study could be incorrect,
1918        or the function might crash, depending on the pattern.
1920    4.  Caseless matching was not working for non-ASCII characters in back
1921        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
1922        It now works when Unicode Property Support is available.
1924    5.  In pcretest, an escape such as \x{de} in the data was always generating
1925        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
1926        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
1927        truncation.
1929    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
1931    7.  Added two (int) casts to pcregrep when printing the difference of two
1932        pointers, in case they are 64-bit values.
1934    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
1935        test 2 if it fails.
1937    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
1938        and a #define of that name to empty if it is not externally set. This is to
1939        allow users of MSVC to set it if necessary.
1941    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
1942        the convenience functions in the pcre_get.c source file.
1944    11. An option change at the start of a pattern that had top-level alternatives
1945        could cause overwriting and/or a crash. This command provoked a crash in
1946        some environments:
1948          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
1950        This potential security problem was recorded as CVE-2008-2371.
1952    12. For a pattern where the match had to start at the beginning or immediately
1953        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
1954        pcre_dfa_exec() could read past the end of the passed subject if there was
1955        no match. To help with detecting such bugs (e.g. with valgrind), I modified
1956        pcretest so that it places the subject at the end of its malloc-ed buffer.
1958    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
1959        exec() might read past the end of the data buffer in UTF-8 mode.
1961    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
1962        the data contained the byte 0x85 as part of a UTF-8 character within its
1963        first line. This applied both to normal and DFA matching.
1965    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
1966        /^[^d]*?$/8 failed to match "abc".
1968    16. Added a missing copyright notice to pcrecpp_internal.h.
1970    17. Make it more clear in the documentation that values returned from
1971        pcre_exec() in ovector are byte offsets, not character counts.
1973    18. Tidied a few places to stop certain compilers from issuing warnings.
1975    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
1976        supplied by Stefan Weber. I made a further small update for 7.8 because
1977        there is a change of source arrangements: the pcre_searchfuncs.c module is
1978        replaced by pcre_ucd.c.
1981    Version 7.7 07-May-08
1982    ---------------------
1984    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
1985        a string to a long long, pretend we don't even have a long long." This is
1986        done by checking for the strtoq, strtoll, and _strtoi64 functions.
1988    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
1989        pre-7.6 versions, which defined a global no_arg variable instead of putting
1990        it in the RE class. (See also #8 below.)
1992    3.  Remove a line of dead code, identified by coverity and reported by Nuno
1993        Lopes.
1995    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
1997        (1) The include/exclude patterns were being applied to the whole pathnames
1998            of files, instead of just to the final components.
2000        (2) If there was more than one level of directory, the subdirectories were
2001            skipped unless they satisfied the include/exclude conditions. This is
2002            inconsistent with GNU grep (and could even be seen as contrary to the
2003            pcregrep specification - which I improved to make it absolutely clear).
2004            The action now is always to scan all levels of directory, and just
2005            apply the include/exclude patterns to regular files.
2007    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
2008        --exclude_dir in the tests to avoid scanning .svn directories.
2010    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
2011        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
2012        doesn't support NULs in patterns.
2014    7.  Added some missing "const"s to declarations of static tables in
2015        pcre_compile.c and pcre_dfa_exec.c.
2017    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
2018        caused by fix #2  above. (Subsequently also a second patch to fix the
2019        first patch. And a third patch - this was a messy problem.)
2021    9.  Applied Craig's patch to remove the use of push_back().
2023    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
2024        matching function regexec().
2026    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
2027        which, however, unlike Perl's \g{...}, are subroutine calls, not back
2028        references. PCRE supports relative numbers with this syntax (I don't think
2029        Oniguruma does).
2031    12. Previously, a group with a zero repeat such as (...){0} was completely
2032        omitted from the compiled regex. However, this means that if the group
2033        was called as a subroutine from elsewhere in the pattern, things went wrong
2034        (an internal error was given). Such groups are now left in the compiled
2035        pattern, with a new opcode that causes them to be skipped at execution
2036        time.
2038    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
2039        to the way PCRE behaves:
2041        (a) A lone ] character is dis-allowed (Perl treats it as data).
2043        (b) A back reference to an unmatched subpattern matches an empty string
2044            (Perl fails the current match path).
2046        (c) A data ] in a character class must be notated as \] because if the
2047            first data character in a class is ], it defines an empty class. (In
2048            Perl it is not possible to have an empty class.) The empty class []
2049            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
2050            The negative empty class [^] matches any one character, independently
2051            of the DOTALL setting.
2053    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
2054        non-existent subpattern following a character class starting with ']' and
2055        containing () gave an internal compiling error instead of "reference to
2056        non-existent subpattern". Fortunately, when the pattern did exist, the
2057        compiled code was correct. (When scanning forwards to check for the
2058        existence of the subpattern, it was treating the data ']' as terminating
2059        the class, so got the count wrong. When actually compiling, the reference
2060        was subsequently set up correctly.)
2062    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
2063        it was being rejected as not supported by pcre_dfa_exec(), even though
2064        other assertions are supported. I have made pcre_dfa_exec() support
2065        (*FAIL).
2067    16. The implementation of 13c above involved the invention of a new opcode,
2068        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
2069        cannot be changed at match time, I realized I could make a small
2070        improvement to matching performance by compiling OP_ALLANY instead of
2071        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
2072        on the OP_ANY path.
2074    17. Compiling pcretest on Windows with readline support failed without the
2075        following two fixes: (1) Make the unistd.h include conditional on
2076        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
2078    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
2079        ncurses library to be included for pcretest when ReadLine support is
2080        requested, but also to allow for it to be overridden. This patch came from
2081        Daniel Bergström.
2083    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
2084        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
2085        any errors with the current Unicode tables. Thanks to Peter Kankowski for
2086        spotting this.
2089    Version 7.6 28-Jan-08
2090    ---------------------
2092    1.  A character class containing a very large number of characters with
2093        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
2094        overflow.
2096    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
2097        HAVE_LONG_LONG is not defined.
2099    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
2100        bring them up to date and include new features. This patch includes:
2102        - Fixed PH's badly added libz and libbz2 support.
2103        - Fixed a problem with static linking.
2104        - Added pcredemo. [But later removed - see 7 below.]
2105        - Fixed dftables problem and added an option.
2106        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
2107            HAVE_LONG_LONG.
2108        - Added readline support for pcretest.
2109        - Added an listing of the option settings after cmake has run.
2111    4.  A user submitted a patch to Makefile that makes it easy to create
2112        "pcre.dll" under mingw when using Configure/Make. I added stuff to
2113        Makefile.am that cause it to include this special target, without
2114        affecting anything else. Note that the same mingw target plus all
2115        the other distribution libraries and programs are now supported
2116        when configuring with CMake (see 6 below) instead of with
2117        Configure/Make.
2119    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
2120        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
2121        exported in the Windows port". It has not yet been confirmed that the patch
2122        solves the problem, but it does no harm.
2124    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
2125        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
2126        with CMake, and also correct the comment about stack recursion.
2128    7.  Remove the automatic building of pcredemo from the ./configure system and
2129        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
2130        of a program that users should build themselves after PCRE is installed, so
2131        building it automatically is not really right. What is more, it gave
2132        trouble in some build environments.
2134    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
2137    Version 7.5 10-Jan-08
2138    ---------------------
2140    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
2141        values in parens when parsing an RE using the C++ wrapper."
2143    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
2144        Characters greater than 255 were excluded from the class instead of being
2145        included.
