ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 87 by nigel, Sat Feb 24 21:41:21 2007 UTC revision 515 by ph10, Tue May 4 09:12:25 2010 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
4    Version 8.10 03 May-2010
5    ------------------------
7    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
8        THEN.
10    2.  (*ACCEPT) was not working when inside an atomic group.
12    3.  Inside a character class, \B is treated as a literal by default, but
13        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
14        causes the error). The code is unchanged, but I tidied the documentation.
16    4.  Inside a character class, PCRE always treated \R and \X as literals,
17        whereas Perl faults them if its -w option is set. I have changed PCRE so
18        that it faults them when PCRE_EXTRA is set.
20    5.  Added support for \N, which always matches any character other than
21        newline. (It is the same as "." when PCRE_DOTALL is not set.)
23    6.  When compiling pcregrep with newer versions of gcc which may have
24        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
25        declared with attribute warn_unused_result" were given. Just casting the
26        result to (void) does not stop the warnings; a more elaborate fudge is
27        needed. I've used a macro to implement this.
30    Version 8.02 19-Mar-2010
31    ------------------------
33    1.  The Unicode data tables have been updated to Unicode 5.2.0.
35    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
36        configured.
38    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
39        original author of that file, following a query about its status.
41    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
42        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
44    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
45        quantifier applied to a forward-referencing subroutine call, could compile
46        incorrect code or give the error "internal error: previously-checked
47        referenced subpattern not found".
49    6.  Both MS Visual Studio and Symbian OS have problems with initializing
50        variables to point to external functions. For these systems, therefore,
51        pcre_malloc etc. are now initialized to local functions that call the
52        relevant global functions.
54    7.  There were two entries missing in the vectors called coptable and poptable
55        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
56        I've fixed the data, and added a kludgy way of testing at compile time that
57        the lengths are correct (equal to the number of opcodes).
59    8.  Following on from 7, I added a similar kludge to check the length of the
60        eint vector in pcreposix.c.
62    9.  Error texts for pcre_compile() are held as one long string to avoid too
63        much relocation at load time. To find a text, the string is searched,
64        counting zeros. There was no check for running off the end of the string,
65        which could happen if a new error number was added without updating the
66        string.
68    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
70    11. \K was not working if it appeared in an atomic group or in a group that
71        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
72        \K is "not well defined" if used in an assertion. PCRE now accepts it if
73        the assertion is positive, but not if it is negative.
75    12. Change 11 fortuitously reduced the size of the stack frame used in the
76        "match()" function of pcre_exec.c by one pointer. Forthcoming
77        implementation of support for (*MARK) will need an extra pointer on the
78        stack; I have reserved it now, so that the stack frame size does not
79        decrease.
81    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
82        item in branch that calls a recursion is a subroutine call - as in the
83        second branch in the above example - was incorrectly given the compile-
84        time error "recursive call could loop indefinitely" because pcre_compile()
85        was not correctly checking the subroutine for matching a non-empty string.
87    14. The checks for overrunning compiling workspace could trigger after an
88        overrun had occurred. This is a "should never occur" error, but it can be
89        triggered by pathological patterns such as hundreds of nested parentheses.
90        The checks now trigger 100 bytes before the end of the workspace.
92    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
95    Version 8.01 19-Jan-2010
96    ------------------------
98    1.  If a pattern contained a conditional subpattern with only one branch (in
99        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
100        computed the wrong minimum data length (which is of course zero for such
101        subpatterns). This could cause incorrect "no match" results.
103    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
104        the pattern is reset in the first branch, pcre_compile() failed with
105        "internal error: code overflow at offset...". This happened only when
106        the reset was to the original external option setting. (An optimization
107        abstracts leading options settings into an external setting, which was the
108        cause of this.)
110    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
111        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
112        assertion pattern did not match (meaning that the assertion was true), it
113        was incorrectly treated as false if the SKIP had been reached during the
114        matching. This also applied to assertions used as conditions.
116    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
117        assertion subpattern, including such a pattern used as a condition,
118        unpredictable results occurred, instead of the error return
121    5.  The C++ GlobalReplace function was not working like Perl for the special
122        situation when an empty string is matched. It now does the fancy magic
123        stuff that is necessary.
125    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
126        removed. (These were left over from very, very early versions of PCRE.)
128    7.  Some cosmetic changes to the code to make life easier when compiling it
129        as part of something else:
131        (a) Change DEBUG to PCRE_DEBUG.
133        (b) In pcre_compile(), rename the member of the "branch_chain" structure
134            called "current" as "current_branch", to prevent a collision with the
135            Linux macro when compiled as a kernel module.
137        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
138            prevent a collision with the Linux macro when compiled as a kernel
139            module.
141    8.  In pcre_compile() there are some checks for integer overflows that used to
142        cast potentially large values to (double). This has been changed to that
143        when building, a check for int64_t is made, and if it is found, it is used
144        instead, thus avoiding the use of floating point arithmetic. (There is no
145        other use of FP in PCRE.) If int64_t is not found, the fallback is to
146        double.
148    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
149        2005 (difference between two addresses compared to an unsigned value).
151    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
152        custom one, because of the following reported problem in Windows:
154          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
155              under Win32.
156          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
157              therefore missing the function definition.
158          - The compiler thus generates a "C" signature for the test function.
159          - The linker fails to find the "C" function.
160          - PCRE fails to configure if asked to do so against libbz2.
162    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
163        messages were output:
165          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
166          rerunning libtoolize, to keep the correct libtool macros in-tree.
167          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
169        I have done both of these things.
171    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
172        most of the time, it *can* run out if it is given a pattern that contains a
173        runaway infinite recursion. I updated the discussion in the pcrestack man
174        page.
176    13. Now that we have gone to the x.xx style of version numbers, the minor
177        version may start with zero. Using 08 or 09 is a bad idea because users
178        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
179        interpreted as invalid octal numbers. I've updated the previous comment in
180        configure.ac, and also added a check that gives an error if 08 or 09 are
181        used.
183    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
184        causing partial matching to fail when the end of the subject matched \W
185        in a UTF-8 pattern where \W was quantified with a minimum of 3.
187    15. There were some discrepancies between the declarations in pcre_internal.h
188        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
189        their definitions. The declarations used "const uschar *" and the
190        definitions used USPTR. Even though USPTR is normally defined as "const
191        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
192        reported that: "This difference in casting confuses some C++ compilers, for
193        example, SunCC recognizes above declarations as different functions and
194        generates broken code for hbpcre." I have changed the declarations to use
195        USPTR.
197    16. GNU libtool is named differently on some systems. The autogen.sh script now
198        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
199        (FreeBSD).
201    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
202        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
203        comment: "Figure out how to create a longlong from a string: strtoll and
204        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
205        instance, but it only takes 2 args instead of 3!"
207    18. A subtle bug concerned with back references has been fixed by a change of
208        specification, with a corresponding code fix. A pattern such as
209        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
210        refers, was giving matches when it shouldn't. For example, xa=xaaa would
211        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
212        same bug. Such groups have to be quantified to be useful, or contained
213        inside another quantified group. (If there's no repetition, the reference
214        can never match.) The problem arises because, having left the group and
215        moved on to the rest of the pattern, a later failure that backtracks into
216        the group uses the captured value from the final iteration of the group
217        rather than the correct earlier one. I have fixed this in PCRE by forcing
218        any group that contains a reference to itself to be an atomic group; that
219        is, there cannot be any backtracking into it once it has completed. This is
220        similar to recursive and subroutine calls.
223    Version 8.00 19-Oct-09
224    ----------------------
226    1.  The table for translating pcre_compile() error codes into POSIX error codes
227        was out-of-date, and there was no check on the pcre_compile() error code
228        being within the table. This could lead to an OK return being given in
229        error.
231    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
232        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
233        in a Windows environment.
