ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 61 by nigel, Sat Feb 24 21:39:58 2007 UTC revision 484 by ph10, Mon Jan 4 17:11:54 2010 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
4  Version 3.0 02-Jan-02  Version 8.01 11-Dec-09
5    ----------------------
7    1.  If a pattern contained a conditional subpattern with only one branch (in
8        particular, this includes all (DEFINE) patterns), a call to pcre_study()
9        computed the wrong minimum data length (which is of course zero for such
10        subpatterns).
12    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
13        the pattern is reset in the first branch, pcre_compile() failed with
14        "internal error: code overflow at offset...". This happened only when
15        the reset was to the original external option setting. (An optimization
16        abstracts leading options settings into an external setting, which was the
17        cause of this.)
19    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
20        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
21        assertion pattern did not match (meaning that the assertion was true), it
22        was incorrectly treated as false if the SKIP had been reached during the
23        matching. This also applied to assertions used as conditions.
25    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
26        assertion subpattern, including such a pattern used as a condition,
27        unpredictable results occurred, instead of the error return
30    5.  The C++ GlobalReplace function was not working like Perl for the special
31        situation when an empty string is matched. It now does the fancy magic
32        stuff that is necessary.
34    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
35        removed. (These were left over from very, very early versions of PCRE.)
37    7.  Some cosmetic changes to the code to make life easier when compiling it
38        as part of something else:
40        (a) Change DEBUG to PCRE_DEBUG.
42        (b) In pcre_compile(), rename the member of the "branch_chain" structure
43            called "current" as "current_branch", to prevent a collision with the
44            Linux macro when compiled as a kernel module.
46        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
47            prevent a collision with the Linux macro when compiled as a kernel
48            module.
50    8.  In pcre_compile() there are some checks for integer overflows that used to
51        cast potentially large values to (double). This has been changed to that
52        when building, a check for int64_t is made, and if it is found, it is used
53        instead, thus avoiding the use of floating point arithmetic. (There is no
54        other use of FP in PCRE.) If int64_t is not found, the fallback is to
55        double.
57    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
58        2005 (difference between two addresses compared to an unsigned value).
60    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
61        custom one, because of the following reported problem in Windows:
63          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
64            under Win32.
65          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
66            therefore missing the function definition.
67          - The compiler thus generates a "C" signature for the test function.
68          - The linker fails to find the "C" function.
69          - PCRE fails to configure if asked to do so against libbz2.
71    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
72        messages were output:
74          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
75          rerunning libtoolize, to keep the correct libtool macros in-tree.
76          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
78        I have done both of these things.
80    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
81        most of the time, it *can* run out if it is given a pattern that contains a
82        runaway infinite recursion. I updated the discussion in the pcrestack man
83        page.
85    13. Now that we have gone to the x.xx style of version numbers, the minor
86        version may start with zero. Using 08 or 09 is a bad idea because users
87        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
88        interpreted as invalid octal numbers. I've updated the previous comment in
89        configure.ac, and also added a check that gives an error if 08 or 09 are
90        used.
92    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
93        causing partial matching to fail when the end of the subject matched \W
94        in a UTF-8 pattern where \W was quantified with a minimum of 3.
96    15. There were some discrepancies between the declarations in pcre_internal.h
97        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
98        their definitions. The declarations used "const uschar *" and the
99        definitions used USPTR. Even though USPTR is normally defined as "const
100        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
101        reported that: "This difference in casting confuses some C++ compilers, for
102        example, SunCC recognizes above declarations as different functions and
103        generates broken code for hbpcre." I have changed the declarations to use
104        USPTR.
106    16. GNU libtool is named differently on some systems. The autogen.sh script now
107        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
108        (FreeBSD).
112    Version 8.00 19-Oct-09
113    ----------------------
115    1.  The table for translating pcre_compile() error codes into POSIX error codes
116        was out-of-date, and there was no check on the pcre_compile() error code
117        being within the table. This could lead to an OK return being given in
118        error.
120    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
121        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
122        in a Windows environment.
124    3.  The pcregrep --count option prints the count for each file even when it is
125        zero, as does GNU grep. However, pcregrep was also printing all files when
126        --files-with-matches was added. Now, when both options are given, it prints
127        counts only for those files that have at least one match. (GNU grep just
128        prints the file name in this circumstance, but including the count seems
129        more useful - otherwise, why use --count?) Also ensured that the
130        combination -clh just lists non-zero counts, with no names.
132    4.  The long form of the pcregrep -F option was incorrectly implemented as
133        --fixed_strings instead of --fixed-strings. This is an incompatible change,
134        but it seems right to fix it, and I didn't think it was worth preserving
135        the old behaviour.
137    5.  The command line items --regex=pattern and --regexp=pattern were not
138        recognized by pcregrep, which required --regex pattern or --regexp pattern
139        (with a space rather than an '='). The man page documented the '=' forms,
140        which are compatible with GNU grep; these now work.
142    6.  No libpcreposix.pc file was created for pkg-config; there was just
143        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
145    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
146        when UCP support is not needed, by modifying the Python script that
147        generates it from Unicode data files. This should not matter if the module
148        is correctly used as a library, but I received one complaint about 50K of
149        unwanted data. My guess is that the person linked everything into his
150        program rather than using a library. Anyway, it does no harm.
152    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
153        was a minimum greater than 1 for a wide character in a possessive
154        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
155        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
156        character. Chaos in the form of incorrect output or a compiling loop could
157        result.
159    9.  The restrictions on what a pattern can contain when partial matching is
160        requested for pcre_exec() have been removed. All patterns can now be
161        partially matched by this function. In addition, if there are at least two
162        slots in the offset vector, the offset of the earliest inspected character
163        for the match and the offset of the end of the subject are set in them when
164        PCRE_ERROR_PARTIAL is returned.
166    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
167        synonymous with PCRE_PARTIAL, for backwards compatibility, and
168        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
169        and may be more useful for multi-segment matching.
171    11. Partial matching with pcre_exec() is now more intuitive. A partial match
172        used to be given if ever the end of the subject was reached; now it is
173        given only if matching could not proceed because another character was
174        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
175        string "Z", which now yields "no match" instead of "partial match". In the
176        case of pcre_dfa_exec(), "no match" is given if every matching path for the
177        final character ended with (*FAIL).
179    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
180        if the pattern had a "must contain" character that was already found in the
181        earlier partial match, unless partial matching was again requested. For
182        example, with the pattern /dog.(body)?/, the "must contain" character is
183        "g". If the first part-match was for the string "dog", restarting with
184        "sbody" failed. This bug has been fixed.
186    13. The string returned by pcre_dfa_exec() after a partial match has been
187        changed so that it starts at the first inspected character rather than the
188        first character of the match. This makes a difference only if the pattern
189        starts with a lookbehind assertion or \b or \B (\K is not supported by
190        pcre_dfa_exec()). It's an incompatible change, but it makes the two
191        matching functions compatible, and I think it's the right thing to do.
193    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
194        so that the demonstration program is easily available in environments where
195        PCRE has not been installed from source.
197    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
198        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
199        library.
201    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
202        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
203        is not the first non-POSIX option to be added. Clearly some people find
204        these options useful.
206    17. If a caller to the POSIX matching function regexec() passes a non-zero
207        value for nmatch with a NULL value for pmatch, the value of
208        nmatch is forced to zero.
