ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 19 by nigel, Sat Feb 24 21:38:33 2007 UTC revision 517 by ph10, Wed May 5 10:44:20 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.
29    7.  Minor change to pcretest.c to avoid a compiler warning.
31    8.  Added four artifical Unicode properties to help with an option to make
32        \s etc use properties. The new properties are: Xan (alphanumeric), Xsp
33        (Perl space), Xps (POSIX space), and Xwd (word).
36    Version 8.02 19-Mar-2010
37    ------------------------
39    1.  The Unicode data tables have been updated to Unicode 5.2.0.
41    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
42        configured.
44    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
45        original author of that file, following a query about its status.
47    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
48        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
50    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
51        quantifier applied to a forward-referencing subroutine call, could compile
52        incorrect code or give the error "internal error: previously-checked
53        referenced subpattern not found".
55    6.  Both MS Visual Studio and Symbian OS have problems with initializing
56        variables to point to external functions. For these systems, therefore,
57        pcre_malloc etc. are now initialized to local functions that call the
58        relevant global functions.
60    7.  There were two entries missing in the vectors called coptable and poptable
61        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
62        I've fixed the data, and added a kludgy way of testing at compile time that
63        the lengths are correct (equal to the number of opcodes).
65    8.  Following on from 7, I added a similar kludge to check the length of the
66        eint vector in pcreposix.c.
68    9.  Error texts for pcre_compile() are held as one long string to avoid too
69        much relocation at load time. To find a text, the string is searched,
70        counting zeros. There was no check for running off the end of the string,
71        which could happen if a new error number was added without updating the
72        string.
74    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
76    11. \K was not working if it appeared in an atomic group or in a group that
77        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
78        \K is "not well defined" if used in an assertion. PCRE now accepts it if
79        the assertion is positive, but not if it is negative.
81    12. Change 11 fortuitously reduced the size of the stack frame used in the
82        "match()" function of pcre_exec.c by one pointer. Forthcoming
83        implementation of support for (*MARK) will need an extra pointer on the
84        stack; I have reserved it now, so that the stack frame size does not
85        decrease.
87    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
88        item in branch that calls a recursion is a subroutine call - as in the
89        second branch in the above example - was incorrectly given the compile-
90        time error "recursive call could loop indefinitely" because pcre_compile()
91        was not correctly checking the subroutine for matching a non-empty string.
93    14. The checks for overrunning compiling workspace could trigger after an
94        overrun had occurred. This is a "should never occur" error, but it can be
95        triggered by pathological patterns such as hundreds of nested parentheses.
96        The checks now trigger 100 bytes before the end of the workspace.
98    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
101    Version 8.01 19-Jan-2010
102    ------------------------
104    1.  If a pattern contained a conditional subpattern with only one branch (in
105        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
106        computed the wrong minimum data length (which is of course zero for such
107        subpatterns). This could cause incorrect "no match" results.
109    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
110        the pattern is reset in the first branch, pcre_compile() failed with
111        "internal error: code overflow at offset...". This happened only when
112        the reset was to the original external option setting. (An optimization
113        abstracts leading options settings into an external setting, which was the
114        cause of this.)
116    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
117        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
118        assertion pattern did not match (meaning that the assertion was true), it
119        was incorrectly treated as false if the SKIP had been reached during the
120        matching. This also applied to assertions used as conditions.
122    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
123        assertion subpattern, including such a pattern used as a condition,
124        unpredictable results occurred, instead of the error return
127    5.  The C++ GlobalReplace function was not working like Perl for the special
128        situation when an empty string is matched. It now does the fancy magic
129        stuff that is necessary.
131    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
132        removed. (These were left over from very, very early versions of PCRE.)
134    7.  Some cosmetic changes to the code to make life easier when compiling it
135        as part of something else:
137        (a) Change DEBUG to PCRE_DEBUG.
139        (b) In pcre_compile(), rename the member of the "branch_chain" structure
140            called "current" as "current_branch", to prevent a collision with the
141            Linux macro when compiled as a kernel module.
143        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
144            prevent a collision with the Linux macro when compiled as a kernel
145            module.
147    8.  In pcre_compile() there are some checks for integer overflows that used to
148        cast potentially large values to (double). This has been changed to that
149        when building, a check for int64_t is made, and if it is found, it is used
150        instead, thus avoiding the use of floating point arithmetic. (There is no
151        other use of FP in PCRE.) If int64_t is not found, the fallback is to
152        double.
154    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
155        2005 (difference between two addresses compared to an unsigned value).
157    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
158        custom one, because of the following reported problem in Windows:
160          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
161              under Win32.
162          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
163              therefore missing the function definition.
164          - The compiler thus generates a "C" signature for the test function.
165          - The linker fails to find the "C" function.
166          - PCRE fails to configure if asked to do so against libbz2.
168    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
169        messages were output:
171          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
172          rerunning libtoolize, to keep the correct libtool macros in-tree.
173          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
175        I have done both of these things.
177    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
178        most of the time, it *can* run out if it is given a pattern that contains a
179        runaway infinite recursion. I updated the discussion in the pcrestack man
180        page.
182    13. Now that we have gone to the x.xx style of version numbers, the minor
183        version may start with zero. Using 08 or 09 is a bad idea because users
184        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
185        interpreted as invalid octal numbers. I've updated the previous comment in
186        configure.ac, and also added a check that gives an error if 08 or 09 are
187        used.
189    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
190        causing partial matching to fail when the end of the subject matched \W
191        in a UTF-8 pattern where \W was quantified with a minimum of 3.
193    15. There were some discrepancies between the declarations in pcre_internal.h
194        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
195        their definitions. The declarations used "const uschar *" and the
196        definitions used USPTR. Even though USPTR is normally defined as "const
197        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
198        reported that: "This difference in casting confuses some C++ compilers, for
199        example, SunCC recognizes above declarations as different functions and
200        generates broken code for hbpcre." I have changed the declarations to use
201        USPTR.
203    16. GNU libtool is named differently on some systems. The autogen.sh script now
204        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
205        (FreeBSD).
207    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
208        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
209        comment: "Figure out how to create a longlong from a string: strtoll and
210        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
211        instance, but it only takes 2 args instead of 3!"
213    18. A subtle bug concerned with back references has been fixed by a change of
214        specification, with a corresponding code fix. A pattern such as
215        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
216        refers, was giving matches when it shouldn't. For example, xa=xaaa would
217        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
218        same bug. Such groups have to be quantified to be useful, or contained
219        inside another quantified group. (If there's no repetition, the reference
220        can never match.) The problem arises because, having left the group and
221        moved on to the rest of the pattern, a later failure that backtracks into
222        the group uses the captured value from the final iteration of the group
223        rather than the correct earlier one. I have fixed this in PCRE by forcing
224        any group that contains a reference to itself to be an atomic group; that
225        is, there cannot be any backtracking into it once it has completed. This is
226        similar to recursive and subroutine calls.
229    Version 8.00 19-Oct-09
230    ----------------------
232    1.  The table for translating pcre_compile() error codes into POSIX error codes
233        was out-of-date, and there was no check on the pcre_compile() error code
234        being within the table. This could lead to an OK return being given in
235        error.
237    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
238        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
239        in a Windows environment.
241    3.  The pcregrep --count option prints the count for each file even when it is
242        zero, as does GNU grep. However, pcregrep was also printing all files when
243        --files-with-matches was added. Now, when both options are given, it prints
244        counts only for those files that have at least one match. (GNU grep just
245        prints the file name in this circumstance, but including the count seems
246        more useful - otherwise, why use --count?) Also ensured that the
247        combination -clh just lists non-zero counts, with no names.
249    4.  The long form of the pcregrep -F option was incorrectly implemented as
250        --fixed_strings instead of --fixed-strings. This is an incompatible change,
251        but it seems right to fix it, and I didn't think it was worth preserving
252        the old behaviour.
254    5.  The command line items --regex=pattern and --regexp=pattern were not
255        recognized by pcregrep, which required --regex pattern or --regexp pattern
256        (with a space rather than an '='). The man page documented the '=' forms,
257        which are compatible with GNU grep; these now work.
259    6.  No libpcreposix.pc file was created for pkg-config; there was just
260        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
262    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
263        when UCP support is not needed, by modifying the Python script that
264        generates it from Unicode data files. This should not matter if the module
265        is correctly used as a library, but I received one complaint about 50K of
266        unwanted data. My guess is that the person linked everything into his
267        program rather than using a library. Anyway, it does no harm.
269    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
270        was a minimum greater than 1 for a wide character in a possessive
271        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
272        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
273        character. Chaos in the form of incorrect output or a compiling loop could
274        result.
276    9.  The restrictions on what a pattern can contain when partial matching is
277        requested for pcre_exec() have been removed. All patterns can now be
278        partially matched by this function. In addition, if there are at least two
279        slots in the offset vector, the offset of the earliest inspected character
280        for the match and the offset of the end of the subject are set in them when
281        PCRE_ERROR_PARTIAL is returned.
283    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
284        synonymous with PCRE_PARTIAL, for backwards compatibility, and
285        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
286        and may be more useful for multi-segment matching.
288    11. Partial matching with pcre_exec() is now more intuitive. A partial match
289        used to be given if ever the end of the subject was reached; now it is
290        given only if matching could not proceed because another character was
291        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
292        string "Z", which now yields "no match" instead of "partial match". In the
293        case of pcre_dfa_exec(), "no match" is given if every matching path for the
294        final character ended with (*FAIL).
296    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
297        if the pattern had a "must contain" character that was already found in the
298        earlier partial match, unless partial matching was again requested. For
299        example, with the pattern /dog.(body)?/, the "must contain" character is
300        "g". If the first part-match was for the string "dog", restarting with
301        "sbody" failed. This bug has been fixed.
303    13. The string returned by pcre_dfa_exec() after a partial match has been
304        changed so that it starts at the first inspected character rather than the
305        first character of the match. This makes a difference only if the pattern
306        starts with a lookbehind assertion or \b or \B (\K is not supported by
307        pcre_dfa_exec()). It's an incompatible change, but it makes the two
308        matching functions compatible, and I think it's the right thing to do.
310    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
311        so that the demonstration program is easily available in environments where
312        PCRE has not been installed from source.
314    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
315        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
316        library.
318    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
319        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
320        is not the first non-POSIX option to be added. Clearly some people find
321        these options useful.
323    17. If a caller to the POSIX matching function regexec() passes a non-zero
324        value for nmatch with a NULL value for pmatch, the value of
325        nmatch is forced to zero.
327    18. RunGrepTest did not have a test for the availability of the -u option of
328        the diff command, as RunTest does. It now checks in the same way as
329        RunTest, and also checks for the -b option.
331    19. If an odd number of negated classes containing just a single character
332        interposed, within parentheses, between a forward reference to a named
333        subpattern and the definition of the subpattern, compilation crashed with
334        an internal error, complaining that it could not find the referenced
335        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
336        [The bug was that it was starting one character too far in when skipping
337        over the character class, thus treating the ] as data rather than
338        terminating the class. This meant it could skip too much.]
340    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
341        /g option in pcretest when the pattern contains \K, which makes it possible
342        to have an empty string match not at the start, even when the pattern is
343        anchored. Updated pcretest and pcredemo to use this option.
345    21. If the maximum number of capturing subpatterns in a recursion was greater
346        than the maximum at the outer level, the higher number was returned, but
347        with unset values at the outer level. The correct (outer level) value is
348        now given.
350    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
351        PCRE did not set those parentheses (unlike Perl). I have now found a way to
352        make it do so. The string so far is captured, making this feature
353        compatible with Perl.
355    23. The tests have been re-organized, adding tests 11 and 12, to make it
356        possible to check the Perl 5.10 features against Perl 5.10.
358    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
359        pattern matches a fixed length string. PCRE did not allow this; now it
360        does. Neither allows recursion.
362    25. I finally figured out how to implement a request to provide the minimum
363        length of subject string that was needed in order to match a given pattern.
364        (It was back references and recursion that I had previously got hung up
365        on.) This code has now been added to pcre_study(); it finds a lower bound
366        to the length of subject needed. It is not necessarily the greatest lower
367        bound, but using it to avoid searching strings that are too short does give
368        some useful speed-ups. The value is available to calling programs via
369        pcre_fullinfo().
371    26. While implementing 25, I discovered to my embarrassment that pcretest had
372        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
373        study optimizations had never been tested with that matching function.
374        Oops. What is worse, even when it was passed study data, there was a bug in
375        pcre_dfa_exec() that meant it never actually used it. Double oops. There
376        were also very few tests of studied patterns with pcre_dfa_exec().
378    27. If (?| is used to create subpatterns with duplicate numbers, they are now
379        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
380        on the other side of the coin, they are no longer allowed to have different
381        names, because these cannot be distinguished in PCRE, and this has caused
382        confusion. (This is a difference from Perl.)
384    28. When duplicate subpattern names are present (necessarily with different
385        numbers, as required by 27 above), and a test is made by name in a
386        conditional pattern, either for a subpattern having been matched, or for
387        recursion in such a pattern, all the associated numbered subpatterns are
388        tested, and the overall condition is true if the condition is true for any
389        one of them. This is the way Perl works, and is also more like the way
390        testing by number works.
393    Version 7.9 11-Apr-09
394    ---------------------
396    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
397        (pcretest), all targets were linked against these libraries. This included
398        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
399        libraries. This caused unwanted dependencies to be created. This problem
400        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
401        pcretest is linked with readline.
403    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
404        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
405        moved outside it again, because FALSE and TRUE are already defined in AIX,
406        but BOOL is not.
408    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
409        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
411    4.  The pcregrep documentation said spaces were inserted as well as colons (or
412        hyphens) following file names and line numbers when outputting matching
413        lines. This is not true; no spaces are inserted. I have also clarified the
414        wording for the --colour (or --color) option.
416    5.  In pcregrep, when --colour was used with -o, the list of matching strings
417        was not coloured; this is different to GNU grep, so I have changed it to be
418        the same.
420    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
421        each matching line was coloured. Now it goes on to look for further matches
422        of any of the test patterns, which is the same behaviour as GNU grep.
424    7.  A pattern that could match an empty string could cause pcregrep to loop; it
425        doesn't make sense to accept an empty string match in pcregrep, so I have
426        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
427        seems to be how GNU grep behaves.
429    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
430        start or after a newline", because the conditional assertion was not being
431        correctly handled. The rule now is that both the assertion and what follows
432        in the first alternative must satisfy the test.
434    9.  If auto-callout was enabled in a pattern with a conditional group whose
435        condition was an assertion, PCRE could crash during matching, both with
436        pcre_exec() and pcre_dfa_exec().
438    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
439        used for matching.
441    11. Unicode property support in character classes was not working for
442        characters (bytes) greater than 127 when not in UTF-8 mode.
444    12. Added the -M command line option to pcretest.
446    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
448    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
450    16. Added comments and documentation about mis-use of no_arg in the C++
451        wrapper.
453    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
454        from Martin Jerabek that uses macro names for all relevant character and
455        string constants.
457    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
458        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
459        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
460        these, but not everybody uses configure.
462    19. A conditional group that had only one branch was not being correctly
463        recognized as an item that could match an empty string. This meant that an
464        enclosing group might also not be so recognized, causing infinite looping
465        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
466        with the subject "ab", where knowledge that the repeated group can match
467        nothing is needed in order to break the loop.
469    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
470        exec(), but without supplying a callout function, matching went wrong.