235    3.  The pcregrep --count option prints the count for each file even when it is
236        zero, as does GNU grep. However, pcregrep was also printing all files when
237        --files-with-matches was added. Now, when both options are given, it prints
238        counts only for those files that have at least one match. (GNU grep just
239        prints the file name in this circumstance, but including the count seems
240        more useful - otherwise, why use --count?) Also ensured that the
241        combination -clh just lists non-zero counts, with no names.
243    4.  The long form of the pcregrep -F option was incorrectly implemented as
244        --fixed_strings instead of --fixed-strings. This is an incompatible change,
245        but it seems right to fix it, and I didn't think it was worth preserving
246        the old behaviour.
248    5.  The command line items --regex=pattern and --regexp=pattern were not
249        recognized by pcregrep, which required --regex pattern or --regexp pattern
250        (with a space rather than an '='). The man page documented the '=' forms,
251        which are compatible with GNU grep; these now work.
253    6.  No libpcreposix.pc file was created for pkg-config; there was just
254        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
256    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
257        when UCP support is not needed, by modifying the Python script that
258        generates it from Unicode data files. This should not matter if the module
259        is correctly used as a library, but I received one complaint about 50K of
260        unwanted data. My guess is that the person linked everything into his
261        program rather than using a library. Anyway, it does no harm.
263    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
264        was a minimum greater than 1 for a wide character in a possessive
265        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
266        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
267        character. Chaos in the form of incorrect output or a compiling loop could
268        result.
270    9.  The restrictions on what a pattern can contain when partial matching is
271        requested for pcre_exec() have been removed. All patterns can now be
272        partially matched by this function. In addition, if there are at least two
273        slots in the offset vector, the offset of the earliest inspected character
274        for the match and the offset of the end of the subject are set in them when
275        PCRE_ERROR_PARTIAL is returned.
277    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
278        synonymous with PCRE_PARTIAL, for backwards compatibility, and
279        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
280        and may be more useful for multi-segment matching.
282    11. Partial matching with pcre_exec() is now more intuitive. A partial match
283        used to be given if ever the end of the subject was reached; now it is
284        given only if matching could not proceed because another character was
285        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
286        string "Z", which now yields "no match" instead of "partial match". In the
287        case of pcre_dfa_exec(), "no match" is given if every matching path for the
288        final character ended with (*FAIL).
290    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
291        if the pattern had a "must contain" character that was already found in the
292        earlier partial match, unless partial matching was again requested. For
293        example, with the pattern /dog.(body)?/, the "must contain" character is
294        "g". If the first part-match was for the string "dog", restarting with
295        "sbody" failed. This bug has been fixed.
297    13. The string returned by pcre_dfa_exec() after a partial match has been
298        changed so that it starts at the first inspected character rather than the
299        first character of the match. This makes a difference only if the pattern
300        starts with a lookbehind assertion or \b or \B (\K is not supported by
301        pcre_dfa_exec()). It's an incompatible change, but it makes the two
302        matching functions compatible, and I think it's the right thing to do.
304    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
305        so that the demonstration program is easily available in environments where
306        PCRE has not been installed from source.
308    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
309        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
310        library.
312    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
313        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
314        is not the first non-POSIX option to be added. Clearly some people find
315        these options useful.
317    17. If a caller to the POSIX matching function regexec() passes a non-zero
318        value for nmatch with a NULL value for pmatch, the value of
319        nmatch is forced to zero.
321    18. RunGrepTest did not have a test for the availability of the -u option of
322        the diff command, as RunTest does. It now checks in the same way as
323        RunTest, and also checks for the -b option.
325    19. If an odd number of negated classes containing just a single character
326        interposed, within parentheses, between a forward reference to a named
327        subpattern and the definition of the subpattern, compilation crashed with
328        an internal error, complaining that it could not find the referenced
329        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
330        [The bug was that it was starting one character too far in when skipping
331        over the character class, thus treating the ] as data rather than
332        terminating the class. This meant it could skip too much.]
334    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
335        /g option in pcretest when the pattern contains \K, which makes it possible
336        to have an empty string match not at the start, even when the pattern is
337        anchored. Updated pcretest and pcredemo to use this option.
339    21. If the maximum number of capturing subpatterns in a recursion was greater
340        than the maximum at the outer level, the higher number was returned, but
341        with unset values at the outer level. The correct (outer level) value is
342        now given.
344    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
345        PCRE did not set those parentheses (unlike Perl). I have now found a way to
346        make it do so. The string so far is captured, making this feature
347        compatible with Perl.
349    23. The tests have been re-organized, adding tests 11 and 12, to make it
350        possible to check the Perl 5.10 features against Perl 5.10.
352    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
353        pattern matches a fixed length string. PCRE did not allow this; now it
354        does. Neither allows recursion.
356    25. I finally figured out how to implement a request to provide the minimum
357        length of subject string that was needed in order to match a given pattern.
358        (It was back references and recursion that I had previously got hung up
359        on.) This code has now been added to pcre_study(); it finds a lower bound
360        to the length of subject needed. It is not necessarily the greatest lower
361        bound, but using it to avoid searching strings that are too short does give
362        some useful speed-ups. The value is available to calling programs via
363        pcre_fullinfo().
365    26. While implementing 25, I discovered to my embarrassment that pcretest had
366        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
367        study optimizations had never been tested with that matching function.
368        Oops. What is worse, even when it was passed study data, there was a bug in
369        pcre_dfa_exec() that meant it never actually used it. Double oops. There
370        were also very few tests of studied patterns with pcre_dfa_exec().
372    27. If (?| is used to create subpatterns with duplicate numbers, they are now
373        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
374        on the other side of the coin, they are no longer allowed to have different
375        names, because these cannot be distinguished in PCRE, and this has caused
376        confusion. (This is a difference from Perl.)
378    28. When duplicate subpattern names are present (necessarily with different
379        numbers, as required by 27 above), and a test is made by name in a
380        conditional pattern, either for a subpattern having been matched, or for
381        recursion in such a pattern, all the associated numbered subpatterns are
382        tested, and the overall condition is true if the condition is true for any
383        one of them. This is the way Perl works, and is also more like the way
384        testing by number works.
387    Version 7.9 11-Apr-09
388    ---------------------
390    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
391        (pcretest), all targets were linked against these libraries. This included
392        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
393        libraries. This caused unwanted dependencies to be created. This problem
394        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
395        pcretest is linked with readline.
397    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
398        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
399        moved outside it again, because FALSE and TRUE are already defined in AIX,
400        but BOOL is not.
402    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
403        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
405    4.  The pcregrep documentation said spaces were inserted as well as colons (or
406        hyphens) following file names and line numbers when outputting matching
407        lines. This is not true; no spaces are inserted. I have also clarified the
408        wording for the --colour (or --color) option.
410    5.  In pcregrep, when --colour was used with -o, the list of matching strings
411        was not coloured; this is different to GNU grep, so I have changed it to be
412        the same.
414    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
415        each matching line was coloured. Now it goes on to look for further matches
416        of any of the test patterns, which is the same behaviour as GNU grep.
418    7.  A pattern that could match an empty string could cause pcregrep to loop; it
419        doesn't make sense to accept an empty string match in pcregrep, so I have
420        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
421        seems to be how GNU grep behaves.
423    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
424        start or after a newline", because the conditional assertion was not being
425        correctly handled. The rule now is that both the assertion and what follows
426        in the first alternative must satisfy the test.
428    9.  If auto-callout was enabled in a pattern with a conditional group whose
429        condition was an assertion, PCRE could crash during matching, both with
430        pcre_exec() and pcre_dfa_exec().
432    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
433        used for matching.
435    11. Unicode property support in character classes was not working for
436        characters (bytes) greater than 127 when not in UTF-8 mode.
438    12. Added the -M command line option to pcretest.
440    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
442    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
444    16. Added comments and documentation about mis-use of no_arg in the C++
445        wrapper.