210    18. RunGrepTest did not have a test for the availability of the -u option of
211        the diff command, as RunTest does. It now checks in the same way as
212        RunTest, and also checks for the -b option.
214    19. If an odd number of negated classes containing just a single character
215        interposed, within parentheses, between a forward reference to a named
216        subpattern and the definition of the subpattern, compilation crashed with
217        an internal error, complaining that it could not find the referenced
218        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
219        [The bug was that it was starting one character too far in when skipping
220        over the character class, thus treating the ] as data rather than
221        terminating the class. This meant it could skip too much.]
223    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
224        /g option in pcretest when the pattern contains \K, which makes it possible
225        to have an empty string match not at the start, even when the pattern is
226        anchored. Updated pcretest and pcredemo to use this option.
228    21. If the maximum number of capturing subpatterns in a recursion was greater
229        than the maximum at the outer level, the higher number was returned, but
230        with unset values at the outer level. The correct (outer level) value is
231        now given.
233    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
234        PCRE did not set those parentheses (unlike Perl). I have now found a way to
235        make it do so. The string so far is captured, making this feature
236        compatible with Perl.
238    23. The tests have been re-organized, adding tests 11 and 12, to make it
239        possible to check the Perl 5.10 features against Perl 5.10.
241    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
242        pattern matches a fixed length string. PCRE did not allow this; now it
243        does. Neither allows recursion.
245    25. I finally figured out how to implement a request to provide the minimum
246        length of subject string that was needed in order to match a given pattern.
247        (It was back references and recursion that I had previously got hung up
248        on.) This code has now been added to pcre_study(); it finds a lower bound
249        to the length of subject needed. It is not necessarily the greatest lower
250        bound, but using it to avoid searching strings that are too short does give
251        some useful speed-ups. The value is available to calling programs via
252        pcre_fullinfo().
254    26. While implementing 25, I discovered to my embarrassment that pcretest had
255        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
256        study optimizations had never been tested with that matching function.
257        Oops. What is worse, even when it was passed study data, there was a bug in
258        pcre_dfa_exec() that meant it never actually used it. Double oops. There
259        were also very few tests of studied patterns with pcre_dfa_exec().
261    27. If (?| is used to create subpatterns with duplicate numbers, they are now
262        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
263        on the other side of the coin, they are no longer allowed to have different
264        names, because these cannot be distinguished in PCRE, and this has caused
265        confusion. (This is a difference from Perl.)
267    28. When duplicate subpattern names are present (necessarily with different
268        numbers, as required by 27 above), and a test is made by name in a
269        conditional pattern, either for a subpattern having been matched, or for
270        recursion in such a pattern, all the associated numbered subpatterns are
271        tested, and the overall condition is true if the condition is true for any
272        one of them. This is the way Perl works, and is also more like the way
273        testing by number works.
276    Version 7.9 11-Apr-09
277    ---------------------
279    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
280        (pcretest), all targets were linked against these libraries. This included
281        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
282        libraries. This caused unwanted dependencies to be created. This problem
283        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
284        pcretest is linked with readline.
286    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
287        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
288        moved outside it again, because FALSE and TRUE are already defined in AIX,
289        but BOOL is not.
291    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
292        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
294    4.  The pcregrep documentation said spaces were inserted as well as colons (or
295        hyphens) following file names and line numbers when outputting matching
296        lines. This is not true; no spaces are inserted. I have also clarified the
297        wording for the --colour (or --color) option.
299    5.  In pcregrep, when --colour was used with -o, the list of matching strings
300        was not coloured; this is different to GNU grep, so I have changed it to be
301        the same.
303    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
304        each matching line was coloured. Now it goes on to look for further matches
305        of any of the test patterns, which is the same behaviour as GNU grep.
307    7.  A pattern that could match an empty string could cause pcregrep to loop; it
308        doesn't make sense to accept an empty string match in pcregrep, so I have
309        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
310        seems to be how GNU grep behaves.
312    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
313        start or after a newline", because the conditional assertion was not being
314        correctly handled. The rule now is that both the assertion and what follows
315        in the first alternative must satisfy the test.
317    9.  If auto-callout was enabled in a pattern with a conditional group whose
318        condition was an assertion, PCRE could crash during matching, both with
319        pcre_exec() and pcre_dfa_exec().
321    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
322        used for matching.
324    11. Unicode property support in character classes was not working for
325        characters (bytes) greater than 127 when not in UTF-8 mode.
327    12. Added the -M command line option to pcretest.
329    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
331    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
333    16. Added comments and documentation about mis-use of no_arg in the C++
334        wrapper.
336    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
337        from Martin Jerabek that uses macro names for all relevant character and
338        string constants.
340    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
341        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
342        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
343        these, but not everybody uses configure.
345    19. A conditional group that had only one branch was not being correctly
346        recognized as an item that could match an empty string. This meant that an
347        enclosing group might also not be so recognized, causing infinite looping
348        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
349        with the subject "ab", where knowledge that the repeated group can match
350        nothing is needed in order to break the loop.
352    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
353        exec(), but without supplying a callout function, matching went wrong.
355    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
356        leak if the size of the offset vector was greater than 30. When the vector
357        is smaller, the saved offsets during recursion go onto a local stack
358        vector, but for larger vectors malloc() is used. It was failing to free
359        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
360        error, in fact).
362    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
363        heapframe that is used only when UTF-8 support is enabled. This caused no
364        problem, but was untidy.
366    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
367        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
368        included within another project.
370    24. Steven Van Ingelgem's patches to add more options to the CMake support,
371        slightly modified by me:
373          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
374              not building pcregrep.
376          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
377              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
379    25. Forward references, both numeric and by name, in patterns that made use of
380        duplicate group numbers, could behave incorrectly or give incorrect errors,
381        because when scanning forward to find the reference group, PCRE was not
382        taking into account the duplicate group numbers. A pattern such as
383        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
385    26. Changed a few more instances of "const unsigned char *" to USPTR, making
386        the feature of a custom pointer more persuasive (as requested by a user).
388    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
389        pcretest.c, inside #ifndefs, because it seems they are sometimes already
390        pre-defined.
392    28. Added support for (*UTF8) at the start of a pattern.
394    29. Arrange for flags added by the "release type" setting in CMake to be shown
395        in the configuration summary.
398    Version 7.8 05-Sep-08
399    ---------------------
401    1.  Replaced UCP searching code with optimized version as implemented for Ad
402        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
403        stage table and inline lookup instead of a function, giving speed ups of 2
404        to 5 times on some simple patterns that I tested. Permission was given to
405        distribute the MultiStage2.py script that generates the tables (it's not in
406        the tarball, but is in the Subversion repository).
408    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
409        scripts.
411    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
412        a group with a zero qualifier. The result of the study could be incorrect,
413        or the function might crash, depending on the pattern.
415    4.  Caseless matching was not working for non-ASCII characters in back
416        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
417        It now works when Unicode Property Support is available.
419    5.  In pcretest, an escape such as \x{de} in the data was always generating
420        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
421        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
422        truncation.
424    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
426    7.  Added two (int) casts to pcregrep when printing the difference of two
427        pointers, in case they are 64-bit values.
429    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
430        test 2 if it fails.
432    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
433        and a #define of that name to empty if it is not externally set. This is to
434        allow users of MSVC to set it if necessary.
436    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
437        the convenience functions in the pcre_get.c source file.
439    11. An option change at the start of a pattern that had top-level alternatives
440        could cause overwriting and/or a crash. This command provoked a crash in
441        some environments:
443          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
445        This potential security problem was recorded as CVE-2008-2371.
447    12. For a pattern where the match had to start at the beginning or immediately
448        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
449        pcre_dfa_exec() could read past the end of the passed subject if there was
450        no match. To help with detecting such bugs (e.g. with valgrind), I modified
451        pcretest so that it places the subject at the end of its malloc-ed buffer.