472    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
473        leak if the size of the offset vector was greater than 30. When the vector
474        is smaller, the saved offsets during recursion go onto a local stack
475        vector, but for larger vectors malloc() is used. It was failing to free
476        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
477        error, in fact).
479    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
480        heapframe that is used only when UTF-8 support is enabled. This caused no
481        problem, but was untidy.
483    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
484        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
485        included within another project.
487    24. Steven Van Ingelgem's patches to add more options to the CMake support,
488        slightly modified by me:
490          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
491              not building pcregrep.
493          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
494              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
496    25. Forward references, both numeric and by name, in patterns that made use of
497        duplicate group numbers, could behave incorrectly or give incorrect errors,
498        because when scanning forward to find the reference group, PCRE was not
499        taking into account the duplicate group numbers. A pattern such as
500        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
502    26. Changed a few more instances of "const unsigned char *" to USPTR, making
503        the feature of a custom pointer more persuasive (as requested by a user).
505    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
506        pcretest.c, inside #ifndefs, because it seems they are sometimes already
507        pre-defined.
509    28. Added support for (*UTF8) at the start of a pattern.
511    29. Arrange for flags added by the "release type" setting in CMake to be shown
512        in the configuration summary.
515    Version 7.8 05-Sep-08
516    ---------------------
518    1.  Replaced UCP searching code with optimized version as implemented for Ad
519        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
520        stage table and inline lookup instead of a function, giving speed ups of 2
521        to 5 times on some simple patterns that I tested. Permission was given to
522        distribute the MultiStage2.py script that generates the tables (it's not in
523        the tarball, but is in the Subversion repository).
525    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
526        scripts.
528    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
529        a group with a zero qualifier. The result of the study could be incorrect,
530        or the function might crash, depending on the pattern.
532    4.  Caseless matching was not working for non-ASCII characters in back
533        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
534        It now works when Unicode Property Support is available.
536    5.  In pcretest, an escape such as \x{de} in the data was always generating
537        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
538        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
539        truncation.
541    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
543    7.  Added two (int) casts to pcregrep when printing the difference of two
544        pointers, in case they are 64-bit values.
546    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
547        test 2 if it fails.
549    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
550        and a #define of that name to empty if it is not externally set. This is to
551        allow users of MSVC to set it if necessary.
553    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
554        the convenience functions in the pcre_get.c source file.
556    11. An option change at the start of a pattern that had top-level alternatives
557        could cause overwriting and/or a crash. This command provoked a crash in
558        some environments:
560          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
562        This potential security problem was recorded as CVE-2008-2371.
564    12. For a pattern where the match had to start at the beginning or immediately
565        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
566        pcre_dfa_exec() could read past the end of the passed subject if there was
567        no match. To help with detecting such bugs (e.g. with valgrind), I modified
568        pcretest so that it places the subject at the end of its malloc-ed buffer.
570    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
571        exec() might read past the end of the data buffer in UTF-8 mode.
573    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
574        the data contained the byte 0x85 as part of a UTF-8 character within its
575        first line. This applied both to normal and DFA matching.
577    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
578        /^[^d]*?$/8 failed to match "abc".
580    16. Added a missing copyright notice to pcrecpp_internal.h.
582    17. Make it more clear in the documentation that values returned from
583        pcre_exec() in ovector are byte offsets, not character counts.
585    18. Tidied a few places to stop certain compilers from issuing warnings.
587    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
588        supplied by Stefan Weber. I made a further small update for 7.8 because
589        there is a change of source arrangements: the pcre_searchfuncs.c module is
590        replaced by pcre_ucd.c.
593    Version 7.7 07-May-08
594    ---------------------
596    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
597        a string to a long long, pretend we don't even have a long long." This is
598        done by checking for the strtoq, strtoll, and _strtoi64 functions.
600    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
601        pre-7.6 versions, which defined a global no_arg variable instead of putting
602        it in the RE class. (See also #8 below.)
604    3.  Remove a line of dead code, identified by coverity and reported by Nuno
605        Lopes.
607    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
609        (1) The include/exclude patterns were being applied to the whole pathnames
610            of files, instead of just to the final components.
612        (2) If there was more than one level of directory, the subdirectories were
613            skipped unless they satisfied the include/exclude conditions. This is
614            inconsistent with GNU grep (and could even be seen as contrary to the
615            pcregrep specification - which I improved to make it absolutely clear).
616            The action now is always to scan all levels of directory, and just
617            apply the include/exclude patterns to regular files.
619    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
620        --exclude_dir in the tests to avoid scanning .svn directories.
622    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
623        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
624        doesn't support NULs in patterns.
626    7.  Added some missing "const"s to declarations of static tables in
627        pcre_compile.c and pcre_dfa_exec.c.
629    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
630        caused by fix #2  above. (Subsequently also a second patch to fix the
631        first patch. And a third patch - this was a messy problem.)
633    9.  Applied Craig's patch to remove the use of push_back().
635    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
636        matching function regexec().
638    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
639        which, however, unlike Perl's \g{...}, are subroutine calls, not back
640        references. PCRE supports relative numbers with this syntax (I don't think
641        Oniguruma does).
643    12. Previously, a group with a zero repeat such as (...){0} was completely
644        omitted from the compiled regex. However, this means that if the group
645        was called as a subroutine from elsewhere in the pattern, things went wrong
646        (an internal error was given). Such groups are now left in the compiled
647        pattern, with a new opcode that causes them to be skipped at execution
648        time.
650    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
651        to the way PCRE behaves:
653        (a) A lone ] character is dis-allowed (Perl treats it as data).
655        (b) A back reference to an unmatched subpattern matches an empty string
656            (Perl fails the current match path).
658        (c) A data ] in a character class must be notated as \] because if the
659            first data character in a class is ], it defines an empty class. (In
660            Perl it is not possible to have an empty class.) The empty class []
661            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
662            The negative empty class [^] matches any one character, independently
663            of the DOTALL setting.
665    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
666        non-existent subpattern following a character class starting with ']' and
667        containing () gave an internal compiling error instead of "reference to
668        non-existent subpattern". Fortunately, when the pattern did exist, the
669        compiled code was correct. (When scanning forwards to check for the
670        existencd of the subpattern, it was treating the data ']' as terminating
671        the class, so got the count wrong. When actually compiling, the reference
672        was subsequently set up correctly.)
674    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
675        it was being rejected as not supported by pcre_dfa_exec(), even though
676        other assertions are supported. I have made pcre_dfa_exec() support
677        (*FAIL).
679    16. The implementation of 13c above involved the invention of a new opcode,
680        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
681        cannot be changed at match time, I realized I could make a small
682        improvement to matching performance by compiling OP_ALLANY instead of
683        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
684        on the OP_ANY path.
686    17. Compiling pcretest on Windows with readline support failed without the
687        following two fixes: (1) Make the unistd.h include conditional on
688        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
690    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
691        ncurses library to be included for pcretest when ReadLine support is
692        requested, but also to allow for it to be overridden. This patch came from
693        Daniel Bergström.
695    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
696        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
697        any errors with the current Unicode tables. Thanks to Peter Kankowski for
698        spotting this.
701    Version 7.6 28-Jan-08
702    ---------------------
704    1.  A character class containing a very large number of characters with
705        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
706        overflow.
708    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
709        HAVE_LONG_LONG is not defined.
711    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
712        bring them up to date and include new features. This patch includes:
714        - Fixed PH's badly added libz and libbz2 support.
715        - Fixed a problem with static linking.
716        - Added pcredemo. [But later removed - see 7 below.]
717        - Fixed dftables problem and added an option.
718        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
719            HAVE_LONG_LONG.
720        - Added readline support for pcretest.
721        - Added an listing of the option settings after cmake has run.
723    4.  A user submitted a patch to Makefile that makes it easy to create
724        "pcre.dll" under mingw when using Configure/Make. I added stuff to
725        Makefile.am that cause it to include this special target, without
726        affecting anything else. Note that the same mingw target plus all
727        the other distribution libraries and programs are now supported
728        when configuring with CMake (see 6 below) instead of with
729        Configure/Make.
731    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
732        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
733        exported in the Windows port". It has not yet been confirmed that the patch
734        solves the problem, but it does no harm.
736    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
737        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
738        with CMake, and also correct the comment about stack recursion.
740    7.  Remove the automatic building of pcredemo from the ./configure system and
741        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
742        of a program that users should build themselves after PCRE is installed, so
743        building it automatically is not really right. What is more, it gave
744        trouble in some build environments.
746    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
749    Version 7.5 10-Jan-08
750    ---------------------
752    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
753        values in parens when parsing an RE using the C++ wrapper."
755    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
756        Characters greater than 255 were excluded from the class instead of being
757        included.
759    3.  The same bug as (2) above applied to negated POSIX classes such as
760        [:^space:].
762    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
763        defined or documented. It seems to have been a typo for PCRE_STATIC, so
764        I have changed it.
766    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
767        first named subpattern) and a construct such as (?&a) would reference the
768        first named subpattern whose name started with "a" (in other words, the
769        length check was missing). Both these problems are fixed. "Subpattern name
770        expected" is now given for (?&) (a zero-length name), and this patch also
771        makes it give the same error for \k'' (previously it complained that that
772        was a reference to a non-existent subpattern).
774    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
775        this is right because (?- can be followed by option settings as well as by
776        digits. I have, however, made the messages clearer.
778    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
779        than the number used in the conditional) now cause a compile-time error.
780        This is actually not compatible with Perl, which accepts such patterns, but
781        treats the conditional as always being FALSE (as PCRE used to), but it
782        seems to me that giving a diagnostic is better.
784    8.  Change "alphameric" to the more common word "alphanumeric" in comments
785        and messages.
787    9.  Fix two occurrences of "backslash" in comments that should have been
788        "backspace".
790    10. Remove two redundant lines of code that can never be obeyed (their function
791        was moved elsewhere).
793    11. The program that makes PCRE's Unicode character property table had a bug
794        which caused it to generate incorrect table entries for sequences of
795        characters that have the same character type, but are in different scripts.
796        It amalgamated them into a single range, with the script of the first of
797        them. In other words, some characters were in the wrong script. There were
798        thirteen such cases, affecting characters in the following ranges:
800          U+002b0 - U+002c1
801          U+0060c - U+0060d
802          U+0061e - U+00612
803          U+0064b - U+0065e
804          U+0074d - U+0076d
805          U+01800 - U+01805
806          U+01d00 - U+01d77
807          U+01d9b - U+01dbf
808          U+0200b - U+0200f
809          U+030fc - U+030fe
810          U+03260 - U+0327f
811          U+0fb46 - U+0fbb1
812          U+10450 - U+1049d
814    12. The -o option (show only the matching part of a line) for pcregrep was not
815        compatible with GNU grep in that, if there was more than one match in a
816        line, it showed only the first of them. It now behaves in the same way as
817        GNU grep.
819    13. If the -o and -v options were combined for pcregrep, it printed a blank
820        line for every non-matching line. GNU grep prints nothing, and pcregrep now
821        does the same. The return code can be used to tell if there were any
822        non-matching lines.
824    14. Added --file-offsets and --line-offsets to pcregrep.
826    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
827        infinitely looping recursion. The bug was that positive lookaheads were not
828        being skipped when checking for a possible empty match (negative lookaheads
829        and both kinds of lookbehind were skipped).
831    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
832        inclusion of <windows.h> to before rather than after the definition of
833        INVALID_FILE_ATTRIBUTES (patch from David Byron).
835    17. Specifying a possessive quantifier with a specific limit for a Unicode
836        character property caused pcre_compile() to compile bad code, which led at
837        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
838        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
839        caused the error; without that there was no problem.
841    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
843    19. Added --enable-pcretest-libreadline.
845    20. In pcrecpp.cc, the variable 'count' was incremented twice in
846        RE::GlobalReplace(). As a result, the number of replacements returned was
847        double what it should be. I removed one of the increments, but Craig sent a
848        later patch that removed the other one (the right fix) and added unit tests
849        that check the return values (which was not done before).
851    21. Several CMake things:
853        (1) Arranged that, when cmake is used on Unix, the libraries end up with
854            the names libpcre and libpcreposix, not just pcre and pcreposix.
856        (2) The above change means that pcretest and pcregrep are now correctly
857            linked with the newly-built libraries, not previously installed ones.
861    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
862        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
863        UTF-8 newline character). The key issue is that the pattern starts .*;
864        this means that the match must be either at the beginning, or after a
865        newline. The bug was in the code for advancing after a failed match and
866        checking that the new position followed a newline. It was not taking
867        account of UTF-8 characters correctly.
869    23. PCRE was behaving differently from Perl in the way it recognized POSIX
870        character classes. PCRE was not treating the sequence [:...:] as a
871        character class unless the ... were all letters. Perl, however, seems to
872        allow any characters between [: and :], though of course it rejects as
873        unknown any "names" that contain non-letters, because all the known class
874        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
875        for example, whereas PCRE did not - it did not recognize a POSIX character
876        class. This seemed a bit dangerous, so the code has been changed to be
877        closer to Perl. The behaviour is not identical to Perl, because PCRE will
878        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
879        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
880        Perl does, and where it didn't before.
882    24. Rewrite so as to remove the single use of %n from pcregrep because in some
883        Windows environments %n is disabled by default.
886    Version 7.4 21-Sep-07
887    ---------------------
889    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
890        means that a class such as [\s] counted as "explicit reference to CR or
891        LF". That isn't really right - the whole point of the change was to try to
892        help when there was an actual mention of one of the two characters. So now
893        the change happens only if \r or \n (or a literal CR or LF) character is
894        encountered.
896    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
897        of both had grown to the point where there were only 3 bits left.
898        Fortunately, there was spare space in the data structure, and so I have
899        moved the internal flags into a new 16-bit field to free up more option
900        bits.
902    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
903        but did not set the internal JCHANGED flag - either of these is enough to
904        control the way the "get" function works - but the PCRE_INFO_JCHANGED
905        facility is supposed to tell if (?J) was ever used, so now (?J) at the
906        start sets both bits.
908    4.  Added options (at build time, compile time, exec time) to change \R from
909        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
911    5.  doc/pcresyntax.html was missing from the distribution.
913    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
914        compatibility, even though it is no longer used.
916    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
917        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
918        windows.h file is present (where different names are used). [This was
919        reversed later after testing - see 16 below.]
921    8.  Changed all #include <config.h> to #include "config.h". There were also
922        some further <pcre.h> cases that I changed to "pcre.h".
924    9.  When pcregrep was used with the --colour option, it missed the line ending
925        sequence off the lines that it output.
927    10. It was pointed out to me that arrays of string pointers cause lots of
928        relocations when a shared library is dynamically loaded. A technique of
929        using a single long string with a table of offsets can drastically reduce
930        these. I have refactored PCRE in four places to do this. The result is
931        dramatic:
933          Originally:                          290
934          After changing UCP table:            187
935          After changing error message table:   43
936          After changing table of "verbs"       36
937          After changing table of Posix names   22
939        Thanks to the folks working on Gregex for glib for this insight.
941    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
942        unicode-properties was also set.
944    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
946    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
947        checked only for CRLF.
949    14. Added casts to pcretest.c to avoid compiler warnings.
951    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
953    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
954        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
955        entirely. This removes changes made in 7 above.
957    17. The CMake files have been updated, and there is now more information about
958        building with CMake in the NON-UNIX-USE document.
961    Version 7.3 28-Aug-07
962    ---------------------
964     1. In the rejigging of the build system that eventually resulted in 7.1, the
965        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
966        brackets there is not right, since it causes compilers to look for an
967        installed pcre.h, not the version that is in the source that is being
968        compiled (which of course may be different). I have changed it back to:
970          #include "pcre.h"
972        I have a vague recollection that the change was concerned with compiling in
973        different directories, but in the new build system, that is taken care of
974        by the VPATH setting the Makefile.