447    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
448        from Martin Jerabek that uses macro names for all relevant character and
449        string constants.
451    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
452        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
453        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
454        these, but not everybody uses configure.
456    19. A conditional group that had only one branch was not being correctly
457        recognized as an item that could match an empty string. This meant that an
458        enclosing group might also not be so recognized, causing infinite looping
459        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
460        with the subject "ab", where knowledge that the repeated group can match
461        nothing is needed in order to break the loop.
463    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
464        exec(), but without supplying a callout function, matching went wrong.
466    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
467        leak if the size of the offset vector was greater than 30. When the vector
468        is smaller, the saved offsets during recursion go onto a local stack
469        vector, but for larger vectors malloc() is used. It was failing to free
470        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
471        error, in fact).
473    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
474        heapframe that is used only when UTF-8 support is enabled. This caused no
475        problem, but was untidy.
477    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
478        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
479        included within another project.
481    24. Steven Van Ingelgem's patches to add more options to the CMake support,
482        slightly modified by me:
484          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
485              not building pcregrep.
487          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
488              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
490    25. Forward references, both numeric and by name, in patterns that made use of
491        duplicate group numbers, could behave incorrectly or give incorrect errors,
492        because when scanning forward to find the reference group, PCRE was not
493        taking into account the duplicate group numbers. A pattern such as
494        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
496    26. Changed a few more instances of "const unsigned char *" to USPTR, making
497        the feature of a custom pointer more persuasive (as requested by a user).
499    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
500        pcretest.c, inside #ifndefs, because it seems they are sometimes already
501        pre-defined.
503    28. Added support for (*UTF8) at the start of a pattern.
505    29. Arrange for flags added by the "release type" setting in CMake to be shown
506        in the configuration summary.
509    Version 7.8 05-Sep-08
510    ---------------------
512    1.  Replaced UCP searching code with optimized version as implemented for Ad
513        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
514        stage table and inline lookup instead of a function, giving speed ups of 2
515        to 5 times on some simple patterns that I tested. Permission was given to
516        distribute the MultiStage2.py script that generates the tables (it's not in
517        the tarball, but is in the Subversion repository).
519    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
520        scripts.
522    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
523        a group with a zero qualifier. The result of the study could be incorrect,
524        or the function might crash, depending on the pattern.
526    4.  Caseless matching was not working for non-ASCII characters in back
527        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
528        It now works when Unicode Property Support is available.
530    5.  In pcretest, an escape such as \x{de} in the data was always generating
531        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
532        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
533        truncation.
535    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
537    7.  Added two (int) casts to pcregrep when printing the difference of two
538        pointers, in case they are 64-bit values.
540    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
541        test 2 if it fails.
543    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
544        and a #define of that name to empty if it is not externally set. This is to
545        allow users of MSVC to set it if necessary.
547    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
548        the convenience functions in the pcre_get.c source file.
550    11. An option change at the start of a pattern that had top-level alternatives
551        could cause overwriting and/or a crash. This command provoked a crash in
552        some environments:
554          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
556        This potential security problem was recorded as CVE-2008-2371.
558    12. For a pattern where the match had to start at the beginning or immediately
559        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
560        pcre_dfa_exec() could read past the end of the passed subject if there was
561        no match. To help with detecting such bugs (e.g. with valgrind), I modified
562        pcretest so that it places the subject at the end of its malloc-ed buffer.
564    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
565        exec() might read past the end of the data buffer in UTF-8 mode.
567    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
568        the data contained the byte 0x85 as part of a UTF-8 character within its
569        first line. This applied both to normal and DFA matching.
571    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
572        /^[^d]*?$/8 failed to match "abc".
574    16. Added a missing copyright notice to pcrecpp_internal.h.
576    17. Make it more clear in the documentation that values returned from
577        pcre_exec() in ovector are byte offsets, not character counts.
579    18. Tidied a few places to stop certain compilers from issuing warnings.
581    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
582        supplied by Stefan Weber. I made a further small update for 7.8 because
583        there is a change of source arrangements: the pcre_searchfuncs.c module is
584        replaced by pcre_ucd.c.
587    Version 7.7 07-May-08
588    ---------------------
590    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
591        a string to a long long, pretend we don't even have a long long." This is
592        done by checking for the strtoq, strtoll, and _strtoi64 functions.
594    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
595        pre-7.6 versions, which defined a global no_arg variable instead of putting
596        it in the RE class. (See also #8 below.)
598    3.  Remove a line of dead code, identified by coverity and reported by Nuno
599        Lopes.
601    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
603        (1) The include/exclude patterns were being applied to the whole pathnames
604            of files, instead of just to the final components.
606        (2) If there was more than one level of directory, the subdirectories were
607            skipped unless they satisfied the include/exclude conditions. This is
608            inconsistent with GNU grep (and could even be seen as contrary to the
609            pcregrep specification - which I improved to make it absolutely clear).
610            The action now is always to scan all levels of directory, and just
611            apply the include/exclude patterns to regular files.
613    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
614        --exclude_dir in the tests to avoid scanning .svn directories.
616    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
617        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
618        doesn't support NULs in patterns.
620    7.  Added some missing "const"s to declarations of static tables in
621        pcre_compile.c and pcre_dfa_exec.c.
623    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
624        caused by fix #2  above. (Subsequently also a second patch to fix the
625        first patch. And a third patch - this was a messy problem.)
627    9.  Applied Craig's patch to remove the use of push_back().
629    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
630        matching function regexec().
632    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
633        which, however, unlike Perl's \g{...}, are subroutine calls, not back
634        references. PCRE supports relative numbers with this syntax (I don't think
635        Oniguruma does).
637    12. Previously, a group with a zero repeat such as (...){0} was completely
638        omitted from the compiled regex. However, this means that if the group
639        was called as a subroutine from elsewhere in the pattern, things went wrong
640        (an internal error was given). Such groups are now left in the compiled
641        pattern, with a new opcode that causes them to be skipped at execution
642        time.
644    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
645        to the way PCRE behaves:
647        (a) A lone ] character is dis-allowed (Perl treats it as data).
649        (b) A back reference to an unmatched subpattern matches an empty string
650            (Perl fails the current match path).
652        (c) A data ] in a character class must be notated as \] because if the
653            first data character in a class is ], it defines an empty class. (In
654            Perl it is not possible to have an empty class.) The empty class []
655            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
656            The negative empty class [^] matches any one character, independently
657            of the DOTALL setting.
659    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
660        non-existent subpattern following a character class starting with ']' and
661        containing () gave an internal compiling error instead of "reference to
662        non-existent subpattern". Fortunately, when the pattern did exist, the
663        compiled code was correct. (When scanning forwards to check for the
664        existencd of the subpattern, it was treating the data ']' as terminating
665        the class, so got the count wrong. When actually compiling, the reference
666        was subsequently set up correctly.)
668    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
669        it was being rejected as not supported by pcre_dfa_exec(), even though
670        other assertions are supported. I have made pcre_dfa_exec() support
671        (*FAIL).
673    16. The implementation of 13c above involved the invention of a new opcode,
674        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
675        cannot be changed at match time, I realized I could make a small
676        improvement to matching performance by compiling OP_ALLANY instead of
677        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
678        on the OP_ANY path.
680    17. Compiling pcretest on Windows with readline support failed without the
681        following two fixes: (1) Make the unistd.h include conditional on
682        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
684    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
685        ncurses library to be included for pcretest when ReadLine support is
686        requested, but also to allow for it to be overridden. This patch came from
687        Daniel Bergström.
689    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
690        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
691        any errors with the current Unicode tables. Thanks to Peter Kankowski for
692        spotting this.
695    Version 7.6 28-Jan-08
696    ---------------------
698    1.  A character class containing a very large number of characters with
699        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
700        overflow.
702    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
703        HAVE_LONG_LONG is not defined.
705    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
706        bring them up to date and include new features. This patch includes:
708        - Fixed PH's badly added libz and libbz2 support.