453    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
454        exec() might read past the end of the data buffer in UTF-8 mode.
456    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
457        the data contained the byte 0x85 as part of a UTF-8 character within its
458        first line. This applied both to normal and DFA matching.
460    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
461        /^[^d]*?$/8 failed to match "abc".
463    16. Added a missing copyright notice to pcrecpp_internal.h.
465    17. Make it more clear in the documentation that values returned from
466        pcre_exec() in ovector are byte offsets, not character counts.
468    18. Tidied a few places to stop certain compilers from issuing warnings.
470    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
471        supplied by Stefan Weber. I made a further small update for 7.8 because
472        there is a change of source arrangements: the pcre_searchfuncs.c module is
473        replaced by pcre_ucd.c.
476    Version 7.7 07-May-08
477    ---------------------
479    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
480        a string to a long long, pretend we don't even have a long long." This is
481        done by checking for the strtoq, strtoll, and _strtoi64 functions.
483    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
484        pre-7.6 versions, which defined a global no_arg variable instead of putting
485        it in the RE class. (See also #8 below.)
487    3.  Remove a line of dead code, identified by coverity and reported by Nuno
488        Lopes.
490    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
492        (1) The include/exclude patterns were being applied to the whole pathnames
493            of files, instead of just to the final components.
495        (2) If there was more than one level of directory, the subdirectories were
496            skipped unless they satisfied the include/exclude conditions. This is
497            inconsistent with GNU grep (and could even be seen as contrary to the
498            pcregrep specification - which I improved to make it absolutely clear).
499            The action now is always to scan all levels of directory, and just
500            apply the include/exclude patterns to regular files.
502    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
503        --exclude_dir in the tests to avoid scanning .svn directories.
505    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
506        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
507        doesn't support NULs in patterns.
509    7.  Added some missing "const"s to declarations of static tables in
510        pcre_compile.c and pcre_dfa_exec.c.
512    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
513        caused by fix #2  above. (Subsequently also a second patch to fix the
514        first patch. And a third patch - this was a messy problem.)
516    9.  Applied Craig's patch to remove the use of push_back().
518    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
519        matching function regexec().
521    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
522        which, however, unlike Perl's \g{...}, are subroutine calls, not back
523        references. PCRE supports relative numbers with this syntax (I don't think
524        Oniguruma does).
526    12. Previously, a group with a zero repeat such as (...){0} was completely
527        omitted from the compiled regex. However, this means that if the group
528        was called as a subroutine from elsewhere in the pattern, things went wrong
529        (an internal error was given). Such groups are now left in the compiled
530        pattern, with a new opcode that causes them to be skipped at execution
531        time.
533    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
534        to the way PCRE behaves:
536        (a) A lone ] character is dis-allowed (Perl treats it as data).
538        (b) A back reference to an unmatched subpattern matches an empty string
539            (Perl fails the current match path).
541        (c) A data ] in a character class must be notated as \] because if the
542            first data character in a class is ], it defines an empty class. (In
543            Perl it is not possible to have an empty class.) The empty class []
544            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
545            The negative empty class [^] matches any one character, independently
546            of the DOTALL setting.
548    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
549        non-existent subpattern following a character class starting with ']' and
550        containing () gave an internal compiling error instead of "reference to
551        non-existent subpattern". Fortunately, when the pattern did exist, the
552        compiled code was correct. (When scanning forwards to check for the
553        existencd of the subpattern, it was treating the data ']' as terminating
554        the class, so got the count wrong. When actually compiling, the reference
555        was subsequently set up correctly.)
557    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
558        it was being rejected as not supported by pcre_dfa_exec(), even though
559        other assertions are supported. I have made pcre_dfa_exec() support
560        (*FAIL).
562    16. The implementation of 13c above involved the invention of a new opcode,
563        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
564        cannot be changed at match time, I realized I could make a small
565        improvement to matching performance by compiling OP_ALLANY instead of
566        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
567        on the OP_ANY path.
569    17. Compiling pcretest on Windows with readline support failed without the
570        following two fixes: (1) Make the unistd.h include conditional on
571        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
573    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
574        ncurses library to be included for pcretest when ReadLine support is
575        requested, but also to allow for it to be overridden. This patch came from
576        Daniel Bergström.
578    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
579        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
580        any errors with the current Unicode tables. Thanks to Peter Kankowski for
581        spotting this.
584    Version 7.6 28-Jan-08
585    ---------------------
587    1.  A character class containing a very large number of characters with
588        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
589        overflow.
591    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
592        HAVE_LONG_LONG is not defined.
594    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
595        bring them up to date and include new features. This patch includes:
597        - Fixed PH's badly added libz and libbz2 support.
598        - Fixed a problem with static linking.
599        - Added pcredemo. [But later removed - see 7 below.]
600        - Fixed dftables problem and added an option.
601        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
602            HAVE_LONG_LONG.
603        - Added readline support for pcretest.
604        - Added an listing of the option settings after cmake has run.
606    4.  A user submitted a patch to Makefile that makes it easy to create
607        "pcre.dll" under mingw when using Configure/Make. I added stuff to
608        Makefile.am that cause it to include this special target, without
609        affecting anything else. Note that the same mingw target plus all
610        the other distribution libraries and programs are now supported
611        when configuring with CMake (see 6 below) instead of with
612        Configure/Make.
614    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
615        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
616        exported in the Windows port". It has not yet been confirmed that the patch
617        solves the problem, but it does no harm.
619    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
620        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
621        with CMake, and also correct the comment about stack recursion.
623    7.  Remove the automatic building of pcredemo from the ./configure system and
624        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
625        of a program that users should build themselves after PCRE is installed, so
626        building it automatically is not really right. What is more, it gave
627        trouble in some build environments.
629    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
632    Version 7.5 10-Jan-08
633    ---------------------
635    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
636        values in parens when parsing an RE using the C++ wrapper."
638    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
639        Characters greater than 255 were excluded from the class instead of being
640        included.
642    3.  The same bug as (2) above applied to negated POSIX classes such as
643        [:^space:].
645    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
646        defined or documented. It seems to have been a typo for PCRE_STATIC, so
647        I have changed it.
649    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
650        first named subpattern) and a construct such as (?&a) would reference the
651        first named subpattern whose name started with "a" (in other words, the
652        length check was missing). Both these problems are fixed. "Subpattern name
653        expected" is now given for (?&) (a zero-length name), and this patch also
654        makes it give the same error for \k'' (previously it complained that that
655        was a reference to a non-existent subpattern).
657    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
658        this is right because (?- can be followed by option settings as well as by
659        digits. I have, however, made the messages clearer.
661    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
662        than the number used in the conditional) now cause a compile-time error.
663        This is actually not compatible with Perl, which accepts such patterns, but
664        treats the conditional as always being FALSE (as PCRE used to), but it
665        seems to me that giving a diagnostic is better.
667    8.  Change "alphameric" to the more common word "alphanumeric" in comments
668        and messages.
670    9.  Fix two occurrences of "backslash" in comments that should have been
671        "backspace".
673    10. Remove two redundant lines of code that can never be obeyed (their function
674        was moved elsewhere).
676    11. The program that makes PCRE's Unicode character property table had a bug
677        which caused it to generate incorrect table entries for sequences of
678        characters that have the same character type, but are in different scripts.
