ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 53 by nigel, Sat Feb 24 21:39:42 2007 UTC revision 497 by ph10, Wed Mar 3 12:01:00 2010 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
4    Version 8.02 01-Mar-2010
5    ------------------------
7    1.  The Unicode data tables have been updated to Unicode 5.2.0.
9    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
10        configured.
12    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
13        original author of that file, following a query about its status.
15    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
16        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
18    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
19        quantifier applied to a forward-referencing subroutine call, could compile
20        incorrect code or give the error "internal error: previously-checked
21        referenced subpattern not found".
23    6.  Both MS Visual Studio and Symbian OS have problems with initializing
24        variables to point to external functions. For these systems, therefore,
25        pcre_malloc etc. are now initialized to local functions that call the
26        relevant global functions.
29    Version 8.01 19-Jan-2010
30    ------------------------
32    1.  If a pattern contained a conditional subpattern with only one branch (in
33        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
34        computed the wrong minimum data length (which is of course zero for such
35        subpatterns). This could cause incorrect "no match" results.
37    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
38        the pattern is reset in the first branch, pcre_compile() failed with
39        "internal error: code overflow at offset...". This happened only when
40        the reset was to the original external option setting. (An optimization
41        abstracts leading options settings into an external setting, which was the
42        cause of this.)
44    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
45        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
46        assertion pattern did not match (meaning that the assertion was true), it
47        was incorrectly treated as false if the SKIP had been reached during the
48        matching. This also applied to assertions used as conditions.
50    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
51        assertion subpattern, including such a pattern used as a condition,
52        unpredictable results occurred, instead of the error return
55    5.  The C++ GlobalReplace function was not working like Perl for the special
56        situation when an empty string is matched. It now does the fancy magic
57        stuff that is necessary.
59    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
60        removed. (These were left over from very, very early versions of PCRE.)
62    7.  Some cosmetic changes to the code to make life easier when compiling it
63        as part of something else:
65        (a) Change DEBUG to PCRE_DEBUG.
67        (b) In pcre_compile(), rename the member of the "branch_chain" structure
68            called "current" as "current_branch", to prevent a collision with the
69            Linux macro when compiled as a kernel module.
71        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
72            prevent a collision with the Linux macro when compiled as a kernel
73            module.
75    8.  In pcre_compile() there are some checks for integer overflows that used to
76        cast potentially large values to (double). This has been changed to that
77        when building, a check for int64_t is made, and if it is found, it is used
78        instead, thus avoiding the use of floating point arithmetic. (There is no
79        other use of FP in PCRE.) If int64_t is not found, the fallback is to
80        double.
82    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
83        2005 (difference between two addresses compared to an unsigned value).
85    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
86        custom one, because of the following reported problem in Windows:
88          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
89              under Win32.
90          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
91              therefore missing the function definition.
92          - The compiler thus generates a "C" signature for the test function.
93          - The linker fails to find the "C" function.
94          - PCRE fails to configure if asked to do so against libbz2.
96    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
97        messages were output:
99          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
100          rerunning libtoolize, to keep the correct libtool macros in-tree.
101          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
103        I have done both of these things.
105    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
106        most of the time, it *can* run out if it is given a pattern that contains a
107        runaway infinite recursion. I updated the discussion in the pcrestack man
108        page.
110    13. Now that we have gone to the x.xx style of version numbers, the minor
111        version may start with zero. Using 08 or 09 is a bad idea because users
112        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
113        interpreted as invalid octal numbers. I've updated the previous comment in
114        configure.ac, and also added a check that gives an error if 08 or 09 are
115        used.
117    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
118        causing partial matching to fail when the end of the subject matched \W
119        in a UTF-8 pattern where \W was quantified with a minimum of 3.
121    15. There were some discrepancies between the declarations in pcre_internal.h
122        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
123        their definitions. The declarations used "const uschar *" and the
124        definitions used USPTR. Even though USPTR is normally defined as "const
125        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
126        reported that: "This difference in casting confuses some C++ compilers, for
127        example, SunCC recognizes above declarations as different functions and
128        generates broken code for hbpcre." I have changed the declarations to use
129        USPTR.
131    16. GNU libtool is named differently on some systems. The autogen.sh script now
132        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
133        (FreeBSD).
135    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
136        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
137        comment: "Figure out how to create a longlong from a string: strtoll and
138        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
139        instance, but it only takes 2 args instead of 3!"
141    18. A subtle bug concerned with back references has been fixed by a change of
142        specification, with a corresponding code fix. A pattern such as
143        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
144        refers, was giving matches when it shouldn't. For example, xa=xaaa would
145        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
146        same bug. Such groups have to be quantified to be useful, or contained
147        inside another quantified group. (If there's no repetition, the reference
148        can never match.) The problem arises because, having left the group and
149        moved on to the rest of the pattern, a later failure that backtracks into
150        the group uses the captured value from the final iteration of the group
151        rather than the correct earlier one. I have fixed this in PCRE by forcing
152        any group that contains a reference to itself to be an atomic group; that
153        is, there cannot be any backtracking into it once it has completed. This is
154        similar to recursive and subroutine calls.
157    Version 8.00 19-Oct-09
158    ----------------------
160    1.  The table for translating pcre_compile() error codes into POSIX error codes
161        was out-of-date, and there was no check on the pcre_compile() error code
162        being within the table. This could lead to an OK return being given in
163        error.
165    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
166        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
167        in a Windows environment.
169    3.  The pcregrep --count option prints the count for each file even when it is
170        zero, as does GNU grep. However, pcregrep was also printing all files when
171        --files-with-matches was added. Now, when both options are given, it prints
172        counts only for those files that have at least one match. (GNU grep just
173        prints the file name in this circumstance, but including the count seems
174        more useful - otherwise, why use --count?) Also ensured that the
175        combination -clh just lists non-zero counts, with no names.
177    4.  The long form of the pcregrep -F option was incorrectly implemented as
178        --fixed_strings instead of --fixed-strings. This is an incompatible change,
179        but it seems right to fix it, and I didn't think it was worth preserving
180        the old behaviour.
182    5.  The command line items --regex=pattern and --regexp=pattern were not
183        recognized by pcregrep, which required --regex pattern or --regexp pattern
184        (with a space rather than an '='). The man page documented the '=' forms,
185        which are compatible with GNU grep; these now work.
187    6.  No libpcreposix.pc file was created for pkg-config; there was just
188        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
190    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
191        when UCP support is not needed, by modifying the Python script that
192        generates it from Unicode data files. This should not matter if the module
193        is correctly used as a library, but I received one complaint about 50K of
194        unwanted data. My guess is that the person linked everything into his
195        program rather than using a library. Anyway, it does no harm.
197    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
198        was a minimum greater than 1 for a wide character in a possessive
199        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
200        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
201        character. Chaos in the form of incorrect output or a compiling loop could
202        result.
204    9.  The restrictions on what a pattern can contain when partial matching is
205        requested for pcre_exec() have been removed. All patterns can now be
206        partially matched by this function. In addition, if there are at least two
207        slots in the offset vector, the offset of the earliest inspected character
208        for the match and the offset of the end of the subject are set in them when
209        PCRE_ERROR_PARTIAL is returned.
211    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
212        synonymous with PCRE_PARTIAL, for backwards compatibility, and
213        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
214        and may be more useful for multi-segment matching.
216    11. Partial matching with pcre_exec() is now more intuitive. A partial match
217        used to be given if ever the end of the subject was reached; now it is
218        given only if matching could not proceed because another character was
219        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
220        string "Z", which now yields "no match" instead of "partial match". In the
221        case of pcre_dfa_exec(), "no match" is given if every matching path for the
222        final character ended with (*FAIL).
224    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
225        if the pattern had a "must contain" character that was already found in the
226        earlier partial match, unless partial matching was again requested. For
227        example, with the pattern /dog.(body)?/, the "must contain" character is
228        "g". If the first part-match was for the string "dog", restarting with
229        "sbody" failed. This bug has been fixed.
231    13. The string returned by pcre_dfa_exec() after a partial match has been
232        changed so that it starts at the first inspected character rather than the
233        first character of the match. This makes a difference only if the pattern
234        starts with a lookbehind assertion or \b or \B (\K is not supported by
235        pcre_dfa_exec()). It's an incompatible change, but it makes the two
236        matching functions compatible, and I think it's the right thing to do.
238    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
239        so that the demonstration program is easily available in environments where
240        PCRE has not been installed from source.
242    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
243        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
244        library.
246    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
247        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
248        is not the first non-POSIX option to be added. Clearly some people find
249        these options useful.
251    17. If a caller to the POSIX matching function regexec() passes a non-zero
252        value for nmatch with a NULL value for pmatch, the value of
253        nmatch is forced to zero.
255    18. RunGrepTest did not have a test for the availability of the -u option of
256        the diff command, as RunTest does. It now checks in the same way as
257        RunTest, and also checks for the -b option.
259    19. If an odd number of negated classes containing just a single character
260        interposed, within parentheses, between a forward reference to a named
261        subpattern and the definition of the subpattern, compilation crashed with
262        an internal error, complaining that it could not find the referenced
263        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
264        [The bug was that it was starting one character too far in when skipping
265        over the character class, thus treating the ] as data rather than
266        terminating the class. This meant it could skip too much.]
268    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
269        /g option in pcretest when the pattern contains \K, which makes it possible
270        to have an empty string match not at the start, even when the pattern is
271        anchored. Updated pcretest and pcredemo to use this option.
273    21. If the maximum number of capturing subpatterns in a recursion was greater
274        than the maximum at the outer level, the higher number was returned, but
275        with unset values at the outer level. The correct (outer level) value is
276        now given.
278    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
279        PCRE did not set those parentheses (unlike Perl). I have now found a way to
280        make it do so. The string so far is captured, making this feature
281        compatible with Perl.
283    23. The tests have been re-organized, adding tests 11 and 12, to make it
284        possible to check the Perl 5.10 features against Perl 5.10.
286    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
287        pattern matches a fixed length string. PCRE did not allow this; now it
288        does. Neither allows recursion.
290    25. I finally figured out how to implement a request to provide the minimum
291        length of subject string that was needed in order to match a given pattern.
292        (It was back references and recursion that I had previously got hung up
293        on.) This code has now been added to pcre_study(); it finds a lower bound
294        to the length of subject needed. It is not necessarily the greatest lower
295        bound, but using it to avoid searching strings that are too short does give
296        some useful speed-ups. The value is available to calling programs via
297        pcre_fullinfo().
299    26. While implementing 25, I discovered to my embarrassment that pcretest had
300        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
301        study optimizations had never been tested with that matching function.
302        Oops. What is worse, even when it was passed study data, there was a bug in
303        pcre_dfa_exec() that meant it never actually used it. Double oops. There
304        were also very few tests of studied patterns with pcre_dfa_exec().
306    27. If (?| is used to create subpatterns with duplicate numbers, they are now
307        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
308        on the other side of the coin, they are no longer allowed to have different
309        names, because these cannot be distinguished in PCRE, and this has caused
310        confusion. (This is a difference from Perl.)
312    28. When duplicate subpattern names are present (necessarily with different
313        numbers, as required by 27 above), and a test is made by name in a
314        conditional pattern, either for a subpattern having been matched, or for
315        recursion in such a pattern, all the associated numbered subpatterns are
316        tested, and the overall condition is true if the condition is true for any
317        one of them. This is the way Perl works, and is also more like the way
318        testing by number works.
321    Version 7.9 11-Apr-09
322    ---------------------
324    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
325        (pcretest), all targets were linked against these libraries. This included
326        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
327        libraries. This caused unwanted dependencies to be created. This problem
328        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
329        pcretest is linked with readline.
331    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
332        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
333        moved outside it again, because FALSE and TRUE are already defined in AIX,
334        but BOOL is not.
336    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
337        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
339    4.  The pcregrep documentation said spaces were inserted as well as colons (or
340        hyphens) following file names and line numbers when outputting matching
341        lines. This is not true; no spaces are inserted. I have also clarified the
342        wording for the --colour (or --color) option.
344    5.  In pcregrep, when --colour was used with -o, the list of matching strings
345        was not coloured; this is different to GNU grep, so I have changed it to be
346        the same.
348    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
349        each matching line was coloured. Now it goes on to look for further matches
350        of any of the test patterns, which is the same behaviour as GNU grep.
352    7.  A pattern that could match an empty string could cause pcregrep to loop; it
353        doesn't make sense to accept an empty string match in pcregrep, so I have
354        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
355        seems to be how GNU grep behaves.
357    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
358        start or after a newline", because the conditional assertion was not being
359        correctly handled. The rule now is that both the assertion and what follows
360        in the first alternative must satisfy the test.
362    9.  If auto-callout was enabled in a pattern with a conditional group whose
363        condition was an assertion, PCRE could crash during matching, both with
364        pcre_exec() and pcre_dfa_exec().
366    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
367        used for matching.
369    11. Unicode property support in character classes was not working for
370        characters (bytes) greater than 127 when not in UTF-8 mode.
372    12. Added the -M command line option to pcretest.
374    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
376    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
378    16. Added comments and documentation about mis-use of no_arg in the C++
379        wrapper.
381    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
382        from Martin Jerabek that uses macro names for all relevant character and
383        string constants.
385    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
386        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
387        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
388        these, but not everybody uses configure.
390    19. A conditional group that had only one branch was not being correctly
391        recognized as an item that could match an empty string. This meant that an
392        enclosing group might also not be so recognized, causing infinite looping
393        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
394        with the subject "ab", where knowledge that the repeated group can match
395        nothing is needed in order to break the loop.
397    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
398        exec(), but without supplying a callout function, matching went wrong.