709        - Fixed a problem with static linking.
710        - Added pcredemo. [But later removed - see 7 below.]
711        - Fixed dftables problem and added an option.
712        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
713            HAVE_LONG_LONG.
714        - Added readline support for pcretest.
715        - Added an listing of the option settings after cmake has run.
717    4.  A user submitted a patch to Makefile that makes it easy to create
718        "pcre.dll" under mingw when using Configure/Make. I added stuff to
719        Makefile.am that cause it to include this special target, without
720        affecting anything else. Note that the same mingw target plus all
721        the other distribution libraries and programs are now supported
722        when configuring with CMake (see 6 below) instead of with
723        Configure/Make.
725    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
726        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
727        exported in the Windows port". It has not yet been confirmed that the patch
728        solves the problem, but it does no harm.
730    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
731        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
732        with CMake, and also correct the comment about stack recursion.
734    7.  Remove the automatic building of pcredemo from the ./configure system and
735        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
736        of a program that users should build themselves after PCRE is installed, so
737        building it automatically is not really right. What is more, it gave
738        trouble in some build environments.
740    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
743    Version 7.5 10-Jan-08
744    ---------------------
746    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
747        values in parens when parsing an RE using the C++ wrapper."
749    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
750        Characters greater than 255 were excluded from the class instead of being
751        included.
753    3.  The same bug as (2) above applied to negated POSIX classes such as
754        [:^space:].
756    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
757        defined or documented. It seems to have been a typo for PCRE_STATIC, so
758        I have changed it.
760    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
761        first named subpattern) and a construct such as (?&a) would reference the
762        first named subpattern whose name started with "a" (in other words, the
763        length check was missing). Both these problems are fixed. "Subpattern name
764        expected" is now given for (?&) (a zero-length name), and this patch also
765        makes it give the same error for \k'' (previously it complained that that
766        was a reference to a non-existent subpattern).
768    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
769        this is right because (?- can be followed by option settings as well as by
770        digits. I have, however, made the messages clearer.
772    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
773        than the number used in the conditional) now cause a compile-time error.
774        This is actually not compatible with Perl, which accepts such patterns, but
775        treats the conditional as always being FALSE (as PCRE used to), but it
776        seems to me that giving a diagnostic is better.
778    8.  Change "alphameric" to the more common word "alphanumeric" in comments
779        and messages.
781    9.  Fix two occurrences of "backslash" in comments that should have been
782        "backspace".
784    10. Remove two redundant lines of code that can never be obeyed (their function
785        was moved elsewhere).
787    11. The program that makes PCRE's Unicode character property table had a bug
788        which caused it to generate incorrect table entries for sequences of
789        characters that have the same character type, but are in different scripts.
790        It amalgamated them into a single range, with the script of the first of
791        them. In other words, some characters were in the wrong script. There were
792        thirteen such cases, affecting characters in the following ranges:
794          U+002b0 - U+002c1
795          U+0060c - U+0060d
796          U+0061e - U+00612
797          U+0064b - U+0065e
798          U+0074d - U+0076d
799          U+01800 - U+01805
800          U+01d00 - U+01d77
801          U+01d9b - U+01dbf
802          U+0200b - U+0200f
803          U+030fc - U+030fe
804          U+03260 - U+0327f
805          U+0fb46 - U+0fbb1
806          U+10450 - U+1049d
808    12. The -o option (show only the matching part of a line) for pcregrep was not
809        compatible with GNU grep in that, if there was more than one match in a
810        line, it showed only the first of them. It now behaves in the same way as
811        GNU grep.
813    13. If the -o and -v options were combined for pcregrep, it printed a blank
814        line for every non-matching line. GNU grep prints nothing, and pcregrep now
815        does the same. The return code can be used to tell if there were any
816        non-matching lines.
818    14. Added --file-offsets and --line-offsets to pcregrep.
820    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
821        infinitely looping recursion. The bug was that positive lookaheads were not
822        being skipped when checking for a possible empty match (negative lookaheads
823        and both kinds of lookbehind were skipped).
825    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
826        inclusion of <windows.h> to before rather than after the definition of
827        INVALID_FILE_ATTRIBUTES (patch from David Byron).
829    17. Specifying a possessive quantifier with a specific limit for a Unicode
830        character property caused pcre_compile() to compile bad code, which led at
831        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
832        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
833        caused the error; without that there was no problem.
835    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
837    19. Added --enable-pcretest-libreadline.
839    20. In pcrecpp.cc, the variable 'count' was incremented twice in
840        RE::GlobalReplace(). As a result, the number of replacements returned was
841        double what it should be. I removed one of the increments, but Craig sent a
842        later patch that removed the other one (the right fix) and added unit tests
843        that check the return values (which was not done before).
845    21. Several CMake things:
847        (1) Arranged that, when cmake is used on Unix, the libraries end up with
848            the names libpcre and libpcreposix, not just pcre and pcreposix.
850        (2) The above change means that pcretest and pcregrep are now correctly
851            linked with the newly-built libraries, not previously installed ones.
855    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
856        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
857        UTF-8 newline character). The key issue is that the pattern starts .*;
858        this means that the match must be either at the beginning, or after a
859        newline. The bug was in the code for advancing after a failed match and
860        checking that the new position followed a newline. It was not taking
861        account of UTF-8 characters correctly.
863    23. PCRE was behaving differently from Perl in the way it recognized POSIX
864        character classes. PCRE was not treating the sequence [:...:] as a
865        character class unless the ... were all letters. Perl, however, seems to
866        allow any characters between [: and :], though of course it rejects as
867        unknown any "names" that contain non-letters, because all the known class
868        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
869        for example, whereas PCRE did not - it did not recognize a POSIX character
870        class. This seemed a bit dangerous, so the code has been changed to be
871        closer to Perl. The behaviour is not identical to Perl, because PCRE will
872        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
873        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
874        Perl does, and where it didn't before.
876    24. Rewrite so as to remove the single use of %n from pcregrep because in some
877        Windows environments %n is disabled by default.
880    Version 7.4 21-Sep-07
881    ---------------------
883    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
884        means that a class such as [\s] counted as "explicit reference to CR or
885        LF". That isn't really right - the whole point of the change was to try to
886        help when there was an actual mention of one of the two characters. So now
887        the change happens only if \r or \n (or a literal CR or LF) character is
888        encountered.
890    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
891        of both had grown to the point where there were only 3 bits left.
892        Fortunately, there was spare space in the data structure, and so I have
893        moved the internal flags into a new 16-bit field to free up more option
894        bits.
896    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
897        but did not set the internal JCHANGED flag - either of these is enough to
898        control the way the "get" function works - but the PCRE_INFO_JCHANGED
899        facility is supposed to tell if (?J) was ever used, so now (?J) at the
900        start sets both bits.
902    4.  Added options (at build time, compile time, exec time) to change \R from
903        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
905    5.  doc/pcresyntax.html was missing from the distribution.
907    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
908        compatibility, even though it is no longer used.
910    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
911        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
912        windows.h file is present (where different names are used). [This was
913        reversed later after testing - see 16 below.]
915    8.  Changed all #include <config.h> to #include "config.h". There were also
916        some further <pcre.h> cases that I changed to "pcre.h".
918    9.  When pcregrep was used with the --colour option, it missed the line ending
919        sequence off the lines that it output.
921    10. It was pointed out to me that arrays of string pointers cause lots of
922        relocations when a shared library is dynamically loaded. A technique of
923        using a single long string with a table of offsets can drastically reduce
924        these. I have refactored PCRE in four places to do this. The result is
925        dramatic:
927          Originally:                          290
928          After changing UCP table:            187
929          After changing error message table:   43
930          After changing table of "verbs"       36
931          After changing table of Posix names   22
933        Thanks to the folks working on Gregex for glib for this insight.
935    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
936        unicode-properties was also set.