976     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
977        when the subject happened to end in the byte 0x85 (e.g. if the last
978        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
979        characters but of course it shouldn't be taken as a newline when it is part
980        of another character. The bug was that, for an unlimited repeat of . in
981        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
982        characters when looking for a newline.
984     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
986     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
987        in debug output.
989     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
990        long printing in the pcrecpp unittest when running under MinGW.
992     6. ESC_K was left out of the EBCDIC table.
994     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
995        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
996        limit also applies to "virtual nesting" when a pattern is recursive, and in
997        this case 1000 isn't so big. I have been able to remove this limit at the
998        expense of backing off one optimization in certain circumstances. Normally,
999        when pcre_exec() would call its internal match() function recursively and
1000        immediately return the result unconditionally, it uses a "tail recursion"
1001        feature to save stack. However, when a subpattern that can match an empty
1002        string has an unlimited repetition quantifier, it no longer makes this
1003        optimization. That gives it a stack frame in which to save the data for
1004        checking that an empty string has been matched. Previously this was taken
1005        from the 1000-entry workspace that had been reserved. So now there is no
1006        explicit limit, but more stack is used.
1008     8. Applied Daniel's patches to solve problems with the import/export magic
1009        syntax that is required for Windows, and which was going wrong for the
1010        pcreposix and pcrecpp parts of the library. These were overlooked when this
1011        problem was solved for the main library.
1013     9. There were some crude static tests to avoid integer overflow when computing
1014        the size of patterns that contain repeated groups with explicit upper
1015        limits. As the maximum quantifier is 65535, the maximum group length was
1016        set at 30,000 so that the product of these two numbers did not overflow a
1017        32-bit integer. However, it turns out that people want to use groups that
1018        are longer than 30,000 bytes (though not repeat them that many times).
1019        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1020        made it possible to implement the integer overflow checks in a much more
1021        dynamic way, which I have now done. The artificial limitation on group
1022        length has been removed - we now have only the limit on the total length of
1023        the compiled pattern, which depends on the LINK_SIZE setting.
1025    10. Fixed a bug in the documentation for get/copy named substring when
1026        duplicate names are permitted. If none of the named substrings are set, the
1027        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1028        empty string.
1030    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1031        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1032        because the ] is interpreted as the first data character and the
1033        terminating ] is not found. PCRE has been made compatible with Perl in this
1034        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1035        cause memory overwriting.
1037    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1038        string has been matched (to stop an infinite loop). It was not recognizing
1039        a conditional subpattern that could match an empty string if that
1040        subpattern was within another subpattern. For example, it looped when
1041        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
1042        condition was not nested. This bug has been fixed.
1044    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1045        past the start of the subject in the presence of bytes with the top bit
1046        set, for example "\x8aBCD".
1048    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1049        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1051    14. Optimized (?!) to (*FAIL).
1053    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1054        This restricts code points to be within the range 0 to 0x10FFFF, excluding
1055        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1056        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1057        does: it's just the validity check that is more restrictive.
1059    16. Inserted checks for integer overflows during escape sequence (backslash)
1060        processing, and also fixed erroneous offset values for syntax errors during
1061        backslash processing.
1063    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1064        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1066    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1067        caused an overrun.
1069    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1070        something other than just ASCII characters) inside a group that had an
1071        unlimited repeat caused a loop at compile time (while checking to see
1072        whether the group could match an empty string).
1074    20. Debugging a pattern containing \p or \P could cause a crash. For example,
1075        [\P{Any}] did so. (Error in the code for printing property names.)
1077    21. An orphan \E inside a character class could cause a crash.
1079    22. A repeated capturing bracket such as (A)? could cause a wild memory
1080        reference during compilation.
1082    23. There are several functions in pcre_compile() that scan along a compiled
1083        expression for various reasons (e.g. to see if it's fixed length for look
1084        behind). There were bugs in these functions when a repeated \p or \P was
1085        present in the pattern. These operators have additional parameters compared
1086        with \d, etc, and these were not being taken into account when moving along
1087        the compiled data. Specifically:
1089        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1090            length.
1092        (b) An item such as \pL+ within a repeated group could cause crashes or
1093            loops.
1095        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1096            "reference to non-existent subpattern" error.
1098        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1100    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1101        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1103    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1105    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1106        character were causing crashes (broken optimization).
1108    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1109        \p or \P) caused a compile-time loop.
1111    28. More problems have arisen in unanchored patterns when CRLF is a valid line
1112        break. For example, the unstudied pattern [\r\n]A does not match the string
1113        "\r\nA" because change 7.0/46 below moves the current point on by two
1114        characters after failing to match at the start. However, the pattern \nA
1115        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
1116        the same is true. There doesn't seem any very clean way out of this, but
1117        what I have chosen to do makes the common cases work: PCRE now takes note
1118        of whether there can be an explicit match for \r or \n anywhere in the
1119        pattern, and if so, 7.0/46 no longer applies. As part of this change,
1120        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1121        pattern has explicit CR or LF references.
1123    29. Added (*CR) etc for changing newline setting at start of pattern.
1126    Version 7.2 19-Jun-07
1127    ---------------------
1129     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
1130        which is apparently normally available under Windows.
1132     2. Re-jig the pcregrep tests with different newline settings in an attempt
1133        to make them independent of the local environment's newline setting.
1135     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
1137     4. Some of the "internals" tests were previously cut out when the link size
1138        was not 2, because the output contained actual offsets. The recent new
1139        "Z" feature of pcretest means that these can be cut out, making the tests
1140        usable with all link sizes.
1142     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
1143        stack recursion. This gives a massive performance boost under BSD, but just
1144        a small improvement under Linux. However, it saves one field in the frame
1145        in all cases.
1147     6. Added more features from the forthcoming Perl 5.10:
1149        (a) (?-n) (where n is a string of digits) is a relative subroutine or
1150            recursion call. It refers to the nth most recently opened parentheses.
1152        (b) (?+n) is also a relative subroutine call; it refers to the nth next
1153            to be opened parentheses.
1155        (c) Conditions that refer to capturing parentheses can be specified
1156            relatively, for example, (?(-2)... or (?(+3)...
1158        (d) \K resets the start of the current match so that everything before
1159            is not part of it.
1161        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1163        (f) \g{name} is another synonym - part of Perl 5.10's unification of
1164            reference syntax.
1166        (g) (?| introduces a group in which the numbering of parentheses in each
1167            alternative starts with the same number.
1169        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
1171     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
1174     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
1175        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
1176        for detecting groups that can match an empty string.
1178     9. A pattern with a very large number of alternatives (more than several
1179        hundred) was running out of internal workspace during the pre-compile
1180        phase, where pcre_compile() figures out how much memory will be needed. A
1181        bit of new cunning has reduced the workspace needed for groups with
1182        alternatives. The 1000-alternative test pattern now uses 12 bytes of
1183        workspace instead of running out of the 4096 that are available.
1185    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
1187    11. Applied patch from Google to remove an optimization that didn't quite work.
1188        The report of the bug said:
1190          pcrecpp::RE("a*").FullMatch("aaa") matches, while
1191          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
1192          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
1194    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
1195        it matched the wrong number of bytes.
1198    Version 7.1 24-Apr-07
1199    ---------------------
1201     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
1202        that is more "standard", making use of automake and other Autotools. There
1203        is some re-arrangement of the files and adjustment of comments consequent
1204        on this.
1206     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
1207        for recursive directory scanning broke on some systems because the files
1208        are not scanned in any specific order and on different systems the order
1209        was different. A call to "sort" has been inserted into RunGrepTest for the
1210        approprate test as a short-term fix. In the longer term there may be an
1211        alternative.
1213     3. I had an email from Eric Raymond about problems translating some of PCRE's
1214        man pages to HTML (despite the fact that I distribute HTML pages, some
1215        people do their own conversions for various reasons). The problems
1216        concerned the use of low-level troff macros .br and .in. I have therefore
1217        removed all such uses from the man pages (some were redundant, some could
1218        be replaced by .nf/.fi pairs). The 132html script that I use to generate
1219        HTML has been updated to handle .nf/.fi and to complain if it encounters
1220        .br or .in.
1222     4. Updated comments in configure.ac that get placed in config.h.in and also
1223        arranged for config.h to be included in the distribution, with the name
1224        config.h.generic, for the benefit of those who have to compile without
1225        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
1227     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
1228        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
1229        makevp.bat for the current PCRE, using the additional files
1230        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
1232     6. A Windows user reported a minor discrepancy with test 2, which turned out
1233        to be caused by a trailing space on an input line that had got lost in his
1234        copy. The trailing space was an accident, so I've just removed it.
1236     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
1237        that is needed.
1239     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
1240        as "const" (a) because they are and (b) because it helps the PHP
1241        maintainers who have recently made a script to detect big data structures
1242        in the php code that should be moved to the .rodata section. I remembered
1243        to update Builducptable as well, so it won't revert if ucptable.h is ever
1244        re-created.
1246     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
1247        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
1248        order to be able to cut out the UTF-8 tables in the latter when UTF-8
1249        support is not required. This saves 1.5-2K of code, which is important in
1250        some applications.
1252        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
1253        so as not to refer to the tables, even though these functions will never be
1254        called when UTF-8 support is disabled. Otherwise there are problems with a
1255        shared library.
1257    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
1259        (a) It was defining its arguments as char * instead of void *.
1261        (b) It was assuming that all moves were upwards in memory; this was true
1262            a long time ago when I wrote it, but is no longer the case.
1264        The emulated memove() is provided for those environments that have neither
1265        memmove() nor bcopy(). I didn't think anyone used it these days, but that
1266        is clearly not the case, as these two bugs were recently reported.
1268    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
1269        and Detrail to create the HTML documentation, the .txt form of the man
1270        pages, and it removes trailing spaces from listed files. It also creates
1271        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
1272        case, it wraps all the #defines with #ifndefs. This script should be run
1273        before "make dist".
1275    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1276        with Unicode property support.
1278        (a) For a maximizing quantifier, if the two different cases of the
1279            character were of different lengths in their UTF-8 codings (there are
1280            some cases like this - I found 11), and the matching function had to
1281            back up over a mixture of the two cases, it incorrectly assumed they
1282            were both the same length.
1284        (b) When PCRE was configured to use the heap rather than the stack for
1285            recursion during matching, it was not correctly preserving the data for
1286            the other case of a UTF-8 character when checking ahead for a match
1287            while processing a minimizing repeat. If the check also involved
1288            matching a wide character, but failed, corruption could cause an
1289            erroneous result when trying to check for a repeat of the original
1290            character.
1292    13. Some tidying changes to the testing mechanism:
1294        (a) The RunTest script now detects the internal link size and whether there
1295            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1296            values substituted by "configure". (The RunGrepTest script already did
1297            this for UTF-8.) The configure.ac script no longer substitutes the
1298            relevant variables.
1300        (b) The debugging options /B and /D in pcretest show the compiled bytecode
1301            with length and offset values. This means that the output is different
1302            for different internal link sizes. Test 2 is skipped for link sizes
1303            other than 2 because of this, bypassing the problem. Unfortunately,
1304            there was also a test in test 3 (the locale tests) that used /B and
1305            failed for link sizes other than 2. Rather than cut the whole test out,
1306            I have added a new /Z option to pcretest that replaces the length and
1307            offset values with spaces. This is now used to make test 3 independent
1308            of link size. (Test 2 will be tidied up later.)
1310    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1311        segmentation fault instead of returning the appropriate error message.
1313    15. In multiline mode when the newline sequence was set to "any", the pattern
1314        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1315        This doesn't seem right; it now treats the CRLF combination as the line
1316        ending, and so does not match in that case. It's only a pattern such as ^$
1317        that would hit this one: something like ^ABC$ would have failed after \r
1318        and then tried again after \r\n.
1320    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1321        in an attempt to make files that differ only in their line terminators
1322        compare equal. This works on Linux.
1324    17. Under certain error circumstances pcregrep might try to free random memory
1325        as it exited. This is now fixed, thanks to valgrind.
1327    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1328        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1329        was because its rules for how to advance for /g after matching an empty
1330        string at the end of a line did not allow for this case. They now check for
1331        it specially.
1333    20. pcretest is supposed to handle patterns and data of any length, by
1334        extending its buffers when necessary. It was getting this wrong when the
1335        buffer for a data line had to be extended.
1337    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1338        CRLF as a newline sequence.
1340    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1341        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1342        I have nevertheless tidied it up.
1344    23. Added some casts to kill warnings from HP-UX ia64 compiler.
1346    24. Added a man page for pcre-config.
1349    Version 7.0 19-Dec-06
1350    ---------------------
1352     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1353        moving to gcc 4.1.1.
1355     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1356        sys/time.h, which is documented as needed for this function. It doesn't
1357        seem to matter on Linux, but it showed up on some releases of OS X.
1359     3. It seems that there are systems where bytes whose values are greater than
1360        127 match isprint() in the "C" locale. The "C" locale should be the
1361        default when a C program starts up. In most systems, only ASCII printing
1362        characters match isprint(). This difference caused the output from pcretest
1363        to vary, making some of the tests fail. I have changed pcretest so that:
1365        (a) When it is outputting text in the compiled version of a pattern, bytes
1366            other than 32-126 are always shown as hex escapes.
1368        (b) When it is outputting text that is a matched part of a subject string,
1369            it does the same, unless a different locale has been set for the match
1370            (using the /L modifier). In this case, it uses isprint() to decide.
1372     4. Fixed a major bug that caused incorrect computation of the amount of memory
1373        required for a compiled pattern when options that changed within the
1374        pattern affected the logic of the preliminary scan that determines the
1375        length. The relevant options are -x, and -i in UTF-8 mode. The result was
1376        that the computed length was too small. The symptoms of this bug were
1377        either the PCRE error "internal error: code overflow" from pcre_compile(),
1378        or a glibc crash with a message such as "pcretest: free(): invalid next
1379        size (fast)". Examples of patterns that provoked this bug (shown in
1380        pcretest format) are:
1382          /(?-x: )/x
1383          /(?x)(?-x: \s*#\s*)/
1384          /((?i)[\x{c0}])/8
1385          /(?i:[\x{c0}])/8
1387        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1388        is now done differently.
1390     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1391        wrapper classes; (b) implement a new function in the C++ scanner that is
1392        more efficient than the old way of doing things because it avoids levels of
1393        recursion in the regex matching; (c) add a paragraph to the documentation
1394        for the FullMatch() function.
1396     6. The escape sequence \n was being treated as whatever was defined as
1397        "newline". Not only was this contrary to the documentation, which states
1398        that \n is character 10 (hex 0A), but it also went horribly wrong when
1399        "newline" was defined as CRLF. This has been fixed.
1401     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1402        was being set to -1 for the "end of line" case (supposedly a value that no
1403        character can have). Though this value is never used (the check for end of
1404        line is "zero bytes in current character"), it caused compiler complaints.
1405        I've changed it to 0xffffffff.
1407     8. In pcre_version.c, the version string was being built by a sequence of
1408        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1409        string (as it is for production releases) called a macro with an empty
1410        argument. The C standard says the result of this is undefined. The gcc
1411        compiler treats it as an empty string (which was what was wanted) but it is
1412        reported that Visual C gives an error. The source has been hacked around to
1413        avoid this problem.
1415     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1416        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1417        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1418        of them did).