400    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
401        leak if the size of the offset vector was greater than 30. When the vector
402        is smaller, the saved offsets during recursion go onto a local stack
403        vector, but for larger vectors malloc() is used. It was failing to free
404        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
405        error, in fact).
407    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
408        heapframe that is used only when UTF-8 support is enabled. This caused no
409        problem, but was untidy.
411    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
412        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
413        included within another project.
415    24. Steven Van Ingelgem's patches to add more options to the CMake support,
416        slightly modified by me:
418          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
419              not building pcregrep.
421          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
422              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
424    25. Forward references, both numeric and by name, in patterns that made use of
425        duplicate group numbers, could behave incorrectly or give incorrect errors,
426        because when scanning forward to find the reference group, PCRE was not
427        taking into account the duplicate group numbers. A pattern such as
428        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
430    26. Changed a few more instances of "const unsigned char *" to USPTR, making
431        the feature of a custom pointer more persuasive (as requested by a user).
433    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
434        pcretest.c, inside #ifndefs, because it seems they are sometimes already
435        pre-defined.
437    28. Added support for (*UTF8) at the start of a pattern.
439    29. Arrange for flags added by the "release type" setting in CMake to be shown
440        in the configuration summary.
443    Version 7.8 05-Sep-08
444    ---------------------
446    1.  Replaced UCP searching code with optimized version as implemented for Ad
447        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
448        stage table and inline lookup instead of a function, giving speed ups of 2
449        to 5 times on some simple patterns that I tested. Permission was given to
450        distribute the MultiStage2.py script that generates the tables (it's not in
451        the tarball, but is in the Subversion repository).
453    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
454        scripts.
456    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
457        a group with a zero qualifier. The result of the study could be incorrect,
458        or the function might crash, depending on the pattern.
460    4.  Caseless matching was not working for non-ASCII characters in back
461        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
462        It now works when Unicode Property Support is available.
464    5.  In pcretest, an escape such as \x{de} in the data was always generating
465        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
466        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
467        truncation.
469    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
471    7.  Added two (int) casts to pcregrep when printing the difference of two
472        pointers, in case they are 64-bit values.
474    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
475        test 2 if it fails.
477    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
478        and a #define of that name to empty if it is not externally set. This is to
479        allow users of MSVC to set it if necessary.
481    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
482        the convenience functions in the pcre_get.c source file.
484    11. An option change at the start of a pattern that had top-level alternatives
485        could cause overwriting and/or a crash. This command provoked a crash in
486        some environments:
488          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
490        This potential security problem was recorded as CVE-2008-2371.
492    12. For a pattern where the match had to start at the beginning or immediately
493        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
494        pcre_dfa_exec() could read past the end of the passed subject if there was
495        no match. To help with detecting such bugs (e.g. with valgrind), I modified
496        pcretest so that it places the subject at the end of its malloc-ed buffer.
498    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
499        exec() might read past the end of the data buffer in UTF-8 mode.
501    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
502        the data contained the byte 0x85 as part of a UTF-8 character within its
503        first line. This applied both to normal and DFA matching.
505    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
506        /^[^d]*?$/8 failed to match "abc".
508    16. Added a missing copyright notice to pcrecpp_internal.h.
510    17. Make it more clear in the documentation that values returned from
511        pcre_exec() in ovector are byte offsets, not character counts.
513    18. Tidied a few places to stop certain compilers from issuing warnings.
515    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
516        supplied by Stefan Weber. I made a further small update for 7.8 because
517        there is a change of source arrangements: the pcre_searchfuncs.c module is
518        replaced by pcre_ucd.c.
521    Version 7.7 07-May-08
522    ---------------------
524    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
525        a string to a long long, pretend we don't even have a long long." This is
526        done by checking for the strtoq, strtoll, and _strtoi64 functions.
528    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
529        pre-7.6 versions, which defined a global no_arg variable instead of putting
530        it in the RE class. (See also #8 below.)
532    3.  Remove a line of dead code, identified by coverity and reported by Nuno
533        Lopes.
535    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
537        (1) The include/exclude patterns were being applied to the whole pathnames
538            of files, instead of just to the final components.
540        (2) If there was more than one level of directory, the subdirectories were
541            skipped unless they satisfied the include/exclude conditions. This is
542            inconsistent with GNU grep (and could even be seen as contrary to the
543            pcregrep specification - which I improved to make it absolutely clear).
544            The action now is always to scan all levels of directory, and just
545            apply the include/exclude patterns to regular files.
547    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
548        --exclude_dir in the tests to avoid scanning .svn directories.
550    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
551        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
552        doesn't support NULs in patterns.
554    7.  Added some missing "const"s to declarations of static tables in
555        pcre_compile.c and pcre_dfa_exec.c.
557    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
558        caused by fix #2  above. (Subsequently also a second patch to fix the
559        first patch. And a third patch - this was a messy problem.)
561    9.  Applied Craig's patch to remove the use of push_back().
563    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
564        matching function regexec().
566    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
567        which, however, unlike Perl's \g{...}, are subroutine calls, not back
568        references. PCRE supports relative numbers with this syntax (I don't think
569        Oniguruma does).
571    12. Previously, a group with a zero repeat such as (...){0} was completely
572        omitted from the compiled regex. However, this means that if the group
573        was called as a subroutine from elsewhere in the pattern, things went wrong
574        (an internal error was given). Such groups are now left in the compiled
575        pattern, with a new opcode that causes them to be skipped at execution
576        time.
578    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
579        to the way PCRE behaves:
581        (a) A lone ] character is dis-allowed (Perl treats it as data).
583        (b) A back reference to an unmatched subpattern matches an empty string
584            (Perl fails the current match path).
586        (c) A data ] in a character class must be notated as \] because if the
587            first data character in a class is ], it defines an empty class. (In
588            Perl it is not possible to have an empty class.) The empty class []
589            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
590            The negative empty class [^] matches any one character, independently
591            of the DOTALL setting.
593    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
594        non-existent subpattern following a character class starting with ']' and
595        containing () gave an internal compiling error instead of "reference to
596        non-existent subpattern". Fortunately, when the pattern did exist, the
597        compiled code was correct. (When scanning forwards to check for the
598        existencd of the subpattern, it was treating the data ']' as terminating
599        the class, so got the count wrong. When actually compiling, the reference
600        was subsequently set up correctly.)
602    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
603        it was being rejected as not supported by pcre_dfa_exec(), even though
604        other assertions are supported. I have made pcre_dfa_exec() support
605        (*FAIL).
607    16. The implementation of 13c above involved the invention of a new opcode,
608        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
609        cannot be changed at match time, I realized I could make a small
610        improvement to matching performance by compiling OP_ALLANY instead of
611        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
612        on the OP_ANY path.
614    17. Compiling pcretest on Windows with readline support failed without the
615        following two fixes: (1) Make the unistd.h include conditional on
616        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
618    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
619        ncurses library to be included for pcretest when ReadLine support is
620        requested, but also to allow for it to be overridden. This patch came from
621        Daniel Bergström.
623    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
624        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
625        any errors with the current Unicode tables. Thanks to Peter Kankowski for
626        spotting this.
629    Version 7.6 28-Jan-08
630    ---------------------
632    1.  A character class containing a very large number of characters with
633        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
634        overflow.
636    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
637        HAVE_LONG_LONG is not defined.
639    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
640        bring them up to date and include new features. This patch includes:
642        - Fixed PH's badly added libz and libbz2 support.
643        - Fixed a problem with static linking.
644        - Added pcredemo. [But later removed - see 7 below.]
645        - Fixed dftables problem and added an option.
646        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
647            HAVE_LONG_LONG.
648        - Added readline support for pcretest.
649        - Added an listing of the option settings after cmake has run.
651    4.  A user submitted a patch to Makefile that makes it easy to create
652        "pcre.dll" under mingw when using Configure/Make. I added stuff to
653        Makefile.am that cause it to include this special target, without
654        affecting anything else. Note that the same mingw target plus all
655        the other distribution libraries and programs are now supported
656        when configuring with CMake (see 6 below) instead of with
657        Configure/Make.
659    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
660        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
661        exported in the Windows port". It has not yet been confirmed that the patch
662        solves the problem, but it does no harm.
664    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
665        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
666        with CMake, and also correct the comment about stack recursion.
668    7.  Remove the automatic building of pcredemo from the ./configure system and
669        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
670        of a program that users should build themselves after PCRE is installed, so
671        building it automatically is not really right. What is more, it gave
672        trouble in some build environments.
674    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
677    Version 7.5 10-Jan-08
678    ---------------------
680    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
681        values in parens when parsing an RE using the C++ wrapper."
683    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
684        Characters greater than 255 were excluded from the class instead of being
685        included.
687    3.  The same bug as (2) above applied to negated POSIX classes such as
688        [:^space:].
690    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
691        defined or documented. It seems to have been a typo for PCRE_STATIC, so
692        I have changed it.
694    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
695        first named subpattern) and a construct such as (?&a) would reference the
696        first named subpattern whose name started with "a" (in other words, the
697        length check was missing). Both these problems are fixed. "Subpattern name
698        expected" is now given for (?&) (a zero-length name), and this patch also
699        makes it give the same error for \k'' (previously it complained that that
700        was a reference to a non-existent subpattern).
702    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
703        this is right because (?- can be followed by option settings as well as by
704        digits. I have, however, made the messages clearer.
706    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
707        than the number used in the conditional) now cause a compile-time error.
708        This is actually not compatible with Perl, which accepts such patterns, but
709        treats the conditional as always being FALSE (as PCRE used to), but it
710        seems to me that giving a diagnostic is better.
712    8.  Change "alphameric" to the more common word "alphanumeric" in comments
713        and messages.
715    9.  Fix two occurrences of "backslash" in comments that should have been
716        "backspace".
718    10. Remove two redundant lines of code that can never be obeyed (their function
719        was moved elsewhere).
721    11. The program that makes PCRE's Unicode character property table had a bug
722        which caused it to generate incorrect table entries for sequences of
723        characters that have the same character type, but are in different scripts.
724        It amalgamated them into a single range, with the script of the first of
725        them. In other words, some characters were in the wrong script. There were
726        thirteen such cases, affecting characters in the following ranges:
728          U+002b0 - U+002c1
729          U+0060c - U+0060d
730          U+0061e - U+00612
731          U+0064b - U+0065e
732          U+0074d - U+0076d
733          U+01800 - U+01805
734          U+01d00 - U+01d77
735          U+01d9b - U+01dbf
736          U+0200b - U+0200f
737          U+030fc - U+030fe
738          U+03260 - U+0327f
739          U+0fb46 - U+0fbb1
740          U+10450 - U+1049d
742    12. The -o option (show only the matching part of a line) for pcregrep was not
743        compatible with GNU grep in that, if there was more than one match in a
744        line, it showed only the first of them. It now behaves in the same way as
745        GNU grep.
747    13. If the -o and -v options were combined for pcregrep, it printed a blank
748        line for every non-matching line. GNU grep prints nothing, and pcregrep now
749        does the same. The return code can be used to tell if there were any
750        non-matching lines.
752    14. Added --file-offsets and --line-offsets to pcregrep.
754    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
755        infinitely looping recursion. The bug was that positive lookaheads were not
756        being skipped when checking for a possible empty match (negative lookaheads
757        and both kinds of lookbehind were skipped).
759    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
760        inclusion of <windows.h> to before rather than after the definition of
761        INVALID_FILE_ATTRIBUTES (patch from David Byron).
763    17. Specifying a possessive quantifier with a specific limit for a Unicode
764        character property caused pcre_compile() to compile bad code, which led at
765        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
766        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
767        caused the error; without that there was no problem.
769    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
771    19. Added --enable-pcretest-libreadline.
773    20. In pcrecpp.cc, the variable 'count' was incremented twice in
774        RE::GlobalReplace(). As a result, the number of replacements returned was
775        double what it should be. I removed one of the increments, but Craig sent a
776        later patch that removed the other one (the right fix) and added unit tests
777        that check the return values (which was not done before).
779    21. Several CMake things:
781        (1) Arranged that, when cmake is used on Unix, the libraries end up with
782            the names libpcre and libpcreposix, not just pcre and pcreposix.
784        (2) The above change means that pcretest and pcregrep are now correctly
785            linked with the newly-built libraries, not previously installed ones.
789    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
790        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
791        UTF-8 newline character). The key issue is that the pattern starts .*;
792        this means that the match must be either at the beginning, or after a
793        newline. The bug was in the code for advancing after a failed match and
794        checking that the new position followed a newline. It was not taking
795        account of UTF-8 characters correctly.
797    23. PCRE was behaving differently from Perl in the way it recognized POSIX
798        character classes. PCRE was not treating the sequence [:...:] as a
799        character class unless the ... were all letters. Perl, however, seems to
800        allow any characters between [: and :], though of course it rejects as
801        unknown any "names" that contain non-letters, because all the known class
802        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
803        for example, whereas PCRE did not - it did not recognize a POSIX character
804        class. This seemed a bit dangerous, so the code has been changed to be
805        closer to Perl. The behaviour is not identical to Perl, because PCRE will
806        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
807        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
808        Perl does, and where it didn't before.
810    24. Rewrite so as to remove the single use of %n from pcregrep because in some
811        Windows environments %n is disabled by default.
814    Version 7.4 21-Sep-07
815    ---------------------
817    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
818        means that a class such as [\s] counted as "explicit reference to CR or
819        LF". That isn't really right - the whole point of the change was to try to
820        help when there was an actual mention of one of the two characters. So now
821        the change happens only if \r or \n (or a literal CR or LF) character is
822        encountered.