679        It amalgamated them into a single range, with the script of the first of
680        them. In other words, some characters were in the wrong script. There were
681        thirteen such cases, affecting characters in the following ranges:
683          U+002b0 - U+002c1
684          U+0060c - U+0060d
685          U+0061e - U+00612
686          U+0064b - U+0065e
687          U+0074d - U+0076d
688          U+01800 - U+01805
689          U+01d00 - U+01d77
690          U+01d9b - U+01dbf
691          U+0200b - U+0200f
692          U+030fc - U+030fe
693          U+03260 - U+0327f
694          U+0fb46 - U+0fbb1
695          U+10450 - U+1049d
697    12. The -o option (show only the matching part of a line) for pcregrep was not
698        compatible with GNU grep in that, if there was more than one match in a
699        line, it showed only the first of them. It now behaves in the same way as
700        GNU grep.
702    13. If the -o and -v options were combined for pcregrep, it printed a blank
703        line for every non-matching line. GNU grep prints nothing, and pcregrep now
704        does the same. The return code can be used to tell if there were any
705        non-matching lines.
707    14. Added --file-offsets and --line-offsets to pcregrep.
709    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
710        infinitely looping recursion. The bug was that positive lookaheads were not
711        being skipped when checking for a possible empty match (negative lookaheads
712        and both kinds of lookbehind were skipped).
714    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
715        inclusion of <windows.h> to before rather than after the definition of
716        INVALID_FILE_ATTRIBUTES (patch from David Byron).
718    17. Specifying a possessive quantifier with a specific limit for a Unicode
719        character property caused pcre_compile() to compile bad code, which led at
720        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
721        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
722        caused the error; without that there was no problem.
724    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
726    19. Added --enable-pcretest-libreadline.
728    20. In pcrecpp.cc, the variable 'count' was incremented twice in
729        RE::GlobalReplace(). As a result, the number of replacements returned was
730        double what it should be. I removed one of the increments, but Craig sent a
731        later patch that removed the other one (the right fix) and added unit tests
732        that check the return values (which was not done before).
734    21. Several CMake things:
736        (1) Arranged that, when cmake is used on Unix, the libraries end up with
737            the names libpcre and libpcreposix, not just pcre and pcreposix.
739        (2) The above change means that pcretest and pcregrep are now correctly
740            linked with the newly-built libraries, not previously installed ones.
744    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
745        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
746        UTF-8 newline character). The key issue is that the pattern starts .*;
747        this means that the match must be either at the beginning, or after a
748        newline. The bug was in the code for advancing after a failed match and
749        checking that the new position followed a newline. It was not taking
750        account of UTF-8 characters correctly.
752    23. PCRE was behaving differently from Perl in the way it recognized POSIX
753        character classes. PCRE was not treating the sequence [:...:] as a
754        character class unless the ... were all letters. Perl, however, seems to
755        allow any characters between [: and :], though of course it rejects as
756        unknown any "names" that contain non-letters, because all the known class
757        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
758        for example, whereas PCRE did not - it did not recognize a POSIX character
759        class. This seemed a bit dangerous, so the code has been changed to be
760        closer to Perl. The behaviour is not identical to Perl, because PCRE will
761        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
762        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
763        Perl does, and where it didn't before.
765    24. Rewrite so as to remove the single use of %n from pcregrep because in some
766        Windows environments %n is disabled by default.
769    Version 7.4 21-Sep-07
770    ---------------------
772    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
773        means that a class such as [\s] counted as "explicit reference to CR or
774        LF". That isn't really right - the whole point of the change was to try to
775        help when there was an actual mention of one of the two characters. So now
776        the change happens only if \r or \n (or a literal CR or LF) character is
777        encountered.
779    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
780        of both had grown to the point where there were only 3 bits left.
781        Fortunately, there was spare space in the data structure, and so I have
782        moved the internal flags into a new 16-bit field to free up more option
783        bits.
785    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
786        but did not set the internal JCHANGED flag - either of these is enough to
787        control the way the "get" function works - but the PCRE_INFO_JCHANGED
788        facility is supposed to tell if (?J) was ever used, so now (?J) at the
789        start sets both bits.
791    4.  Added options (at build time, compile time, exec time) to change \R from
792        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
794    5.  doc/pcresyntax.html was missing from the distribution.
796    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
797        compatibility, even though it is no longer used.
799    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
800        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
801        windows.h file is present (where different names are used). [This was
802        reversed later after testing - see 16 below.]
804    8.  Changed all #include <config.h> to #include "config.h". There were also
805        some further <pcre.h> cases that I changed to "pcre.h".
807    9.  When pcregrep was used with the --colour option, it missed the line ending
808        sequence off the lines that it output.
810    10. It was pointed out to me that arrays of string pointers cause lots of
811        relocations when a shared library is dynamically loaded. A technique of
812        using a single long string with a table of offsets can drastically reduce
813        these. I have refactored PCRE in four places to do this. The result is
814        dramatic:
816          Originally:                          290
817          After changing UCP table:            187
818          After changing error message table:   43
819          After changing table of "verbs"       36
820          After changing table of Posix names   22
822        Thanks to the folks working on Gregex for glib for this insight.
824    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
825        unicode-properties was also set.
827    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
829    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
830        checked only for CRLF.
832    14. Added casts to pcretest.c to avoid compiler warnings.
834    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
836    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
837        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
838        entirely. This removes changes made in 7 above.
840    17. The CMake files have been updated, and there is now more information about
841        building with CMake in the NON-UNIX-USE document.
844    Version 7.3 28-Aug-07
845    ---------------------
847     1. In the rejigging of the build system that eventually resulted in 7.1, the
848        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
849        brackets there is not right, since it causes compilers to look for an
850        installed pcre.h, not the version that is in the source that is being
851        compiled (which of course may be different). I have changed it back to:
853          #include "pcre.h"
855        I have a vague recollection that the change was concerned with compiling in
856        different directories, but in the new build system, that is taken care of
857        by the VPATH setting the Makefile.
859     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
860        when the subject happened to end in the byte 0x85 (e.g. if the last
861        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
862        characters but of course it shouldn't be taken as a newline when it is part
863        of another character. The bug was that, for an unlimited repeat of . in
864        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
865        characters when looking for a newline.
867     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
869     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
870        in debug output.
872     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
873        long printing in the pcrecpp unittest when running under MinGW.
875     6. ESC_K was left out of the EBCDIC table.
877     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
878        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
879        limit also applies to "virtual nesting" when a pattern is recursive, and in
880        this case 1000 isn't so big. I have been able to remove this limit at the
881        expense of backing off one optimization in certain circumstances. Normally,
882        when pcre_exec() would call its internal match() function recursively and
883        immediately return the result unconditionally, it uses a "tail recursion"
884        feature to save stack. However, when a subpattern that can match an empty
885        string has an unlimited repetition quantifier, it no longer makes this
886        optimization. That gives it a stack frame in which to save the data for
887        checking that an empty string has been matched. Previously this was taken
888        from the 1000-entry workspace that had been reserved. So now there is no
889        explicit limit, but more stack is used.
891     8. Applied Daniel's patches to solve problems with the import/export magic
892        syntax that is required for Windows, and which was going wrong for the
893        pcreposix and pcrecpp parts of the library. These were overlooked when this
894        problem was solved for the main library.
896     9. There were some crude static tests to avoid integer overflow when computing
897        the size of patterns that contain repeated groups with explicit upper
898        limits. As the maximum quantifier is 65535, the maximum group length was
899        set at 30,000 so that the product of these two numbers did not overflow a
900        32-bit integer. However, it turns out that people want to use groups that
901        are longer than 30,000 bytes (though not repeat them that many times).