1420    10. Originally, pcretest opened its input and output without "b"; then I was
1421        told that "b" was needed in some environments, so it was added for release
1422        5.0 to both the input and output. (It makes no difference on Unix-like
1423        systems.) Later I was told that it is wrong for the input on Windows. I've
1424        now abstracted the modes into two macros, to make it easier to fiddle with
1425        them, and removed "b" from the input mode under Windows.
1427    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1429    12. Added -help and --help to pcretest as an official way of being reminded
1430        of the options.
1432    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1433        and pcrecpp.cc because they annoy compilers at high warning levels.
1435    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1437    15. Fixed an occurrence of == in configure.ac that should have been = (shell
1438        scripts are not C programs :-) and which was not noticed because it works
1439        on Linux.
1441    16. pcretest is supposed to handle any length of pattern and data line (as one
1442        line or as a continued sequence of lines) by extending its input buffer if
1443        necessary. This feature was broken for very long pattern lines, leading to
1444        a string of junk being passed to pcre_compile() if the pattern was longer
1445        than about 50K.
1447    17. I have done a major re-factoring of the way pcre_compile() computes the
1448        amount of memory needed for a compiled pattern. Previously, there was code
1449        that made a preliminary scan of the pattern in order to do this. That was
1450        OK when PCRE was new, but as the facilities have expanded, it has become
1451        harder and harder to keep it in step with the real compile phase, and there
1452        have been a number of bugs (see for example, 4 above). I have now found a
1453        cunning way of running the real compile function in a "fake" mode that
1454        enables it to compute how much memory it would need, while actually only
1455        ever using a few hundred bytes of working memory and without too many
1456        tests of the mode. This should make future maintenance and development
1457        easier. A side effect of this work is that the limit of 200 on the nesting
1458        depth of parentheses has been removed (though this was never a serious
1459        limitation, I suspect). However, there is a downside: pcre_compile() now
1460        runs more slowly than before (30% or more, depending on the pattern). I
1461        hope this isn't a big issue. There is no effect on runtime performance.
1463    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1464        newline (only possible for the last line of a file) and it was a
1465        pattern that set a locale (followed by /Lsomething), pcretest crashed.
1467    19. Added additional timing features to pcretest. (1) The -tm option now times
1468        matching only, not compiling. (2) Both -t and -tm can be followed, as a
1469        separate command line item, by a number that specifies the number of
1470        repeats to use when timing. The default is 50000; this gives better
1471        precision, but takes uncomfortably long for very large patterns.
1473    20. Extended pcre_study() to be more clever in cases where a branch of a
1474        subpattern has no definite first character. For example, (a*|b*)[cd] would
1475        previously give no result from pcre_study(). Now it recognizes that the
1476        first character must be a, b, c, or d.
1478    21. There was an incorrect error "recursive call could loop indefinitely" if
1479        a subpattern (or the entire pattern) that was being tested for matching an
1480        empty string contained only one non-empty item after a nested subpattern.
1481        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1482        incorrectly, because the \d was being skipped in the check.
1484    22. The pcretest program now has a new pattern option /B and a command line
1485        option -b, which is equivalent to adding /B to every pattern. This causes
1486        it to show the compiled bytecode, without the additional information that
1487        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1488        is the same as /B/I).
1490    23. A new optimization is now able automatically to treat some sequences such
1491        as a*b as a*+b. More specifically, if something simple (such as a character
1492        or a simple class like \d) has an unlimited quantifier, and is followed by
1493        something that cannot possibly match the quantified thing, the quantifier
1494        is automatically "possessified".
1496    24. A recursive reference to a subpattern whose number was greater than 39
1497        went wrong under certain circumstances in UTF-8 mode. This bug could also
1498        have affected the operation of pcre_study().
1500    25. Realized that a little bit of performance could be had by replacing
1501        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1503    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1505    27. Possessive quantifiers such as a++ were previously implemented by turning
1506        them into atomic groups such as ($>a+). Now they have their own opcodes,
1507        which improves performance. This includes the automatically created ones
1508        from 23 above.
1510    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1511        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1512        the first matched character to be a colon. This applied both to named and
1513        numbered groups.
1515    29. The ucpinternal.h header file was missing its idempotency #ifdef.
1517    30. I was sent a "project" file called libpcre.a.dev which I understand makes
1518        building PCRE on Windows easier, so I have included it in the distribution.
1520    31. There is now a check in pcretest against a ridiculously large number being
1521        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1522        loop, the loop is abandoned.
1524    32. Forward references to subpatterns in conditions such as (?(2)...) where
1525        subpattern 2 is defined later cause pcre_compile() to search forwards in
1526        the pattern for the relevant set of parentheses. This search went wrong
1527        when there were unescaped parentheses in a character class, parentheses
1528        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1530    33. "Subroutine" calls and backreferences were previously restricted to
1531        referencing subpatterns earlier in the regex. This restriction has now
1532        been removed.
1534    34. Added a number of extra features that are going to be in Perl 5.10. On the
1535        whole, these are just syntactic alternatives for features that PCRE had
1536        previously implemented using the Python syntax or my own invention. The
1537        other formats are all retained for compatibility.
1539        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1540            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1541            also .NET compatible.
1543        (b) A recursion or subroutine call to a named group can now be defined as
1544            (?&name) as well as (?P>name).
1546        (c) A backreference to a named group can now be defined as \k<name> or
1547            \k'name' as well as (?P=name). The new forms, as well as being in Perl
1548            5.10, are also .NET compatible.
1550        (d) A conditional reference to a named group can now use the syntax
1551            (?(<name>) or (?('name') as well as (?(name).
1553        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1554            groups (named and numbered) that are never evaluated inline, but can be
1555            called as "subroutines" from elsewhere. In effect, the DEFINE condition
1556            is always false. There may be only one alternative in such a group.
1558        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1559            as the simple (?(R). The condition is true only if the most recent
1560            recursion is that of the given number or name. It does not search out
1561            through the entire recursion stack.
1563        (g) The escape \gN or \g{N} has been added, where N is a positive or
1564            negative number, specifying an absolute or relative reference.
1566    35. Tidied to get rid of some further signed/unsigned compiler warnings and
1567        some "unreachable code" warnings.
1569    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1570        things, this adds five new scripts.
1572    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1573        There were also incompatibilities regarding the handling of \Q..\E inside
1574        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1575        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1577    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1578        matches an empty string, and forcibly breaks the loop. There were bugs in
1579        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
1580        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
1581        separate and independent bugs (that affected different cases) have been
1582        fixed.
1584    39. Refactored the code to abolish the use of different opcodes for small
1585        capturing bracket numbers. This is a tidy that I avoided doing when I
1586        removed the limit on the number of capturing brackets for 3.5 back in 2001.
1587        The new approach is not only tidier, it makes it possible to reduce the
1588        memory needed to fix the previous bug (38).
1590    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1591        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1592        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1593        mode.
1595    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1596        report.
1598    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1599        copy construction and assignment in the C++ wrapper.
1601    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1602        couple of bugs that could have given wrong results in the "--newline=crlf"
1603        case.
1605    44. Added a number of casts and did some reorganization of signed/unsigned int
1606        variables following suggestions from Dair Grant. Also renamed the variable
1607        "this" as "item" because it is a C++ keyword.
1609    45. Arranged for dftables to add
1611          #include "pcre_internal.h"
1613        to pcre_chartables.c because without it, gcc 4.x may remove the array
1614        definition from the final binary if PCRE is built into a static library and
1615        dead code stripping is activated.
1617    46. For an unanchored pattern, if a match attempt fails at the start of a
1618        newline sequence, and the newline setting is CRLF or ANY, and the next two
1619        characters are CRLF, advance by two characters instead of one.
1622    Version 6.7 04-Jul-06
1623    ---------------------
1625     1. In order to handle tests when input lines are enormously long, pcretest has
1626        been re-factored so that it automatically extends its buffers when
1627        necessary. The code is crude, but this _is_ just a test program. The
1628        default size has been increased from 32K to 50K.
1630     2. The code in pcre_study() was using the value of the re argument before
1631        testing it for NULL. (Of course, in any sensible call of the function, it
1632        won't be NULL.)
1634     3. The memmove() emulation function in pcre_internal.h, which is used on
1635        systems that lack both memmove() and bcopy() - that is, hardly ever -
1636        was missing a "static" storage class specifier.
1638     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1639        containing an extended class (one that cannot be represented by a bitmap
1640        because it contains high-valued characters or Unicode property items, e.g.
1641        [\pZ]). Almost always one would set UTF-8 mode when processing such a
1642        pattern, but PCRE should not loop if you do not (it no longer does).
1643        [Detail: two cases were found: (a) a repeated subpattern containing an
1644        extended class; (b) a recursive reference to a subpattern that followed a
1645        previous extended class. It wasn't skipping over the extended class
1646        correctly when UTF-8 mode was not set.]
1648     5. A negated single-character class was not being recognized as fixed-length
1649        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1650        compile error "lookbehind assertion is not fixed length".
1652     6. The RunPerlTest auxiliary script was showing an unexpected difference
1653        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1654        write a Perl script that can interpret lines of an input file either as
1655        byte characters or as UTF-8, which is what "perltest" was being required to
1656        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1657        can't do is switch easily at run time between having the "use utf8;" pragma
1658        or not. In the end, I fudged it by using the RunPerlTest script to insert
1659        "use utf8;" explicitly for the UTF-8 tests.
1661     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1662        the end of the subject string, contrary to the documentation and to what
1663        Perl does. This was true of both matching functions. Now it matches only at
1664        the start of the subject and immediately after *internal* newlines.
1666     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1667        a pointer to an int instead of a pointer to an unsigned long int. This
1668        caused problems on 64-bit systems.
1670     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1671        instance of the 'standard' template library not being so standard".
1673    10. There was no check on the number of named subpatterns nor the maximum
1674        length of a subpattern name. The product of these values is used to compute
1675        the size of the memory block for a compiled pattern. By supplying a very
1676        long subpattern name and a large number of named subpatterns, the size
1677        computation could be caused to overflow. This is now prevented by limiting
1678        the length of names to 32 characters, and the number of named subpatterns
1679        to 10,000.
1681    11. Subpatterns that are repeated with specific counts have to be replicated in
1682        the compiled pattern. The size of memory for this was computed from the
1683        length of the subpattern and the repeat count. The latter is limited to
1684        65535, but there was no limit on the former, meaning that integer overflow
1685        could in principle occur. The compiled length of a repeated subpattern is
1686        now limited to 30,000 bytes in order to prevent this.
1688    12. Added the optional facility to have named substrings with the same name.
1690    13. Added the ability to use a named substring as a condition, using the
1691        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1692        are numbers (not recommended). Forward references are permitted.
1694    14. Added forward references in named backreferences (if you see what I mean).
1696    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1697        pattern could run off the end of the subject. For example, the pattern
1698        "(?s)(.{1,5})"8 did this with the subject "ab".
1700    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1701        PCRE_CASELESS was set when matching characters that were quantified with ?
1702        or *.
1704    17. A character class other than a single negated character that had a minimum
1705        but no maximum quantifier - for example [ab]{6,} - was not handled
1706        correctly by pce_dfa_exec(). It would match only one character.
1708    18. A valid (though odd) pattern that looked like a POSIX character
1709        class but used an invalid character after [ (for example [[,abc,]]) caused
1710        pcre_compile() to give the error "Failed: internal error: code overflow" or
1711        in some cases to crash with a glibc free() error. This could even happen if
1712        the pattern terminated after [[ but there just happened to be a sequence of
1713        letters, a binary zero, and a closing ] in the memory that followed.
1715    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1716        over the years. Originally (before any Unicode support), just the bottom 8
1717        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1718        output from "man perlunicode" includes this:
1720          The regular expression compiler produces polymorphic opcodes.  That
1721          is, the pattern adapts to the data and automatically switches to
1722          the Unicode character scheme when presented with Unicode data--or
1723          instead uses a traditional byte scheme when presented with byte
1724          data.
1726        Sadly, a wide octal escape does not cause a switch, and in a string with
1727        no other multibyte characters, these octal escapes are treated as before.
1728        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
1729        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1730        Unicode string.
1732        I have not perpetrated such confusion in PCRE. Up till now, it took just
1733        the bottom 8 bits, as in old Perl. I have now made octal escapes with
1734        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1735        translate to the appropriate multibyte character.
1737    29. Applied some refactoring to reduce the number of warnings from Microsoft
1738        and Borland compilers. This has included removing the fudge introduced
1739        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1740        a warning about an unused variable.
1742    21. PCRE has not included VT (character 0x0b) in the set of whitespace
1743        characters since release 4.0, because Perl (from release 5.004) does not.
1744        [Or at least, is documented not to: some releases seem to be in conflict
1745        with the documentation.] However, when a pattern was studied with
1746        pcre_study() and all its branches started with \s, PCRE still included VT
1747        as a possible starting character. Of course, this did no harm; it just
1748        caused an unnecessary match attempt.
1750    22. Removed a now-redundant internal flag bit that recorded the fact that case
1751        dependency changed within the pattern. This was once needed for "required
1752        byte" processing, but is no longer used. This recovers a now-scarce options
1753        bit. Also moved the least significant internal flag bit to the most-
1754        significant bit of the word, which was not previously used (hangover from
1755        the days when it was an int rather than a uint) to free up another bit for
1756        the future.
1758    23. Added support for CRLF line endings as well as CR and LF. As well as the
1759        default being selectable at build time, it can now be changed at runtime
1760        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1761        specify that it is scanning data with non-default line endings.
1763    24. Changed the definition of CXXLINK to make it agree with the definition of
1764        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1766    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1767        recursions. This makes a big different to stack usage for some patterns.
1769    26. If a subpattern containing a named recursion or subroutine reference such
1770        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1771        the space required for the compiled pattern went wrong and gave too small a
1772        value. Depending on the environment, this could lead to "Failed: internal
1773        error: code overflow at offset 49" or "glibc detected double free or
1774        corruption" errors.
1776    27. Applied patches from Google (a) to support the new newline modes and (b) to
1777        advance over multibyte UTF-8 characters in GlobalReplace.
1779    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1780        difference for some implementation of PCRE in some Windows version.
1782    29. Added some extra testing facilities to pcretest:
1784        \q<number>   in a data line sets the "match limit" value
1785        \Q<number>   in a data line sets the "match recursion limt" value
1786        -S <number>  sets the stack size, where <number> is in megabytes
1788        The -S option isn't available for Windows.
1791    Version 6.6 06-Feb-06
1792    ---------------------
1794     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1795        in pcreposix.h. I have copied the definition from pcre.h.
1797     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1798        because pcre.h is no longer a built file.
1800     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1801        not normally included in the compiled code.
1804    Version 6.5 01-Feb-06
1805    ---------------------
1807     1. When using the partial match feature with pcre_dfa_exec(), it was not
1808        anchoring the second and subsequent partial matches at the new starting
1809        point. This could lead to incorrect results. For example, with the pattern
1810        /1234/, partially matching against "123" and then "a4" gave a match.
1812     2. Changes to pcregrep:
1814        (a) All non-match returns from pcre_exec() were being treated as failures
1815            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1816            error message is output. Some extra information is given for the
1817            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
1818            probably the only errors that are likely to be caused by users (by
1819            specifying a regex that has nested indefinite repeats, for instance).
1820            If there are more than 20 of these errors, pcregrep is abandoned.
1822        (b) A binary zero was treated as data while matching, but terminated the
1823            output line if it was written out. This has been fixed: binary zeroes
1824            are now no different to any other data bytes.