824    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
825        of both had grown to the point where there were only 3 bits left.
826        Fortunately, there was spare space in the data structure, and so I have
827        moved the internal flags into a new 16-bit field to free up more option
828        bits.
830    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
831        but did not set the internal JCHANGED flag - either of these is enough to
832        control the way the "get" function works - but the PCRE_INFO_JCHANGED
833        facility is supposed to tell if (?J) was ever used, so now (?J) at the
834        start sets both bits.
836    4.  Added options (at build time, compile time, exec time) to change \R from
837        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
839    5.  doc/pcresyntax.html was missing from the distribution.
841    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
842        compatibility, even though it is no longer used.
844    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
845        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
846        windows.h file is present (where different names are used). [This was
847        reversed later after testing - see 16 below.]
849    8.  Changed all #include <config.h> to #include "config.h". There were also
850        some further <pcre.h> cases that I changed to "pcre.h".
852    9.  When pcregrep was used with the --colour option, it missed the line ending
853        sequence off the lines that it output.
855    10. It was pointed out to me that arrays of string pointers cause lots of
856        relocations when a shared library is dynamically loaded. A technique of
857        using a single long string with a table of offsets can drastically reduce
858        these. I have refactored PCRE in four places to do this. The result is
859        dramatic:
861          Originally:                          290
862          After changing UCP table:            187
863          After changing error message table:   43
864          After changing table of "verbs"       36
865          After changing table of Posix names   22
867        Thanks to the folks working on Gregex for glib for this insight.
869    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
870        unicode-properties was also set.
872    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
874    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
875        checked only for CRLF.
877    14. Added casts to pcretest.c to avoid compiler warnings.
879    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
881    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
882        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
883        entirely. This removes changes made in 7 above.
885    17. The CMake files have been updated, and there is now more information about
886        building with CMake in the NON-UNIX-USE document.
889    Version 7.3 28-Aug-07
890    ---------------------
892     1. In the rejigging of the build system that eventually resulted in 7.1, the
893        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
894        brackets there is not right, since it causes compilers to look for an
895        installed pcre.h, not the version that is in the source that is being
896        compiled (which of course may be different). I have changed it back to:
898          #include "pcre.h"
900        I have a vague recollection that the change was concerned with compiling in
901        different directories, but in the new build system, that is taken care of
902        by the VPATH setting the Makefile.
904     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
905        when the subject happened to end in the byte 0x85 (e.g. if the last
906        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
907        characters but of course it shouldn't be taken as a newline when it is part
908        of another character. The bug was that, for an unlimited repeat of . in
909        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
910        characters when looking for a newline.
912     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
914     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
915        in debug output.
917     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
918        long printing in the pcrecpp unittest when running under MinGW.
920     6. ESC_K was left out of the EBCDIC table.
922     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
923        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
924        limit also applies to "virtual nesting" when a pattern is recursive, and in
925        this case 1000 isn't so big. I have been able to remove this limit at the
926        expense of backing off one optimization in certain circumstances. Normally,
927        when pcre_exec() would call its internal match() function recursively and
928        immediately return the result unconditionally, it uses a "tail recursion"
929        feature to save stack. However, when a subpattern that can match an empty
930        string has an unlimited repetition quantifier, it no longer makes this
931        optimization. That gives it a stack frame in which to save the data for
932        checking that an empty string has been matched. Previously this was taken
933        from the 1000-entry workspace that had been reserved. So now there is no
934        explicit limit, but more stack is used.
936     8. Applied Daniel's patches to solve problems with the import/export magic
937        syntax that is required for Windows, and which was going wrong for the
938        pcreposix and pcrecpp parts of the library. These were overlooked when this
939        problem was solved for the main library.
941     9. There were some crude static tests to avoid integer overflow when computing
942        the size of patterns that contain repeated groups with explicit upper
943        limits. As the maximum quantifier is 65535, the maximum group length was
944        set at 30,000 so that the product of these two numbers did not overflow a
945        32-bit integer. However, it turns out that people want to use groups that
946        are longer than 30,000 bytes (though not repeat them that many times).
947        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
948        made it possible to implement the integer overflow checks in a much more
949        dynamic way, which I have now done. The artificial limitation on group
950        length has been removed - we now have only the limit on the total length of
951        the compiled pattern, which depends on the LINK_SIZE setting.
953    10. Fixed a bug in the documentation for get/copy named substring when
954        duplicate names are permitted. If none of the named substrings are set, the
955        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
956        empty string.
958    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
959        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
960        because the ] is interpreted as the first data character and the
961        terminating ] is not found. PCRE has been made compatible with Perl in this
962        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
963        cause memory overwriting.
965    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
966        string has been matched (to stop an infinite loop). It was not recognizing
967        a conditional subpattern that could match an empty string if that
968        subpattern was within another subpattern. For example, it looped when
969        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
970        condition was not nested. This bug has been fixed.
972    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
973        past the start of the subject in the presence of bytes with the top bit
974        set, for example "\x8aBCD".
976    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
977        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
979    14. Optimized (?!) to (*FAIL).
981    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
982        This restricts code points to be within the range 0 to 0x10FFFF, excluding
983        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
984        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
985        does: it's just the validity check that is more restrictive.
987    16. Inserted checks for integer overflows during escape sequence (backslash)
988        processing, and also fixed erroneous offset values for syntax errors during
989        backslash processing.
991    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
992        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
994    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
995        caused an overrun.
997    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
998        something other than just ASCII characters) inside a group that had an
999        unlimited repeat caused a loop at compile time (while checking to see
1000        whether the group could match an empty string).
1002    20. Debugging a pattern containing \p or \P could cause a crash. For example,
1003        [\P{Any}] did so. (Error in the code for printing property names.)
1005    21. An orphan \E inside a character class could cause a crash.
1007    22. A repeated capturing bracket such as (A)? could cause a wild memory
1008        reference during compilation.
1010    23. There are several functions in pcre_compile() that scan along a compiled
1011        expression for various reasons (e.g. to see if it's fixed length for look
1012        behind). There were bugs in these functions when a repeated \p or \P was
1013        present in the pattern. These operators have additional parameters compared
1014        with \d, etc, and these were not being taken into account when moving along
1015        the compiled data. Specifically:
1017        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1018            length.
1020        (b) An item such as \pL+ within a repeated group could cause crashes or
1021            loops.
1023        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1024            "reference to non-existent subpattern" error.
1026        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1028    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1029        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1031    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1033    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1034        character were causing crashes (broken optimization).
1036    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1037        \p or \P) caused a compile-time loop.
1039    28. More problems have arisen in unanchored patterns when CRLF is a valid line
1040        break. For example, the unstudied pattern [\r\n]A does not match the string
1041        "\r\nA" because change 7.0/46 below moves the current point on by two
1042        characters after failing to match at the start. However, the pattern \nA
1043        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
1044        the same is true. There doesn't seem any very clean way out of this, but
1045        what I have chosen to do makes the common cases work: PCRE now takes note
1046        of whether there can be an explicit match for \r or \n anywhere in the
1047        pattern, and if so, 7.0/46 no longer applies. As part of this change,
1048        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1049        pattern has explicit CR or LF references.
1051    29. Added (*CR) etc for changing newline setting at start of pattern.
1054    Version 7.2 19-Jun-07
1055    ---------------------
1057     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
1058        which is apparently normally available under Windows.
1060     2. Re-jig the pcregrep tests with different newline settings in an attempt
1061        to make them independent of the local environment's newline setting.
1063     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
1065     4. Some of the "internals" tests were previously cut out when the link size
1066        was not 2, because the output contained actual offsets. The recent new
1067        "Z" feature of pcretest means that these can be cut out, making the tests
1068        usable with all link sizes.
1070     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
1071        stack recursion. This gives a massive performance boost under BSD, but just
1072        a small improvement under Linux. However, it saves one field in the frame
1073        in all cases.
1075     6. Added more features from the forthcoming Perl 5.10:
1077        (a) (?-n) (where n is a string of digits) is a relative subroutine or
1078            recursion call. It refers to the nth most recently opened parentheses.
1080        (b) (?+n) is also a relative subroutine call; it refers to the nth next
1081            to be opened parentheses.
1083        (c) Conditions that refer to capturing parentheses can be specified
1084            relatively, for example, (?(-2)... or (?(+3)...
1086        (d) \K resets the start of the current match so that everything before
1087            is not part of it.
1089        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1091        (f) \g{name} is another synonym - part of Perl 5.10's unification of
1092            reference syntax.
1094        (g) (?| introduces a group in which the numbering of parentheses in each
1095            alternative starts with the same number.
1097        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
1099     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
1102     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
1103        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
1104        for detecting groups that can match an empty string.
1106     9. A pattern with a very large number of alternatives (more than several
1107        hundred) was running out of internal workspace during the pre-compile
1108        phase, where pcre_compile() figures out how much memory will be needed. A
1109        bit of new cunning has reduced the workspace needed for groups with
1110        alternatives. The 1000-alternative test pattern now uses 12 bytes of
1111        workspace instead of running out of the 4096 that are available.
1113    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
1115    11. Applied patch from Google to remove an optimization that didn't quite work.
1116        The report of the bug said:
1118          pcrecpp::RE("a*").FullMatch("aaa") matches, while
1119          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
1120          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
1122    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
1123        it matched the wrong number of bytes.
1126    Version 7.1 24-Apr-07
1127    ---------------------
1129     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
1130        that is more "standard", making use of automake and other Autotools. There
1131        is some re-arrangement of the files and adjustment of comments consequent
1132        on this.
1134     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
1135        for recursive directory scanning broke on some systems because the files
1136        are not scanned in any specific order and on different systems the order
1137        was different. A call to "sort" has been inserted into RunGrepTest for the
1138        approprate test as a short-term fix. In the longer term there may be an
1139        alternative.
1141     3. I had an email from Eric Raymond about problems translating some of PCRE's
1142        man pages to HTML (despite the fact that I distribute HTML pages, some
1143        people do their own conversions for various reasons). The problems
1144        concerned the use of low-level troff macros .br and .in. I have therefore
1145        removed all such uses from the man pages (some were redundant, some could
1146        be replaced by .nf/.fi pairs). The 132html script that I use to generate
1147        HTML has been updated to handle .nf/.fi and to complain if it encounters
1148        .br or .in.
1150     4. Updated comments in configure.ac that get placed in config.h.in and also
1151        arranged for config.h to be included in the distribution, with the name
1152        config.h.generic, for the benefit of those who have to compile without
1153        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
1155     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
1156        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
1157        makevp.bat for the current PCRE, using the additional files
1158        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
1160     6. A Windows user reported a minor discrepancy with test 2, which turned out
1161        to be caused by a trailing space on an input line that had got lost in his
1162        copy. The trailing space was an accident, so I've just removed it.
1164     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
1165        that is needed.
1167     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
1168        as "const" (a) because they are and (b) because it helps the PHP
1169        maintainers who have recently made a script to detect big data structures
1170        in the php code that should be moved to the .rodata section. I remembered
1171        to update Builducptable as well, so it won't revert if ucptable.h is ever
1172        re-created.
1174     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
1175        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
1176        order to be able to cut out the UTF-8 tables in the latter when UTF-8
1177        support is not required. This saves 1.5-2K of code, which is important in
1178        some applications.
1180        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
1181        so as not to refer to the tables, even though these functions will never be
1182        called when UTF-8 support is disabled. Otherwise there are problems with a
1183        shared library.
1185    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
1187        (a) It was defining its arguments as char * instead of void *.
1189        (b) It was assuming that all moves were upwards in memory; this was true
1190            a long time ago when I wrote it, but is no longer the case.
1192        The emulated memove() is provided for those environments that have neither
1193        memmove() nor bcopy(). I didn't think anyone used it these days, but that
1194        is clearly not the case, as these two bugs were recently reported.
1196    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
1197        and Detrail to create the HTML documentation, the .txt form of the man
1198        pages, and it removes trailing spaces from listed files. It also creates
1199        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
1200        case, it wraps all the #defines with #ifndefs. This script should be run
1201        before "make dist".
1203    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1204        with Unicode property support.
1206        (a) For a maximizing quantifier, if the two different cases of the
1207            character were of different lengths in their UTF-8 codings (there are
1208            some cases like this - I found 11), and the matching function had to
1209            back up over a mixture of the two cases, it incorrectly assumed they
1210            were both the same length.
1212        (b) When PCRE was configured to use the heap rather than the stack for
1213            recursion during matching, it was not correctly preserving the data for
1214            the other case of a UTF-8 character when checking ahead for a match
1215            while processing a minimizing repeat. If the check also involved
1216            matching a wide character, but failed, corruption could cause an
1217            erroneous result when trying to check for a repeat of the original
1218            character.
1220    13. Some tidying changes to the testing mechanism:
1222        (a) The RunTest script now detects the internal link size and whether there
1223            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1224            values substituted by "configure". (The RunGrepTest script already did
1225            this for UTF-8.) The configure.ac script no longer substitutes the
1226            relevant variables.
1228        (b) The debugging options /B and /D in pcretest show the compiled bytecode
1229            with length and offset values. This means that the output is different
1230            for different internal link sizes. Test 2 is skipped for link sizes
1231            other than 2 because of this, bypassing the problem. Unfortunately,
1232            there was also a test in test 3 (the locale tests) that used /B and
1233            failed for link sizes other than 2. Rather than cut the whole test out,
1234            I have added a new /Z option to pcretest that replaces the length and
1235            offset values with spaces. This is now used to make test 3 independent
1236            of link size. (Test 2 will be tidied up later.)
1238    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1239        segmentation fault instead of returning the appropriate error message.