2147    3.  The same bug as (2) above applied to negated POSIX classes such as
2148        [:^space:].
2150    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
2151        defined or documented. It seems to have been a typo for PCRE_STATIC, so
2152        I have changed it.
2154    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
2155        first named subpattern) and a construct such as (?&a) would reference the
2156        first named subpattern whose name started with "a" (in other words, the
2157        length check was missing). Both these problems are fixed. "Subpattern name
2158        expected" is now given for (?&) (a zero-length name), and this patch also
2159        makes it give the same error for \k'' (previously it complained that that
2160        was a reference to a non-existent subpattern).
2162    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
2163        this is right because (?- can be followed by option settings as well as by
2164        digits. I have, however, made the messages clearer.
2166    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
2167        than the number used in the conditional) now cause a compile-time error.
2168        This is actually not compatible with Perl, which accepts such patterns, but
2169        treats the conditional as always being FALSE (as PCRE used to), but it
2170        seems to me that giving a diagnostic is better.
2172    8.  Change "alphameric" to the more common word "alphanumeric" in comments
2173        and messages.
2175    9.  Fix two occurrences of "backslash" in comments that should have been
2176        "backspace".
2178    10. Remove two redundant lines of code that can never be obeyed (their function
2179        was moved elsewhere).
2181    11. The program that makes PCRE's Unicode character property table had a bug
2182        which caused it to generate incorrect table entries for sequences of
2183        characters that have the same character type, but are in different scripts.
2184        It amalgamated them into a single range, with the script of the first of
2185        them. In other words, some characters were in the wrong script. There were
2186        thirteen such cases, affecting characters in the following ranges:
2188          U+002b0 - U+002c1
2189          U+0060c - U+0060d
2190          U+0061e - U+00612
2191          U+0064b - U+0065e
2192          U+0074d - U+0076d
2193          U+01800 - U+01805
2194          U+01d00 - U+01d77
2195          U+01d9b - U+01dbf
2196          U+0200b - U+0200f
2197          U+030fc - U+030fe
2198          U+03260 - U+0327f
2199          U+0fb46 - U+0fbb1
2200          U+10450 - U+1049d
2202    12. The -o option (show only the matching part of a line) for pcregrep was not
2203        compatible with GNU grep in that, if there was more than one match in a
2204        line, it showed only the first of them. It now behaves in the same way as
2205        GNU grep.
2207    13. If the -o and -v options were combined for pcregrep, it printed a blank
2208        line for every non-matching line. GNU grep prints nothing, and pcregrep now
2209        does the same. The return code can be used to tell if there were any
2210        non-matching lines.
2212    14. Added --file-offsets and --line-offsets to pcregrep.
2214    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
2215        infinitely looping recursion. The bug was that positive lookaheads were not
2216        being skipped when checking for a possible empty match (negative lookaheads
2217        and both kinds of lookbehind were skipped).
2219    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
2220        inclusion of <windows.h> to before rather than after the definition of
2221        INVALID_FILE_ATTRIBUTES (patch from David Byron).
2223    17. Specifying a possessive quantifier with a specific limit for a Unicode
2224        character property caused pcre_compile() to compile bad code, which led at
2225        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
2226        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
2227        caused the error; without that there was no problem.
2229    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
2231    19. Added --enable-pcretest-libreadline.
2233    20. In pcrecpp.cc, the variable 'count' was incremented twice in
2234        RE::GlobalReplace(). As a result, the number of replacements returned was
2235        double what it should be. I removed one of the increments, but Craig sent a
2236        later patch that removed the other one (the right fix) and added unit tests
2237        that check the return values (which was not done before).
2239    21. Several CMake things:
2241        (1) Arranged that, when cmake is used on Unix, the libraries end up with
2242            the names libpcre and libpcreposix, not just pcre and pcreposix.
2244        (2) The above change means that pcretest and pcregrep are now correctly
2245            linked with the newly-built libraries, not previously installed ones.
2249    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
2250        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
2251        UTF-8 newline character). The key issue is that the pattern starts .*;
2252        this means that the match must be either at the beginning, or after a
2253        newline. The bug was in the code for advancing after a failed match and
2254        checking that the new position followed a newline. It was not taking
2255        account of UTF-8 characters correctly.
2257    23. PCRE was behaving differently from Perl in the way it recognized POSIX
2258        character classes. PCRE was not treating the sequence [:...:] as a
2259        character class unless the ... were all letters. Perl, however, seems to
2260        allow any characters between [: and :], though of course it rejects as
2261        unknown any "names" that contain non-letters, because all the known class
2262        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
2263        for example, whereas PCRE did not - it did not recognize a POSIX character
2264        class. This seemed a bit dangerous, so the code has been changed to be
2265        closer to Perl. The behaviour is not identical to Perl, because PCRE will
2266        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
2267        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
2268        Perl does, and where it didn't before.
2270    24. Rewrite so as to remove the single use of %n from pcregrep because in some
2271        Windows environments %n is disabled by default.
2274    Version 7.4 21-Sep-07
2275    ---------------------
2277    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
2278        means that a class such as [\s] counted as "explicit reference to CR or
2279        LF". That isn't really right - the whole point of the change was to try to
2280        help when there was an actual mention of one of the two characters. So now
2281        the change happens only if \r or \n (or a literal CR or LF) character is
2282        encountered.
2284    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
2285        of both had grown to the point where there were only 3 bits left.
2286        Fortunately, there was spare space in the data structure, and so I have
2287        moved the internal flags into a new 16-bit field to free up more option
2288        bits.
2290    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
2291        but did not set the internal JCHANGED flag - either of these is enough to
2292        control the way the "get" function works - but the PCRE_INFO_JCHANGED
2293        facility is supposed to tell if (?J) was ever used, so now (?J) at the
2294        start sets both bits.
2296    4.  Added options (at build time, compile time, exec time) to change \R from
2297        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
2299    5.  doc/pcresyntax.html was missing from the distribution.
2301    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
2302        compatibility, even though it is no longer used.
2304    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
2305        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
2306        windows.h file is present (where different names are used). [This was
2307        reversed later after testing - see 16 below.]
2309    8.  Changed all #include <config.h> to #include "config.h". There were also
2310        some further <pcre.h> cases that I changed to "pcre.h".
2312    9.  When pcregrep was used with the --colour option, it missed the line ending
2313        sequence off the lines that it output.
2315    10. It was pointed out to me that arrays of string pointers cause lots of
2316        relocations when a shared library is dynamically loaded. A technique of
2317        using a single long string with a table of offsets can drastically reduce
2318        these. I have refactored PCRE in four places to do this. The result is
2319        dramatic:
2321          Originally:                          290
2322          After changing UCP table:            187
2323          After changing error message table:   43
2324          After changing table of "verbs"       36
2325          After changing table of Posix names   22
2327        Thanks to the folks working on Gregex for glib for this insight.
2329    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
2330        unicode-properties was also set.
2332    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
2334    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
2335        checked only for CRLF.
2337    14. Added casts to pcretest.c to avoid compiler warnings.
2339    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
2341    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
2342        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
2343        entirely. This removes changes made in 7 above.
2345    17. The CMake files have been updated, and there is now more information about
2346        building with CMake in the NON-UNIX-USE document.
2349    Version 7.3 28-Aug-07
2350    ---------------------
2352     1. In the rejigging of the build system that eventually resulted in 7.1, the
2353        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
2354        brackets there is not right, since it causes compilers to look for an
2355        installed pcre.h, not the version that is in the source that is being
2356        compiled (which of course may be different). I have changed it back to:
2358          #include "pcre.h"
2360        I have a vague recollection that the change was concerned with compiling in
2361        different directories, but in the new build system, that is taken care of
2362        by the VPATH setting the Makefile.