1826        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1827            used to set a locale for matching. The --locale=xxxx long option has
1828            been added (no short equivalent) to specify a locale explicitly on the
1829            pcregrep command, overriding the environment variables.
1831        (d) When -B was used with -n, some line numbers in the output were one less
1832            than they should have been.
1834        (e) Added the -o (--only-matching) option.
1836        (f) If -A or -C was used with -c (count only), some lines of context were
1837            accidentally printed for the final match.
1839        (g) Added the -H (--with-filename) option.
1841        (h) The combination of options -rh failed to suppress file names for files
1842            that were found from directory arguments.
1844        (i) Added the -D (--devices) and -d (--directories) options.
1846        (j) Added the -F (--fixed-strings) option.
1848        (k) Allow "-" to be used as a file name for -f as well as for a data file.
1850        (l) Added the --colo(u)r option.
1852        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1853            is not present by default.
1855     3. A nasty bug was discovered in the handling of recursive patterns, that is,
1856        items such as (?R) or (?1), when the recursion could match a number of
1857        alternatives. If it matched one of the alternatives, but subsequently,
1858        outside the recursion, there was a failure, the code tried to back up into
1859        the recursion. However, because of the way PCRE is implemented, this is not
1860        possible, and the result was an incorrect result from the match.
1862        In order to prevent this happening, the specification of recursion has
1863        been changed so that all such subpatterns are automatically treated as
1864        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1866     4. I had overlooked the fact that, in some locales, there are characters for
1867        which isalpha() is true but neither isupper() nor islower() are true. In
1868        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1869        and ordfeminine) are like this. This affected the treatment of \w and \W
1870        when they appeared in character classes, but not when they appeared outside
1871        a character class. The bit map for "word" characters is now created
1872        separately from the results of isalnum() instead of just taking it from the
1873        upper, lower, and digit maps. (Plus the underscore character, of course.)
1875     5. The above bug also affected the handling of POSIX character classes such as
1876        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1877        permanent tables. Instead, the bit maps for such a class were previously
1878        created as the appropriate unions of the upper, lower, and digit bitmaps.
1879        Now they are created by subtraction from the [[:word:]] class, which has
1880        its own bitmap.
1882     6. The [[:blank:]] character class matches horizontal, but not vertical space.
1883        It is created by subtracting the vertical space characters (\x09, \x0a,
1884        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1885        subtraction was done in the overall bitmap for a character class, meaning
1886        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1887        be recognized. This bug has been fixed.
1889     7. Patches from the folks at Google:
1891          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1892          real life, but is still worth protecting against".
1894          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1895          regular expressions".
1897          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1898          have it.
1900          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1901          "configure" and the latter not, in order to fix a problem somebody had
1902          with compiling the Arg class on HP-UX.
1904          (e) Improve the error-handling of the C++ wrapper a little bit.
1906          (f) New tests for checking recursion limiting.
1908     8. The pcre_memmove() function, which is used only if the environment does not
1909        have a standard memmove() function (and is therefore rarely compiled),
1910        contained two bugs: (a) use of int instead of size_t, and (b) it was not
1911        returning a result (though PCRE never actually uses the result).
1913     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1914        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1915        returned instead of calling malloc() with an overflowing number that would
1916        most likely cause subsequent chaos.
1918    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1920    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1921        with this option is matched, the nmatch and pmatch options of regexec() are
1922        ignored.
1924    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1925        provided in case anyone wants to the the POSIX interface with UTF-8
1926        strings.
1928    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1929        C++ linking (needed for some HP-UX environments).
1931    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1932        (unused parameter) and in the pcre_printint() function (omitted "default"
1933        switch label when the default is to do nothing).
1935    15. Added some code to make it possible, when PCRE is compiled as a C++
1936        library, to replace subject pointers for pcre_exec() with a smart pointer
1937        class, thus making it possible to process discontinuous strings.
1939    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1940        much the same function. They were added by different people who were trying
1941        to make PCRE easy to compile on non-Unix systems. It has been suggested
1942        that PCRE_EXPORT be abolished now that there is more automatic apparatus
1943        for compiling on Windows systems. I have therefore replaced it with
1944        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1945        defaults to "extern" for C or "extern C" for C++, which works fine on
1946        Unix-like systems. It is now possible to override the value of PCRE_DATA_
1947        SCOPE with something explicit in config.h. In addition:
1949        (a) pcreposix.h still had just "extern" instead of either of these macros;
1950            I have replaced it with PCRE_DATA_SCOPE.
1952        (b) Functions such as _pcre_xclass(), which are internal to the library,
1953            but external in the C sense, all had PCRE_EXPORT in their definitions.
1954            This is apparently wrong for the Windows case, so I have removed it.
1955            (It makes no difference on Unix-like systems.)
1957    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1958        of recursive calls to match(). This is different to MATCH_LIMIT because
1959        that limits the total number of calls to match(), not all of which increase
1960        the depth of recursion. Limiting the recursion depth limits the amount of
1961        stack (or heap if NO_RECURSE is set) that is used. The default can be set
1962        when PCRE is compiled, and changed at run time. A patch from Google adds
1963        this functionality to the C++ interface.
1965    18. Changes to the handling of Unicode character properties:
1967        (a) Updated the table to Unicode 4.1.0.
1969        (b) Recognize characters that are not in the table as "Cn" (undefined).
1971        (c) I revised the way the table is implemented to a much improved format
1972            which includes recognition of ranges. It now supports the ranges that
1973            are defined in UnicodeData.txt, and it also amalgamates other
1974            characters into ranges. This has reduced the number of entries in the
1975            table from around 16,000 to around 3,000, thus reducing its size
1976            considerably. I realized I did not need to use a tree structure after
1977            all - a binary chop search is just as efficient. Having reduced the
1978            number of entries, I extended their size from 6 bytes to 8 bytes to
1979            allow for more data.
1981        (d) Added support for Unicode script names via properties such as \p{Han}.
1983    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1984        matching that character.
1986    20. When matching a repeated Unicode property with a minimum greater than zero,
1987        (for example \pL{2,}), PCRE could look past the end of the subject if it
1988        reached it while seeking the minimum number of characters. This could
1989        happen only if some of the characters were more than one byte long, because
1990        there is a check for at least the minimum number of bytes.
1992    21. Refactored the implementation of \p and \P so as to be more general, to
1993        allow for more different types of property in future. This has changed the
1994        compiled form incompatibly. Anybody with saved compiled patterns that use
1995        \p or \P will have to recompile them.
1997    22. Added "Any" and "L&" to the supported property types.
1999    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
2000        but give a compile time error if the value is greater than 0xff.
2002    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
2003        accidentally not being installed or uninstalled.
2005    25. The pcre.h file was built from pcre.h.in, but the only changes that were
2006        made were to insert the current release number. This seemed silly, because
2007        it made things harder for people building PCRE on systems that don't run
2008        "configure". I have turned pcre.h into a distributed file, no longer built
2009        by "configure", with the version identification directly included. There is
2010        no longer a pcre.h.in file.
2012        However, this change necessitated a change to the pcre-config script as
2013        well. It is built from pcre-config.in, and one of the substitutions was the
2014        release number. I have updated configure.ac so that ./configure now finds
2015        the release number by grepping pcre.h.
2017    26. Added the ability to run the tests under valgrind.
2020    Version 6.4 05-Sep-05
2021    ---------------------
2023     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
2024        "--" to be printed when multiple files were scanned, even when none of the
2025        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
2026        consider it to be a bug, and have restored the previous behaviour.
2028     2. A couple of code tidies to get rid of compiler warnings.
2030     3. The pcretest program used to cheat by referring to symbols in the library
2031        whose names begin with _pcre_. These are internal symbols that are not
2032        really supposed to be visible externally, and in some environments it is
2033        possible to suppress them. The cheating is now confined to including
2034        certain files from the library's source, which is a bit cleaner.
2036     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
2037        file's purpose clearer.
2039     5. Reorganized pcre_ucp_findchar().
2042    Version 6.3 15-Aug-05
2043    ---------------------
2045     1. The file libpcre.pc.in did not have general read permission in the tarball.
2047     2. There were some problems when building without C++ support:
2049        (a) If C++ support was not built, "make install" and "make test" still
2050            tried to test it.
2052        (b) There were problems when the value of CXX was explicitly set. Some
2053            changes have been made to try to fix these, and ...
2055        (c) --disable-cpp can now be used to explicitly disable C++ support.
2057        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
2058            backslash in a target when C++ was disabled. This confuses some
2059            versions of "make", apparently. Using an intermediate variable solves
2060            this. (Same for CPP_LOBJ.)
2062     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
2063        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
2064        necessary on certain architectures.
2066     4. Added a setting of -export-symbols-regex to the link command to remove
2067        those symbols that are exported in the C sense, but actually are local
2068        within the library, and not documented. Their names all begin with
2069        "_pcre_". This is not a perfect job, because (a) we have to except some
2070        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
2071        available (and never for static libraries). I have made a note to try to
2072        find a way round (a) in the future.
2075    Version 6.2 01-Aug-05
2076    ---------------------
2078     1. There was no test for integer overflow of quantifier values. A construction
2079        such as {1111111111111111} would give undefined results. What is worse, if
2080        a minimum quantifier for a parenthesized subpattern overflowed and became
2081        negative, the calculation of the memory size went wrong. This could have
2082        led to memory overwriting.
2084     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
2086     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
2087        operating environments where this matters.
2089     4. Applied Giuseppe Maxia's patch to add additional features for controlling
2090        PCRE options from within the C++ wrapper.
2092     5. Named capturing subpatterns were not being correctly counted when a pattern
2093        was compiled. This caused two problems: (a) If there were more than 100
2094        such subpatterns, the calculation of the memory needed for the whole
2095        compiled pattern went wrong, leading to an overflow error. (b) Numerical
2096        back references of the form \12, where the number was greater than 9, were
2097        not recognized as back references, even though there were sufficient
2098        previous subpatterns.
2100     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
2101        versions of gcc, e.g. 2.95.4.
2104    Version 6.1 21-Jun-05
2105    ---------------------
2107     1. There was one reference to the variable "posix" in pcretest.c that was not
2108        surrounded by "#if !defined NOPOSIX".
2110     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
2111        the cross-check on the old pcre_info() function, for the benefit of the
2112        cut-down version of PCRE that is currently imported into Exim.
2114     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
2115        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
2116        patterns that start (?i) but I don't think that matters. The use of (?i) is
2117        just an example; this all applies to the other options as well.
2119     4. Since libtool seems to echo the compile commands it is issuing, the output
2120        from "make" can be reduced a bit by putting "@" in front of each libtool
2121        compile command.
2123     5. Patch from the folks at Google for configure.in to be a bit more thorough
2124        in checking for a suitable C++ installation before trying to compile the
2125        C++ stuff. This should fix a reported problem when a compiler was present,
2126        but no suitable headers.
2128     6. The man pages all had just "PCRE" as their title. I have changed them to
2129        be the relevant file name. I have also arranged that these names are
2130        retained in the file doc/pcre.txt, which is a concatenation in text format
2131        of all the man pages except the little individual ones for each function.
2133     7. The NON-UNIX-USE file had not been updated for the different set of source
2134        files that come with release 6. I also added a few comments about the C++
2135        wrapper.
2138    Version 6.0 07-Jun-05
2139    ---------------------
2141     1. Some minor internal re-organization to help with my DFA experiments.
2143     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
2144        didn't matter for the library itself when fully configured, but did matter
2145        when compiling without UCP support, or within Exim, where the ucp files are
2146        not imported.
2148     3. Refactoring of the library code to split up the various functions into
2149        different source modules. The addition of the new DFA matching code (see
2150        below) to a single monolithic source would have made it really too
2151        unwieldy, quite apart from causing all the code to be include in a
2152        statically linked application, when only some functions are used. This is
2153        relevant even without the DFA addition now that patterns can be compiled in
2154        one application and matched in another.
2156        The downside of splitting up is that there have to be some external
2157        functions and data tables that are used internally in different modules of
2158        the library but which are not part of the API. These have all had their
2159        names changed to start with "_pcre_" so that they are unlikely to clash
2160        with other external names.
2162     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
2163        a different (DFA) algorithm. Although it is slower than the original
2164        function, it does have some advantages for certain types of matching
2165        problem.
2167     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
2168        including restarting after a partial match.
2170     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
2171        defined when compiling for Windows was sent to me. I have put it into the
2172        code, though I have no means of testing or verifying it.
2174     7. Added the pcre_refcount() auxiliary function.
2176     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
2177        match before or at the first newline in the subject string. In pcretest,
2178        the /f option on a pattern can be used to set this.
2180     9. A repeated \w when used in UTF-8 mode with characters greater than 256
2181        would behave wrongly. This has been present in PCRE since release 4.0.
2183    10. A number of changes to the pcregrep command:
2185        (a) Refactored how -x works; insert ^(...)$ instead of setting
2186            PCRE_ANCHORED and checking the length, in preparation for adding
2187            something similar for -w.
2189        (b) Added the -w (match as a word) option.
2191        (c) Refactored the way lines are read and buffered so as to have more
2192            than one at a time available.
2194        (d) Implemented a pcregrep test script.
2196        (e) Added the -M (multiline match) option. This allows patterns to match
2197            over several lines of the subject. The buffering ensures that at least
2198            8K, or the rest of the document (whichever is the shorter) is available
2199            for matching (and similarly the previous 8K for lookbehind assertions).
2201        (f) Changed the --help output so that it now says
2203              -w, --word-regex(p)
2205            instead of two lines, one with "regex" and the other with "regexp"
2206            because that confused at least one person since the short forms are the
2207            same. (This required a bit of code, as the output is generated
2208            automatically from a table. It wasn't just a text change.)
2210        (g) -- can be used to terminate pcregrep options if the next thing isn't an
2211            option but starts with a hyphen. Could be a pattern or a path name
2212            starting with a hyphen, for instance.
2214        (h) "-" can be given as a file name to represent stdin.
2216        (i) When file names are being printed, "(standard input)" is used for
2217            the standard input, for compatibility with GNU grep. Previously
2218            "<stdin>" was used.
2220        (j) The option --label=xxx can be used to supply a name to be used for
2221            stdin when file names are being printed. There is no short form.
2223        (k) Re-factored the options decoding logic because we are going to add
2224            two more options that take data. Such options can now be given in four
2225            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
2227        (l) Added the -A, -B, and -C options for requesting that lines of context
2228            around matches be printed.
2230        (m) Added the -L option to print the names of files that do not contain
2231            any matching lines, that is, the complement of -l.
2233        (n) The return code is 2 if any file cannot be opened, but pcregrep does
2234            continue to scan other files.
2236        (o) The -s option was incorrectly implemented. For compatibility with other
2237            greps, it now suppresses the error message for a non-existent or non-
2238            accessible file (but not the return code). There is a new option called
2239            -q that suppresses the output of matching lines, which was what -s was
2240            previously doing.
2242        (p) Added --include and --exclude options to specify files for inclusion
2243            and exclusion when recursing.
2245    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
2246        Hopefully, it now does.
2248    12. Missing cast in pcre_study().
2250    13. Added an "uninstall" target to the makefile.
2252    14. Replaced "extern" in the function prototypes in Makefile.in with
2253        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
2254        world, but is set differently for Windows.