902        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
903        made it possible to implement the integer overflow checks in a much more
904        dynamic way, which I have now done. The artificial limitation on group
905        length has been removed - we now have only the limit on the total length of
906        the compiled pattern, which depends on the LINK_SIZE setting.
908    10. Fixed a bug in the documentation for get/copy named substring when
909        duplicate names are permitted. If none of the named substrings are set, the
910        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
911        empty string.
913    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
914        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
915        because the ] is interpreted as the first data character and the
916        terminating ] is not found. PCRE has been made compatible with Perl in this
917        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
918        cause memory overwriting.
920    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
921        string has been matched (to stop an infinite loop). It was not recognizing
922        a conditional subpattern that could match an empty string if that
923        subpattern was within another subpattern. For example, it looped when
924        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
925        condition was not nested. This bug has been fixed.
927    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
928        past the start of the subject in the presence of bytes with the top bit
929        set, for example "\x8aBCD".
931    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
932        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
934    14. Optimized (?!) to (*FAIL).
936    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
937        This restricts code points to be within the range 0 to 0x10FFFF, excluding
938        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
939        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
940        does: it's just the validity check that is more restrictive.
942    16. Inserted checks for integer overflows during escape sequence (backslash)
943        processing, and also fixed erroneous offset values for syntax errors during
944        backslash processing.
946    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
947        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
949    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
950        caused an overrun.
952    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
953        something other than just ASCII characters) inside a group that had an
954        unlimited repeat caused a loop at compile time (while checking to see
955        whether the group could match an empty string).
957    20. Debugging a pattern containing \p or \P could cause a crash. For example,
958        [\P{Any}] did so. (Error in the code for printing property names.)
960    21. An orphan \E inside a character class could cause a crash.
962    22. A repeated capturing bracket such as (A)? could cause a wild memory
963        reference during compilation.
965    23. There are several functions in pcre_compile() that scan along a compiled
966        expression for various reasons (e.g. to see if it's fixed length for look
967        behind). There were bugs in these functions when a repeated \p or \P was
968        present in the pattern. These operators have additional parameters compared
969        with \d, etc, and these were not being taken into account when moving along
970        the compiled data. Specifically:
972        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
973            length.
975        (b) An item such as \pL+ within a repeated group could cause crashes or
976            loops.
978        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
979            "reference to non-existent subpattern" error.
981        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
983    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
984        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
986    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
988    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
989        character were causing crashes (broken optimization).
991    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
992        \p or \P) caused a compile-time loop.
994    28. More problems have arisen in unanchored patterns when CRLF is a valid line
995        break. For example, the unstudied pattern [\r\n]A does not match the string
996        "\r\nA" because change 7.0/46 below moves the current point on by two
997        characters after failing to match at the start. However, the pattern \nA
998        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
999        the same is true. There doesn't seem any very clean way out of this, but
1000        what I have chosen to do makes the common cases work: PCRE now takes note
1001        of whether there can be an explicit match for \r or \n anywhere in the
1002        pattern, and if so, 7.0/46 no longer applies. As part of this change,
1003        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1004        pattern has explicit CR or LF references.
1006    29. Added (*CR) etc for changing newline setting at start of pattern.
1009    Version 7.2 19-Jun-07
1010    ---------------------
1012     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
1013        which is apparently normally available under Windows.
1015     2. Re-jig the pcregrep tests with different newline settings in an attempt
1016        to make them independent of the local environment's newline setting.
1018     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
1020     4. Some of the "internals" tests were previously cut out when the link size
1021        was not 2, because the output contained actual offsets. The recent new
1022        "Z" feature of pcretest means that these can be cut out, making the tests
1023        usable with all link sizes.
1025     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
1026        stack recursion. This gives a massive performance boost under BSD, but just
1027        a small improvement under Linux. However, it saves one field in the frame
1028        in all cases.
1030     6. Added more features from the forthcoming Perl 5.10:
1032        (a) (?-n) (where n is a string of digits) is a relative subroutine or
1033            recursion call. It refers to the nth most recently opened parentheses.
1035        (b) (?+n) is also a relative subroutine call; it refers to the nth next
1036            to be opened parentheses.
1038        (c) Conditions that refer to capturing parentheses can be specified
1039            relatively, for example, (?(-2)... or (?(+3)...
1041        (d) \K resets the start of the current match so that everything before
1042            is not part of it.
1044        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1046        (f) \g{name} is another synonym - part of Perl 5.10's unification of
1047            reference syntax.
1049        (g) (?| introduces a group in which the numbering of parentheses in each
1050            alternative starts with the same number.
1052        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
1054     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
1057     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
1058        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
1059        for detecting groups that can match an empty string.
1061     9. A pattern with a very large number of alternatives (more than several
1062        hundred) was running out of internal workspace during the pre-compile
1063        phase, where pcre_compile() figures out how much memory will be needed. A
1064        bit of new cunning has reduced the workspace needed for groups with
1065        alternatives. The 1000-alternative test pattern now uses 12 bytes of
1066        workspace instead of running out of the 4096 that are available.
1068    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
1070    11. Applied patch from Google to remove an optimization that didn't quite work.
1071        The report of the bug said:
1073          pcrecpp::RE("a*").FullMatch("aaa") matches, while
1074          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
1075          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
1077    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
1078        it matched the wrong number of bytes.
1081    Version 7.1 24-Apr-07
1082    ---------------------
1084     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
1085        that is more "standard", making use of automake and other Autotools. There
1086        is some re-arrangement of the files and adjustment of comments consequent
1087        on this.
1089     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
1090        for recursive directory scanning broke on some systems because the files
1091        are not scanned in any specific order and on different systems the order
1092        was different. A call to "sort" has been inserted into RunGrepTest for the
1093        approprate test as a short-term fix. In the longer term there may be an
1094        alternative.
1096     3. I had an email from Eric Raymond about problems translating some of PCRE's
1097        man pages to HTML (despite the fact that I distribute HTML pages, some
1098        people do their own conversions for various reasons). The problems
1099        concerned the use of low-level troff macros .br and .in. I have therefore
1100        removed all such uses from the man pages (some were redundant, some could
1101        be replaced by .nf/.fi pairs). The 132html script that I use to generate
1102        HTML has been updated to handle .nf/.fi and to complain if it encounters
1103        .br or .in.
1105     4. Updated comments in configure.ac that get placed in config.h.in and also
1106        arranged for config.h to be included in the distribution, with the name
1107        config.h.generic, for the benefit of those who have to compile without
1108        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
1110     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
1111        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
1112        makevp.bat for the current PCRE, using the additional files
1113        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
1115     6. A Windows user reported a minor discrepancy with test 2, which turned out
1116        to be caused by a trailing space on an input line that had got lost in his
1117        copy. The trailing space was an accident, so I've just removed it.
1119     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
1120        that is needed.
1122     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
1123        as "const" (a) because they are and (b) because it helps the PHP
1124        maintainers who have recently made a script to detect big data structures
1125        in the php code that should be moved to the .rodata section. I remembered
1126        to update Builducptable as well, so it won't revert if ucptable.h is ever
1127        re-created.
1129     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
1130        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
1131        order to be able to cut out the UTF-8 tables in the latter when UTF-8
1132        support is not required. This saves 1.5-2K of code, which is important in
1133        some applications.
1135        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
1136        so as not to refer to the tables, even though these functions will never be
1137        called when UTF-8 support is disabled. Otherwise there are problems with a
1138        shared library.
1140    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
1142        (a) It was defining its arguments as char * instead of void *.