2364     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
2365        when the subject happened to end in the byte 0x85 (e.g. if the last
2366        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
2367        characters but of course it shouldn't be taken as a newline when it is part
2368        of another character. The bug was that, for an unlimited repeat of . in
2369        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
2370        characters when looking for a newline.
2372     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
2374     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
2375        in debug output.
2377     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
2378        long printing in the pcrecpp unittest when running under MinGW.
2380     6. ESC_K was left out of the EBCDIC table.
2382     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
2383        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
2384        limit also applies to "virtual nesting" when a pattern is recursive, and in
2385        this case 1000 isn't so big. I have been able to remove this limit at the
2386        expense of backing off one optimization in certain circumstances. Normally,
2387        when pcre_exec() would call its internal match() function recursively and
2388        immediately return the result unconditionally, it uses a "tail recursion"
2389        feature to save stack. However, when a subpattern that can match an empty
2390        string has an unlimited repetition quantifier, it no longer makes this
2391        optimization. That gives it a stack frame in which to save the data for
2392        checking that an empty string has been matched. Previously this was taken
2393        from the 1000-entry workspace that had been reserved. So now there is no
2394        explicit limit, but more stack is used.
2396     8. Applied Daniel's patches to solve problems with the import/export magic
2397        syntax that is required for Windows, and which was going wrong for the
2398        pcreposix and pcrecpp parts of the library. These were overlooked when this
2399        problem was solved for the main library.
2401     9. There were some crude static tests to avoid integer overflow when computing
2402        the size of patterns that contain repeated groups with explicit upper
2403        limits. As the maximum quantifier is 65535, the maximum group length was
2404        set at 30,000 so that the product of these two numbers did not overflow a
2405        32-bit integer. However, it turns out that people want to use groups that
2406        are longer than 30,000 bytes (though not repeat them that many times).
2407        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
2408        made it possible to implement the integer overflow checks in a much more
2409        dynamic way, which I have now done. The artificial limitation on group
2410        length has been removed - we now have only the limit on the total length of
2411        the compiled pattern, which depends on the LINK_SIZE setting.
2413    10. Fixed a bug in the documentation for get/copy named substring when
2414        duplicate names are permitted. If none of the named substrings are set, the
2415        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
2416        empty string.
2418    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
2419        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
2420        because the ] is interpreted as the first data character and the
2421        terminating ] is not found. PCRE has been made compatible with Perl in this
2422        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
2423        cause memory overwriting.
2425    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
2426        string has been matched (to stop an infinite loop). It was not recognizing
2427        a conditional subpattern that could match an empty string if that
2428        subpattern was within another subpattern. For example, it looped when
2429        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
2430        condition was not nested. This bug has been fixed.
2432    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
2433        past the start of the subject in the presence of bytes with the top bit
2434        set, for example "\x8aBCD".
2436    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
2437        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
2439    14. Optimized (?!) to (*FAIL).
2441    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
2442        This restricts code points to be within the range 0 to 0x10FFFF, excluding
2443        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
2444        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
2445        does: it's just the validity check that is more restrictive.
2447    16. Inserted checks for integer overflows during escape sequence (backslash)
2448        processing, and also fixed erroneous offset values for syntax errors during
2449        backslash processing.
2451    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
2452        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
2454    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
2455        caused an overrun.
2457    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
2458        something other than just ASCII characters) inside a group that had an
2459        unlimited repeat caused a loop at compile time (while checking to see
2460        whether the group could match an empty string).
2462    20. Debugging a pattern containing \p or \P could cause a crash. For example,
2463        [\P{Any}] did so. (Error in the code for printing property names.)
2465    21. An orphan \E inside a character class could cause a crash.
2467    22. A repeated capturing bracket such as (A)? could cause a wild memory
2468        reference during compilation.
2470    23. There are several functions in pcre_compile() that scan along a compiled
2471        expression for various reasons (e.g. to see if it's fixed length for look
2472        behind). There were bugs in these functions when a repeated \p or \P was
2473        present in the pattern. These operators have additional parameters compared
2474        with \d, etc, and these were not being taken into account when moving along
2475        the compiled data. Specifically:
2477        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
2478            length.
2480        (b) An item such as \pL+ within a repeated group could cause crashes or
2481            loops.
2483        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
2484            "reference to non-existent subpattern" error.
2486        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
2488    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
2489        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
2491    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
2493    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
2494        character were causing crashes (broken optimization).
2496    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
2497        \p or \P) caused a compile-time loop.
2499    28. More problems have arisen in unanchored patterns when CRLF is a valid line
2500        break. For example, the unstudied pattern [\r\n]A does not match the string
2501        "\r\nA" because change 7.0/46 below moves the current point on by two
2502        characters after failing to match at the start. However, the pattern \nA
2503        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
2504        the same is true. There doesn't seem any very clean way out of this, but
2505        what I have chosen to do makes the common cases work: PCRE now takes note
2506        of whether there can be an explicit match for \r or \n anywhere in the
2507        pattern, and if so, 7.0/46 no longer applies. As part of this change,
2508        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
2509        pattern has explicit CR or LF references.
2511    29. Added (*CR) etc for changing newline setting at start of pattern.
2514    Version 7.2 19-Jun-07
2515    ---------------------
2517     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
2518        which is apparently normally available under Windows.
2520     2. Re-jig the pcregrep tests with different newline settings in an attempt
2521        to make them independent of the local environment's newline setting.
2523     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
2525     4. Some of the "internals" tests were previously cut out when the link size
2526        was not 2, because the output contained actual offsets. The recent new
2527        "Z" feature of pcretest means that these can be cut out, making the tests
2528        usable with all link sizes.
2530     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
2531        stack recursion. This gives a massive performance boost under BSD, but just
2532        a small improvement under Linux. However, it saves one field in the frame
2533        in all cases.
2535     6. Added more features from the forthcoming Perl 5.10:
2537        (a) (?-n) (where n is a string of digits) is a relative subroutine or
2538            recursion call. It refers to the nth most recently opened parentheses.
2540        (b) (?+n) is also a relative subroutine call; it refers to the nth next
2541            to be opened parentheses.
2543        (c) Conditions that refer to capturing parentheses can be specified
2544            relatively, for example, (?(-2)... or (?(+3)...
2546        (d) \K resets the start of the current match so that everything before
2547            is not part of it.
2549        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
2551        (f) \g{name} is another synonym - part of Perl 5.10's unification of
2552            reference syntax.
2554        (g) (?| introduces a group in which the numbering of parentheses in each
2555            alternative starts with the same number.
2557        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
2559     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
2562     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
2563        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
2564        for detecting groups that can match an empty string.
2566     9. A pattern with a very large number of alternatives (more than several
2567        hundred) was running out of internal workspace during the pre-compile
2568        phase, where pcre_compile() figures out how much memory will be needed. A
2569        bit of new cunning has reduced the workspace needed for groups with
2570        alternatives. The 1000-alternative test pattern now uses 12 bytes of
2571        workspace instead of running out of the 4096 that are available.
2573    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
2575    11. Applied patch from Google to remove an optimization that didn't quite work.
2576        The report of the bug said:
2578          pcrecpp::RE("a*").FullMatch("aaa") matches, while
2579          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
2580          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
2582    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
2583        it matched the wrong number of bytes.