2256    15. Added a second compiling function called pcre_compile2(). The only
2257        difference is that it has an extra argument, which is a pointer to an
2258        integer error code. When there is a compile-time failure, this is set
2259        non-zero, in addition to the error test pointer being set to point to an
2260        error message. The new argument may be NULL if no error number is required
2261        (but then you may as well call pcre_compile(), which is now just a
2262        wrapper). This facility is provided because some applications need a
2263        numeric error indication, but it has also enabled me to tidy up the way
2264        compile-time errors are handled in the POSIX wrapper.
2266    16. Added VPATH=.libs to the makefile; this should help when building with one
2267        prefix path and installing with another. (Or so I'm told by someone who
2268        knows more about this stuff than I do.)
2270    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
2271        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
2272        match everything, including newlines. This is not POSIX-compatible, but
2273        somebody wanted the feature. From pcretest it can be activated by using
2274        both the P and the s flags.
2276    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
2278    19. libpcre.pc was being incorrectly installed as executable.
2280    20. A couple of places in pcretest check for end-of-line by looking for '\n';
2281        it now also looks for '\r' so that it will work unmodified on Windows.
2283    21. Added Google's contributed C++ wrapper to the distribution.
2285    22. Added some untidy missing memory free() calls in pcretest, to keep
2286        Electric Fence happy when testing.
2290    Version 5.0 13-Sep-04
2291    ---------------------
2293     1. Internal change: literal characters are no longer packed up into items
2294        containing multiple characters in a single byte-string. Each character
2295        is now matched using a separate opcode. However, there may be more than one
2296        byte in the character in UTF-8 mode.
2298     2. The pcre_callout_block structure has two new fields: pattern_position and
2299        next_item_length. These contain the offset in the pattern to the next match
2300        item, and its length, respectively.
2302     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
2303        insertion of callouts before each pattern item. Added the /C option to
2304        pcretest to make use of this.
2306     4. On the advice of a Windows user, the lines
2308          #if defined(_WIN32) || defined(WIN32)
2309          _setmode( _fileno( stdout ), 0x8000 );
2310          #endif  /* defined(_WIN32) || defined(WIN32) */
2312        have been added to the source of pcretest. This apparently does useful
2313        magic in relation to line terminators.
2315     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
2316        for the benefit of those environments where the "b" makes a difference.
2318     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
2319        to know about it. I have put a hack into configure.in that adds in code
2320        to set GCC=yes if CC=icc. This seems to end up at a point in the
2321        generated configure script that is early enough to affect the setting of
2322        compiler options, which is what is needed, but I have no means of testing
2323        whether it really works. (The user who reported this had patched the
2324        generated configure script, which of course I cannot do.)
2326        LATER: After change 22 below (new libtool files), the configure script
2327        seems to know about icc (and also ecc). Therefore, I have commented out
2328        this hack in configure.in.
2330     7. Added support for pkg-config (2 patches were sent in).
2332     8. Negated POSIX character classes that used a combination of internal tables
2333        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2334        [[:^ascii]]. Typically, they would match almost any characters. The other
2335        POSIX classes were not broken in this way.
2337     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2338        to find the match, as PCRE was deluded into thinking that the match had to
2339        start at the start point or following a newline. The same bug applied to
2340        patterns with negative forward assertions or any backward assertions
2341        preceding ".*" at the start, unless the pattern required a fixed first
2342        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2344    10. In UTF-8 mode, when moving forwards in the subject after a failed match
2345        starting at the last subject character, bytes beyond the end of the subject
2346        string were read.
2348    11. Renamed the variable "class" as "classbits" to make life easier for C++
2349        users. (Previously there was a macro definition, but it apparently wasn't
2350        enough.)
2352    12. Added the new field "tables" to the extra data so that tables can be passed
2353        in at exec time, or the internal tables can be re-selected. This allows
2354        a compiled regex to be saved and re-used at a later time by a different
2355        program that might have everything at different addresses.
2357    13. Modified the pcre-config script so that, when run on Solaris, it shows a
2358        -R library as well as a -L library.
2360    14. The debugging options of pcretest (-d on the command line or D on a
2361        pattern) showed incorrect output for anything following an extended class
2362        that contained multibyte characters and which was followed by a quantifier.
2364    15. Added optional support for general category Unicode character properties
2365        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2366        support. It adds about 90K to the size of the library. The meanings of the
2367        inbuilt class escapes such as \d and \s have NOT been changed.
2369    16. Updated pcredemo.c to include calls to free() to release the memory for the
2370        compiled pattern.
2372    17. The generated file chartables.c was being created in the source directory
2373        instead of in the building directory. This caused the build to fail if the
2374        source directory was different from the building directory, and was
2375        read-only.
2377    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2378        file. No doubt somebody will tell me if they don't make sense... Also added
2379        Dan Mooney's comments about building on OpenVMS.
2381    19. Added support for partial matching via the PCRE_PARTIAL option for
2382        pcre_exec() and the \P data escape in pcretest.
2384    20. Extended pcretest with 3 new pattern features:
2386        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
2387              write the compiled pattern to the file whose name is "rest-of-line".
2388              This is a straight binary dump of the data, with the saved pointer to
2389              the character tables forced to be NULL. The study data, if any, is
2390              written too. After writing, pcretest reads a new pattern.
2392        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2393              compiled pattern from the given file. There must not be any
2394              occurrences of "<" in the file name (pretty unlikely); if there are,
2395              pcretest will instead treat the initial "<" as a pattern delimiter.
2396              After reading in the pattern, pcretest goes on to read data lines as
2397              usual.
2399        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2400              and 16-bit fields in a compiled pattern, to simulate a pattern that
2401              was compiled on a host of opposite endianness.
2403    21. The pcre-exec() function can now cope with patterns that were compiled on
2404        hosts of opposite endianness, with this restriction:
2406          As for any compiled expression that is saved and used later, the tables
2407          pointer field cannot be preserved; the extra_data field in the arguments
2408          to pcre_exec() should be used to pass in a tables address if a value
2409          other than the default internal tables were used at compile time.
2411    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2412        now diagnosed as an error. Previously, most of the time, a negative number
2413        would have been treated as zero, but if in addition "ovector" was passed as
2414        NULL, a crash could occur.
2416    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2417        new versions from the libtool 1.5 distribution (the last one is a copy of
2418        a file called libtool.m4). This seems to have fixed the need to patch
2419        "configure" to support Darwin 1.3 (which I used to do). However, I still
2420        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2421        workstation).
2423    24. Changed the PCRE licence to be the more standard "BSD" licence.
2426    Version 4.5 01-Dec-03
2427    ---------------------
2429     1. There has been some re-arrangement of the code for the match() function so
2430        that it can be compiled in a version that does not call itself recursively.
2431        Instead, it keeps those local variables that need separate instances for
2432        each "recursion" in a frame on the heap, and gets/frees frames whenever it
2433        needs to "recurse". Keeping track of where control must go is done by means
2434        of setjmp/longjmp. The whole thing is implemented by a set of macros that
2435        hide most of the details from the main code, and operates only if
2436        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2437        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2438        operating.
2440        To make it easier for callers to provide specially tailored get/free
2441        functions for this usage, two new functions, pcre_stack_malloc, and
2442        pcre_stack_free, are used. They are always called in strict stacking order,
2443        and the size of block requested is always the same.
2445        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2446        PCRE has been compiled to use the stack or the heap for recursion. The
2447        -C option of pcretest uses this to show which version is compiled.
2449        A new data escape \S, is added to pcretest; it causes the amounts of store
2450        obtained and freed by both kinds of malloc/free at match time to be added
2451        to the output.
2453     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2454        what's available on my current Linux desktop machine.
2456     3. When matching a UTF-8 string, the test for a valid string at the start has
2457        been extended. If start_offset is not zero, PCRE now checks that it points
2458        to a byte that is the start of a UTF-8 character. If not, it returns
2459        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2460        this is necessary because there may be backward assertions in the pattern.
2461        When matching the same subject several times, it may save resources to use
2462        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2464     4. The code for checking the validity of UTF-8 strings has been tightened so
2465        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2466        containing "overlong sequences".
2468     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2469        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2470        should have been "&", but it just so happened that all the cases this let
2471        through by mistake were picked up later in the function.
2473     6. I had used a variable called "isblank" - this is a C99 function, causing
2474        some compilers to warn. To avoid this, I renamed it (as "blankclass").
2476     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2477        prompting; (b) run "./pcretest /dev/null" at the start of the test script
2478        so the version is shown; (c) stop "make test" echoing "./RunTest".
2480     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2482     9. The prototype for memmove() for systems that don't have it was using
2483        size_t, but the inclusion of the header that defines size_t was later. I've
2484        moved the #includes for the C headers earlier to avoid this.
2486    10. Added some adjustments to the code to make it easier to compiler on certain
2487        special systems:
2489          (a) Some "const" qualifiers were missing.
2490          (b) Added the macro EXPORT before all exported functions; by default this
2491              is defined to be empty.
2492          (c) Changed the dftables auxiliary program (that builds chartables.c) so
2493              that it reads its output file name as an argument instead of writing
2494              to the standard output and assuming this can be redirected.
2496    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2497        class containing characters with values greater than 255, PCRE compilation
2498        went into a loop.
2500    12. A recursive reference to a subpattern that was within another subpattern
2501        that had a minimum quantifier of zero caused PCRE to crash. For example,
2502        (x(y(?2))z)? provoked this bug with a subject that got as far as the
2503        recursion. If the recursively-called subpattern itself had a zero repeat,
2504        that was OK.
2506    13. In pcretest, the buffer for reading a data line was set at 30K, but the
2507        buffer into which it was copied (for escape processing) was still set at
2508        1024, so long lines caused crashes.
2510    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2511        "internal error: code overflow...". This applied to any character class
2512        that was followed by a possessive quantifier.
2514    15. Modified the Makefile to add libpcre.la as a prerequisite for
2515        libpcreposix.la because I was told this is needed for a parallel build to
2516        work.
2518    16. If a pattern that contained .* following optional items at the start was
2519        studied, the wrong optimizing data was generated, leading to matching
2520        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2521        matching string must start with a or b or c. The correct conclusion for
2522        this pattern is that a match can start with any character.
2525    Version 4.4 13-Aug-03
2526    ---------------------
2528     1. In UTF-8 mode, a character class containing characters with values between
2529        127 and 255 was not handled correctly if the compiled pattern was studied.
2530        In fixing this, I have also improved the studying algorithm for such
2531        classes (slightly).
2533     2. Three internal functions had redundant arguments passed to them. Removal
2534        might give a very teeny performance improvement.
2536     3. Documentation bug: the value of the capture_top field in a callout is *one
2537        more than* the number of the hightest numbered captured substring.
2539     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2540        in incorrectly linking with a previously installed version. They now link
2541        explicitly with libpcre.la.
2543     5. configure.in no longer needs to recognize Cygwin specially.
2545     6. A problem in pcre.in for Windows platforms is fixed.
2547     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2548        pcretest, it used to include the size of the study block as part of its
2549        output. Unfortunately, the structure contains a field that has a different
2550        size on different hardware architectures. This meant that the tests that
2551        showed this size failed. As the block is currently always of a fixed size,
2552        this information isn't actually particularly useful in pcretest output, so
2553        I have just removed it.
2555     8. Three pre-processor statements accidentally did not start in column 1.
2556        Sadly, there are *still* compilers around that complain, even though
2557        standard C has not required this for well over a decade. Sigh.
2559     9. In pcretest, the code for checking callouts passed small integers in the
2560        callout_data field, which is a void * field. However, some picky compilers
2561        complained about the casts involved for this on 64-bit systems. Now
2562        pcretest passes the address of the small integer instead, which should get
2563        rid of the warnings.
2565    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2566        both compile and run time, and gives an error if an invalid UTF-8 sequence
2567        is found. There is a option for disabling this check in cases where the
2568        string is known to be correct and/or the maximum performance is wanted.
2570    11. In response to a bug report, I changed one line in Makefile.in from
2572            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2573        to
2574            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2576        to look similar to other lines, but I have no way of telling whether this
2577        is the right thing to do, as I do not use Windows. No doubt I'll get told
2578        if it's wrong...
2581    Version 4.3 21-May-03
2582    ---------------------
2584    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2585       Makefile.
2587    2. Some refactoring to improve the quality of the code:
2589       (i)   The utf8_table... variables are now declared "const".
2591       (ii)  The code for \cx, which used the "case flipping" table to upper case
2592             lower case letters, now just substracts 32. This is ASCII-specific,
2593             but the whole concept of \cx is ASCII-specific, so it seems
2594             reasonable.
2596       (iii) PCRE was using its character types table to recognize decimal and
2597             hexadecimal digits in the pattern. This is silly, because it handles
2598             only 0-9, a-f, and A-F, but the character types table is locale-
2599             specific, which means strange things might happen. A private
2600             table is now used for this - though it costs 256 bytes, a table is
2601             much faster than multiple explicit tests. Of course, the standard
2602             character types table is still used for matching digits in subject
2603             strings against \d.
2605       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2606             ending in _t are). So I've renamed it as ESC_tee.
2608    3. The first argument for regexec() in the POSIX wrapper should have been
2609       defined as "const".
2611    4. Changed pcretest to use malloc() for its buffers so that they can be
2612       Electric Fenced for debugging.
2614    5. There were several places in the code where, in UTF-8 mode, PCRE would try
2615       to read one or more bytes before the start of the subject string. Often this
2616       had no effect on PCRE's behaviour, but in some circumstances it could
2617       provoke a segmentation fault.
2619    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2620       to try to read one or more bytes before the start of the subject string.
2622    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
2623       UTF-8 support could misbehave in various ways if the subject string
2624       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
2625       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
2626       back over UTF-8 characters.)
2629    Version 4.2 14-Apr-03
2630    ---------------------
2632    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
2634    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
2635         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
2636         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
2637         [WIN_PREFIX]: new variable, "cyg" for Cygwin
2638         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
2639           and BUILD_EXEEXT
2640         Note: automatic setting of the BUILD variables is not yet working
2641         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
2642           compile-time but not at link-time
2643         [LINK]: use for linking executables only
2644         make different versions for Windows and non-Windows
2645         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
2646           libraries
2647         [LINK_FOR_BUILD]: new variable
2648         [OBJEXT]: use throughout
2649         [EXEEXT]: use throughout
2650         <winshared>: new target
2651         <wininstall>: new target
2652         <dftables.o>: use native compiler
2653         <dftables>: use native linker
2654         <install>: handle Windows platform correctly
2655         <clean>: ditto
2656         <check>: ditto
2657         copy DLL to top builddir before testing
2659       As part of these changes, -no-undefined was removed again. This was reported
2660       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
2661       in any case.
2663    3. Some tidies to get rid of compiler warnings:
2665       . In the match_data structure, match_limit was an unsigned long int, whereas
2666         match_call_count was an int. I've made them both unsigned long ints.
2668       . In pcretest the fact that a const uschar * doesn't automatically cast to
2669         a void * provoked a warning.
2671       . Turning on some more compiler warnings threw up some "shadow" variables
2672         and a few more missing casts.
2674    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
2675       option, a class that contained a single character with a value between 128
2676       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
2678    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
2679       option, a class that contained several characters, but with at least one
2680       whose value was between 128 and 255 caused PCRE to crash.
2683    Version 4.1 12-Mar-03
2684    ---------------------
2686    1. Compiling with gcc -pedantic found a couple of places where casts were
2687    needed, and a string in dftables.c that was longer than standard compilers are
2688    required to support.
2690    2. Compiling with Sun's compiler found a few more places where the code could
2691    be tidied up in order to avoid warnings.
2693    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
2694    first of these names is deprecated in the latest Autoconf in favour of the name
2695    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
2696    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2697    analogy I have changed it to CFLAGS_FOR_BUILD.