938    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
940    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
941        checked only for CRLF.
943    14. Added casts to pcretest.c to avoid compiler warnings.
945    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
947    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
948        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
949        entirely. This removes changes made in 7 above.
951    17. The CMake files have been updated, and there is now more information about
952        building with CMake in the NON-UNIX-USE document.
955    Version 7.3 28-Aug-07
956    ---------------------
958     1. In the rejigging of the build system that eventually resulted in 7.1, the
959        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
960        brackets there is not right, since it causes compilers to look for an
961        installed pcre.h, not the version that is in the source that is being
962        compiled (which of course may be different). I have changed it back to:
964          #include "pcre.h"
966        I have a vague recollection that the change was concerned with compiling in
967        different directories, but in the new build system, that is taken care of
968        by the VPATH setting the Makefile.
970     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
971        when the subject happened to end in the byte 0x85 (e.g. if the last
972        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
973        characters but of course it shouldn't be taken as a newline when it is part
974        of another character. The bug was that, for an unlimited repeat of . in
975        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
976        characters when looking for a newline.
978     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
980     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
981        in debug output.
983     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
984        long printing in the pcrecpp unittest when running under MinGW.
986     6. ESC_K was left out of the EBCDIC table.
988     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
989        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
990        limit also applies to "virtual nesting" when a pattern is recursive, and in
991        this case 1000 isn't so big. I have been able to remove this limit at the
992        expense of backing off one optimization in certain circumstances. Normally,
993        when pcre_exec() would call its internal match() function recursively and
994        immediately return the result unconditionally, it uses a "tail recursion"
995        feature to save stack. However, when a subpattern that can match an empty
996        string has an unlimited repetition quantifier, it no longer makes this
997        optimization. That gives it a stack frame in which to save the data for
998        checking that an empty string has been matched. Previously this was taken
999        from the 1000-entry workspace that had been reserved. So now there is no
1000        explicit limit, but more stack is used.
1002     8. Applied Daniel's patches to solve problems with the import/export magic
1003        syntax that is required for Windows, and which was going wrong for the
1004        pcreposix and pcrecpp parts of the library. These were overlooked when this
1005        problem was solved for the main library.
1007     9. There were some crude static tests to avoid integer overflow when computing
1008        the size of patterns that contain repeated groups with explicit upper
1009        limits. As the maximum quantifier is 65535, the maximum group length was
1010        set at 30,000 so that the product of these two numbers did not overflow a
1011        32-bit integer. However, it turns out that people want to use groups that
1012        are longer than 30,000 bytes (though not repeat them that many times).
1013        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1014        made it possible to implement the integer overflow checks in a much more
1015        dynamic way, which I have now done. The artificial limitation on group
1016        length has been removed - we now have only the limit on the total length of
1017        the compiled pattern, which depends on the LINK_SIZE setting.
1019    10. Fixed a bug in the documentation for get/copy named substring when
1020        duplicate names are permitted. If none of the named substrings are set, the
1021        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1022        empty string.
1024    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1025        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1026        because the ] is interpreted as the first data character and the
1027        terminating ] is not found. PCRE has been made compatible with Perl in this
1028        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1029        cause memory overwriting.
1031    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1032        string has been matched (to stop an infinite loop). It was not recognizing
1033        a conditional subpattern that could match an empty string if that
1034        subpattern was within another subpattern. For example, it looped when
1035        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
1036        condition was not nested. This bug has been fixed.
1038    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1039        past the start of the subject in the presence of bytes with the top bit
1040        set, for example "\x8aBCD".
1042    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1043        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1045    14. Optimized (?!) to (*FAIL).
1047    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1048        This restricts code points to be within the range 0 to 0x10FFFF, excluding
1049        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1050        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1051        does: it's just the validity check that is more restrictive.
1053    16. Inserted checks for integer overflows during escape sequence (backslash)
1054        processing, and also fixed erroneous offset values for syntax errors during
1055        backslash processing.
1057    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1058        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1060    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1061        caused an overrun.
1063    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1064        something other than just ASCII characters) inside a group that had an
1065        unlimited repeat caused a loop at compile time (while checking to see
1066        whether the group could match an empty string).
1068    20. Debugging a pattern containing \p or \P could cause a crash. For example,
1069        [\P{Any}] did so. (Error in the code for printing property names.)
1071    21. An orphan \E inside a character class could cause a crash.
1073    22. A repeated capturing bracket such as (A)? could cause a wild memory
1074        reference during compilation.
1076    23. There are several functions in pcre_compile() that scan along a compiled
1077        expression for various reasons (e.g. to see if it's fixed length for look
1078        behind). There were bugs in these functions when a repeated \p or \P was
1079        present in the pattern. These operators have additional parameters compared
1080        with \d, etc, and these were not being taken into account when moving along
1081        the compiled data. Specifically:
1083        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1084            length.
1086        (b) An item such as \pL+ within a repeated group could cause crashes or
1087            loops.
1089        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1090            "reference to non-existent subpattern" error.
1092        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1094    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1095        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1097    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1099    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1100        character were causing crashes (broken optimization).
1102    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1103        \p or \P) caused a compile-time loop.
1105    28. More problems have arisen in unanchored patterns when CRLF is a valid line
1106        break. For example, the unstudied pattern [\r\n]A does not match the string
1107        "\r\nA" because change 7.0/46 below moves the current point on by two
1108        characters after failing to match at the start. However, the pattern \nA
1109        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
1110        the same is true. There doesn't seem any very clean way out of this, but
1111        what I have chosen to do makes the common cases work: PCRE now takes note
1112        of whether there can be an explicit match for \r or \n anywhere in the
1113        pattern, and if so, 7.0/46 no longer applies. As part of this change,
1114        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1115        pattern has explicit CR or LF references.
1117    29. Added (*CR) etc for changing newline setting at start of pattern.
1120    Version 7.2 19-Jun-07
1121    ---------------------
1123     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
1124        which is apparently normally available under Windows.
1126     2. Re-jig the pcregrep tests with different newline settings in an attempt
1127        to make them independent of the local environment's newline setting.
1129     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
1131     4. Some of the "internals" tests were previously cut out when the link size
1132        was not 2, because the output contained actual offsets. The recent new
1133        "Z" feature of pcretest means that these can be cut out, making the tests
1134        usable with all link sizes.
1136     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
1137        stack recursion. This gives a massive performance boost under BSD, but just
1138        a small improvement under Linux. However, it saves one field in the frame
1139        in all cases.
1141     6. Added more features from the forthcoming Perl 5.10:
1143        (a) (?-n) (where n is a string of digits) is a relative subroutine or
1144            recursion call. It refers to the nth most recently opened parentheses.
1146        (b) (?+n) is also a relative subroutine call; it refers to the nth next
1147            to be opened parentheses.
1149        (c) Conditions that refer to capturing parentheses can be specified
1150            relatively, for example, (?(-2)... or (?(+3)...
1152        (d) \K resets the start of the current match so that everything before
1153            is not part of it.
1155        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1157        (f) \g{name} is another synonym - part of Perl 5.10's unification of
1158            reference syntax.
1160        (g) (?| introduces a group in which the numbering of parentheses in each
1161            alternative starts with the same number.
1163        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
1165     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
1168     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
1169        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
1170        for detecting groups that can match an empty string.
1172     9. A pattern with a very large number of alternatives (more than several
1173        hundred) was running out of internal workspace during the pre-compile
1174        phase, where pcre_compile() figures out how much memory will be needed. A
1175        bit of new cunning has reduced the workspace needed for groups with
1176        alternatives. The 1000-alternative test pattern now uses 12 bytes of
1177        workspace instead of running out of the 4096 that are available.
1179    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
1181    11. Applied patch from Google to remove an optimization that didn't quite work.