2586    Version 7.1 24-Apr-07
2587    ---------------------
2589     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
2590        that is more "standard", making use of automake and other Autotools. There
2591        is some re-arrangement of the files and adjustment of comments consequent
2592        on this.
2594     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
2595        for recursive directory scanning broke on some systems because the files
2596        are not scanned in any specific order and on different systems the order
2597        was different. A call to "sort" has been inserted into RunGrepTest for the
2598        approprate test as a short-term fix. In the longer term there may be an
2599        alternative.
2601     3. I had an email from Eric Raymond about problems translating some of PCRE's
2602        man pages to HTML (despite the fact that I distribute HTML pages, some
2603        people do their own conversions for various reasons). The problems
2604        concerned the use of low-level troff macros .br and .in. I have therefore
2605        removed all such uses from the man pages (some were redundant, some could
2606        be replaced by .nf/.fi pairs). The 132html script that I use to generate
2607        HTML has been updated to handle .nf/.fi and to complain if it encounters
2608        .br or .in.
2610     4. Updated comments in configure.ac that get placed in config.h.in and also
2611        arranged for config.h to be included in the distribution, with the name
2612        config.h.generic, for the benefit of those who have to compile without
2613        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
2615     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
2616        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
2617        makevp.bat for the current PCRE, using the additional files
2618        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
2620     6. A Windows user reported a minor discrepancy with test 2, which turned out
2621        to be caused by a trailing space on an input line that had got lost in his
2622        copy. The trailing space was an accident, so I've just removed it.
2624     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
2625        that is needed.
2627     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
2628        as "const" (a) because they are and (b) because it helps the PHP
2629        maintainers who have recently made a script to detect big data structures
2630        in the php code that should be moved to the .rodata section. I remembered
2631        to update Builducptable as well, so it won't revert if ucptable.h is ever
2632        re-created.
2634     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
2635        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
2636        order to be able to cut out the UTF-8 tables in the latter when UTF-8
2637        support is not required. This saves 1.5-2K of code, which is important in
2638        some applications.
2640        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
2641        so as not to refer to the tables, even though these functions will never be
2642        called when UTF-8 support is disabled. Otherwise there are problems with a
2643        shared library.
2645    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
2647        (a) It was defining its arguments as char * instead of void *.
2649        (b) It was assuming that all moves were upwards in memory; this was true
2650            a long time ago when I wrote it, but is no longer the case.
2652        The emulated memove() is provided for those environments that have neither
2653        memmove() nor bcopy(). I didn't think anyone used it these days, but that
2654        is clearly not the case, as these two bugs were recently reported.
2656    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
2657        and Detrail to create the HTML documentation, the .txt form of the man
2658        pages, and it removes trailing spaces from listed files. It also creates
2659        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
2660        case, it wraps all the #defines with #ifndefs. This script should be run
2661        before "make dist".
2663    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
2664        with Unicode property support.
2666        (a) For a maximizing quantifier, if the two different cases of the
2667            character were of different lengths in their UTF-8 codings (there are
2668            some cases like this - I found 11), and the matching function had to
2669            back up over a mixture of the two cases, it incorrectly assumed they
2670            were both the same length.
2672        (b) When PCRE was configured to use the heap rather than the stack for
2673            recursion during matching, it was not correctly preserving the data for
2674            the other case of a UTF-8 character when checking ahead for a match
2675            while processing a minimizing repeat. If the check also involved
2676            matching a wide character, but failed, corruption could cause an
2677            erroneous result when trying to check for a repeat of the original
2678            character.
2680    13. Some tidying changes to the testing mechanism:
2682        (a) The RunTest script now detects the internal link size and whether there
2683            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
2684            values substituted by "configure". (The RunGrepTest script already did
2685            this for UTF-8.) The configure.ac script no longer substitutes the
2686            relevant variables.
2688        (b) The debugging options /B and /D in pcretest show the compiled bytecode
2689            with length and offset values. This means that the output is different
2690            for different internal link sizes. Test 2 is skipped for link sizes
2691            other than 2 because of this, bypassing the problem. Unfortunately,
2692            there was also a test in test 3 (the locale tests) that used /B and
2693            failed for link sizes other than 2. Rather than cut the whole test out,
2694            I have added a new /Z option to pcretest that replaces the length and
2695            offset values with spaces. This is now used to make test 3 independent
2696            of link size. (Test 2 will be tidied up later.)
2698    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
2699        segmentation fault instead of returning the appropriate error message.
2701    15. In multiline mode when the newline sequence was set to "any", the pattern
2702        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
2703        This doesn't seem right; it now treats the CRLF combination as the line
2704        ending, and so does not match in that case. It's only a pattern such as ^$
2705        that would hit this one: something like ^ABC$ would have failed after \r
2706        and then tried again after \r\n.
2708    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
2709        in an attempt to make files that differ only in their line terminators
2710        compare equal. This works on Linux.
2712    17. Under certain error circumstances pcregrep might try to free random memory
2713        as it exited. This is now fixed, thanks to valgrind.
2715    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
2716        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
2717        was because its rules for how to advance for /g after matching an empty
2718        string at the end of a line did not allow for this case. They now check for
2719        it specially.
2721    20. pcretest is supposed to handle patterns and data of any length, by
2722        extending its buffers when necessary. It was getting this wrong when the
2723        buffer for a data line had to be extended.
2725    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
2726        CRLF as a newline sequence.
2728    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
2729        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
2730        I have nevertheless tidied it up.
2732    23. Added some casts to kill warnings from HP-UX ia64 compiler.
2734    24. Added a man page for pcre-config.
2737    Version 7.0 19-Dec-06
2738    ---------------------
2740     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
2741        moving to gcc 4.1.1.
2743     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
2744        sys/time.h, which is documented as needed for this function. It doesn't
2745        seem to matter on Linux, but it showed up on some releases of OS X.
2747     3. It seems that there are systems where bytes whose values are greater than
2748        127 match isprint() in the "C" locale. The "C" locale should be the
2749        default when a C program starts up. In most systems, only ASCII printing
2750        characters match isprint(). This difference caused the output from pcretest
2751        to vary, making some of the tests fail. I have changed pcretest so that:
2753        (a) When it is outputting text in the compiled version of a pattern, bytes
2754            other than 32-126 are always shown as hex escapes.
2756        (b) When it is outputting text that is a matched part of a subject string,
2757            it does the same, unless a different locale has been set for the match
2758            (using the /L modifier). In this case, it uses isprint() to decide.
2760     4. Fixed a major bug that caused incorrect computation of the amount of memory
2761        required for a compiled pattern when options that changed within the
2762        pattern affected the logic of the preliminary scan that determines the
2763        length. The relevant options are -x, and -i in UTF-8 mode. The result was
2764        that the computed length was too small. The symptoms of this bug were
2765        either the PCRE error "internal error: code overflow" from pcre_compile(),
2766        or a glibc crash with a message such as "pcretest: free(): invalid next
2767        size (fast)". Examples of patterns that provoked this bug (shown in
2768        pcretest format) are:
2770          /(?-x: )/x
2771          /(?x)(?-x: \s*#\s*)/
2772          /((?i)[\x{c0}])/8
2773          /(?i:[\x{c0}])/8
2775        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
2776        is now done differently.
2778     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
2779        wrapper classes; (b) implement a new function in the C++ scanner that is
2780        more efficient than the old way of doing things because it avoids levels of
2781        recursion in the regex matching; (c) add a paragraph to the documentation
2782        for the FullMatch() function.