1144        (b) It was assuming that all moves were upwards in memory; this was true
1145            a long time ago when I wrote it, but is no longer the case.
1147        The emulated memove() is provided for those environments that have neither
1148        memmove() nor bcopy(). I didn't think anyone used it these days, but that
1149        is clearly not the case, as these two bugs were recently reported.
1151    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
1152        and Detrail to create the HTML documentation, the .txt form of the man
1153        pages, and it removes trailing spaces from listed files. It also creates
1154        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
1155        case, it wraps all the #defines with #ifndefs. This script should be run
1156        before "make dist".
1158    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1159        with Unicode property support.
1161        (a) For a maximizing quantifier, if the two different cases of the
1162            character were of different lengths in their UTF-8 codings (there are
1163            some cases like this - I found 11), and the matching function had to
1164            back up over a mixture of the two cases, it incorrectly assumed they
1165            were both the same length.
1167        (b) When PCRE was configured to use the heap rather than the stack for
1168            recursion during matching, it was not correctly preserving the data for
1169            the other case of a UTF-8 character when checking ahead for a match
1170            while processing a minimizing repeat. If the check also involved
1171            matching a wide character, but failed, corruption could cause an
1172            erroneous result when trying to check for a repeat of the original
1173            character.
1175    13. Some tidying changes to the testing mechanism:
1177        (a) The RunTest script now detects the internal link size and whether there
1178            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1179            values substituted by "configure". (The RunGrepTest script already did
1180            this for UTF-8.) The configure.ac script no longer substitutes the
1181            relevant variables.
1183        (b) The debugging options /B and /D in pcretest show the compiled bytecode
1184            with length and offset values. This means that the output is different
1185            for different internal link sizes. Test 2 is skipped for link sizes
1186            other than 2 because of this, bypassing the problem. Unfortunately,
1187            there was also a test in test 3 (the locale tests) that used /B and
1188            failed for link sizes other than 2. Rather than cut the whole test out,
1189            I have added a new /Z option to pcretest that replaces the length and
1190            offset values with spaces. This is now used to make test 3 independent
1191            of link size. (Test 2 will be tidied up later.)
1193    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1194        segmentation fault instead of returning the appropriate error message.
1196    15. In multiline mode when the newline sequence was set to "any", the pattern
1197        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1198        This doesn't seem right; it now treats the CRLF combination as the line
1199        ending, and so does not match in that case. It's only a pattern such as ^$
1200        that would hit this one: something like ^ABC$ would have failed after \r
1201        and then tried again after \r\n.
1203    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1204        in an attempt to make files that differ only in their line terminators
1205        compare equal. This works on Linux.
1207    17. Under certain error circumstances pcregrep might try to free random memory
1208        as it exited. This is now fixed, thanks to valgrind.
1210    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1211        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1212        was because its rules for how to advance for /g after matching an empty
1213        string at the end of a line did not allow for this case. They now check for
1214        it specially.
1216    20. pcretest is supposed to handle patterns and data of any length, by
1217        extending its buffers when necessary. It was getting this wrong when the
1218        buffer for a data line had to be extended.
1220    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1221        CRLF as a newline sequence.
1223    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1224        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1225        I have nevertheless tidied it up.
1227    23. Added some casts to kill warnings from HP-UX ia64 compiler.
1229    24. Added a man page for pcre-config.
1232    Version 7.0 19-Dec-06
1233    ---------------------
1235     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1236        moving to gcc 4.1.1.
1238     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1239        sys/time.h, which is documented as needed for this function. It doesn't
1240        seem to matter on Linux, but it showed up on some releases of OS X.
1242     3. It seems that there are systems where bytes whose values are greater than
1243        127 match isprint() in the "C" locale. The "C" locale should be the
1244        default when a C program starts up. In most systems, only ASCII printing
1245        characters match isprint(). This difference caused the output from pcretest
1246        to vary, making some of the tests fail. I have changed pcretest so that:
1248        (a) When it is outputting text in the compiled version of a pattern, bytes
1249            other than 32-126 are always shown as hex escapes.
1251        (b) When it is outputting text that is a matched part of a subject string,
1252            it does the same, unless a different locale has been set for the match
1253            (using the /L modifier). In this case, it uses isprint() to decide.
1255     4. Fixed a major bug that caused incorrect computation of the amount of memory
1256        required for a compiled pattern when options that changed within the
1257        pattern affected the logic of the preliminary scan that determines the
1258        length. The relevant options are -x, and -i in UTF-8 mode. The result was
1259        that the computed length was too small. The symptoms of this bug were
1260        either the PCRE error "internal error: code overflow" from pcre_compile(),
1261        or a glibc crash with a message such as "pcretest: free(): invalid next
1262        size (fast)". Examples of patterns that provoked this bug (shown in
1263        pcretest format) are:
1265          /(?-x: )/x
1266          /(?x)(?-x: \s*#\s*)/
1267          /((?i)[\x{c0}])/8
1268          /(?i:[\x{c0}])/8
1270        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1271        is now done differently.
1273     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1274        wrapper classes; (b) implement a new function in the C++ scanner that is
1275        more efficient than the old way of doing things because it avoids levels of
1276        recursion in the regex matching; (c) add a paragraph to the documentation
1277        for the FullMatch() function.
1279     6. The escape sequence \n was being treated as whatever was defined as
1280        "newline". Not only was this contrary to the documentation, which states
1281        that \n is character 10 (hex 0A), but it also went horribly wrong when
1282        "newline" was defined as CRLF. This has been fixed.
1284     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1285        was being set to -1 for the "end of line" case (supposedly a value that no
1286        character can have). Though this value is never used (the check for end of
1287        line is "zero bytes in current character"), it caused compiler complaints.
1288        I've changed it to 0xffffffff.
1290     8. In pcre_version.c, the version string was being built by a sequence of
1291        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1292        string (as it is for production releases) called a macro with an empty
1293        argument. The C standard says the result of this is undefined. The gcc
1294        compiler treats it as an empty string (which was what was wanted) but it is
1295        reported that Visual C gives an error. The source has been hacked around to
1296        avoid this problem.
1298     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1299        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1300        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1301        of them did).
1303    10. Originally, pcretest opened its input and output without "b"; then I was
1304        told that "b" was needed in some environments, so it was added for release
1305        5.0 to both the input and output. (It makes no difference on Unix-like
1306        systems.) Later I was told that it is wrong for the input on Windows. I've
1307        now abstracted the modes into two macros, to make it easier to fiddle with
1308        them, and removed "b" from the input mode under Windows.
1310    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1312    12. Added -help and --help to pcretest as an official way of being reminded
1313        of the options.
1315    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1316        and pcrecpp.cc because they annoy compilers at high warning levels.
1318    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1320    15. Fixed an occurrence of == in configure.ac that should have been = (shell
1321        scripts are not C programs :-) and which was not noticed because it works
1322        on Linux.
1324    16. pcretest is supposed to handle any length of pattern and data line (as one
1325        line or as a continued sequence of lines) by extending its input buffer if
1326        necessary. This feature was broken for very long pattern lines, leading to
1327        a string of junk being passed to pcre_compile() if the pattern was longer
1328        than about 50K.