1241    15. In multiline mode when the newline sequence was set to "any", the pattern
1242        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1243        This doesn't seem right; it now treats the CRLF combination as the line
1244        ending, and so does not match in that case. It's only a pattern such as ^$
1245        that would hit this one: something like ^ABC$ would have failed after \r
1246        and then tried again after \r\n.
1248    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1249        in an attempt to make files that differ only in their line terminators
1250        compare equal. This works on Linux.
1252    17. Under certain error circumstances pcregrep might try to free random memory
1253        as it exited. This is now fixed, thanks to valgrind.
1255    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1256        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1257        was because its rules for how to advance for /g after matching an empty
1258        string at the end of a line did not allow for this case. They now check for
1259        it specially.
1261    20. pcretest is supposed to handle patterns and data of any length, by
1262        extending its buffers when necessary. It was getting this wrong when the
1263        buffer for a data line had to be extended.
1265    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1266        CRLF as a newline sequence.
1268    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1269        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1270        I have nevertheless tidied it up.
1272    23. Added some casts to kill warnings from HP-UX ia64 compiler.
1274    24. Added a man page for pcre-config.
1277    Version 7.0 19-Dec-06
1278    ---------------------
1280     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1281        moving to gcc 4.1.1.
1283     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1284        sys/time.h, which is documented as needed for this function. It doesn't
1285        seem to matter on Linux, but it showed up on some releases of OS X.
1287     3. It seems that there are systems where bytes whose values are greater than
1288        127 match isprint() in the "C" locale. The "C" locale should be the
1289        default when a C program starts up. In most systems, only ASCII printing
1290        characters match isprint(). This difference caused the output from pcretest
1291        to vary, making some of the tests fail. I have changed pcretest so that:
1293        (a) When it is outputting text in the compiled version of a pattern, bytes
1294            other than 32-126 are always shown as hex escapes.
1296        (b) When it is outputting text that is a matched part of a subject string,
1297            it does the same, unless a different locale has been set for the match
1298            (using the /L modifier). In this case, it uses isprint() to decide.
1300     4. Fixed a major bug that caused incorrect computation of the amount of memory
1301        required for a compiled pattern when options that changed within the
1302        pattern affected the logic of the preliminary scan that determines the
1303        length. The relevant options are -x, and -i in UTF-8 mode. The result was
1304        that the computed length was too small. The symptoms of this bug were
1305        either the PCRE error "internal error: code overflow" from pcre_compile(),
1306        or a glibc crash with a message such as "pcretest: free(): invalid next
1307        size (fast)". Examples of patterns that provoked this bug (shown in
1308        pcretest format) are:
1310          /(?-x: )/x
1311          /(?x)(?-x: \s*#\s*)/
1312          /((?i)[\x{c0}])/8
1313          /(?i:[\x{c0}])/8
1315        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1316        is now done differently.
1318     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1319        wrapper classes; (b) implement a new function in the C++ scanner that is
1320        more efficient than the old way of doing things because it avoids levels of
1321        recursion in the regex matching; (c) add a paragraph to the documentation
1322        for the FullMatch() function.
1324     6. The escape sequence \n was being treated as whatever was defined as
1325        "newline". Not only was this contrary to the documentation, which states
1326        that \n is character 10 (hex 0A), but it also went horribly wrong when
1327        "newline" was defined as CRLF. This has been fixed.
1329     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1330        was being set to -1 for the "end of line" case (supposedly a value that no
1331        character can have). Though this value is never used (the check for end of
1332        line is "zero bytes in current character"), it caused compiler complaints.
1333        I've changed it to 0xffffffff.
1335     8. In pcre_version.c, the version string was being built by a sequence of
1336        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1337        string (as it is for production releases) called a macro with an empty
1338        argument. The C standard says the result of this is undefined. The gcc
1339        compiler treats it as an empty string (which was what was wanted) but it is
1340        reported that Visual C gives an error. The source has been hacked around to
1341        avoid this problem.
1343     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1344        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1345        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1346        of them did).
1348    10. Originally, pcretest opened its input and output without "b"; then I was
1349        told that "b" was needed in some environments, so it was added for release
1350        5.0 to both the input and output. (It makes no difference on Unix-like
1351        systems.) Later I was told that it is wrong for the input on Windows. I've
1352        now abstracted the modes into two macros, to make it easier to fiddle with
1353        them, and removed "b" from the input mode under Windows.
1355    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1357    12. Added -help and --help to pcretest as an official way of being reminded
1358        of the options.
1360    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1361        and pcrecpp.cc because they annoy compilers at high warning levels.
1363    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1365    15. Fixed an occurrence of == in configure.ac that should have been = (shell
1366        scripts are not C programs :-) and which was not noticed because it works
1367        on Linux.
1369    16. pcretest is supposed to handle any length of pattern and data line (as one
1370        line or as a continued sequence of lines) by extending its input buffer if
1371        necessary. This feature was broken for very long pattern lines, leading to
1372        a string of junk being passed to pcre_compile() if the pattern was longer
1373        than about 50K.
1375    17. I have done a major re-factoring of the way pcre_compile() computes the
1376        amount of memory needed for a compiled pattern. Previously, there was code
1377        that made a preliminary scan of the pattern in order to do this. That was
1378        OK when PCRE was new, but as the facilities have expanded, it has become
1379        harder and harder to keep it in step with the real compile phase, and there
1380        have been a number of bugs (see for example, 4 above). I have now found a
1381        cunning way of running the real compile function in a "fake" mode that
1382        enables it to compute how much memory it would need, while actually only
1383        ever using a few hundred bytes of working memory and without too many
1384        tests of the mode. This should make future maintenance and development
1385        easier. A side effect of this work is that the limit of 200 on the nesting
1386        depth of parentheses has been removed (though this was never a serious
1387        limitation, I suspect). However, there is a downside: pcre_compile() now
1388        runs more slowly than before (30% or more, depending on the pattern). I
1389        hope this isn't a big issue. There is no effect on runtime performance.
1391    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1392        newline (only possible for the last line of a file) and it was a
1393        pattern that set a locale (followed by /Lsomething), pcretest crashed.
1395    19. Added additional timing features to pcretest. (1) The -tm option now times
1396        matching only, not compiling. (2) Both -t and -tm can be followed, as a
1397        separate command line item, by a number that specifies the number of
1398        repeats to use when timing. The default is 50000; this gives better
1399        precision, but takes uncomfortably long for very large patterns.
1401    20. Extended pcre_study() to be more clever in cases where a branch of a
1402        subpattern has no definite first character. For example, (a*|b*)[cd] would
1403        previously give no result from pcre_study(). Now it recognizes that the
1404        first character must be a, b, c, or d.
1406    21. There was an incorrect error "recursive call could loop indefinitely" if
1407        a subpattern (or the entire pattern) that was being tested for matching an
1408        empty string contained only one non-empty item after a nested subpattern.
1409        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1410        incorrectly, because the \d was being skipped in the check.
1412    22. The pcretest program now has a new pattern option /B and a command line
1413        option -b, which is equivalent to adding /B to every pattern. This causes
1414        it to show the compiled bytecode, without the additional information that
1415        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1416        is the same as /B/I).
1418    23. A new optimization is now able automatically to treat some sequences such
1419        as a*b as a*+b. More specifically, if something simple (such as a character
1420        or a simple class like \d) has an unlimited quantifier, and is followed by
1421        something that cannot possibly match the quantified thing, the quantifier
1422        is automatically "possessified".
1424    24. A recursive reference to a subpattern whose number was greater than 39
1425        went wrong under certain circumstances in UTF-8 mode. This bug could also
1426        have affected the operation of pcre_study().
1428    25. Realized that a little bit of performance could be had by replacing
1429        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1431    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1433    27. Possessive quantifiers such as a++ were previously implemented by turning
1434        them into atomic groups such as ($>a+). Now they have their own opcodes,
1435        which improves performance. This includes the automatically created ones
1436        from 23 above.
1438    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1439        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1440        the first matched character to be a colon. This applied both to named and
1441        numbered groups.
1443    29. The ucpinternal.h header file was missing its idempotency #ifdef.
1445    30. I was sent a "project" file called libpcre.a.dev which I understand makes
1446        building PCRE on Windows easier, so I have included it in the distribution.
1448    31. There is now a check in pcretest against a ridiculously large number being
1449        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1450        loop, the loop is abandoned.
1452    32. Forward references to subpatterns in conditions such as (?(2)...) where
1453        subpattern 2 is defined later cause pcre_compile() to search forwards in
1454        the pattern for the relevant set of parentheses. This search went wrong
1455        when there were unescaped parentheses in a character class, parentheses
1456        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1458    33. "Subroutine" calls and backreferences were previously restricted to
1459        referencing subpatterns earlier in the regex. This restriction has now
1460        been removed.
1462    34. Added a number of extra features that are going to be in Perl 5.10. On the
1463        whole, these are just syntactic alternatives for features that PCRE had
1464        previously implemented using the Python syntax or my own invention. The
1465        other formats are all retained for compatibility.
1467        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1468            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1469            also .NET compatible.
1471        (b) A recursion or subroutine call to a named group can now be defined as
1472            (?&name) as well as (?P>name).
1474        (c) A backreference to a named group can now be defined as \k<name> or
1475            \k'name' as well as (?P=name). The new forms, as well as being in Perl
1476            5.10, are also .NET compatible.
1478        (d) A conditional reference to a named group can now use the syntax
1479            (?(<name>) or (?('name') as well as (?(name).
1481        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1482            groups (named and numbered) that are never evaluated inline, but can be
1483            called as "subroutines" from elsewhere. In effect, the DEFINE condition
1484            is always false. There may be only one alternative in such a group.
1486        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1487            as the simple (?(R). The condition is true only if the most recent
1488            recursion is that of the given number or name. It does not search out
1489            through the entire recursion stack.
1491        (g) The escape \gN or \g{N} has been added, where N is a positive or
1492            negative number, specifying an absolute or relative reference.
1494    35. Tidied to get rid of some further signed/unsigned compiler warnings and
1495        some "unreachable code" warnings.
1497    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1498        things, this adds five new scripts.
1500    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1501        There were also incompatibilities regarding the handling of \Q..\E inside
1502        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1503        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1505    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1506        matches an empty string, and forcibly breaks the loop. There were bugs in
1507        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
1508        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
1509        separate and independent bugs (that affected different cases) have been
1510        fixed.
1512    39. Refactored the code to abolish the use of different opcodes for small
1513        capturing bracket numbers. This is a tidy that I avoided doing when I
1514        removed the limit on the number of capturing brackets for 3.5 back in 2001.
1515        The new approach is not only tidier, it makes it possible to reduce the
1516        memory needed to fix the previous bug (38).
1518    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1519        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1520        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1521        mode.
1523    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1524        report.
1526    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1527        copy construction and assignment in the C++ wrapper.
1529    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1530        couple of bugs that could have given wrong results in the "--newline=crlf"
1531        case.
1533    44. Added a number of casts and did some reorganization of signed/unsigned int
1534        variables following suggestions from Dair Grant. Also renamed the variable
1535        "this" as "item" because it is a C++ keyword.
1537    45. Arranged for dftables to add
1539          #include "pcre_internal.h"
1541        to pcre_chartables.c because without it, gcc 4.x may remove the array
1542        definition from the final binary if PCRE is built into a static library and
1543        dead code stripping is activated.
1545    46. For an unanchored pattern, if a match attempt fails at the start of a
1546        newline sequence, and the newline setting is CRLF or ANY, and the next two
1547        characters are CRLF, advance by two characters instead of one.
1550    Version 6.7 04-Jul-06
1551    ---------------------
1553     1. In order to handle tests when input lines are enormously long, pcretest has
1554        been re-factored so that it automatically extends its buffers when
1555        necessary. The code is crude, but this _is_ just a test program. The
1556        default size has been increased from 32K to 50K.
1558     2. The code in pcre_study() was using the value of the re argument before
1559        testing it for NULL. (Of course, in any sensible call of the function, it
1560        won't be NULL.)
1562     3. The memmove() emulation function in pcre_internal.h, which is used on
1563        systems that lack both memmove() and bcopy() - that is, hardly ever -
1564        was missing a "static" storage class specifier.
1566     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1567        containing an extended class (one that cannot be represented by a bitmap
1568        because it contains high-valued characters or Unicode property items, e.g.
1569        [\pZ]). Almost always one would set UTF-8 mode when processing such a
1570        pattern, but PCRE should not loop if you do not (it no longer does).
1571        [Detail: two cases were found: (a) a repeated subpattern containing an
1572        extended class; (b) a recursive reference to a subpattern that followed a
1573        previous extended class. It wasn't skipping over the extended class
1574        correctly when UTF-8 mode was not set.]
1576     5. A negated single-character class was not being recognized as fixed-length
1577        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1578        compile error "lookbehind assertion is not fixed length".
1580     6. The RunPerlTest auxiliary script was showing an unexpected difference
1581        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1582        write a Perl script that can interpret lines of an input file either as
1583        byte characters or as UTF-8, which is what "perltest" was being required to
1584        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1585        can't do is switch easily at run time between having the "use utf8;" pragma
1586        or not. In the end, I fudged it by using the RunPerlTest script to insert
1587        "use utf8;" explicitly for the UTF-8 tests.
1589     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1590        the end of the subject string, contrary to the documentation and to what
1591        Perl does. This was true of both matching functions. Now it matches only at
1592        the start of the subject and immediately after *internal* newlines.
1594     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1595        a pointer to an int instead of a pointer to an unsigned long int. This
1596        caused problems on 64-bit systems.
1598     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1599        instance of the 'standard' template library not being so standard".