2784     6. The escape sequence \n was being treated as whatever was defined as
2785        "newline". Not only was this contrary to the documentation, which states
2786        that \n is character 10 (hex 0A), but it also went horribly wrong when
2787        "newline" was defined as CRLF. This has been fixed.
2789     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
2790        was being set to -1 for the "end of line" case (supposedly a value that no
2791        character can have). Though this value is never used (the check for end of
2792        line is "zero bytes in current character"), it caused compiler complaints.
2793        I've changed it to 0xffffffff.
2795     8. In pcre_version.c, the version string was being built by a sequence of
2796        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
2797        string (as it is for production releases) called a macro with an empty
2798        argument. The C standard says the result of this is undefined. The gcc
2799        compiler treats it as an empty string (which was what was wanted) but it is
2800        reported that Visual C gives an error. The source has been hacked around to
2801        avoid this problem.
2803     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
2804        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
2805        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
2806        of them did).
2808    10. Originally, pcretest opened its input and output without "b"; then I was
2809        told that "b" was needed in some environments, so it was added for release
2810        5.0 to both the input and output. (It makes no difference on Unix-like
2811        systems.) Later I was told that it is wrong for the input on Windows. I've
2812        now abstracted the modes into two macros, to make it easier to fiddle with
2813        them, and removed "b" from the input mode under Windows.
2815    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
2817    12. Added -help and --help to pcretest as an official way of being reminded
2818        of the options.
2820    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
2821        and pcrecpp.cc because they annoy compilers at high warning levels.
2823    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
2825    15. Fixed an occurrence of == in configure.ac that should have been = (shell
2826        scripts are not C programs :-) and which was not noticed because it works
2827        on Linux.
2829    16. pcretest is supposed to handle any length of pattern and data line (as one
2830        line or as a continued sequence of lines) by extending its input buffer if
2831        necessary. This feature was broken for very long pattern lines, leading to
2832        a string of junk being passed to pcre_compile() if the pattern was longer
2833        than about 50K.
2835    17. I have done a major re-factoring of the way pcre_compile() computes the
2836        amount of memory needed for a compiled pattern. Previously, there was code
2837        that made a preliminary scan of the pattern in order to do this. That was
2838        OK when PCRE was new, but as the facilities have expanded, it has become
2839        harder and harder to keep it in step with the real compile phase, and there
2840        have been a number of bugs (see for example, 4 above). I have now found a
2841        cunning way of running the real compile function in a "fake" mode that
2842        enables it to compute how much memory it would need, while actually only
2843        ever using a few hundred bytes of working memory and without too many
2844        tests of the mode. This should make future maintenance and development
2845        easier. A side effect of this work is that the limit of 200 on the nesting
2846        depth of parentheses has been removed (though this was never a serious
2847        limitation, I suspect). However, there is a downside: pcre_compile() now
2848        runs more slowly than before (30% or more, depending on the pattern). I
2849        hope this isn't a big issue. There is no effect on runtime performance.
2851    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
2852        newline (only possible for the last line of a file) and it was a
2853        pattern that set a locale (followed by /Lsomething), pcretest crashed.
2855    19. Added additional timing features to pcretest. (1) The -tm option now times
2856        matching only, not compiling. (2) Both -t and -tm can be followed, as a
2857        separate command line item, by a number that specifies the number of
2858        repeats to use when timing. The default is 50000; this gives better
2859        precision, but takes uncomfortably long for very large patterns.
2861    20. Extended pcre_study() to be more clever in cases where a branch of a
2862        subpattern has no definite first character. For example, (a*|b*)[cd] would
2863        previously give no result from pcre_study(). Now it recognizes that the
2864        first character must be a, b, c, or d.
2866    21. There was an incorrect error "recursive call could loop indefinitely" if
2867        a subpattern (or the entire pattern) that was being tested for matching an
2868        empty string contained only one non-empty item after a nested subpattern.
2869        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
2870        incorrectly, because the \d was being skipped in the check.
2872    22. The pcretest program now has a new pattern option /B and a command line
2873        option -b, which is equivalent to adding /B to every pattern. This causes
2874        it to show the compiled bytecode, without the additional information that
2875        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
2876        is the same as /B/I).
2878    23. A new optimization is now able automatically to treat some sequences such
2879        as a*b as a*+b. More specifically, if something simple (such as a character
2880        or a simple class like \d) has an unlimited quantifier, and is followed by
2881        something that cannot possibly match the quantified thing, the quantifier
2882        is automatically "possessified".
2884    24. A recursive reference to a subpattern whose number was greater than 39
2885        went wrong under certain circumstances in UTF-8 mode. This bug could also
2886        have affected the operation of pcre_study().
2888    25. Realized that a little bit of performance could be had by replacing
2889        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
2891    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
2893    27. Possessive quantifiers such as a++ were previously implemented by turning
2894        them into atomic groups such as ($>a+). Now they have their own opcodes,
2895        which improves performance. This includes the automatically created ones
2896        from 23 above.
2898    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
2899        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
2900        the first matched character to be a colon. This applied both to named and
2901        numbered groups.
2903    29. The ucpinternal.h header file was missing its idempotency #ifdef.
2905    30. I was sent a "project" file called libpcre.a.dev which I understand makes
2906        building PCRE on Windows easier, so I have included it in the distribution.
2908    31. There is now a check in pcretest against a ridiculously large number being
2909        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
2910        loop, the loop is abandoned.
2912    32. Forward references to subpatterns in conditions such as (?(2)...) where
2913        subpattern 2 is defined later cause pcre_compile() to search forwards in
2914        the pattern for the relevant set of parentheses. This search went wrong
2915        when there were unescaped parentheses in a character class, parentheses
2916        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
2918    33. "Subroutine" calls and backreferences were previously restricted to
2919        referencing subpatterns earlier in the regex. This restriction has now
2920        been removed.
2922    34. Added a number of extra features that are going to be in Perl 5.10. On the
2923        whole, these are just syntactic alternatives for features that PCRE had
2924        previously implemented using the Python syntax or my own invention. The
2925        other formats are all retained for compatibility.
2927        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
2928            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
2929            also .NET compatible.
2931        (b) A recursion or subroutine call to a named group can now be defined as
2932            (?&name) as well as (?P>name).
2934        (c) A backreference to a named group can now be defined as \k<name> or
2935            \k'name' as well as (?P=name). The new forms, as well as being in Perl
2936            5.10, are also .NET compatible.
2938        (d) A conditional reference to a named group can now use the syntax
2939            (?(<name>) or (?('name') as well as (?(name).
2941        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
2942            groups (named and numbered) that are never evaluated inline, but can be
2943            called as "subroutines" from elsewhere. In effect, the DEFINE condition
2944            is always false. There may be only one alternative in such a group.
2946        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
2947            as the simple (?(R). The condition is true only if the most recent
2948            recursion is that of the given number or name. It does not search out
2949            through the entire recursion stack.
2951        (g) The escape \gN or \g{N} has been added, where N is a positive or
2952            negative number, specifying an absolute or relative reference.
2954    35. Tidied to get rid of some further signed/unsigned compiler warnings and
2955        some "unreachable code" warnings.
2957    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
2958        things, this adds five new scripts.
2960    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
2961        There were also incompatibilities regarding the handling of \Q..\E inside
2962        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
2963        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
2965    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
2966        matches an empty string, and forcibly breaks the loop. There were bugs in
2967        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
2968        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
2969        separate and independent bugs (that affected different cases) have been
2970        fixed.
2972    39. Refactored the code to abolish the use of different opcodes for small
2973        capturing bracket numbers. This is a tidy that I avoided doing when I
2974        removed the limit on the number of capturing brackets for 3.5 back in 2001.