1330    17. I have done a major re-factoring of the way pcre_compile() computes the
1331        amount of memory needed for a compiled pattern. Previously, there was code
1332        that made a preliminary scan of the pattern in order to do this. That was
1333        OK when PCRE was new, but as the facilities have expanded, it has become
1334        harder and harder to keep it in step with the real compile phase, and there
1335        have been a number of bugs (see for example, 4 above). I have now found a
1336        cunning way of running the real compile function in a "fake" mode that
1337        enables it to compute how much memory it would need, while actually only
1338        ever using a few hundred bytes of working memory and without too many
1339        tests of the mode. This should make future maintenance and development
1340        easier. A side effect of this work is that the limit of 200 on the nesting
1341        depth of parentheses has been removed (though this was never a serious
1342        limitation, I suspect). However, there is a downside: pcre_compile() now
1343        runs more slowly than before (30% or more, depending on the pattern). I
1344        hope this isn't a big issue. There is no effect on runtime performance.
1346    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1347        newline (only possible for the last line of a file) and it was a
1348        pattern that set a locale (followed by /Lsomething), pcretest crashed.
1350    19. Added additional timing features to pcretest. (1) The -tm option now times
1351        matching only, not compiling. (2) Both -t and -tm can be followed, as a
1352        separate command line item, by a number that specifies the number of
1353        repeats to use when timing. The default is 50000; this gives better
1354        precision, but takes uncomfortably long for very large patterns.
1356    20. Extended pcre_study() to be more clever in cases where a branch of a
1357        subpattern has no definite first character. For example, (a*|b*)[cd] would
1358        previously give no result from pcre_study(). Now it recognizes that the
1359        first character must be a, b, c, or d.
1361    21. There was an incorrect error "recursive call could loop indefinitely" if
1362        a subpattern (or the entire pattern) that was being tested for matching an
1363        empty string contained only one non-empty item after a nested subpattern.
1364        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1365        incorrectly, because the \d was being skipped in the check.
1367    22. The pcretest program now has a new pattern option /B and a command line
1368        option -b, which is equivalent to adding /B to every pattern. This causes
1369        it to show the compiled bytecode, without the additional information that
1370        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1371        is the same as /B/I).
1373    23. A new optimization is now able automatically to treat some sequences such
1374        as a*b as a*+b. More specifically, if something simple (such as a character
1375        or a simple class like \d) has an unlimited quantifier, and is followed by
1376        something that cannot possibly match the quantified thing, the quantifier
1377        is automatically "possessified".
1379    24. A recursive reference to a subpattern whose number was greater than 39
1380        went wrong under certain circumstances in UTF-8 mode. This bug could also
1381        have affected the operation of pcre_study().
1383    25. Realized that a little bit of performance could be had by replacing
1384        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1386    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1388    27. Possessive quantifiers such as a++ were previously implemented by turning
1389        them into atomic groups such as ($>a+). Now they have their own opcodes,
1390        which improves performance. This includes the automatically created ones
1391        from 23 above.
1393    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1394        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1395        the first matched character to be a colon. This applied both to named and
1396        numbered groups.
1398    29. The ucpinternal.h header file was missing its idempotency #ifdef.
1400    30. I was sent a "project" file called libpcre.a.dev which I understand makes
1401        building PCRE on Windows easier, so I have included it in the distribution.
1403    31. There is now a check in pcretest against a ridiculously large number being
1404        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1405        loop, the loop is abandoned.
1407    32. Forward references to subpatterns in conditions such as (?(2)...) where
1408        subpattern 2 is defined later cause pcre_compile() to search forwards in
1409        the pattern for the relevant set of parentheses. This search went wrong
1410        when there were unescaped parentheses in a character class, parentheses
1411        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1413    33. "Subroutine" calls and backreferences were previously restricted to
1414        referencing subpatterns earlier in the regex. This restriction has now
1415        been removed.
1417    34. Added a number of extra features that are going to be in Perl 5.10. On the
1418        whole, these are just syntactic alternatives for features that PCRE had
1419        previously implemented using the Python syntax or my own invention. The
1420        other formats are all retained for compatibility.
1422        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1423            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1424            also .NET compatible.
1426        (b) A recursion or subroutine call to a named group can now be defined as
1427            (?&name) as well as (?P>name).
1429        (c) A backreference to a named group can now be defined as \k<name> or
1430            \k'name' as well as (?P=name). The new forms, as well as being in Perl
1431            5.10, are also .NET compatible.
1433        (d) A conditional reference to a named group can now use the syntax
1434            (?(<name>) or (?('name') as well as (?(name).
1436        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1437            groups (named and numbered) that are never evaluated inline, but can be
1438            called as "subroutines" from elsewhere. In effect, the DEFINE condition
1439            is always false. There may be only one alternative in such a group.
1441        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1442            as the simple (?(R). The condition is true only if the most recent
1443            recursion is that of the given number or name. It does not search out
1444            through the entire recursion stack.
1446        (g) The escape \gN or \g{N} has been added, where N is a positive or
1447            negative number, specifying an absolute or relative reference.
1449    35. Tidied to get rid of some further signed/unsigned compiler warnings and
1450        some "unreachable code" warnings.
1452    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1453        things, this adds five new scripts.
1455    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1456        There were also incompatibilities regarding the handling of \Q..\E inside
1457        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1458        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1460    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1461        matches an empty string, and forcibly breaks the loop. There were bugs in
1462        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
1463        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
1464        separate and independent bugs (that affected different cases) have been
1465        fixed.
1467    39. Refactored the code to abolish the use of different opcodes for small
1468        capturing bracket numbers. This is a tidy that I avoided doing when I
1469        removed the limit on the number of capturing brackets for 3.5 back in 2001.
1470        The new approach is not only tidier, it makes it possible to reduce the
1471        memory needed to fix the previous bug (38).
1473    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1474        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1475        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1476        mode.
1478    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1479        report.
1481    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1482        copy construction and assignment in the C++ wrapper.
1484    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1485        couple of bugs that could have given wrong results in the "--newline=crlf"
1486        case.
1488    44. Added a number of casts and did some reorganization of signed/unsigned int
1489        variables following suggestions from Dair Grant. Also renamed the variable
1490        "this" as "item" because it is a C++ keyword.
1492    45. Arranged for dftables to add
1494          #include "pcre_internal.h"
1496        to pcre_chartables.c because without it, gcc 4.x may remove the array
1497        definition from the final binary if PCRE is built into a static library and
1498        dead code stripping is activated.
1500    46. For an unanchored pattern, if a match attempt fails at the start of a
1501        newline sequence, and the newline setting is CRLF or ANY, and the next two
1502        characters are CRLF, advance by two characters instead of one.
1505    Version 6.7 04-Jul-06
1506    ---------------------
1508     1. In order to handle tests when input lines are enormously long, pcretest has
1509        been re-factored so that it automatically extends its buffers when
1510        necessary. The code is crude, but this _is_ just a test program. The
1511        default size has been increased from 32K to 50K.
1513     2. The code in pcre_study() was using the value of the re argument before
1514        testing it for NULL. (Of course, in any sensible call of the function, it
1515        won't be NULL.)
1517     3. The memmove() emulation function in pcre_internal.h, which is used on
1518        systems that lack both memmove() and bcopy() - that is, hardly ever -
1519        was missing a "static" storage class specifier.
1521     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1522        containing an extended class (one that cannot be represented by a bitmap
1523        because it contains high-valued characters or Unicode property items, e.g.
1524        [\pZ]). Almost always one would set UTF-8 mode when processing such a
1525        pattern, but PCRE should not loop if you do not (it no longer does).
1526        [Detail: two cases were found: (a) a repeated subpattern containing an
1527        extended class; (b) a recursive reference to a subpattern that followed a
1528        previous extended class. It wasn't skipping over the extended class
1529        correctly when UTF-8 mode was not set.]
1531     5. A negated single-character class was not being recognized as fixed-length
1532        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1533        compile error "lookbehind assertion is not fixed length".