1182        The report of the bug said:
1184          pcrecpp::RE("a*").FullMatch("aaa") matches, while
1185          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
1186          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
1188    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
1189        it matched the wrong number of bytes.
1192    Version 7.1 24-Apr-07
1193    ---------------------
1195     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
1196        that is more "standard", making use of automake and other Autotools. There
1197        is some re-arrangement of the files and adjustment of comments consequent
1198        on this.
1200     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
1201        for recursive directory scanning broke on some systems because the files
1202        are not scanned in any specific order and on different systems the order
1203        was different. A call to "sort" has been inserted into RunGrepTest for the
1204        approprate test as a short-term fix. In the longer term there may be an
1205        alternative.
1207     3. I had an email from Eric Raymond about problems translating some of PCRE's
1208        man pages to HTML (despite the fact that I distribute HTML pages, some
1209        people do their own conversions for various reasons). The problems
1210        concerned the use of low-level troff macros .br and .in. I have therefore
1211        removed all such uses from the man pages (some were redundant, some could
1212        be replaced by .nf/.fi pairs). The 132html script that I use to generate
1213        HTML has been updated to handle .nf/.fi and to complain if it encounters
1214        .br or .in.
1216     4. Updated comments in configure.ac that get placed in config.h.in and also
1217        arranged for config.h to be included in the distribution, with the name
1218        config.h.generic, for the benefit of those who have to compile without
1219        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
1221     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
1222        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
1223        makevp.bat for the current PCRE, using the additional files
1224        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
1226     6. A Windows user reported a minor discrepancy with test 2, which turned out
1227        to be caused by a trailing space on an input line that had got lost in his
1228        copy. The trailing space was an accident, so I've just removed it.
1230     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
1231        that is needed.
1233     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
1234        as "const" (a) because they are and (b) because it helps the PHP
1235        maintainers who have recently made a script to detect big data structures
1236        in the php code that should be moved to the .rodata section. I remembered
1237        to update Builducptable as well, so it won't revert if ucptable.h is ever
1238        re-created.
1240     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
1241        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
1242        order to be able to cut out the UTF-8 tables in the latter when UTF-8
1243        support is not required. This saves 1.5-2K of code, which is important in
1244        some applications.
1246        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
1247        so as not to refer to the tables, even though these functions will never be
1248        called when UTF-8 support is disabled. Otherwise there are problems with a
1249        shared library.
1251    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
1253        (a) It was defining its arguments as char * instead of void *.
1255        (b) It was assuming that all moves were upwards in memory; this was true
1256            a long time ago when I wrote it, but is no longer the case.
1258        The emulated memove() is provided for those environments that have neither
1259        memmove() nor bcopy(). I didn't think anyone used it these days, but that
1260        is clearly not the case, as these two bugs were recently reported.
1262    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
1263        and Detrail to create the HTML documentation, the .txt form of the man
1264        pages, and it removes trailing spaces from listed files. It also creates
1265        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
1266        case, it wraps all the #defines with #ifndefs. This script should be run
1267        before "make dist".
1269    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1270        with Unicode property support.
1272        (a) For a maximizing quantifier, if the two different cases of the
1273            character were of different lengths in their UTF-8 codings (there are
1274            some cases like this - I found 11), and the matching function had to
1275            back up over a mixture of the two cases, it incorrectly assumed they
1276            were both the same length.
1278        (b) When PCRE was configured to use the heap rather than the stack for
1279            recursion during matching, it was not correctly preserving the data for
1280            the other case of a UTF-8 character when checking ahead for a match
1281            while processing a minimizing repeat. If the check also involved
1282            matching a wide character, but failed, corruption could cause an
1283            erroneous result when trying to check for a repeat of the original
1284            character.
1286    13. Some tidying changes to the testing mechanism:
1288        (a) The RunTest script now detects the internal link size and whether there
1289            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1290            values substituted by "configure". (The RunGrepTest script already did
1291            this for UTF-8.) The configure.ac script no longer substitutes the
1292            relevant variables.
1294        (b) The debugging options /B and /D in pcretest show the compiled bytecode
1295            with length and offset values. This means that the output is different
1296            for different internal link sizes. Test 2 is skipped for link sizes
1297            other than 2 because of this, bypassing the problem. Unfortunately,
1298            there was also a test in test 3 (the locale tests) that used /B and
1299            failed for link sizes other than 2. Rather than cut the whole test out,
1300            I have added a new /Z option to pcretest that replaces the length and
1301            offset values with spaces. This is now used to make test 3 independent
1302            of link size. (Test 2 will be tidied up later.)
1304    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1305        segmentation fault instead of returning the appropriate error message.
1307    15. In multiline mode when the newline sequence was set to "any", the pattern
1308        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1309        This doesn't seem right; it now treats the CRLF combination as the line
1310        ending, and so does not match in that case. It's only a pattern such as ^$
1311        that would hit this one: something like ^ABC$ would have failed after \r
1312        and then tried again after \r\n.
1314    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1315        in an attempt to make files that differ only in their line terminators
1316        compare equal. This works on Linux.
1318    17. Under certain error circumstances pcregrep might try to free random memory
1319        as it exited. This is now fixed, thanks to valgrind.
1321    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1322        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1323        was because its rules for how to advance for /g after matching an empty
1324        string at the end of a line did not allow for this case. They now check for
1325        it specially.
1327    20. pcretest is supposed to handle patterns and data of any length, by
1328        extending its buffers when necessary. It was getting this wrong when the
1329        buffer for a data line had to be extended.
1331    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1332        CRLF as a newline sequence.
1334    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1335        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1336        I have nevertheless tidied it up.
1338    23. Added some casts to kill warnings from HP-UX ia64 compiler.
1340    24. Added a man page for pcre-config.
1343    Version 7.0 19-Dec-06
1344    ---------------------
1346     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1347        moving to gcc 4.1.1.
1349     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1350        sys/time.h, which is documented as needed for this function. It doesn't
1351        seem to matter on Linux, but it showed up on some releases of OS X.
1353     3. It seems that there are systems where bytes whose values are greater than
1354        127 match isprint() in the "C" locale. The "C" locale should be the
1355        default when a C program starts up. In most systems, only ASCII printing
1356        characters match isprint(). This difference caused the output from pcretest
1357        to vary, making some of the tests fail. I have changed pcretest so that:
1359        (a) When it is outputting text in the compiled version of a pattern, bytes
1360            other than 32-126 are always shown as hex escapes.
1362        (b) When it is outputting text that is a matched part of a subject string,
1363            it does the same, unless a different locale has been set for the match
1364            (using the /L modifier). In this case, it uses isprint() to decide.
1366     4. Fixed a major bug that caused incorrect computation of the amount of memory
1367        required for a compiled pattern when options that changed within the
1368        pattern affected the logic of the preliminary scan that determines the
1369        length. The relevant options are -x, and -i in UTF-8 mode. The result was
1370        that the computed length was too small. The symptoms of this bug were
1371        either the PCRE error "internal error: code overflow" from pcre_compile(),
1372        or a glibc crash with a message such as "pcretest: free(): invalid next
1373        size (fast)". Examples of patterns that provoked this bug (shown in
1374        pcretest format) are:
1376          /(?-x: )/x
1377          /(?x)(?-x: \s*#\s*)/
1378          /((?i)[\x{c0}])/8
1379          /(?i:[\x{c0}])/8
1381        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1382        is now done differently.
1384     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1385        wrapper classes; (b) implement a new function in the C++ scanner that is
1386        more efficient than the old way of doing things because it avoids levels of
1387        recursion in the regex matching; (c) add a paragraph to the documentation
1388        for the FullMatch() function.
1390     6. The escape sequence \n was being treated as whatever was defined as
1391        "newline". Not only was this contrary to the documentation, which states
1392        that \n is character 10 (hex 0A), but it also went horribly wrong when
1393        "newline" was defined as CRLF. This has been fixed.