2975        The new approach is not only tidier, it makes it possible to reduce the
2976        memory needed to fix the previous bug (38).
2978    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
2979        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
2980        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
2981        mode.
2983    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
2984        report.
2986    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
2987        copy construction and assignment in the C++ wrapper.
2989    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
2990        couple of bugs that could have given wrong results in the "--newline=crlf"
2991        case.
2993    44. Added a number of casts and did some reorganization of signed/unsigned int
2994        variables following suggestions from Dair Grant. Also renamed the variable
2995        "this" as "item" because it is a C++ keyword.
2997    45. Arranged for dftables to add
2999          #include "pcre_internal.h"
3001        to pcre_chartables.c because without it, gcc 4.x may remove the array
3002        definition from the final binary if PCRE is built into a static library and
3003        dead code stripping is activated.
3005    46. For an unanchored pattern, if a match attempt fails at the start of a
3006        newline sequence, and the newline setting is CRLF or ANY, and the next two
3007        characters are CRLF, advance by two characters instead of one.
3010    Version 6.7 04-Jul-06
3011    ---------------------
3013     1. In order to handle tests when input lines are enormously long, pcretest has
3014        been re-factored so that it automatically extends its buffers when
3015        necessary. The code is crude, but this _is_ just a test program. The
3016        default size has been increased from 32K to 50K.
3018     2. The code in pcre_study() was using the value of the re argument before
3019        testing it for NULL. (Of course, in any sensible call of the function, it
3020        won't be NULL.)
3022     3. The memmove() emulation function in pcre_internal.h, which is used on
3023        systems that lack both memmove() and bcopy() - that is, hardly ever -
3024        was missing a "static" storage class specifier.
3026     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
3027        containing an extended class (one that cannot be represented by a bitmap
3028        because it contains high-valued characters or Unicode property items, e.g.
3029        [\pZ]). Almost always one would set UTF-8 mode when processing such a
3030        pattern, but PCRE should not loop if you do not (it no longer does).
3031        [Detail: two cases were found: (a) a repeated subpattern containing an
3032        extended class; (b) a recursive reference to a subpattern that followed a
3033        previous extended class. It wasn't skipping over the extended class
3034        correctly when UTF-8 mode was not set.]
3036     5. A negated single-character class was not being recognized as fixed-length
3037        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
3038        compile error "lookbehind assertion is not fixed length".
3040     6. The RunPerlTest auxiliary script was showing an unexpected difference
3041        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
3042        write a Perl script that can interpret lines of an input file either as
3043        byte characters or as UTF-8, which is what "perltest" was being required to
3044        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
3045        can't do is switch easily at run time between having the "use utf8;" pragma
3046        or not. In the end, I fudged it by using the RunPerlTest script to insert
3047        "use utf8;" explicitly for the UTF-8 tests.
3049     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
3050        the end of the subject string, contrary to the documentation and to what
3051        Perl does. This was true of both matching functions. Now it matches only at
3052        the start of the subject and immediately after *internal* newlines.
3054     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
3055        a pointer to an int instead of a pointer to an unsigned long int. This
3056        caused problems on 64-bit systems.
3058     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
3059        instance of the 'standard' template library not being so standard".
3061    10. There was no check on the number of named subpatterns nor the maximum
3062        length of a subpattern name. The product of these values is used to compute
3063        the size of the memory block for a compiled pattern. By supplying a very
3064        long subpattern name and a large number of named subpatterns, the size
3065        computation could be caused to overflow. This is now prevented by limiting
3066        the length of names to 32 characters, and the number of named subpatterns
3067        to 10,000.
3069    11. Subpatterns that are repeated with specific counts have to be replicated in
3070        the compiled pattern. The size of memory for this was computed from the
3071        length of the subpattern and the repeat count. The latter is limited to
3072        65535, but there was no limit on the former, meaning that integer overflow
3073        could in principle occur. The compiled length of a repeated subpattern is
3074        now limited to 30,000 bytes in order to prevent this.
3076    12. Added the optional facility to have named substrings with the same name.
3078    13. Added the ability to use a named substring as a condition, using the
3079        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
3080        are numbers (not recommended). Forward references are permitted.
3082    14. Added forward references in named backreferences (if you see what I mean).
3084    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
3085        pattern could run off the end of the subject. For example, the pattern
3086        "(?s)(.{1,5})"8 did this with the subject "ab".
3088    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
3089        PCRE_CASELESS was set when matching characters that were quantified with ?
3090        or *.
3092    17. A character class other than a single negated character that had a minimum
3093        but no maximum quantifier - for example [ab]{6,} - was not handled
3094        correctly by pce_dfa_exec(). It would match only one character.
3096    18. A valid (though odd) pattern that looked like a POSIX character
3097        class but used an invalid character after [ (for example [[,abc,]]) caused
3098        pcre_compile() to give the error "Failed: internal error: code overflow" or
3099        in some cases to crash with a glibc free() error. This could even happen if
3100        the pattern terminated after [[ but there just happened to be a sequence of
3101        letters, a binary zero, and a closing ] in the memory that followed.
3103    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
3104        over the years. Originally (before any Unicode support), just the bottom 8
3105        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
3106        output from "man perlunicode" includes this:
3108          The regular expression compiler produces polymorphic opcodes.  That
3109          is, the pattern adapts to the data and automatically switches to
3110          the Unicode character scheme when presented with Unicode data--or
3111          instead uses a traditional byte scheme when presented with byte
3112          data.
3114        Sadly, a wide octal escape does not cause a switch, and in a string with
3115        no other multibyte characters, these octal escapes are treated as before.
3116        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
3117        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
3118        Unicode string.
3120        I have not perpetrated such confusion in PCRE. Up till now, it took just
3121        the bottom 8 bits, as in old Perl. I have now made octal escapes with
3122        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
3123        translate to the appropriate multibyte character.
3125    29. Applied some refactoring to reduce the number of warnings from Microsoft
3126        and Borland compilers. This has included removing the fudge introduced
3127        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
3128        a warning about an unused variable.
3130    21. PCRE has not included VT (character 0x0b) in the set of whitespace
3131        characters since release 4.0, because Perl (from release 5.004) does not.
3132        [Or at least, is documented not to: some releases seem to be in conflict
3133        with the documentation.] However, when a pattern was studied with
3134        pcre_study() and all its branches started with \s, PCRE still included VT
3135        as a possible starting character. Of course, this did no harm; it just
3136        caused an unnecessary match attempt.
3138    22. Removed a now-redundant internal flag bit that recorded the fact that case
3139        dependency changed within the pattern. This was once needed for "required
3140        byte" processing, but is no longer used. This recovers a now-scarce options
3141        bit. Also moved the least significant internal flag bit to the most-
3142        significant bit of the word, which was not previously used (hangover from
3143        the days when it was an int rather than a uint) to free up another bit for
3144        the future.
3146    23. Added support for CRLF line endings as well as CR and LF. As well as the
3147        default being selectable at build time, it can now be changed at runtime
3148        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
3149        specify that it is scanning data with non-default line endings.
3151    24. Changed the definition of CXXLINK to make it agree with the definition of
3152        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
3154    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
3155        recursions. This makes a big different to stack usage for some patterns.
3157    26. If a subpattern containing a named recursion or subroutine reference such
3158        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
3159        the space required for the compiled pattern went wrong and gave too small a
3160        value. Depending on the environment, this could lead to "Failed: internal
3161        error: code overflow at offset 49" or "glibc detected double free or
3162        corruption" errors.