2699    4. Added -no-undefined to the linking command in the Makefile, because this is
2700    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2701    linking step for the pcreposix library.
2703    5. PCRE was failing to diagnose the case of two named groups with the same
2704    name.
2706    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2707    literal character that is needed in the subject for a match, and scans along to
2708    ensure that it is present before embarking on the full matching process. This
2709    saves time in cases of nested unlimited repeats that are never going to match.
2710    Problem: the scan can take a lot of time if the subject is very long (e.g.
2711    megabytes), thus penalizing straightforward matches. It is now done only if the
2712    amount of subject to be scanned is less than 1000 bytes.
2714    7. A lesser problem with the same optimization is that it was recording the
2715    first character of an anchored pattern as "needed", thus provoking a search
2716    right along the subject, even when the first match of the pattern was going to
2717    fail. The "needed" character is now not set for anchored patterns, unless it
2718    follows something in the pattern that is of non-fixed length. Thus, it still
2719    fulfils its original purpose of finding quick non-matches in cases of nested
2720    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
2723    Version 4.0 17-Feb-03
2724    ---------------------
2726    1. If a comment in an extended regex that started immediately after a meta-item
2727    extended to the end of string, PCRE compiled incorrect data. This could lead to
2728    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
2730    2. Moved to autoconf 2.53 and libtool 1.4.2.
2732    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
2733    the special perltest8 script is no longer needed - all the tests can be run
2734    from a single perltest script.
2736    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
2737    by \s. It has now been removed in PCRE. This means it isn't recognized as
2738    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
2739    class [:space:] *does* include VT, thereby creating a mess.
2741    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
2742    space and tab.
2744    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
2745    its new features into the main test script, reducing the number of scripts.
2747    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
2748    were backward compatible, and made the (?i) apply to the whole pattern, as if
2749    /i were given. Now it behaves more logically, and applies the option setting
2750    only to what follows. PCRE has been changed to follow suit. However, if it
2751    finds options settings right at the start of the pattern, it extracts them into
2752    the global options, as before. Thus, they show up in the info data.
2754    8. Added support for the \Q...\E escape sequence. Characters in between are
2755    treated as literals. This is slightly different from Perl in that $ and @ are
2756    also handled as literals inside the quotes. In Perl, they will cause variable
2757    interpolation. Note the following examples:
2759        Pattern            PCRE matches      Perl matches
2761        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
2762        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
2763        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
2765    For compatibility with Perl, \Q...\E sequences are recognized inside character
2766    classes as well as outside them.
2768    9. Re-organized 3 code statements in pcretest to avoid "overflow in
2769    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2770    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2771    signed/unsigned warnings.
2773    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2774    option for pcretest, so I've replaced it by a simple function that does just
2775    that job.
2777    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2778    "pcregrep -".
2780    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2781    Java package. This provides some syntactic sugar for simple cases of what my
2782    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2783    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2784    item, you can use this simplified notation. Note that only makes sense with
2785    greedy quantifiers. Consequently, the use of the possessive quantifier forces
2786    greediness, whatever the setting of the PCRE_UNGREEDY option.
2788    13. A change of greediness default within a pattern was not taking effect at
2789    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2790    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2791    was abstracted outside.
2793    14. PCRE now supports the \G assertion. It is true when the current matching
2794    position is at the start point of the match. This differs from \A when the
2795    starting offset is non-zero. Used with the /g option of pcretest (or similar
2796    code), it works in the same way as it does for Perl's /g option. If all
2797    alternatives of a regex begin with \G, the expression is anchored to the start
2798    match position, and the "anchored" flag is set in the compiled expression.
2800    15. Some bugs concerning the handling of certain option changes within patterns
2801    have been fixed. These applied to options other than (?ims). For example,
2802    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2803    been the other way round. Some of this was related to change 7 above.
2805    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2806    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2807    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2808    POSIX classes only within a class (e.g. /[[:alpha:]]/).
2810    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2811    mode. Unlike ".", it always matches newline, whatever the setting of
2812    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2813    assertions. Perl allows it, but it doesn't (in general) work because it can't
2814    calculate the length of the lookbehind. At least, that's the case for Perl
2815    5.8.0 - I've been told they are going to document that it doesn't work in
2816    future.
2818    18. Added an error diagnosis for escapes that PCRE does not support: these are
2819    \L, \l, \N, \P, \p, \U, \u, and \X.
2821    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2822    reading past the end of the pattern in cases such as /[abcd/.
2824    20. PCRE was getting more memory than necessary for patterns with classes that
2825    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2827    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2828    compiling PCRE for use with Virtual Pascal.
2830    22. Small fix to the Makefile to make it work properly if the build is done
2831    outside the source tree.
2833    23. Added a new extension: a condition to go with recursion. If a conditional
2834    subpattern starts with (?(R) the "true" branch is used if recursion has
2835    happened, whereas the "false" branch is used only at the top level.
2837    24. When there was a very long string of literal characters (over 255 bytes
2838    without UTF support, over 250 bytes with UTF support), the computation of how
2839    much memory was required could be incorrect, leading to segfaults or other
2840    strange effects.
2842    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2843    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2844    there was a subsequent back reference to those brackets. This meant that, for
2845    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2846    possible to check for precisely this case. All we can do is abandon the
2847    optimization if .* occurs inside capturing brackets when there are any back
2848    references whatsoever. (See below for a better fix that came later.)
2850    26. The handling of the optimization for finding the first character of a
2851    non-anchored pattern, and for finding a character that is required later in the
2852    match were failing in some cases. This didn't break the matching; it just
2853    failed to optimize when it could. The way this is done has been re-implemented.
2855    27. Fixed typo in error message for invalid (?R item (it said "(?p").
2857    28. Added a new feature that provides some of the functionality that Perl
2858    provides with (?{...}). The facility is termed a "callout". The way it is done
2859    in PCRE is for the caller to provide an optional function, by setting
2860    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2861    global variable. By default it is unset, which disables all calling out. To get
2862    the function called, the regex must include (?C) at appropriate points. This
2863    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2864    This provides a means of identifying different callout points. When PCRE
2865    reaches such a point in the regex, if pcre_callout has been set, the external
2866    function is called. It is provided with data in a structure called
2867    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2868    matching continues; if it returns a non-zero value, the match at the current
2869    point fails. However, backtracking will occur if possible. [This was changed
2870    later and other features added - see item 49 below.]
2872    29. pcretest is upgraded to test the callout functionality. It provides a
2873    callout function that displays information. By default, it shows the start of
2874    the match and the current position in the text. There are some new data escapes
2875    to vary what happens:
2877        \C+         in addition, show current contents of captured substrings
2878        \C-         do not supply a callout function
2879        \C!n        return 1 when callout number n is reached
2880        \C!n!m      return 1 when callout number n is reached for the mth time
2882    30. If pcregrep was called with the -l option and just a single file name, it
2883    output "<stdin>" if a match was found, instead of the file name.
2885    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2886    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2887    pcre_exec(). This saves a malloc/free per call. The default value of
2888    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2889    when configuring.
2891    32. The default maximum size of a compiled pattern is 64K. There have been a
2892    few cases of people hitting this limit. The code now uses macros to handle the
2893    storing of links as offsets within the compiled pattern. It defaults to 2-byte
2894    links, but this can be changed to 3 or 4 bytes by --with-link-size when
2895    configuring. Tests 2 and 5 work only with 2-byte links because they output
2896    debugging information about compiled patterns.
2898    33. Internal code re-arrangements:
2900    (a) Moved the debugging function for printing out a compiled regex into
2901        its own source file (printint.c) and used #include to pull it into
2902        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2903        separate copies.
2905    (b) Defined the list of op-code names for debugging as a macro in
2906        internal.h so that it is next to the definition of the opcodes.
2908    (c) Defined a table of op-code lengths for simpler skipping along compiled
2909        code. This is again a macro in internal.h so that it is next to the
2910        definition of the opcodes.
2912    34. Added support for recursive calls to individual subpatterns, along the
2913    lines of Robin Houston's patch (but implemented somewhat differently).
2915    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2916    allow it to read and process whole directories in Win32. This code was
2917    contributed by Lionel Fourquaux; it has not been tested by me.
2919    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2920    used to name a group. Names consist of alphanumerics and underscores, and must
2921    be unique. Back references use the syntax (?P=name) and recursive calls use
2922    (?P>name) which is a PCRE extension to the Python extension. Groups still have
2923    numbers. The function pcre_fullinfo() can be used after compilation to extract
2924    a name/number map. There are three relevant calls:
2926      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
2927      PCRE_INFO_NAMECOUNT            yields the number of entries
2928      PCRE_INFO_NAMETABLE            yields a pointer to the map.
2930    The map is a vector of fixed-size entries. The size of each entry depends on
2931    the length of the longest name used. The first two bytes of each entry are the
2932    group number, most significant byte first. There follows the corresponding
2933    name, zero terminated. The names are in alphabetical order.
2935    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2936    case instead of 255. Making it the same both with and without UTF-8 support
2937    means that the same test output works with both.
2939    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2940    calling malloc() with a zero argument.
2942    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2943    optimization. I've improved things by keeping a bitmap of backreferences with
2944    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2945    fact referenced, the optimization can be applied. It is unlikely that a
2946    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2947    the match to follow \n) will appear inside brackets with a number greater than
2948    31, but if it does, any back reference > 31 suppresses the optimization.
2950    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2951    of disabling numbered capturing parentheses. Any opening parenthesis that is
2952    not followed by ? behaves as if it were followed by ?: but named parentheses
2953    can still be used for capturing (and they will acquire numbers in the usual
2954    way).
2956    41. Redesigned the return codes from the match() function into yes/no/error so
2957    that errors can be passed back from deep inside the nested calls. A malloc
2958    failure while inside a recursive subpattern call now causes the
2959    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2961    42. It is now possible to set a limit on the number of times the match()
2962    function is called in a call to pcre_exec(). This facility makes it possible to
2963    limit the amount of recursion and backtracking, though not in a directly
2964    obvious way, because the match() function is used in a number of different
2965    circumstances. The count starts from zero for each position in the subject
2966    string (for non-anchored patterns). The default limit is, for compatibility, a
2967    large number, namely 10 000 000. You can change this in two ways:
2969    (a) When configuring PCRE before making, you can use --with-match-limit=n
2970        to set a default value for the compiled library.
2972    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2973        a different value is set. See 45 below.
2975    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2977    43. Added a new function pcre_config(int, void *) to enable run-time extraction
2978    of things that can be changed at compile time. The first argument specifies
2979    what is wanted and the second points to where the information is to be placed.
2980    The current list of available information is:
2982      PCRE_CONFIG_UTF8
2984    The output is an integer that is set to one if UTF-8 support is available;
2985    otherwise it is set to zero.
2989    The output is an integer that it set to the value of the code that is used for
2990    newline. It is either LF (10) or CR (13).
2994    The output is an integer that contains the number of bytes used for internal
2995    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2999    The output is an integer that contains the threshold above which the POSIX
3000    interface uses malloc() for output vectors. See item 31 above.
3004    The output is an unsigned integer that contains the default limit of the number
3005    of match() calls in a pcre_exec() execution. See 42 above.
3007    44. pcretest has been upgraded by the addition of the -C option. This causes it
3008    to extract all the available output from the new pcre_config() function, and to
3009    output it. The program then exits immediately.
3011    45. A need has arisen to pass over additional data with calls to pcre_exec() in
3012    order to support additional features. One way would have been to define
3013    pcre_exec2() (for example) with extra arguments, but this would not have been
3014    extensible, and would also have required all calls to the original function to
3015    be mapped to the new one. Instead, I have chosen to extend the mechanism that
3016    is used for passing in "extra" data from pcre_study().
3018    The pcre_extra structure is now exposed and defined in pcre.h. It currently
3019    contains the following fields:
3021      flags         a bitmap indicating which of the following fields are set
3022      study_data    opaque data from pcre_study()
3023      match_limit   a way of specifying a limit on match() calls for a specific
3024                      call to pcre_exec()
3025      callout_data  data for callouts (see 49 below)
3027    The flag bits are also defined in pcre.h, and are
3033    The pcre_study() function now returns one of these new pcre_extra blocks, with
3034    the actual study data pointed to by the study_data field, and the
3035    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
3036    before. That is, this change is entirely upwards-compatible and requires no
3037    change to existing code.
3039    If you want to pass in additional data to pcre_exec(), you can either place it
3040    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
3041    block.
3043    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
3044    data string contains the escape sequence \M, pcretest calls pcre_exec() several
3045    times with different match limits, until it finds the minimum value needed for
3046    pcre_exec() to complete. The value is then output. This can be instructive; for
3047    most simple matches the number is quite small, but for pathological cases it
3048    gets very large very quickly.
3050    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
3051    returns the size of the data block pointed to by the study_data field in a
3052    pcre_extra block, that is, the value that was passed as the argument to
3053    pcre_malloc() when PCRE was getting memory in which to place the information
3054    created by pcre_study(). The fourth argument should point to a size_t variable.
3055    pcretest has been extended so that this information is shown after a successful
3056    pcre_study() call when information about the compiled regex is being displayed.
3058    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
3059    because what follows is always an absolute path. (Later: it turns out that this
3060    is more than cosmetic for MinGW, because it doesn't like empty path
3061    components.)
3063    49. Some changes have been made to the callout feature (see 28 above):
3065    (i)  A callout function now has three choices for what it returns:
3067           0  =>  success, carry on matching
3068         > 0  =>  failure at this point, but backtrack if possible
3069         < 0  =>  serious error, return this value from pcre_exec()
3071         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
3072         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
3073         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
3074         use by callout functions. It will never be used by PCRE itself.
3076    (ii) The pcre_extra structure (see 45 above) has a void * field called
3077         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
3078         pcre_callout_block structure has a field of the same name. The contents of
3079         the field passed in the pcre_extra structure are passed to the callout
3080         function in the corresponding field in the callout block. This makes it
3081         easier to use the same callout-containing regex from multiple threads. For
3082         testing, the pcretest program has a new data escape
3084           \C*n        pass the number n (may be negative) as callout_data
3086         If the callout function in pcretest receives a non-zero value as
3087         callout_data, it returns that value.
3089    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
3090    there were some redundant $(CFLAGS) in commands that are now specified as
3091    $(LINK), which already includes $(CFLAGS).
3093    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
3094    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
3095    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
3096    one-byte characters throughout. Note that case-insensitive matching applies
3097    only to characters whose values are less than 256. PCRE doesn't support the
3098    notion of cases for higher-valued characters.
3100    (i)   A character class whose characters are all within 0-255 is handled as
3101          a bit map, and the map is inverted for negative classes. Previously, a
3102          character > 255 always failed to match such a class; however it should
3103          match if the class was a negative one (e.g. [^ab]). This has been fixed.
3105    (ii)  A negated character class with a single character < 255 is coded as
3106          "not this character" (OP_NOT). This wasn't working properly when the test
3107          character was multibyte, either singly or repeated.
3109    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
3110          mode, for example: \x{100}{2,3}.
3112    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
3113          singly or repeated) now correctly test multibyte characters. However,
3114          PCRE doesn't recognize any characters with values greater than 255 as
3115          digits, spaces, or word characters. Such characters always match \D, \S,
3116          and \W, and never match \d, \s, or \w.
3118    (v)   Classes may now contain characters and character ranges with values
3119          greater than 255. For example: [ab\x{100}-\x{400}].