1601    10. There was no check on the number of named subpatterns nor the maximum
1602        length of a subpattern name. The product of these values is used to compute
1603        the size of the memory block for a compiled pattern. By supplying a very
1604        long subpattern name and a large number of named subpatterns, the size
1605        computation could be caused to overflow. This is now prevented by limiting
1606        the length of names to 32 characters, and the number of named subpatterns
1607        to 10,000.
1609    11. Subpatterns that are repeated with specific counts have to be replicated in
1610        the compiled pattern. The size of memory for this was computed from the
1611        length of the subpattern and the repeat count. The latter is limited to
1612        65535, but there was no limit on the former, meaning that integer overflow
1613        could in principle occur. The compiled length of a repeated subpattern is
1614        now limited to 30,000 bytes in order to prevent this.
1616    12. Added the optional facility to have named substrings with the same name.
1618    13. Added the ability to use a named substring as a condition, using the
1619        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1620        are numbers (not recommended). Forward references are permitted.
1622    14. Added forward references in named backreferences (if you see what I mean).
1624    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1625        pattern could run off the end of the subject. For example, the pattern
1626        "(?s)(.{1,5})"8 did this with the subject "ab".
1628    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1629        PCRE_CASELESS was set when matching characters that were quantified with ?
1630        or *.
1632    17. A character class other than a single negated character that had a minimum
1633        but no maximum quantifier - for example [ab]{6,} - was not handled
1634        correctly by pce_dfa_exec(). It would match only one character.
1636    18. A valid (though odd) pattern that looked like a POSIX character
1637        class but used an invalid character after [ (for example [[,abc,]]) caused
1638        pcre_compile() to give the error "Failed: internal error: code overflow" or
1639        in some cases to crash with a glibc free() error. This could even happen if
1640        the pattern terminated after [[ but there just happened to be a sequence of
1641        letters, a binary zero, and a closing ] in the memory that followed.
1643    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1644        over the years. Originally (before any Unicode support), just the bottom 8
1645        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1646        output from "man perlunicode" includes this:
1648          The regular expression compiler produces polymorphic opcodes.  That
1649          is, the pattern adapts to the data and automatically switches to
1650          the Unicode character scheme when presented with Unicode data--or
1651          instead uses a traditional byte scheme when presented with byte
1652          data.
1654        Sadly, a wide octal escape does not cause a switch, and in a string with
1655        no other multibyte characters, these octal escapes are treated as before.
1656        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
1657        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1658        Unicode string.
1660        I have not perpetrated such confusion in PCRE. Up till now, it took just
1661        the bottom 8 bits, as in old Perl. I have now made octal escapes with
1662        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1663        translate to the appropriate multibyte character.
1665    29. Applied some refactoring to reduce the number of warnings from Microsoft
1666        and Borland compilers. This has included removing the fudge introduced
1667        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1668        a warning about an unused variable.
1670    21. PCRE has not included VT (character 0x0b) in the set of whitespace
1671        characters since release 4.0, because Perl (from release 5.004) does not.
1672        [Or at least, is documented not to: some releases seem to be in conflict
1673        with the documentation.] However, when a pattern was studied with
1674        pcre_study() and all its branches started with \s, PCRE still included VT
1675        as a possible starting character. Of course, this did no harm; it just
1676        caused an unnecessary match attempt.
1678    22. Removed a now-redundant internal flag bit that recorded the fact that case
1679        dependency changed within the pattern. This was once needed for "required
1680        byte" processing, but is no longer used. This recovers a now-scarce options
1681        bit. Also moved the least significant internal flag bit to the most-
1682        significant bit of the word, which was not previously used (hangover from
1683        the days when it was an int rather than a uint) to free up another bit for
1684        the future.
1686    23. Added support for CRLF line endings as well as CR and LF. As well as the
1687        default being selectable at build time, it can now be changed at runtime
1688        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1689        specify that it is scanning data with non-default line endings.
1691    24. Changed the definition of CXXLINK to make it agree with the definition of
1692        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1694    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1695        recursions. This makes a big different to stack usage for some patterns.
1697    26. If a subpattern containing a named recursion or subroutine reference such
1698        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1699        the space required for the compiled pattern went wrong and gave too small a
1700        value. Depending on the environment, this could lead to "Failed: internal
1701        error: code overflow at offset 49" or "glibc detected double free or
1702        corruption" errors.
1704    27. Applied patches from Google (a) to support the new newline modes and (b) to
1705        advance over multibyte UTF-8 characters in GlobalReplace.
1707    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1708        difference for some implementation of PCRE in some Windows version.
1710    29. Added some extra testing facilities to pcretest:
1712        \q<number>   in a data line sets the "match limit" value
1713        \Q<number>   in a data line sets the "match recursion limt" value
1714        -S <number>  sets the stack size, where <number> is in megabytes
1716        The -S option isn't available for Windows.
1719    Version 6.6 06-Feb-06
1720    ---------------------
1722     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1723        in pcreposix.h. I have copied the definition from pcre.h.
1725     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1726        because pcre.h is no longer a built file.
1728     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1729        not normally included in the compiled code.
1732    Version 6.5 01-Feb-06
1733    ---------------------
1735     1. When using the partial match feature with pcre_dfa_exec(), it was not
1736        anchoring the second and subsequent partial matches at the new starting
1737        point. This could lead to incorrect results. For example, with the pattern
1738        /1234/, partially matching against "123" and then "a4" gave a match.
1740     2. Changes to pcregrep:
1742        (a) All non-match returns from pcre_exec() were being treated as failures
1743            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1744            error message is output. Some extra information is given for the
1745            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
1746            probably the only errors that are likely to be caused by users (by
1747            specifying a regex that has nested indefinite repeats, for instance).
1748            If there are more than 20 of these errors, pcregrep is abandoned.
1750        (b) A binary zero was treated as data while matching, but terminated the
1751            output line if it was written out. This has been fixed: binary zeroes
1752            are now no different to any other data bytes.
1754        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1755            used to set a locale for matching. The --locale=xxxx long option has
1756            been added (no short equivalent) to specify a locale explicitly on the
1757            pcregrep command, overriding the environment variables.
1759        (d) When -B was used with -n, some line numbers in the output were one less
1760            than they should have been.
1762        (e) Added the -o (--only-matching) option.
1764        (f) If -A or -C was used with -c (count only), some lines of context were
1765            accidentally printed for the final match.
1767        (g) Added the -H (--with-filename) option.
1769        (h) The combination of options -rh failed to suppress file names for files
1770            that were found from directory arguments.
1772        (i) Added the -D (--devices) and -d (--directories) options.
1774        (j) Added the -F (--fixed-strings) option.
1776        (k) Allow "-" to be used as a file name for -f as well as for a data file.
1778        (l) Added the --colo(u)r option.
1780        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1781            is not present by default.
1783     3. A nasty bug was discovered in the handling of recursive patterns, that is,
1784        items such as (?R) or (?1), when the recursion could match a number of
1785        alternatives. If it matched one of the alternatives, but subsequently,
1786        outside the recursion, there was a failure, the code tried to back up into
1787        the recursion. However, because of the way PCRE is implemented, this is not
1788        possible, and the result was an incorrect result from the match.
1790        In order to prevent this happening, the specification of recursion has
1791        been changed so that all such subpatterns are automatically treated as
1792        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1794     4. I had overlooked the fact that, in some locales, there are characters for
1795        which isalpha() is true but neither isupper() nor islower() are true. In
1796        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1797        and ordfeminine) are like this. This affected the treatment of \w and \W
1798        when they appeared in character classes, but not when they appeared outside
1799        a character class. The bit map for "word" characters is now created
1800        separately from the results of isalnum() instead of just taking it from the
1801        upper, lower, and digit maps. (Plus the underscore character, of course.)
1803     5. The above bug also affected the handling of POSIX character classes such as
1804        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1805        permanent tables. Instead, the bit maps for such a class were previously
1806        created as the appropriate unions of the upper, lower, and digit bitmaps.
1807        Now they are created by subtraction from the [[:word:]] class, which has
1808        its own bitmap.
1810     6. The [[:blank:]] character class matches horizontal, but not vertical space.
1811        It is created by subtracting the vertical space characters (\x09, \x0a,
1812        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1813        subtraction was done in the overall bitmap for a character class, meaning
1814        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1815        be recognized. This bug has been fixed.
1817     7. Patches from the folks at Google:
1819          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1820          real life, but is still worth protecting against".
1822          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1823          regular expressions".
1825          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1826          have it.
1828          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1829          "configure" and the latter not, in order to fix a problem somebody had
1830          with compiling the Arg class on HP-UX.
1832          (e) Improve the error-handling of the C++ wrapper a little bit.
1834          (f) New tests for checking recursion limiting.
1836     8. The pcre_memmove() function, which is used only if the environment does not
1837        have a standard memmove() function (and is therefore rarely compiled),
1838        contained two bugs: (a) use of int instead of size_t, and (b) it was not
1839        returning a result (though PCRE never actually uses the result).
1841     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1842        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1843        returned instead of calling malloc() with an overflowing number that would
1844        most likely cause subsequent chaos.
1846    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1848    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1849        with this option is matched, the nmatch and pmatch options of regexec() are
1850        ignored.
1852    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1853        provided in case anyone wants to the the POSIX interface with UTF-8
1854        strings.
1856    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1857        C++ linking (needed for some HP-UX environments).
1859    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1860        (unused parameter) and in the pcre_printint() function (omitted "default"
1861        switch label when the default is to do nothing).
1863    15. Added some code to make it possible, when PCRE is compiled as a C++
1864        library, to replace subject pointers for pcre_exec() with a smart pointer
1865        class, thus making it possible to process discontinuous strings.
1867    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1868        much the same function. They were added by different people who were trying
1869        to make PCRE easy to compile on non-Unix systems. It has been suggested
1870        that PCRE_EXPORT be abolished now that there is more automatic apparatus
1871        for compiling on Windows systems. I have therefore replaced it with
1872        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1873        defaults to "extern" for C or "extern C" for C++, which works fine on
1874        Unix-like systems. It is now possible to override the value of PCRE_DATA_
1875        SCOPE with something explicit in config.h. In addition:
1877        (a) pcreposix.h still had just "extern" instead of either of these macros;
1878            I have replaced it with PCRE_DATA_SCOPE.
1880        (b) Functions such as _pcre_xclass(), which are internal to the library,
1881            but external in the C sense, all had PCRE_EXPORT in their definitions.
1882            This is apparently wrong for the Windows case, so I have removed it.
1883            (It makes no difference on Unix-like systems.)
1885    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1886        of recursive calls to match(). This is different to MATCH_LIMIT because
1887        that limits the total number of calls to match(), not all of which increase
1888        the depth of recursion. Limiting the recursion depth limits the amount of
1889        stack (or heap if NO_RECURSE is set) that is used. The default can be set
1890        when PCRE is compiled, and changed at run time. A patch from Google adds
1891        this functionality to the C++ interface.
1893    18. Changes to the handling of Unicode character properties:
1895        (a) Updated the table to Unicode 4.1.0.
1897        (b) Recognize characters that are not in the table as "Cn" (undefined).
1899        (c) I revised the way the table is implemented to a much improved format
1900            which includes recognition of ranges. It now supports the ranges that
1901            are defined in UnicodeData.txt, and it also amalgamates other
1902            characters into ranges. This has reduced the number of entries in the
1903            table from around 16,000 to around 3,000, thus reducing its size
1904            considerably. I realized I did not need to use a tree structure after
1905            all - a binary chop search is just as efficient. Having reduced the
1906            number of entries, I extended their size from 6 bytes to 8 bytes to
1907            allow for more data.
1909        (d) Added support for Unicode script names via properties such as \p{Han}.
1911    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1912        matching that character.
1914    20. When matching a repeated Unicode property with a minimum greater than zero,
1915        (for example \pL{2,}), PCRE could look past the end of the subject if it
1916        reached it while seeking the minimum number of characters. This could
1917        happen only if some of the characters were more than one byte long, because
1918        there is a check for at least the minimum number of bytes.
1920    21. Refactored the implementation of \p and \P so as to be more general, to
1921        allow for more different types of property in future. This has changed the
1922        compiled form incompatibly. Anybody with saved compiled patterns that use
1923        \p or \P will have to recompile them.
1925    22. Added "Any" and "L&" to the supported property types.
1927    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1928        but give a compile time error if the value is greater than 0xff.
1930    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1931        accidentally not being installed or uninstalled.
1933    25. The pcre.h file was built from pcre.h.in, but the only changes that were
1934        made were to insert the current release number. This seemed silly, because
1935        it made things harder for people building PCRE on systems that don't run
1936        "configure". I have turned pcre.h into a distributed file, no longer built
1937        by "configure", with the version identification directly included. There is
1938        no longer a pcre.h.in file.
1940        However, this change necessitated a change to the pcre-config script as
1941        well. It is built from pcre-config.in, and one of the substitutions was the
1942        release number. I have updated configure.ac so that ./configure now finds
1943        the release number by grepping pcre.h.
1945    26. Added the ability to run the tests under valgrind.
1948    Version 6.4 05-Sep-05
1949    ---------------------
1951     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1952        "--" to be printed when multiple files were scanned, even when none of the
1953        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1954        consider it to be a bug, and have restored the previous behaviour.
1956     2. A couple of code tidies to get rid of compiler warnings.
1958     3. The pcretest program used to cheat by referring to symbols in the library
1959        whose names begin with _pcre_. These are internal symbols that are not
1960        really supposed to be visible externally, and in some environments it is
1961        possible to suppress them. The cheating is now confined to including
1962        certain files from the library's source, which is a bit cleaner.
1964     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1965        file's purpose clearer.