3164    27. Applied patches from Google (a) to support the new newline modes and (b) to
3165        advance over multibyte UTF-8 characters in GlobalReplace.
3167    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
3168        difference for some implementation of PCRE in some Windows version.
3170    29. Added some extra testing facilities to pcretest:
3172        \q<number>   in a data line sets the "match limit" value
3173        \Q<number>   in a data line sets the "match recursion limt" value
3174        -S <number>  sets the stack size, where <number> is in megabytes
3176        The -S option isn't available for Windows.
3179    Version 6.6 06-Feb-06
3180    ---------------------
3182     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
3183        in pcreposix.h. I have copied the definition from pcre.h.
3185     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
3186        because pcre.h is no longer a built file.
3188     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
3189        not normally included in the compiled code.
3192    Version 6.5 01-Feb-06
3193    ---------------------
3195     1. When using the partial match feature with pcre_dfa_exec(), it was not
3196        anchoring the second and subsequent partial matches at the new starting
3197        point. This could lead to incorrect results. For example, with the pattern
3198        /1234/, partially matching against "123" and then "a4" gave a match.
3200     2. Changes to pcregrep:
3202        (a) All non-match returns from pcre_exec() were being treated as failures
3203            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
3204            error message is output. Some extra information is given for the
3205            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
3206            probably the only errors that are likely to be caused by users (by
3207            specifying a regex that has nested indefinite repeats, for instance).
3208            If there are more than 20 of these errors, pcregrep is abandoned.
3210        (b) A binary zero was treated as data while matching, but terminated the
3211            output line if it was written out. This has been fixed: binary zeroes
3212            are now no different to any other data bytes.
3214        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
3215            used to set a locale for matching. The --locale=xxxx long option has
3216            been added (no short equivalent) to specify a locale explicitly on the
3217            pcregrep command, overriding the environment variables.
3219        (d) When -B was used with -n, some line numbers in the output were one less
3220            than they should have been.
3222        (e) Added the -o (--only-matching) option.
3224        (f) If -A or -C was used with -c (count only), some lines of context were
3225            accidentally printed for the final match.
3227        (g) Added the -H (--with-filename) option.
3229        (h) The combination of options -rh failed to suppress file names for files
3230            that were found from directory arguments.
3232        (i) Added the -D (--devices) and -d (--directories) options.
3234        (j) Added the -F (--fixed-strings) option.
3236        (k) Allow "-" to be used as a file name for -f as well as for a data file.
3238        (l) Added the --colo(u)r option.
3240        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
3241            is not present by default.
3243     3. A nasty bug was discovered in the handling of recursive patterns, that is,
3244        items such as (?R) or (?1), when the recursion could match a number of
3245        alternatives. If it matched one of the alternatives, but subsequently,
3246        outside the recursion, there was a failure, the code tried to back up into
3247        the recursion. However, because of the way PCRE is implemented, this is not
3248        possible, and the result was an incorrect result from the match.
3250        In order to prevent this happening, the specification of recursion has
3251        been changed so that all such subpatterns are automatically treated as
3252        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
3254     4. I had overlooked the fact that, in some locales, there are characters for
3255        which isalpha() is true but neither isupper() nor islower() are true. In
3256        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
3257        and ordfeminine) are like this. This affected the treatment of \w and \W
3258        when they appeared in character classes, but not when they appeared outside
3259        a character class. The bit map for "word" characters is now created
3260        separately from the results of isalnum() instead of just taking it from the
3261        upper, lower, and digit maps. (Plus the underscore character, of course.)
3263     5. The above bug also affected the handling of POSIX character classes such as
3264        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
3265        permanent tables. Instead, the bit maps for such a class were previously
3266        created as the appropriate unions of the upper, lower, and digit bitmaps.
3267        Now they are created by subtraction from the [[:word:]] class, which has
3268        its own bitmap.
3270     6. The [[:blank:]] character class matches horizontal, but not vertical space.
3271        It is created by subtracting the vertical space characters (\x09, \x0a,
3272        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
3273        subtraction was done in the overall bitmap for a character class, meaning
3274        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
3275        be recognized. This bug has been fixed.
3277     7. Patches from the folks at Google:
3279          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
3280          real life, but is still worth protecting against".
3282          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
3283          regular expressions".
3285          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
3286          have it.
3288          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
3289          "configure" and the latter not, in order to fix a problem somebody had
3290          with compiling the Arg class on HP-UX.
3292          (e) Improve the error-handling of the C++ wrapper a little bit.
3294          (f) New tests for checking recursion limiting.
3296     8. The pcre_memmove() function, which is used only if the environment does not
3297        have a standard memmove() function (and is therefore rarely compiled),
3298        contained two bugs: (a) use of int instead of size_t, and (b) it was not
3299        returning a result (though PCRE never actually uses the result).
3301     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
3302        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
3303        returned instead of calling malloc() with an overflowing number that would
3304        most likely cause subsequent chaos.
3306    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
3308    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
3309        with this option is matched, the nmatch and pmatch options of regexec() are
3310        ignored.
3312    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
3313        provided in case anyone wants to the the POSIX interface with UTF-8
3314        strings.
3316    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
3317        C++ linking (needed for some HP-UX environments).
3319    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
3320        (unused parameter) and in the pcre_printint() function (omitted "default"
3321        switch label when the default is to do nothing).
3323    15. Added some code to make it possible, when PCRE is compiled as a C++
3324        library, to replace subject pointers for pcre_exec() with a smart pointer
3325        class, thus making it possible to process discontinuous strings.
3327    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
3328        much the same function. They were added by different people who were trying
3329        to make PCRE easy to compile on non-Unix systems. It has been suggested
3330        that PCRE_EXPORT be abolished now that there is more automatic apparatus
3331        for compiling on Windows systems. I have therefore replaced it with
3332        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
3333        defaults to "extern" for C or "extern C" for C++, which works fine on
3334        Unix-like systems. It is now possible to override the value of PCRE_DATA_
3335        SCOPE with something explicit in config.h. In addition:
3337        (a) pcreposix.h still had just "extern" instead of either of these macros;
3338            I have replaced it with PCRE_DATA_SCOPE.
3340        (b) Functions such as _pcre_xclass(), which are internal to the library,
3341            but external in the C sense, all had PCRE_EXPORT in their definitions.
3342            This is apparently wrong for the Windows case, so I have removed it.
3343            (It makes no difference on Unix-like systems.)
3345    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
3346        of recursive calls to match(). This is different to MATCH_LIMIT because
3347        that limits the total number of calls to match(), not all of which increase
3348        the depth of recursion. Limiting the recursion depth limits the amount of
3349        stack (or heap if NO_RECURSE is set) that is used. The default can be set
3350        when PCRE is compiled, and changed at run time. A patch from Google adds
3351        this functionality to the C++ interface.
3353    18. Changes to the handling of Unicode character properties:
3355        (a) Updated the table to Unicode 4.1.0.
3357        (b) Recognize characters that are not in the table as "Cn" (undefined).
3359        (c) I revised the way the table is implemented to a much improved format
3360            which includes recognition of ranges. It now supports the ranges that
3361            are defined in UnicodeData.txt, and it also amalgamates other
3362            characters into ranges. This has reduced the number of entries in the
3363            table from around 16,000 to around 3,000, thus reducing its size
3364            considerably. I realized I did not need to use a tree structure after
3365            all - a binary chop search is just as efficient. Having reduced the
3366            number of entries, I extended their size from 6 bytes to 8 bytes to
3367            allow for more data.