3121    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
3122          PCRE in UTF-8 mode.
3124    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
3125    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
3126    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
3127    value.)
3129    53. The single man page has become too large. I have therefore split it up into
3130    a number of separate man pages. These also give rise to individual HTML pages;
3131    these are now put in a separate directory, and there is an index.html page that
3132    lists them all. Some hyperlinking between the pages has been installed.
3134    54. Added convenience functions for handling named capturing parentheses.
3136    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
3137    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
3138    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
3139    are faulted.
3141    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
3142    calling configure. These values are used when compiling the dftables.c program
3143    which is run to generate the source of the default character tables. They
3144    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
3145    you will need to set these values.
3147    57. Updated the building process for Windows DLL, as provided by Fred Cox.
3150    Version 3.9 02-Jan-02
3151    ---------------------
3153    1. A bit of extraneous text had somehow crept into the pcregrep documentation.
3155    2. If --disable-static was given, the building process failed when trying to
3156    build pcretest and pcregrep. (For some reason it was using libtool to compile
3157    them, which is not right, as they aren't part of the library.)
3160    Version 3.8 18-Dec-01
3161    ---------------------
3163    1. The experimental UTF-8 code was completely screwed up. It was packing the
3164    bytes in the wrong order. How dumb can you get?
3167    Version 3.7 29-Oct-01
3168    ---------------------
3170    1. In updating pcretest to check change 1 of version 3.6, I screwed up.
3171    This caused pcretest, when used on the test data, to segfault. Unfortunately,
3172    this didn't happen under Solaris 8, where I normally test things.
3174    2. The Makefile had to be changed to make it work on BSD systems, where 'make'
3175    doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
3176    isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
3177    this fix an hour or so after the initial 3.7 release.)
3180    Version 3.6 23-Oct-01
3181    ---------------------
3183    1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
3184    offsets passed as NULL with zero offset count.
3186    2. The config.guess and config.sub files had not been updated when I moved to
3187    the latest autoconf.
3190    Version 3.5 15-Aug-01
3191    ---------------------
3193    1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
3194    had been forgotten.
3196    2. By using declared but undefined structures, we can avoid using "void"
3197    definitions in pcre.h while keeping the internal definitions of the structures
3198    private.
3200    3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
3201    user point of view, this means that both static and shared libraries are built
3202    by default, but this can be individually controlled. More of the work of
3203    handling this static/shared cases is now inside libtool instead of PCRE's make
3204    file.
3206    4. The pcretest utility is now installed along with pcregrep because it is
3207    useful for users (to test regexs) and by doing this, it automatically gets
3208    relinked by libtool. The documentation has been turned into a man page, so
3209    there are now .1, .txt, and .html versions in /doc.
3211    5. Upgrades to pcregrep:
3212       (i)   Added long-form option names like gnu grep.
3213       (ii)  Added --help to list all options with an explanatory phrase.
3214       (iii) Added -r, --recursive to recurse into sub-directories.
3215       (iv)  Added -f, --file to read patterns from a file.
3217    6. pcre_exec() was referring to its "code" argument before testing that
3218    argument for NULL (and giving an error if it was NULL).
3220    7. Upgraded Makefile.in to allow for compiling in a different directory from
3221    the source directory.
3223    8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
3224    options bits, the pointer it was passed was to an int instead of to an unsigned
3225    long int. This mattered only on 64-bit systems.
3227    9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
3228    generated) instead of pcre.in, which it its source. Also made the same change
3229    in several of the .c files.
3231    10. A new release of gcc defines printf() as a macro, which broke pcretest
3232    because it had an ifdef in the middle of a string argument for printf(). Fixed
3233    by using separate calls to printf().
3235    11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
3236    script, to force use of CR or LF instead of \n in the source. On non-Unix
3237    systems, the value can be set in config.h.
3239    12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
3240    absolute limit. Changed the text of the error message to make this clear, and
3241    likewise updated the man page.
3243    13. The limit of 99 on the number of capturing subpatterns has been removed.
3244    The new limit is 65535, which I hope will not be a "real" limit.
3247    Version 3.4 22-Aug-00
3248    ---------------------
3250    1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
3252    2. Diagnose condition (?(0) as an error instead of crashing on matching.
3255    Version 3.3 01-Aug-00
3256    ---------------------
3258    1. If an octal character was given, but the value was greater than \377, it
3259    was not getting masked to the least significant bits, as documented. This could
3260    lead to crashes in some systems.
3262    2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
3263    the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
3265    3. Added the functions pcre_free_substring() and pcre_free_substring_list().
3266    These just pass their arguments on to (pcre_free)(), but they are provided
3267    because some uses of PCRE bind it to non-C systems that can call its functions,
3268    but cannot call free() or pcre_free() directly.
3270    4. Add "make test" as a synonym for "make check". Corrected some comments in
3271    the Makefile.
3273    5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
3274    Makefile.
3276    6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
3277    command called pgrep for grepping around the active processes.
3279    7. Added the beginnings of support for UTF-8 character strings.
3281    8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
3282    RANLIB to ./ltconfig so that they are used by libtool. I think these are all
3283    the relevant ones. (AR is not passed because ./ltconfig does its own figuring
3284    out for the ar command.)
3287    Version 3.2 12-May-00
3288    ---------------------
3290    This is purely a bug fixing release.
3292    1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
3293    of ZA. This was just one example of several cases that could provoke this bug,
3294    which was introduced by change 9 of version 2.00. The code for breaking
3295    infinite loops after an iteration that matches an empty string was't working
3296    correctly.
3298    2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
3299    when matched against abbab (for example). After matching an empty string, it
3300    wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
3301    caused it to match further down the string than it should.
3303    3. The code contained an inclusion of sys/types.h. It isn't clear why this
3304    was there because it doesn't seem to be needed, and it causes trouble on some
3305    systems, as it is not a Standard C header. It has been removed.
3307    4. Made 4 silly changes to the source to avoid stupid compiler warnings that
3308    were reported on the Macintosh. The changes were from
3310      while ((c = *(++ptr)) != 0 && c != '\n');
3311    to
3312      while ((c = *(++ptr)) != 0 && c != '\n') ;
3314    Totally extraordinary, but if that's what it takes...
3316    5. PCRE is being used in one environment where neither memmove() nor bcopy() is
3317    available. Added HAVE_BCOPY and an autoconf test for it; if neither
3318    HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
3319    assumes the way PCRE uses memmove() (always moving upwards).
3321    6. PCRE is being used in one environment where strchr() is not available. There
3322    was only one use in pcre.c, and writing it out to avoid strchr() probably gives
3323    faster code anyway.
3326    Version 3.1 09-Feb-00
3327    ---------------------
3329    The only change in this release is the fixing of some bugs in Makefile.in for
3330    the "install" target:
3332    (1) It was failing to install pcreposix.h.
3334    (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
3337    Version 3.0 01-Feb-00
3338    ---------------------
3340    1. Add support for the /+ modifier to perltest (to output $` like it does in
3341    pcretest).
3343    2. Add support for the /g modifier to perltest.
3345    3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
3346    matches null strings.
3348    4. Fix perltest so that it doesn't do unwanted things when fed an empty
3349    pattern. Perl treats empty patterns specially - it reuses the most recent
3350    pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
3351    effect.
3353    5. The POSIX interface was broken in that it was just handing over the POSIX
3354    captured string vector to pcre_exec(), but (since release 2.00) PCRE has
3355    required a bigger vector, with some working space on the end. This means that
3356    the POSIX wrapper now has to get and free some memory, and copy the results.
3358    6. Added some simple autoconf support, placing the test data and the
3359    documentation in separate directories, re-organizing some of the
3360    information files, and making it build pcre-config (a GNU standard). Also added
3361    libtool support for building PCRE as a shared library, which is now the
3362    default.
3364    7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
3365    09 are not valid octal constants. Single digits will be used for minor values
3366    less than 10.
3368    8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
3369    existing programs that set these in the POSIX interface can use PCRE without
3370    modification.
3372    9. Added a new function, pcre_fullinfo() with an extensible interface. It can
3373    return all that pcre_info() returns, plus additional data. The pcre_info()
3374    function is retained for compatibility, but is considered to be obsolete.
3376    10. Added experimental recursion feature (?R) to handle one common case that
3377    Perl 5.6 will be able to do with (?p{...}).
3379    11. Added support for POSIX character classes like [:alpha:], which Perl is
3380    adopting.
3383    Version 2.08 31-Aug-99
3384    ----------------------
3386    1. When startoffset was not zero and the pattern began with ".*", PCRE was not
3387    trying to match at the startoffset position, but instead was moving forward to
3388    the next newline as if a previous match had failed.
3390    2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
3391    and could get into a loop if a null string was matched other than at the start
3392    of the subject.
3394    3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
3395    be distinguished at compile time, and for completeness also added PCRE_DATE.
3397    5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
3398    in GnuWin32 environments.
3401    Version 2.07 29-Jul-99
3402    ----------------------
3404    1. The documentation is now supplied in plain text form and HTML as well as in
3405    the form of man page sources.
3407    2. C++ compilers don't like assigning (void *) values to other pointer types.
3408    In particular this affects malloc(). Although there is no problem in Standard
3409    C, I've put in casts to keep C++ compilers happy.
3411    3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
3412    should be (const char *).
3414    4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
3415    be useful for non-Unix systems who don't want to bother with the POSIX stuff.
3416    However, I haven't made this a standard facility. The documentation doesn't
3417    mention it, and the Makefile doesn't support it.
3419    5. The Makefile now contains an "install" target, with editable destinations at
3420    the top of the file. The pcretest program is not installed.
3422    6. pgrep -V now gives the PCRE version number and date.
3424    7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
3425    causing the entire string to be ignored, instead of just the last character.
3427    8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
3428    non-matching string, it can take a very, very long time, even for strings of
3429    quite modest length, because of the nested recursion. PCRE now does better in
3430    some of these cases. It does this by remembering the last required literal
3431    character in the pattern, and pre-searching the subject to ensure it is present
3432    before running the real match. In other words, it applies a heuristic to detect
3433    some types of certain failure quickly, and in the above example, if presented
3434    with a string that has no trailing " it gives "no match" very quickly.
3436    9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
3437    other alternatives are tried instead.
3440    Version 2.06 09-Jun-99
3441    ----------------------
3443    1. Change pcretest's output for amount of store used to show just the code
3444    space, because the remainder (the data block) varies in size between 32-bit and
3445    64-bit systems.
3447    2. Added an extra argument to pcre_exec() to supply an offset in the subject to
3448    start matching at. This allows lookbehinds to work when searching for multiple
3449    occurrences in a string.
3451    3. Added additional options to pcretest for testing multiple occurrences:
3453       /+   outputs the rest of the string that follows a match
3454       /g   loops for multiple occurrences, using the new startoffset argument
3455       /G   loops for multiple occurrences by passing an incremented pointer
3457    4. PCRE wasn't doing the "first character" optimization for patterns starting
3458    with \b or \B, though it was doing it for other lookbehind assertions. That is,
3459    it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
3460    the letter 'x'. On long subject strings, this gives a significant speed-up.
3463    Version 2.05 21-Apr-99
3464    ----------------------
3466    1. Changed the type of magic_number from int to long int so that it works
3467    properly on 16-bit systems.
3469    2. Fixed a bug which caused patterns starting with .* not to work correctly
3470    when the subject string contained newline characters. PCRE was assuming
3471    anchoring for such patterns in all cases, which is not correct because .* will
3472    not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
3473    DOTALL is set at top level; otherwise it knows that patterns starting with .*
3474    must be retried after every newline in the subject.
3477    Version 2.04 18-Feb-99
3478    ----------------------
3480    1. For parenthesized subpatterns with repeats whose minimum was zero, the
3481    computation of the store needed to hold the pattern was incorrect (too large).
3482    If such patterns were nested a few deep, this could multiply and become a real
3483    problem.
3485    2. Added /M option to pcretest to show the memory requirement of a specific
3486    pattern. Made -m a synonym of -s (which does this globally) for compatibility.
3488    3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
3489    compiled in such a way that the backtracking after subsequent failure was
3490    pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
3491    ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
3494    Version 2.03 02-Feb-99
3495    ----------------------
3497    1. Fixed typo and small mistake in man page.
3499    2. Added 4th condition (GPL supersedes if conflict) and created separate
3500    LICENCE file containing the conditions.
3502    3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
3503    Perl, that is the internal \ allows the delimiter to be included in the
3504    pattern. Locked out the use of \ as a delimiter. If \ immediately follows
3505    the final delimiter, add \ to the end of the pattern (to test the error).
3507    4. Added the convenience functions for extracting substrings after a successful
3508    match. Updated pcretest to make it able to test these functions.
3511    Version 2.02 14-Jan-99
3512    ----------------------
3514    1. Initialized the working variables associated with each extraction so that
3515    their saving and restoring doesn't refer to uninitialized store.
3517    2. Put dummy code into study.c in order to trick the optimizer of the IBM C
3518    compiler for OS/2 into generating correct code. Apparently IBM isn't going to
3519    fix the problem.
3521    3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
3522    calls, and wasn't printing the correct value for compiling calls. Increased the
3523    default value of LOOPREPEAT, and the number of significant figures in the
3524    times.
3526    4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
3528    5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
3529    a building problem on Windows NT with a FAT file system.
3532    Version 2.01 21-Oct-98
3533    ----------------------
3535    1. Changed the API for pcre_compile() to allow for the provision of a pointer
3536    to character tables built by pcre_maketables() in the current locale. If NULL
3537    is passed, the default tables are used.
3540    Version 2.00 24-Sep-98
3541    ----------------------
3543    1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
3544    it any more.
3546    2. Allow quantification of (?>) groups, and make it work correctly.
3548    3. The first character computation wasn't working for (?>) groups.
3550    4. Correct the implementation of \Z (it is permitted to match on the \n at the
3551    end of the subject) and add 5.005's \z, which really does match only at the
3552    very end of the subject.
3554    5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
3556    6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
3557    DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
3558    localized options. All options to pcre_study() were also removed.
3560    7. Add other new features from 5.005:
3562       $(?<=           positive lookbehind
3563       $(?<!           negative lookbehind
3564       (?imsx-imsx)    added the unsetting capability
3565                       such a setting is global if at outer level; local otherwise
3566       (?imsx-imsx:)   non-capturing groups with option setting
3567       (?(cond)re|re)  conditional pattern matching
3569       A backreference to itself in a repeated group matches the previous
3570       captured string.
3572    8. General tidying up of studying (both automatic and via "study")
3573    consequential on the addition of new assertions.
3575    9. As in 5.005, unlimited repeated groups that could match an empty substring
3576    are no longer faulted at compile time. Instead, the loop is forcibly broken at
3577    runtime if any iteration does actually match an empty substring.
3579    10. Include the RunTest script in the distribution.
3581    11. Added tests from the Perl 5.005_02 distribution. This showed up a few
3582    discrepancies, some of which were old and were also with respect to 5.004. They
3583    have now been fixed.
3586    Version 1.09 28-Apr-98
3587    ----------------------
3589    1. A negated single character class followed by a quantifier with a minimum
3590    value of one (e.g.  [^x]{1,6}  ) was not compiled correctly. This could lead to
3591    program crashes, or just wrong answers. This did not apply to negated classes
3592    containing more than one character, or to minima other than one.
3595  Version 1.08 27-Mar-98  Version 1.08 27-Mar-98
3596  ----------------------  ----------------------

Removed from v.19  
changed lines
  Added in v.517

  ViewVC Help
Powered by ViewVC 1.1.5