1967     5. Reorganized pcre_ucp_findchar().
1970    Version 6.3 15-Aug-05
1971    ---------------------
1973     1. The file libpcre.pc.in did not have general read permission in the tarball.
1975     2. There were some problems when building without C++ support:
1977        (a) If C++ support was not built, "make install" and "make test" still
1978            tried to test it.
1980        (b) There were problems when the value of CXX was explicitly set. Some
1981            changes have been made to try to fix these, and ...
1983        (c) --disable-cpp can now be used to explicitly disable C++ support.
1985        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1986            backslash in a target when C++ was disabled. This confuses some
1987            versions of "make", apparently. Using an intermediate variable solves
1988            this. (Same for CPP_LOBJ.)
1990     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1991        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1992        necessary on certain architectures.
1994     4. Added a setting of -export-symbols-regex to the link command to remove
1995        those symbols that are exported in the C sense, but actually are local
1996        within the library, and not documented. Their names all begin with
1997        "_pcre_". This is not a perfect job, because (a) we have to except some
1998        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1999        available (and never for static libraries). I have made a note to try to
2000        find a way round (a) in the future.
2003    Version 6.2 01-Aug-05
2004    ---------------------
2006     1. There was no test for integer overflow of quantifier values. A construction
2007        such as {1111111111111111} would give undefined results. What is worse, if
2008        a minimum quantifier for a parenthesized subpattern overflowed and became
2009        negative, the calculation of the memory size went wrong. This could have
2010        led to memory overwriting.
2012     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
2014     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
2015        operating environments where this matters.
2017     4. Applied Giuseppe Maxia's patch to add additional features for controlling
2018        PCRE options from within the C++ wrapper.
2020     5. Named capturing subpatterns were not being correctly counted when a pattern
2021        was compiled. This caused two problems: (a) If there were more than 100
2022        such subpatterns, the calculation of the memory needed for the whole
2023        compiled pattern went wrong, leading to an overflow error. (b) Numerical
2024        back references of the form \12, where the number was greater than 9, were
2025        not recognized as back references, even though there were sufficient
2026        previous subpatterns.
2028     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
2029        versions of gcc, e.g. 2.95.4.
2032    Version 6.1 21-Jun-05
2033    ---------------------
2035     1. There was one reference to the variable "posix" in pcretest.c that was not
2036        surrounded by "#if !defined NOPOSIX".
2038     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
2039        the cross-check on the old pcre_info() function, for the benefit of the
2040        cut-down version of PCRE that is currently imported into Exim.
2042     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
2043        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
2044        patterns that start (?i) but I don't think that matters. The use of (?i) is
2045        just an example; this all applies to the other options as well.
2047     4. Since libtool seems to echo the compile commands it is issuing, the output
2048        from "make" can be reduced a bit by putting "@" in front of each libtool
2049        compile command.
2051     5. Patch from the folks at Google for configure.in to be a bit more thorough
2052        in checking for a suitable C++ installation before trying to compile the
2053        C++ stuff. This should fix a reported problem when a compiler was present,
2054        but no suitable headers.
2056     6. The man pages all had just "PCRE" as their title. I have changed them to
2057        be the relevant file name. I have also arranged that these names are
2058        retained in the file doc/pcre.txt, which is a concatenation in text format
2059        of all the man pages except the little individual ones for each function.
2061     7. The NON-UNIX-USE file had not been updated for the different set of source
2062        files that come with release 6. I also added a few comments about the C++
2063        wrapper.
2066    Version 6.0 07-Jun-05
2067    ---------------------
2069     1. Some minor internal re-organization to help with my DFA experiments.
2071     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
2072        didn't matter for the library itself when fully configured, but did matter
2073        when compiling without UCP support, or within Exim, where the ucp files are
2074        not imported.
2076     3. Refactoring of the library code to split up the various functions into
2077        different source modules. The addition of the new DFA matching code (see
2078        below) to a single monolithic source would have made it really too
2079        unwieldy, quite apart from causing all the code to be include in a
2080        statically linked application, when only some functions are used. This is
2081        relevant even without the DFA addition now that patterns can be compiled in
2082        one application and matched in another.
2084        The downside of splitting up is that there have to be some external
2085        functions and data tables that are used internally in different modules of
2086        the library but which are not part of the API. These have all had their
2087        names changed to start with "_pcre_" so that they are unlikely to clash
2088        with other external names.
2090     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
2091        a different (DFA) algorithm. Although it is slower than the original
2092        function, it does have some advantages for certain types of matching
2093        problem.
2095     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
2096        including restarting after a partial match.
2098     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
2099        defined when compiling for Windows was sent to me. I have put it into the
2100        code, though I have no means of testing or verifying it.
2102     7. Added the pcre_refcount() auxiliary function.
2104     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
2105        match before or at the first newline in the subject string. In pcretest,
2106        the /f option on a pattern can be used to set this.
2108     9. A repeated \w when used in UTF-8 mode with characters greater than 256
2109        would behave wrongly. This has been present in PCRE since release 4.0.
2111    10. A number of changes to the pcregrep command:
2113        (a) Refactored how -x works; insert ^(...)$ instead of setting
2114            PCRE_ANCHORED and checking the length, in preparation for adding
2115            something similar for -w.
2117        (b) Added the -w (match as a word) option.
2119        (c) Refactored the way lines are read and buffered so as to have more
2120            than one at a time available.
2122        (d) Implemented a pcregrep test script.
2124        (e) Added the -M (multiline match) option. This allows patterns to match
2125            over several lines of the subject. The buffering ensures that at least
2126            8K, or the rest of the document (whichever is the shorter) is available
2127            for matching (and similarly the previous 8K for lookbehind assertions).
2129        (f) Changed the --help output so that it now says
2131              -w, --word-regex(p)
2133            instead of two lines, one with "regex" and the other with "regexp"
2134            because that confused at least one person since the short forms are the
2135            same. (This required a bit of code, as the output is generated
2136            automatically from a table. It wasn't just a text change.)
2138        (g) -- can be used to terminate pcregrep options if the next thing isn't an
2139            option but starts with a hyphen. Could be a pattern or a path name
2140            starting with a hyphen, for instance.
2142        (h) "-" can be given as a file name to represent stdin.
2144        (i) When file names are being printed, "(standard input)" is used for
2145            the standard input, for compatibility with GNU grep. Previously
2146            "<stdin>" was used.
2148        (j) The option --label=xxx can be used to supply a name to be used for
2149            stdin when file names are being printed. There is no short form.
2151        (k) Re-factored the options decoding logic because we are going to add
2152            two more options that take data. Such options can now be given in four
2153            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
2155        (l) Added the -A, -B, and -C options for requesting that lines of context
2156            around matches be printed.
2158        (m) Added the -L option to print the names of files that do not contain
2159            any matching lines, that is, the complement of -l.
2161        (n) The return code is 2 if any file cannot be opened, but pcregrep does
2162            continue to scan other files.
2164        (o) The -s option was incorrectly implemented. For compatibility with other
2165            greps, it now suppresses the error message for a non-existent or non-
2166            accessible file (but not the return code). There is a new option called
2167            -q that suppresses the output of matching lines, which was what -s was
2168            previously doing.
2170        (p) Added --include and --exclude options to specify files for inclusion
2171            and exclusion when recursing.
2173    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
2174        Hopefully, it now does.
2176    12. Missing cast in pcre_study().
2178    13. Added an "uninstall" target to the makefile.
2180    14. Replaced "extern" in the function prototypes in Makefile.in with
2181        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
2182        world, but is set differently for Windows.
2184    15. Added a second compiling function called pcre_compile2(). The only
2185        difference is that it has an extra argument, which is a pointer to an
2186        integer error code. When there is a compile-time failure, this is set
2187        non-zero, in addition to the error test pointer being set to point to an
2188        error message. The new argument may be NULL if no error number is required
2189        (but then you may as well call pcre_compile(), which is now just a
2190        wrapper). This facility is provided because some applications need a
2191        numeric error indication, but it has also enabled me to tidy up the way
2192        compile-time errors are handled in the POSIX wrapper.
2194    16. Added VPATH=.libs to the makefile; this should help when building with one
2195        prefix path and installing with another. (Or so I'm told by someone who
2196        knows more about this stuff than I do.)
2198    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
2199        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
2200        match everything, including newlines. This is not POSIX-compatible, but
2201        somebody wanted the feature. From pcretest it can be activated by using
2202        both the P and the s flags.
2204    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
2206    19. libpcre.pc was being incorrectly installed as executable.
2208    20. A couple of places in pcretest check for end-of-line by looking for '\n';
2209        it now also looks for '\r' so that it will work unmodified on Windows.
2211    21. Added Google's contributed C++ wrapper to the distribution.
2213    22. Added some untidy missing memory free() calls in pcretest, to keep
2214        Electric Fence happy when testing.
2218    Version 5.0 13-Sep-04
2219    ---------------------
2221     1. Internal change: literal characters are no longer packed up into items
2222        containing multiple characters in a single byte-string. Each character
2223        is now matched using a separate opcode. However, there may be more than one
2224        byte in the character in UTF-8 mode.
2226     2. The pcre_callout_block structure has two new fields: pattern_position and
2227        next_item_length. These contain the offset in the pattern to the next match
2228        item, and its length, respectively.
2230     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
2231        insertion of callouts before each pattern item. Added the /C option to
2232        pcretest to make use of this.
2234     4. On the advice of a Windows user, the lines
2236          #if defined(_WIN32) || defined(WIN32)
2237          _setmode( _fileno( stdout ), 0x8000 );
2238          #endif  /* defined(_WIN32) || defined(WIN32) */
2240        have been added to the source of pcretest. This apparently does useful
2241        magic in relation to line terminators.
2243     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
2244        for the benefit of those environments where the "b" makes a difference.
2246     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
2247        to know about it. I have put a hack into configure.in that adds in code
2248        to set GCC=yes if CC=icc. This seems to end up at a point in the
2249        generated configure script that is early enough to affect the setting of
2250        compiler options, which is what is needed, but I have no means of testing
2251        whether it really works. (The user who reported this had patched the
2252        generated configure script, which of course I cannot do.)
2254        LATER: After change 22 below (new libtool files), the configure script
2255        seems to know about icc (and also ecc). Therefore, I have commented out
2256        this hack in configure.in.
2258     7. Added support for pkg-config (2 patches were sent in).
2260     8. Negated POSIX character classes that used a combination of internal tables
2261        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2262        [[:^ascii]]. Typically, they would match almost any characters. The other
2263        POSIX classes were not broken in this way.
2265     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2266        to find the match, as PCRE was deluded into thinking that the match had to
2267        start at the start point or following a newline. The same bug applied to
2268        patterns with negative forward assertions or any backward assertions
2269        preceding ".*" at the start, unless the pattern required a fixed first
2270        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2272    10. In UTF-8 mode, when moving forwards in the subject after a failed match
2273        starting at the last subject character, bytes beyond the end of the subject
2274        string were read.
2276    11. Renamed the variable "class" as "classbits" to make life easier for C++
2277        users. (Previously there was a macro definition, but it apparently wasn't
2278        enough.)
2280    12. Added the new field "tables" to the extra data so that tables can be passed
2281        in at exec time, or the internal tables can be re-selected. This allows
2282        a compiled regex to be saved and re-used at a later time by a different
2283        program that might have everything at different addresses.
2285    13. Modified the pcre-config script so that, when run on Solaris, it shows a
2286        -R library as well as a -L library.
2288    14. The debugging options of pcretest (-d on the command line or D on a
2289        pattern) showed incorrect output for anything following an extended class
2290        that contained multibyte characters and which was followed by a quantifier.
2292    15. Added optional support for general category Unicode character properties
2293        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2294        support. It adds about 90K to the size of the library. The meanings of the
2295        inbuilt class escapes such as \d and \s have NOT been changed.
2297    16. Updated pcredemo.c to include calls to free() to release the memory for the
2298        compiled pattern.
2300    17. The generated file chartables.c was being created in the source directory
2301        instead of in the building directory. This caused the build to fail if the
2302        source directory was different from the building directory, and was
2303        read-only.
2305    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2306        file. No doubt somebody will tell me if they don't make sense... Also added
2307        Dan Mooney's comments about building on OpenVMS.
2309    19. Added support for partial matching via the PCRE_PARTIAL option for
2310        pcre_exec() and the \P data escape in pcretest.
2312    20. Extended pcretest with 3 new pattern features:
2314        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
2315              write the compiled pattern to the file whose name is "rest-of-line".
2316              This is a straight binary dump of the data, with the saved pointer to
2317              the character tables forced to be NULL. The study data, if any, is
2318              written too. After writing, pcretest reads a new pattern.
2320        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2321              compiled pattern from the given file. There must not be any
2322              occurrences of "<" in the file name (pretty unlikely); if there are,
2323              pcretest will instead treat the initial "<" as a pattern delimiter.
2324              After reading in the pattern, pcretest goes on to read data lines as
2325              usual.
2327        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2328              and 16-bit fields in a compiled pattern, to simulate a pattern that
2329              was compiled on a host of opposite endianness.
2331    21. The pcre-exec() function can now cope with patterns that were compiled on
2332        hosts of opposite endianness, with this restriction:
2334          As for any compiled expression that is saved and used later, the tables
2335          pointer field cannot be preserved; the extra_data field in the arguments
2336          to pcre_exec() should be used to pass in a tables address if a value
2337          other than the default internal tables were used at compile time.