3369        (d) Added support for Unicode script names via properties such as \p{Han}.
3371    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
3372        matching that character.
3374    20. When matching a repeated Unicode property with a minimum greater than zero,
3375        (for example \pL{2,}), PCRE could look past the end of the subject if it
3376        reached it while seeking the minimum number of characters. This could
3377        happen only if some of the characters were more than one byte long, because
3378        there is a check for at least the minimum number of bytes.
3380    21. Refactored the implementation of \p and \P so as to be more general, to
3381        allow for more different types of property in future. This has changed the
3382        compiled form incompatibly. Anybody with saved compiled patterns that use
3383        \p or \P will have to recompile them.
3385    22. Added "Any" and "L&" to the supported property types.
3387    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
3388        but give a compile time error if the value is greater than 0xff.
3390    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
3391        accidentally not being installed or uninstalled.
3393    25. The pcre.h file was built from pcre.h.in, but the only changes that were
3394        made were to insert the current release number. This seemed silly, because
3395        it made things harder for people building PCRE on systems that don't run
3396        "configure". I have turned pcre.h into a distributed file, no longer built
3397        by "configure", with the version identification directly included. There is
3398        no longer a pcre.h.in file.
3400        However, this change necessitated a change to the pcre-config script as
3401        well. It is built from pcre-config.in, and one of the substitutions was the
3402        release number. I have updated configure.ac so that ./configure now finds
3403        the release number by grepping pcre.h.
3405    26. Added the ability to run the tests under valgrind.
3408    Version 6.4 05-Sep-05
3409    ---------------------
3411     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
3412        "--" to be printed when multiple files were scanned, even when none of the
3413        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
3414        consider it to be a bug, and have restored the previous behaviour.
3416     2. A couple of code tidies to get rid of compiler warnings.
3418     3. The pcretest program used to cheat by referring to symbols in the library
3419        whose names begin with _pcre_. These are internal symbols that are not
3420        really supposed to be visible externally, and in some environments it is
3421        possible to suppress them. The cheating is now confined to including
3422        certain files from the library's source, which is a bit cleaner.
3424     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
3425        file's purpose clearer.
3427     5. Reorganized pcre_ucp_findchar().
3430    Version 6.3 15-Aug-05
3431    ---------------------
3433     1. The file libpcre.pc.in did not have general read permission in the tarball.
3435     2. There were some problems when building without C++ support:
3437        (a) If C++ support was not built, "make install" and "make test" still
3438            tried to test it.
3440        (b) There were problems when the value of CXX was explicitly set. Some
3441            changes have been made to try to fix these, and ...
3443        (c) --disable-cpp can now be used to explicitly disable C++ support.
3445        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
3446            backslash in a target when C++ was disabled. This confuses some
3447            versions of "make", apparently. Using an intermediate variable solves
3448            this. (Same for CPP_LOBJ.)
3450     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
3451        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
3452        necessary on certain architectures.
3454     4. Added a setting of -export-symbols-regex to the link command to remove
3455        those symbols that are exported in the C sense, but actually are local
3456        within the library, and not documented. Their names all begin with
3457        "_pcre_". This is not a perfect job, because (a) we have to except some
3458        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
3459        available (and never for static libraries). I have made a note to try to
3460        find a way round (a) in the future.
3463    Version 6.2 01-Aug-05
3464    ---------------------
3466     1. There was no test for integer overflow of quantifier values. A construction
3467        such as {1111111111111111} would give undefined results. What is worse, if
3468        a minimum quantifier for a parenthesized subpattern overflowed and became
3469        negative, the calculation of the memory size went wrong. This could have
3470        led to memory overwriting.
3472     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
3474     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
3475        operating environments where this matters.
3477     4. Applied Giuseppe Maxia's patch to add additional features for controlling
3478        PCRE options from within the C++ wrapper.
3480     5. Named capturing subpatterns were not being correctly counted when a pattern
3481        was compiled. This caused two problems: (a) If there were more than 100
3482        such subpatterns, the calculation of the memory needed for the whole
3483        compiled pattern went wrong, leading to an overflow error. (b) Numerical
3484        back references of the form \12, where the number was greater than 9, were
3485        not recognized as back references, even though there were sufficient
3486        previous subpatterns.
3488     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
3489        versions of gcc, e.g. 2.95.4.
3492    Version 6.1 21-Jun-05
3493    ---------------------
3495     1. There was one reference to the variable "posix" in pcretest.c that was not
3496        surrounded by "#if !defined NOPOSIX".
3498     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
3499        the cross-check on the old pcre_info() function, for the benefit of the
3500        cut-down version of PCRE that is currently imported into Exim.
3502     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
3503        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
3504        patterns that start (?i) but I don't think that matters. The use of (?i) is
3505        just an example; this all applies to the other options as well.
3507     4. Since libtool seems to echo the compile commands it is issuing, the output
3508        from "make" can be reduced a bit by putting "@" in front of each libtool
3509        compile command.
3511     5. Patch from the folks at Google for configure.in to be a bit more thorough
3512        in checking for a suitable C++ installation before trying to compile the
3513        C++ stuff. This should fix a reported problem when a compiler was present,
3514        but no suitable headers.
3516     6. The man pages all had just "PCRE" as their title. I have changed them to
3517        be the relevant file name. I have also arranged that these names are
3518        retained in the file doc/pcre.txt, which is a concatenation in text format
3519        of all the man pages except the little individual ones for each function.
3521     7. The NON-UNIX-USE file had not been updated for the different set of source
3522        files that come with release 6. I also added a few comments about the C++
3523        wrapper.
3526    Version 6.0 07-Jun-05
3527    ---------------------
3529     1. Some minor internal re-organization to help with my DFA experiments.
3531     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
3532        didn't matter for the library itself when fully configured, but did matter
3533        when compiling without UCP support, or within Exim, where the ucp files are
3534        not imported.
3536     3. Refactoring of the library code to split up the various functions into
3537        different source modules. The addition of the new DFA matching code (see
3538        below) to a single monolithic source would have made it really too
3539        unwieldy, quite apart from causing all the code to be include in a
3540        statically linked application, when only some functions are used. This is
3541        relevant even without the DFA addition now that patterns can be compiled in
3542        one application and matched in another.
3544        The downside of splitting up is that there have to be some external
3545        functions and data tables that are used internally in different modules of
3546        the library but which are not part of the API. These have all had their
3547        names changed to start with "_pcre_" so that they are unlikely to clash
3548        with other external names.
3550     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
3551        a different (DFA) algorithm. Although it is slower than the original
3552        function, it does have some advantages for certain types of matching
3553        problem.
3555     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
3556        including restarting after a partial match.
3558     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
3559        defined when compiling for Windows was sent to me. I have put it into the
3560        code, though I have no means of testing or verifying it.
3562     7. Added the pcre_refcount() auxiliary function.
3564     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
3565        match before or at the first newline in the subject string. In pcretest,
3566        the /f option on a pattern can be used to set this.
3568     9. A repeated \w when used in UTF-8 mode with characters greater than 256
3569        would behave wrongly. This has been present in PCRE since release 4.0.
3571    10. A number of changes to the pcregrep command:
3573        (a) Refactored how -x works; insert ^(...)$ instead of setting
3574            PCRE_ANCHORED and checking the length, in preparation for adding
3575            something similar for -w.
3577        (b) Added the -w (match as a word) option.
3579        (c) Refactored the way lines are read and buffered so as to have more
3580            than one at a time available.
3582        (d) Implemented a pcregrep test script.