1395     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1396        was being set to -1 for the "end of line" case (supposedly a value that no
1397        character can have). Though this value is never used (the check for end of
1398        line is "zero bytes in current character"), it caused compiler complaints.
1399        I've changed it to 0xffffffff.
1401     8. In pcre_version.c, the version string was being built by a sequence of
1402        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1403        string (as it is for production releases) called a macro with an empty
1404        argument. The C standard says the result of this is undefined. The gcc
1405        compiler treats it as an empty string (which was what was wanted) but it is
1406        reported that Visual C gives an error. The source has been hacked around to
1407        avoid this problem.
1409     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1410        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1411        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1412        of them did).
1414    10. Originally, pcretest opened its input and output without "b"; then I was
1415        told that "b" was needed in some environments, so it was added for release
1416        5.0 to both the input and output. (It makes no difference on Unix-like
1417        systems.) Later I was told that it is wrong for the input on Windows. I've
1418        now abstracted the modes into two macros, to make it easier to fiddle with
1419        them, and removed "b" from the input mode under Windows.
1421    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1423    12. Added -help and --help to pcretest as an official way of being reminded
1424        of the options.
1426    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1427        and pcrecpp.cc because they annoy compilers at high warning levels.
1429    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1431    15. Fixed an occurrence of == in configure.ac that should have been = (shell
1432        scripts are not C programs :-) and which was not noticed because it works
1433        on Linux.
1435    16. pcretest is supposed to handle any length of pattern and data line (as one
1436        line or as a continued sequence of lines) by extending its input buffer if
1437        necessary. This feature was broken for very long pattern lines, leading to
1438        a string of junk being passed to pcre_compile() if the pattern was longer
1439        than about 50K.
1441    17. I have done a major re-factoring of the way pcre_compile() computes the
1442        amount of memory needed for a compiled pattern. Previously, there was code
1443        that made a preliminary scan of the pattern in order to do this. That was
1444        OK when PCRE was new, but as the facilities have expanded, it has become
1445        harder and harder to keep it in step with the real compile phase, and there
1446        have been a number of bugs (see for example, 4 above). I have now found a
1447        cunning way of running the real compile function in a "fake" mode that
1448        enables it to compute how much memory it would need, while actually only
1449        ever using a few hundred bytes of working memory and without too many
1450        tests of the mode. This should make future maintenance and development
1451        easier. A side effect of this work is that the limit of 200 on the nesting
1452        depth of parentheses has been removed (though this was never a serious
1453        limitation, I suspect). However, there is a downside: pcre_compile() now
1454        runs more slowly than before (30% or more, depending on the pattern). I
1455        hope this isn't a big issue. There is no effect on runtime performance.
1457    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1458        newline (only possible for the last line of a file) and it was a
1459        pattern that set a locale (followed by /Lsomething), pcretest crashed.
1461    19. Added additional timing features to pcretest. (1) The -tm option now times
1462        matching only, not compiling. (2) Both -t and -tm can be followed, as a
1463        separate command line item, by a number that specifies the number of
1464        repeats to use when timing. The default is 50000; this gives better
1465        precision, but takes uncomfortably long for very large patterns.
1467    20. Extended pcre_study() to be more clever in cases where a branch of a
1468        subpattern has no definite first character. For example, (a*|b*)[cd] would
1469        previously give no result from pcre_study(). Now it recognizes that the
1470        first character must be a, b, c, or d.
1472    21. There was an incorrect error "recursive call could loop indefinitely" if
1473        a subpattern (or the entire pattern) that was being tested for matching an
1474        empty string contained only one non-empty item after a nested subpattern.
1475        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1476        incorrectly, because the \d was being skipped in the check.
1478    22. The pcretest program now has a new pattern option /B and a command line
1479        option -b, which is equivalent to adding /B to every pattern. This causes
1480        it to show the compiled bytecode, without the additional information that
1481        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1482        is the same as /B/I).
1484    23. A new optimization is now able automatically to treat some sequences such
1485        as a*b as a*+b. More specifically, if something simple (such as a character
1486        or a simple class like \d) has an unlimited quantifier, and is followed by
1487        something that cannot possibly match the quantified thing, the quantifier
1488        is automatically "possessified".
1490    24. A recursive reference to a subpattern whose number was greater than 39
1491        went wrong under certain circumstances in UTF-8 mode. This bug could also
1492        have affected the operation of pcre_study().
1494    25. Realized that a little bit of performance could be had by replacing
1495        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1497    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1499    27. Possessive quantifiers such as a++ were previously implemented by turning
1500        them into atomic groups such as ($>a+). Now they have their own opcodes,
1501        which improves performance. This includes the automatically created ones
1502        from 23 above.
1504    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1505        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1506        the first matched character to be a colon. This applied both to named and
1507        numbered groups.
1509    29. The ucpinternal.h header file was missing its idempotency #ifdef.
1511    30. I was sent a "project" file called libpcre.a.dev which I understand makes
1512        building PCRE on Windows easier, so I have included it in the distribution.
1514    31. There is now a check in pcretest against a ridiculously large number being
1515        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1516        loop, the loop is abandoned.
1518    32. Forward references to subpatterns in conditions such as (?(2)...) where
1519        subpattern 2 is defined later cause pcre_compile() to search forwards in
1520        the pattern for the relevant set of parentheses. This search went wrong
1521        when there were unescaped parentheses in a character class, parentheses
1522        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1524    33. "Subroutine" calls and backreferences were previously restricted to
1525        referencing subpatterns earlier in the regex. This restriction has now
1526        been removed.
1528    34. Added a number of extra features that are going to be in Perl 5.10. On the
1529        whole, these are just syntactic alternatives for features that PCRE had
1530        previously implemented using the Python syntax or my own invention. The
1531        other formats are all retained for compatibility.
1533        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1534            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1535            also .NET compatible.
1537        (b) A recursion or subroutine call to a named group can now be defined as
1538            (?&name) as well as (?P>name).
1540        (c) A backreference to a named group can now be defined as \k<name> or
1541            \k'name' as well as (?P=name). The new forms, as well as being in Perl
1542            5.10, are also .NET compatible.
1544        (d) A conditional reference to a named group can now use the syntax
1545            (?(<name>) or (?('name') as well as (?(name).
1547        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1548            groups (named and numbered) that are never evaluated inline, but can be
1549            called as "subroutines" from elsewhere. In effect, the DEFINE condition
1550            is always false. There may be only one alternative in such a group.
1552        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1553            as the simple (?(R). The condition is true only if the most recent
1554            recursion is that of the given number or name. It does not search out
1555            through the entire recursion stack.
1557        (g) The escape \gN or \g{N} has been added, where N is a positive or
1558            negative number, specifying an absolute or relative reference.
1560    35. Tidied to get rid of some further signed/unsigned compiler warnings and
1561        some "unreachable code" warnings.
1563    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1564        things, this adds five new scripts.
1566    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1567        There were also incompatibilities regarding the handling of \Q..\E inside
1568        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1569        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1571    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1572        matches an empty string, and forcibly breaks the loop. There were bugs in
1573        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
1574        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
1575        separate and independent bugs (that affected different cases) have been
1576        fixed.
1578    39. Refactored the code to abolish the use of different opcodes for small
1579        capturing bracket numbers. This is a tidy that I avoided doing when I
1580        removed the limit on the number of capturing brackets for 3.5 back in 2001.
1581        The new approach is not only tidier, it makes it possible to reduce the
1582        memory needed to fix the previous bug (38).
1584    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1585        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1586        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1587        mode.
1589    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1590        report.
1592    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1593        copy construction and assignment in the C++ wrapper.
1595    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1596        couple of bugs that could have given wrong results in the "--newline=crlf"
1597        case.
1599    44. Added a number of casts and did some reorganization of signed/unsigned int
1600        variables following suggestions from Dair Grant. Also renamed the variable
1601        "this" as "item" because it is a C++ keyword.