2339    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2340        now diagnosed as an error. Previously, most of the time, a negative number
2341        would have been treated as zero, but if in addition "ovector" was passed as
2342        NULL, a crash could occur.
2344    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2345        new versions from the libtool 1.5 distribution (the last one is a copy of
2346        a file called libtool.m4). This seems to have fixed the need to patch
2347        "configure" to support Darwin 1.3 (which I used to do). However, I still
2348        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2349        workstation).
2351    24. Changed the PCRE licence to be the more standard "BSD" licence.
2354    Version 4.5 01-Dec-03
2355    ---------------------
2357     1. There has been some re-arrangement of the code for the match() function so
2358        that it can be compiled in a version that does not call itself recursively.
2359        Instead, it keeps those local variables that need separate instances for
2360        each "recursion" in a frame on the heap, and gets/frees frames whenever it
2361        needs to "recurse". Keeping track of where control must go is done by means
2362        of setjmp/longjmp. The whole thing is implemented by a set of macros that
2363        hide most of the details from the main code, and operates only if
2364        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2365        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2366        operating.
2368        To make it easier for callers to provide specially tailored get/free
2369        functions for this usage, two new functions, pcre_stack_malloc, and
2370        pcre_stack_free, are used. They are always called in strict stacking order,
2371        and the size of block requested is always the same.
2373        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2374        PCRE has been compiled to use the stack or the heap for recursion. The
2375        -C option of pcretest uses this to show which version is compiled.
2377        A new data escape \S, is added to pcretest; it causes the amounts of store
2378        obtained and freed by both kinds of malloc/free at match time to be added
2379        to the output.
2381     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2382        what's available on my current Linux desktop machine.
2384     3. When matching a UTF-8 string, the test for a valid string at the start has
2385        been extended. If start_offset is not zero, PCRE now checks that it points
2386        to a byte that is the start of a UTF-8 character. If not, it returns
2387        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2388        this is necessary because there may be backward assertions in the pattern.
2389        When matching the same subject several times, it may save resources to use
2390        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2392     4. The code for checking the validity of UTF-8 strings has been tightened so
2393        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2394        containing "overlong sequences".
2396     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2397        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2398        should have been "&", but it just so happened that all the cases this let
2399        through by mistake were picked up later in the function.
2401     6. I had used a variable called "isblank" - this is a C99 function, causing
2402        some compilers to warn. To avoid this, I renamed it (as "blankclass").
2404     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2405        prompting; (b) run "./pcretest /dev/null" at the start of the test script
2406        so the version is shown; (c) stop "make test" echoing "./RunTest".
2408     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2410     9. The prototype for memmove() for systems that don't have it was using
2411        size_t, but the inclusion of the header that defines size_t was later. I've
2412        moved the #includes for the C headers earlier to avoid this.
2414    10. Added some adjustments to the code to make it easier to compiler on certain
2415        special systems:
2417          (a) Some "const" qualifiers were missing.
2418          (b) Added the macro EXPORT before all exported functions; by default this
2419              is defined to be empty.
2420          (c) Changed the dftables auxiliary program (that builds chartables.c) so
2421              that it reads its output file name as an argument instead of writing
2422              to the standard output and assuming this can be redirected.
2424    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2425        class containing characters with values greater than 255, PCRE compilation
2426        went into a loop.
2428    12. A recursive reference to a subpattern that was within another subpattern
2429        that had a minimum quantifier of zero caused PCRE to crash. For example,
2430        (x(y(?2))z)? provoked this bug with a subject that got as far as the
2431        recursion. If the recursively-called subpattern itself had a zero repeat,
2432        that was OK.
2434    13. In pcretest, the buffer for reading a data line was set at 30K, but the
2435        buffer into which it was copied (for escape processing) was still set at
2436        1024, so long lines caused crashes.
2438    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2439        "internal error: code overflow...". This applied to any character class
2440        that was followed by a possessive quantifier.
2442    15. Modified the Makefile to add libpcre.la as a prerequisite for
2443        libpcreposix.la because I was told this is needed for a parallel build to
2444        work.
2446    16. If a pattern that contained .* following optional items at the start was
2447        studied, the wrong optimizing data was generated, leading to matching
2448        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2449        matching string must start with a or b or c. The correct conclusion for
2450        this pattern is that a match can start with any character.
2453    Version 4.4 13-Aug-03
2454    ---------------------
2456     1. In UTF-8 mode, a character class containing characters with values between
2457        127 and 255 was not handled correctly if the compiled pattern was studied.
2458        In fixing this, I have also improved the studying algorithm for such
2459        classes (slightly).
2461     2. Three internal functions had redundant arguments passed to them. Removal
2462        might give a very teeny performance improvement.
2464     3. Documentation bug: the value of the capture_top field in a callout is *one
2465        more than* the number of the hightest numbered captured substring.
2467     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2468        in incorrectly linking with a previously installed version. They now link
2469        explicitly with libpcre.la.
2471     5. configure.in no longer needs to recognize Cygwin specially.
2473     6. A problem in pcre.in for Windows platforms is fixed.
2475     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2476        pcretest, it used to include the size of the study block as part of its
2477        output. Unfortunately, the structure contains a field that has a different
2478        size on different hardware architectures. This meant that the tests that
2479        showed this size failed. As the block is currently always of a fixed size,
2480        this information isn't actually particularly useful in pcretest output, so
2481        I have just removed it.
2483     8. Three pre-processor statements accidentally did not start in column 1.
2484        Sadly, there are *still* compilers around that complain, even though
2485        standard C has not required this for well over a decade. Sigh.
2487     9. In pcretest, the code for checking callouts passed small integers in the
2488        callout_data field, which is a void * field. However, some picky compilers
2489        complained about the casts involved for this on 64-bit systems. Now
2490        pcretest passes the address of the small integer instead, which should get
2491        rid of the warnings.
2493    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2494        both compile and run time, and gives an error if an invalid UTF-8 sequence
2495        is found. There is a option for disabling this check in cases where the
2496        string is known to be correct and/or the maximum performance is wanted.
2498    11. In response to a bug report, I changed one line in Makefile.in from
2500            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2501        to
2502            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2504        to look similar to other lines, but I have no way of telling whether this
2505        is the right thing to do, as I do not use Windows. No doubt I'll get told
2506        if it's wrong...
2509    Version 4.3 21-May-03
2510    ---------------------
2512    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2513       Makefile.
2515    2. Some refactoring to improve the quality of the code:
2517       (i)   The utf8_table... variables are now declared "const".
2519       (ii)  The code for \cx, which used the "case flipping" table to upper case
2520             lower case letters, now just substracts 32. This is ASCII-specific,
2521             but the whole concept of \cx is ASCII-specific, so it seems
2522             reasonable.
2524       (iii) PCRE was using its character types table to recognize decimal and
2525             hexadecimal digits in the pattern. This is silly, because it handles
2526             only 0-9, a-f, and A-F, but the character types table is locale-
2527             specific, which means strange things might happen. A private
2528             table is now used for this - though it costs 256 bytes, a table is
2529             much faster than multiple explicit tests. Of course, the standard
2530             character types table is still used for matching digits in subject
2531             strings against \d.
2533       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2534             ending in _t are). So I've renamed it as ESC_tee.
2536    3. The first argument for regexec() in the POSIX wrapper should have been
2537       defined as "const".
2539    4. Changed pcretest to use malloc() for its buffers so that they can be
2540       Electric Fenced for debugging.
2542    5. There were several places in the code where, in UTF-8 mode, PCRE would try
2543       to read one or more bytes before the start of the subject string. Often this
2544       had no effect on PCRE's behaviour, but in some circumstances it could
2545       provoke a segmentation fault.
2547    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2548       to try to read one or more bytes before the start of the subject string.
2550    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
2551       UTF-8 support could misbehave in various ways if the subject string
2552       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
2553       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
2554       back over UTF-8 characters.)
2557    Version 4.2 14-Apr-03
2558    ---------------------
2560    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
2562    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
2563         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
2564         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
2565         [WIN_PREFIX]: new variable, "cyg" for Cygwin
2566         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
2567           and BUILD_EXEEXT
2568         Note: automatic setting of the BUILD variables is not yet working
2569         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
2570           compile-time but not at link-time
2571         [LINK]: use for linking executables only
2572         make different versions for Windows and non-Windows
2573         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
2574           libraries
2575         [LINK_FOR_BUILD]: new variable
2576         [OBJEXT]: use throughout
2577         [EXEEXT]: use throughout
2578         <winshared>: new target
2579         <wininstall>: new target
2580         <dftables.o>: use native compiler
2581         <dftables>: use native linker
2582         <install>: handle Windows platform correctly
2583         <clean>: ditto
2584         <check>: ditto
2585         copy DLL to top builddir before testing
2587       As part of these changes, -no-undefined was removed again. This was reported
2588       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
2589       in any case.
2591    3. Some tidies to get rid of compiler warnings:
2593       . In the match_data structure, match_limit was an unsigned long int, whereas
2594         match_call_count was an int. I've made them both unsigned long ints.
2596       . In pcretest the fact that a const uschar * doesn't automatically cast to
2597         a void * provoked a warning.
2599       . Turning on some more compiler warnings threw up some "shadow" variables
2600         and a few more missing casts.
2602    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
2603       option, a class that contained a single character with a value between 128
2604       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
2606    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
2607       option, a class that contained several characters, but with at least one
2608       whose value was between 128 and 255 caused PCRE to crash.
2611    Version 4.1 12-Mar-03
2612    ---------------------
2614    1. Compiling with gcc -pedantic found a couple of places where casts were
2615    needed, and a string in dftables.c that was longer than standard compilers are
2616    required to support.
2618    2. Compiling with Sun's compiler found a few more places where the code could
2619    be tidied up in order to avoid warnings.
2621    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
2622    first of these names is deprecated in the latest Autoconf in favour of the name
2623    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
2624    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2625    analogy I have changed it to CFLAGS_FOR_BUILD.
2627    4. Added -no-undefined to the linking command in the Makefile, because this is
2628    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2629    linking step for the pcreposix library.
2631    5. PCRE was failing to diagnose the case of two named groups with the same
2632    name.
2634    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2635    literal character that is needed in the subject for a match, and scans along to
2636    ensure that it is present before embarking on the full matching process. This
2637    saves time in cases of nested unlimited repeats that are never going to match.
2638    Problem: the scan can take a lot of time if the subject is very long (e.g.
2639    megabytes), thus penalizing straightforward matches. It is now done only if the
2640    amount of subject to be scanned is less than 1000 bytes.
2642    7. A lesser problem with the same optimization is that it was recording the
2643    first character of an anchored pattern as "needed", thus provoking a search
2644    right along the subject, even when the first match of the pattern was going to
2645    fail. The "needed" character is now not set for anchored patterns, unless it
2646    follows something in the pattern that is of non-fixed length. Thus, it still
2647    fulfils its original purpose of finding quick non-matches in cases of nested
2648    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
2651    Version 4.0 17-Feb-03
2652    ---------------------
2654    1. If a comment in an extended regex that started immediately after a meta-item
2655    extended to the end of string, PCRE compiled incorrect data. This could lead to
2656    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
2658    2. Moved to autoconf 2.53 and libtool 1.4.2.
2660    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
2661    the special perltest8 script is no longer needed - all the tests can be run
2662    from a single perltest script.
2664    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
2665    by \s. It has now been removed in PCRE. This means it isn't recognized as
2666    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
2667    class [:space:] *does* include VT, thereby creating a mess.
2669    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
2670    space and tab.
2672    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
2673    its new features into the main test script, reducing the number of scripts.
2675    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
2676    were backward compatible, and made the (?i) apply to the whole pattern, as if
2677    /i were given. Now it behaves more logically, and applies the option setting
2678    only to what follows. PCRE has been changed to follow suit. However, if it
2679    finds options settings right at the start of the pattern, it extracts them into
2680    the global options, as before. Thus, they show up in the info data.
2682    8. Added support for the \Q...\E escape sequence. Characters in between are
2683    treated as literals. This is slightly different from Perl in that $ and @ are
2684    also handled as literals inside the quotes. In Perl, they will cause variable
2685    interpolation. Note the following examples:
2687        Pattern            PCRE matches      Perl matches
2689        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
2690        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
2691        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
2693    For compatibility with Perl, \Q...\E sequences are recognized inside character
2694    classes as well as outside them.
2696    9. Re-organized 3 code statements in pcretest to avoid "overflow in
2697    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2698    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2699    signed/unsigned warnings.
2701    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2702    option for pcretest, so I've replaced it by a simple function that does just
2703    that job.
2705    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2706    "pcregrep -".
2708    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2709    Java package. This provides some syntactic sugar for simple cases of what my
2710    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2711    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2712    item, you can use this simplified notation. Note that only makes sense with
2713    greedy quantifiers. Consequently, the use of the possessive quantifier forces
2714    greediness, whatever the setting of the PCRE_UNGREEDY option.
2716    13. A change of greediness default within a pattern was not taking effect at
2717    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2718    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2719    was abstracted outside.
2721    14. PCRE now supports the \G assertion. It is true when the current matching
2722    position is at the start point of the match. This differs from \A when the
2723    starting offset is non-zero. Used with the /g option of pcretest (or similar
2724    code), it works in the same way as it does for Perl's /g option. If all
2725    alternatives of a regex begin with \G, the expression is anchored to the start
2726    match position, and the "anchored" flag is set in the compiled expression.
2728    15. Some bugs concerning the handling of certain option changes within patterns
2729    have been fixed. These applied to options other than (?ims). For example,
2730    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2731    been the other way round. Some of this was related to change 7 above.
2733    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2734    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2735    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2736    POSIX classes only within a class (e.g. /[[:alpha:]]/).
2738    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2739    mode. Unlike ".", it always matches newline, whatever the setting of
2740    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2741    assertions. Perl allows it, but it doesn't (in general) work because it can't
2742    calculate the length of the lookbehind. At least, that's the case for Perl
2743    5.8.0 - I've been told they are going to document that it doesn't work in
2744    future.
2746    18. Added an error diagnosis for escapes that PCRE does not support: these are
2747    \L, \l, \N, \P, \p, \U, \u, and \X.
2749    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2750    reading past the end of the pattern in cases such as /[abcd/.
2752    20. PCRE was getting more memory than necessary for patterns with classes that
2753    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2755    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2756    compiling PCRE for use with Virtual Pascal.
2758    22. Small fix to the Makefile to make it work properly if the build is done
2759    outside the source tree.
2761    23. Added a new extension: a condition to go with recursion. If a conditional
2762    subpattern starts with (?(R) the "true" branch is used if recursion has
2763    happened, whereas the "false" branch is used only at the top level.
2765    24. When there was a very long string of literal characters (over 255 bytes
2766    without UTF support, over 250 bytes with UTF support), the computation of how
2767    much memory was required could be incorrect, leading to segfaults or other
2768    strange effects.
2770    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2771    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2772    there was a subsequent back reference to those brackets. This meant that, for
2773    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2774    possible to check for precisely this case. All we can do is abandon the
2775    optimization if .* occurs inside capturing brackets when there are any back
2776    references whatsoever. (See below for a better fix that came later.)
2778    26. The handling of the optimization for finding the first character of a
2779    non-anchored pattern, and for finding a character that is required later in the
2780    match were failing in some cases. This didn't break the matching; it just
2781    failed to optimize when it could. The way this is done has been re-implemented.
2783    27. Fixed typo in error message for invalid (?R item (it said "(?p").
2785    28. Added a new feature that provides some of the functionality that Perl
2786    provides with (?{...}). The facility is termed a "callout". The way it is done
2787    in PCRE is for the caller to provide an optional function, by setting
2788    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2789    global variable. By default it is unset, which disables all calling out. To get
2790    the function called, the regex must include (?C) at appropriate points. This
2791    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2792    This provides a means of identifying different callout points. When PCRE
2793    reaches such a point in the regex, if pcre_callout has been set, the external
2794    function is called. It is provided with data in a structure called
2795    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2796    matching continues; if it returns a non-zero value, the match at the current
2797    point fails. However, backtracking will occur if possible. [This was changed
2798    later and other features added - see item 49 below.]
2800    29. pcretest is upgraded to test the callout functionality. It provides a
2801    callout function that displays information. By default, it shows the start of
2802    the match and the current position in the text. There are some new data escapes
2803    to vary what happens:
2805        \C+         in addition, show current contents of captured substrings
2806        \C-         do not supply a callout function
2807        \C!n        return 1 when callout number n is reached
2808        \C!n!m      return 1 when callout number n is reached for the mth time
2810    30. If pcregrep was called with the -l option and just a single file name, it
2811    output "<stdin>" if a match was found, instead of the file name.
2813    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2814    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2815    pcre_exec(). This saves a malloc/free per call. The default value of
2816    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2817    when configuring.
2819    32. The default maximum size of a compiled pattern is 64K. There have been a
2820    few cases of people hitting this limit. The code now uses macros to handle the
2821    storing of links as offsets within the compiled pattern. It defaults to 2-byte
2822    links, but this can be changed to 3 or 4 bytes by --with-link-size when
2823    configuring. Tests 2 and 5 work only with 2-byte links because they output
2824    debugging information about compiled patterns.
2826    33. Internal code re-arrangements:
2828    (a) Moved the debugging function for printing out a compiled regex into
2829        its own source file (printint.c) and used #include to pull it into
2830        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2831        separate copies.
2833    (b) Defined the list of op-code names for debugging as a macro in
2834        internal.h so that it is next to the definition of the opcodes.
2836    (c) Defined a table of op-code lengths for simpler skipping along compiled
2837        code. This is again a macro in internal.h so that it is next to the
2838        definition of the opcodes.
2840    34. Added support for recursive calls to individual subpatterns, along the
2841    lines of Robin Houston's patch (but implemented somewhat differently).
2843    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2844    allow it to read and process whole directories in Win32. This code was
2845    contributed by Lionel Fourquaux; it has not been tested by me.
2847    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2848    used to name a group. Names consist of alphanumerics and underscores, and must
2849    be unique. Back references use the syntax (?P=name) and recursive calls use
2850    (?P>name) which is a PCRE extension to the Python extension. Groups still have
2851    numbers. The function pcre_fullinfo() can be used after compilation to extract
2852    a name/number map. There are three relevant calls:
2854      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
2855      PCRE_INFO_NAMECOUNT            yields the number of entries
2856      PCRE_INFO_NAMETABLE            yields a pointer to the map.
2858    The map is a vector of fixed-size entries. The size of each entry depends on
2859    the length of the longest name used. The first two bytes of each entry are the
2860    group number, most significant byte first. There follows the corresponding
2861    name, zero terminated. The names are in alphabetical order.
2863    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2864    case instead of 255. Making it the same both with and without UTF-8 support
2865    means that the same test output works with both.
2867    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2868    calling malloc() with a zero argument.
2870    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2871    optimization. I've improved things by keeping a bitmap of backreferences with
2872    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2873    fact referenced, the optimization can be applied. It is unlikely that a
2874    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2875    the match to follow \n) will appear inside brackets with a number greater than
2876    31, but if it does, any back reference > 31 suppresses the optimization.
2878    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2879    of disabling numbered capturing parentheses. Any opening parenthesis that is
2880    not followed by ? behaves as if it were followed by ?: but named parentheses
2881    can still be used for capturing (and they will acquire numbers in the usual
2882    way).
2884    41. Redesigned the return codes from the match() function into yes/no/error so
2885    that errors can be passed back from deep inside the nested calls. A malloc
2886    failure while inside a recursive subpattern call now causes the
2887    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2889    42. It is now possible to set a limit on the number of times the match()
2890    function is called in a call to pcre_exec(). This facility makes it possible to
2891    limit the amount of recursion and backtracking, though not in a directly
2892    obvious way, because the match() function is used in a number of different
2893    circumstances. The count starts from zero for each position in the subject
2894    string (for non-anchored patterns). The default limit is, for compatibility, a
2895    large number, namely 10 000 000. You can change this in two ways:
2897    (a) When configuring PCRE before making, you can use --with-match-limit=n
2898        to set a default value for the compiled library.
2900    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2901        a different value is set. See 45 below.
2903    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2905    43. Added a new function pcre_config(int, void *) to enable run-time extraction
2906    of things that can be changed at compile time. The first argument specifies
2907    what is wanted and the second points to where the information is to be placed.
2908    The current list of available information is:
2910      PCRE_CONFIG_UTF8
2912    The output is an integer that is set to one if UTF-8 support is available;
2913    otherwise it is set to zero.
2917    The output is an integer that it set to the value of the code that is used for
2918    newline. It is either LF (10) or CR (13).
2922    The output is an integer that contains the number of bytes used for internal
2923    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2927    The output is an integer that contains the threshold above which the POSIX
2928    interface uses malloc() for output vectors. See item 31 above.
2932    The output is an unsigned integer that contains the default limit of the number
2933    of match() calls in a pcre_exec() execution. See 42 above.
2935    44. pcretest has been upgraded by the addition of the -C option. This causes it
2936    to extract all the available output from the new pcre_config() function, and to
2937    output it. The program then exits immediately.
2939    45. A need has arisen to pass over additional data with calls to pcre_exec() in
2940    order to support additional features. One way would have been to define
2941    pcre_exec2() (for example) with extra arguments, but this would not have been
2942    extensible, and would also have required all calls to the original function to
2943    be mapped to the new one. Instead, I have chosen to extend the mechanism that
2944    is used for passing in "extra" data from pcre_study().
2946    The pcre_extra structure is now exposed and defined in pcre.h. It currently
2947    contains the following fields:
2949      flags         a bitmap indicating which of the following fields are set
2950      study_data    opaque data from pcre_study()
2951      match_limit   a way of specifying a limit on match() calls for a specific
2952                      call to pcre_exec()
2953      callout_data  data for callouts (see 49 below)
2955    The flag bits are also defined in pcre.h, and are
2961    The pcre_study() function now returns one of these new pcre_extra blocks, with
2962    the actual study data pointed to by the study_data field, and the
2963    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2964    before. That is, this change is entirely upwards-compatible and requires no
2965    change to existing code.
2967    If you want to pass in additional data to pcre_exec(), you can either place it
2968    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2969    block.
2971    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2972    data string contains the escape sequence \M, pcretest calls pcre_exec() several
2973    times with different match limits, until it finds the minimum value needed for
2974    pcre_exec() to complete. The value is then output. This can be instructive; for
2975    most simple matches the number is quite small, but for pathological cases it
2976    gets very large very quickly.
2978    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2979    returns the size of the data block pointed to by the study_data field in a
2980    pcre_extra block, that is, the value that was passed as the argument to
2981    pcre_malloc() when PCRE was getting memory in which to place the information
2982    created by pcre_study(). The fourth argument should point to a size_t variable.
2983    pcretest has been extended so that this information is shown after a successful
2984    pcre_study() call when information about the compiled regex is being displayed.
2986    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2987    because what follows is always an absolute path. (Later: it turns out that this
2988    is more than cosmetic for MinGW, because it doesn't like empty path
2989    components.)
2991    49. Some changes have been made to the callout feature (see 28 above):
2993    (i)  A callout function now has three choices for what it returns:
2995           0  =>  success, carry on matching
2996         > 0  =>  failure at this point, but backtrack if possible
2997         < 0  =>  serious error, return this value from pcre_exec()
2999         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
3000         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
3001         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
3002         use by callout functions. It will never be used by PCRE itself.
3004    (ii) The pcre_extra structure (see 45 above) has a void * field called
3005         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
3006         pcre_callout_block structure has a field of the same name. The contents of
3007         the field passed in the pcre_extra structure are passed to the callout
3008         function in the corresponding field in the callout block. This makes it
3009         easier to use the same callout-containing regex from multiple threads. For
3010         testing, the pcretest program has a new data escape
3012           \C*n        pass the number n (may be negative) as callout_data
3014         If the callout function in pcretest receives a non-zero value as
3015         callout_data, it returns that value.
3017    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
3018    there were some redundant $(CFLAGS) in commands that are now specified as
3019    $(LINK), which already includes $(CFLAGS).
3021    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
3022    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
3023    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
3024    one-byte characters throughout. Note that case-insensitive matching applies
3025    only to characters whose values are less than 256. PCRE doesn't support the
3026    notion of cases for higher-valued characters.
3028    (i)   A character class whose characters are all within 0-255 is handled as
3029          a bit map, and the map is inverted for negative classes. Previously, a
3030          character > 255 always failed to match such a class; however it should
3031          match if the class was a negative one (e.g. [^ab]). This has been fixed.
3033    (ii)  A negated character class with a single character < 255 is coded as
3034          "not this character" (OP_NOT). This wasn't working properly when the test
3035          character was multibyte, either singly or repeated.
3037    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
3038          mode, for example: \x{100}{2,3}.
3040    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
3041          singly or repeated) now correctly test multibyte characters. However,
3042          PCRE doesn't recognize any characters with values greater than 255 as
3043          digits, spaces, or word characters. Such characters always match \D, \S,
3044          and \W, and never match \d, \s, or \w.
3046    (v)   Classes may now contain characters and character ranges with values
3047          greater than 255. For example: [ab\x{100}-\x{400}].
3049    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
3050          PCRE in UTF-8 mode.
3052    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
3053    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
3054    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
3055    value.)
3057    53. The single man page has become too large. I have therefore split it up into
3058    a number of separate man pages. These also give rise to individual HTML pages;
3059    these are now put in a separate directory, and there is an index.html page that
3060    lists them all. Some hyperlinking between the pages has been installed.
3062    54. Added convenience functions for handling named capturing parentheses.
3064    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
3065    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
3066    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
3067    are faulted.
3069    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
3070    calling configure. These values are used when compiling the dftables.c program
3071    which is run to generate the source of the default character tables. They
3072    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
3073    you will need to set these values.
3075    57. Updated the building process for Windows DLL, as provided by Fred Cox.
3078    Version 3.9 02-Jan-02
3079    ---------------------
3081    1. A bit of extraneous text had somehow crept into the pcregrep documentation.
3083    2. If --disable-static was given, the building process failed when trying to
3084    build pcretest and pcregrep. (For some reason it was using libtool to compile
3085    them, which is not right, as they aren't part of the library.)
3088    Version 3.8 18-Dec-01
3089    ---------------------
3091    1. The experimental UTF-8 code was completely screwed up. It was packing the
3092    bytes in the wrong order. How dumb can you get?
3095    Version 3.7 29-Oct-01
3096    ---------------------
3098    1. In updating pcretest to check change 1 of version 3.6, I screwed up.
3099    This caused pcretest, when used on the test data, to segfault. Unfortunately,
3100    this didn't happen under Solaris 8, where I normally test things.
3102    2. The Makefile had to be changed to make it work on BSD systems, where 'make'
3103    doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
3104    isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
3105    this fix an hour or so after the initial 3.7 release.)
3108    Version 3.6 23-Oct-01
3109    ---------------------
3111    1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
3112    offsets passed as NULL with zero offset count.
3114    2. The config.guess and config.sub files had not been updated when I moved to
3115    the latest autoconf.
3118  Version 3.5 15-Aug-01  Version 3.5 15-Aug-01
3119  ---------------------  ---------------------

Removed from v.53  
changed lines
  Added in v.497

  ViewVC Help
Powered by ViewVC 1.1.5