1603    45. Arranged for dftables to add
1605          #include "pcre_internal.h"
1607        to pcre_chartables.c because without it, gcc 4.x may remove the array
1608        definition from the final binary if PCRE is built into a static library and
1609        dead code stripping is activated.
1611    46. For an unanchored pattern, if a match attempt fails at the start of a
1612        newline sequence, and the newline setting is CRLF or ANY, and the next two
1613        characters are CRLF, advance by two characters instead of one.
1616    Version 6.7 04-Jul-06
1617    ---------------------
1619     1. In order to handle tests when input lines are enormously long, pcretest has
1620        been re-factored so that it automatically extends its buffers when
1621        necessary. The code is crude, but this _is_ just a test program. The
1622        default size has been increased from 32K to 50K.
1624     2. The code in pcre_study() was using the value of the re argument before
1625        testing it for NULL. (Of course, in any sensible call of the function, it
1626        won't be NULL.)
1628     3. The memmove() emulation function in pcre_internal.h, which is used on
1629        systems that lack both memmove() and bcopy() - that is, hardly ever -
1630        was missing a "static" storage class specifier.
1632     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1633        containing an extended class (one that cannot be represented by a bitmap
1634        because it contains high-valued characters or Unicode property items, e.g.
1635        [\pZ]). Almost always one would set UTF-8 mode when processing such a
1636        pattern, but PCRE should not loop if you do not (it no longer does).
1637        [Detail: two cases were found: (a) a repeated subpattern containing an
1638        extended class; (b) a recursive reference to a subpattern that followed a
1639        previous extended class. It wasn't skipping over the extended class
1640        correctly when UTF-8 mode was not set.]
1642     5. A negated single-character class was not being recognized as fixed-length
1643        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1644        compile error "lookbehind assertion is not fixed length".
1646     6. The RunPerlTest auxiliary script was showing an unexpected difference
1647        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1648        write a Perl script that can interpret lines of an input file either as
1649        byte characters or as UTF-8, which is what "perltest" was being required to
1650        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1651        can't do is switch easily at run time between having the "use utf8;" pragma
1652        or not. In the end, I fudged it by using the RunPerlTest script to insert
1653        "use utf8;" explicitly for the UTF-8 tests.
1655     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1656        the end of the subject string, contrary to the documentation and to what
1657        Perl does. This was true of both matching functions. Now it matches only at
1658        the start of the subject and immediately after *internal* newlines.
1660     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1661        a pointer to an int instead of a pointer to an unsigned long int. This
1662        caused problems on 64-bit systems.
1664     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1665        instance of the 'standard' template library not being so standard".
1667    10. There was no check on the number of named subpatterns nor the maximum
1668        length of a subpattern name. The product of these values is used to compute
1669        the size of the memory block for a compiled pattern. By supplying a very
1670        long subpattern name and a large number of named subpatterns, the size
1671        computation could be caused to overflow. This is now prevented by limiting
1672        the length of names to 32 characters, and the number of named subpatterns
1673        to 10,000.
1675    11. Subpatterns that are repeated with specific counts have to be replicated in
1676        the compiled pattern. The size of memory for this was computed from the
1677        length of the subpattern and the repeat count. The latter is limited to
1678        65535, but there was no limit on the former, meaning that integer overflow
1679        could in principle occur. The compiled length of a repeated subpattern is
1680        now limited to 30,000 bytes in order to prevent this.
1682    12. Added the optional facility to have named substrings with the same name.
1684    13. Added the ability to use a named substring as a condition, using the
1685        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1686        are numbers (not recommended). Forward references are permitted.
1688    14. Added forward references in named backreferences (if you see what I mean).
1690    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1691        pattern could run off the end of the subject. For example, the pattern
1692        "(?s)(.{1,5})"8 did this with the subject "ab".
1694    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1695        PCRE_CASELESS was set when matching characters that were quantified with ?
1696        or *.
1698    17. A character class other than a single negated character that had a minimum
1699        but no maximum quantifier - for example [ab]{6,} - was not handled
1700        correctly by pce_dfa_exec(). It would match only one character.
1702    18. A valid (though odd) pattern that looked like a POSIX character
1703        class but used an invalid character after [ (for example [[,abc,]]) caused
1704        pcre_compile() to give the error "Failed: internal error: code overflow" or
1705        in some cases to crash with a glibc free() error. This could even happen if
1706        the pattern terminated after [[ but there just happened to be a sequence of
1707        letters, a binary zero, and a closing ] in the memory that followed.
1709    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1710        over the years. Originally (before any Unicode support), just the bottom 8
1711        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1712        output from "man perlunicode" includes this:
1714          The regular expression compiler produces polymorphic opcodes.  That
1715          is, the pattern adapts to the data and automatically switches to
1716          the Unicode character scheme when presented with Unicode data--or
1717          instead uses a traditional byte scheme when presented with byte
1718          data.
1720        Sadly, a wide octal escape does not cause a switch, and in a string with
1721        no other multibyte characters, these octal escapes are treated as before.
1722        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
1723        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1724        Unicode string.
1726        I have not perpetrated such confusion in PCRE. Up till now, it took just
1727        the bottom 8 bits, as in old Perl. I have now made octal escapes with
1728        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1729        translate to the appropriate multibyte character.
1731    29. Applied some refactoring to reduce the number of warnings from Microsoft
1732        and Borland compilers. This has included removing the fudge introduced
1733        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1734        a warning about an unused variable.
1736    21. PCRE has not included VT (character 0x0b) in the set of whitespace
1737        characters since release 4.0, because Perl (from release 5.004) does not.
1738        [Or at least, is documented not to: some releases seem to be in conflict
1739        with the documentation.] However, when a pattern was studied with
1740        pcre_study() and all its branches started with \s, PCRE still included VT
1741        as a possible starting character. Of course, this did no harm; it just
1742        caused an unnecessary match attempt.
1744    22. Removed a now-redundant internal flag bit that recorded the fact that case
1745        dependency changed within the pattern. This was once needed for "required
1746        byte" processing, but is no longer used. This recovers a now-scarce options
1747        bit. Also moved the least significant internal flag bit to the most-
1748        significant bit of the word, which was not previously used (hangover from
1749        the days when it was an int rather than a uint) to free up another bit for
1750        the future.
1752    23. Added support for CRLF line endings as well as CR and LF. As well as the
1753        default being selectable at build time, it can now be changed at runtime
1754        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1755        specify that it is scanning data with non-default line endings.
1757    24. Changed the definition of CXXLINK to make it agree with the definition of
1758        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1760    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1761        recursions. This makes a big different to stack usage for some patterns.
1763    26. If a subpattern containing a named recursion or subroutine reference such
1764        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1765        the space required for the compiled pattern went wrong and gave too small a
1766        value. Depending on the environment, this could lead to "Failed: internal
1767        error: code overflow at offset 49" or "glibc detected double free or
1768        corruption" errors.
1770    27. Applied patches from Google (a) to support the new newline modes and (b) to
1771        advance over multibyte UTF-8 characters in GlobalReplace.
1773    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1774        difference for some implementation of PCRE in some Windows version.
1776    29. Added some extra testing facilities to pcretest:
1778        \q<number>   in a data line sets the "match limit" value
1779        \Q<number>   in a data line sets the "match recursion limt" value
1780        -S <number>  sets the stack size, where <number> is in megabytes
1782        The -S option isn't available for Windows.
1785    Version 6.6 06-Feb-06
1786    ---------------------
1788     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1789        in pcreposix.h. I have copied the definition from pcre.h.
1791     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1792        because pcre.h is no longer a built file.
1794     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1795        not normally included in the compiled code.
1798  Version 6.5 01-Feb-06  Version 6.5 01-Feb-06
1799  ---------------------  ---------------------

Removed from v.87  
changed lines
  Added in v.515

  ViewVC Help
Powered by ViewVC 1.1.5