1535     6. The RunPerlTest auxiliary script was showing an unexpected difference
1536        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1537        write a Perl script that can interpret lines of an input file either as
1538        byte characters or as UTF-8, which is what "perltest" was being required to
1539        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1540        can't do is switch easily at run time between having the "use utf8;" pragma
1541        or not. In the end, I fudged it by using the RunPerlTest script to insert
1542        "use utf8;" explicitly for the UTF-8 tests.
1544     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1545        the end of the subject string, contrary to the documentation and to what
1546        Perl does. This was true of both matching functions. Now it matches only at
1547        the start of the subject and immediately after *internal* newlines.
1549     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1550        a pointer to an int instead of a pointer to an unsigned long int. This
1551        caused problems on 64-bit systems.
1553     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1554        instance of the 'standard' template library not being so standard".
1556    10. There was no check on the number of named subpatterns nor the maximum
1557        length of a subpattern name. The product of these values is used to compute
1558        the size of the memory block for a compiled pattern. By supplying a very
1559        long subpattern name and a large number of named subpatterns, the size
1560        computation could be caused to overflow. This is now prevented by limiting
1561        the length of names to 32 characters, and the number of named subpatterns
1562        to 10,000.
1564    11. Subpatterns that are repeated with specific counts have to be replicated in
1565        the compiled pattern. The size of memory for this was computed from the
1566        length of the subpattern and the repeat count. The latter is limited to
1567        65535, but there was no limit on the former, meaning that integer overflow
1568        could in principle occur. The compiled length of a repeated subpattern is
1569        now limited to 30,000 bytes in order to prevent this.
1571    12. Added the optional facility to have named substrings with the same name.
1573    13. Added the ability to use a named substring as a condition, using the
1574        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1575        are numbers (not recommended). Forward references are permitted.
1577    14. Added forward references in named backreferences (if you see what I mean).
1579    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1580        pattern could run off the end of the subject. For example, the pattern
1581        "(?s)(.{1,5})"8 did this with the subject "ab".
1583    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1584        PCRE_CASELESS was set when matching characters that were quantified with ?
1585        or *.
1587    17. A character class other than a single negated character that had a minimum
1588        but no maximum quantifier - for example [ab]{6,} - was not handled
1589        correctly by pce_dfa_exec(). It would match only one character.
1591    18. A valid (though odd) pattern that looked like a POSIX character
1592        class but used an invalid character after [ (for example [[,abc,]]) caused
1593        pcre_compile() to give the error "Failed: internal error: code overflow" or
1594        in some cases to crash with a glibc free() error. This could even happen if
1595        the pattern terminated after [[ but there just happened to be a sequence of
1596        letters, a binary zero, and a closing ] in the memory that followed.
1598    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1599        over the years. Originally (before any Unicode support), just the bottom 8
1600        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1601        output from "man perlunicode" includes this:
1603          The regular expression compiler produces polymorphic opcodes.  That
1604          is, the pattern adapts to the data and automatically switches to
1605          the Unicode character scheme when presented with Unicode data--or
1606          instead uses a traditional byte scheme when presented with byte
1607          data.
1609        Sadly, a wide octal escape does not cause a switch, and in a string with
1610        no other multibyte characters, these octal escapes are treated as before.
1611        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
1612        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1613        Unicode string.
1615        I have not perpetrated such confusion in PCRE. Up till now, it took just
1616        the bottom 8 bits, as in old Perl. I have now made octal escapes with
1617        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1618        translate to the appropriate multibyte character.
1620    29. Applied some refactoring to reduce the number of warnings from Microsoft
1621        and Borland compilers. This has included removing the fudge introduced
1622        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1623        a warning about an unused variable.
1625    21. PCRE has not included VT (character 0x0b) in the set of whitespace
1626        characters since release 4.0, because Perl (from release 5.004) does not.
1627        [Or at least, is documented not to: some releases seem to be in conflict
1628        with the documentation.] However, when a pattern was studied with
1629        pcre_study() and all its branches started with \s, PCRE still included VT
1630        as a possible starting character. Of course, this did no harm; it just
1631        caused an unnecessary match attempt.
1633    22. Removed a now-redundant internal flag bit that recorded the fact that case
1634        dependency changed within the pattern. This was once needed for "required
1635        byte" processing, but is no longer used. This recovers a now-scarce options
1636        bit. Also moved the least significant internal flag bit to the most-
1637        significant bit of the word, which was not previously used (hangover from
1638        the days when it was an int rather than a uint) to free up another bit for
1639        the future.
1641    23. Added support for CRLF line endings as well as CR and LF. As well as the
1642        default being selectable at build time, it can now be changed at runtime
1643        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1644        specify that it is scanning data with non-default line endings.
1646    24. Changed the definition of CXXLINK to make it agree with the definition of
1647        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1649    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1650        recursions. This makes a big different to stack usage for some patterns.
1652    26. If a subpattern containing a named recursion or subroutine reference such
1653        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1654        the space required for the compiled pattern went wrong and gave too small a
1655        value. Depending on the environment, this could lead to "Failed: internal
1656        error: code overflow at offset 49" or "glibc detected double free or
1657        corruption" errors.
1659    27. Applied patches from Google (a) to support the new newline modes and (b) to
1660        advance over multibyte UTF-8 characters in GlobalReplace.
1662    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1663        difference for some implementation of PCRE in some Windows version.
1665    29. Added some extra testing facilities to pcretest:
1667        \q<number>   in a data line sets the "match limit" value
1668        \Q<number>   in a data line sets the "match recursion limt" value
1669        -S <number>  sets the stack size, where <number> is in megabytes
1671        The -S option isn't available for Windows.
1674    Version 6.6 06-Feb-06
1675    ---------------------
1677     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1678        in pcreposix.h. I have copied the definition from pcre.h.
1680     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1681        because pcre.h is no longer a built file.
1683     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1684        not normally included in the compiled code.
1687    Version 6.5 01-Feb-06
1688    ---------------------
1690     1. When using the partial match feature with pcre_dfa_exec(), it was not
1691        anchoring the second and subsequent partial matches at the new starting
1692        point. This could lead to incorrect results. For example, with the pattern
1693        /1234/, partially matching against "123" and then "a4" gave a match.
1695     2. Changes to pcregrep:
1697        (a) All non-match returns from pcre_exec() were being treated as failures
1698            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1699            error message is output. Some extra information is given for the
1700            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
1701            probably the only errors that are likely to be caused by users (by
1702            specifying a regex that has nested indefinite repeats, for instance).
1703            If there are more than 20 of these errors, pcregrep is abandoned.
1705        (b) A binary zero was treated as data while matching, but terminated the
1706            output line if it was written out. This has been fixed: binary zeroes
1707            are now no different to any other data bytes.
1709        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1710            used to set a locale for matching. The --locale=xxxx long option has
1711            been added (no short equivalent) to specify a locale explicitly on the
1712            pcregrep command, overriding the environment variables.
1714        (d) When -B was used with -n, some line numbers in the output were one less
1715            than they should have been.
1717        (e) Added the -o (--only-matching) option.
1719        (f) If -A or -C was used with -c (count only), some lines of context were
1720            accidentally printed for the final match.
1722        (g) Added the -H (--with-filename) option.
1724        (h) The combination of options -rh failed to suppress file names for files
1725            that were found from directory arguments.
1727        (i) Added the -D (--devices) and -d (--directories) options.
1729        (j) Added the -F (--fixed-strings) option.
1731        (k) Allow "-" to be used as a file name for -f as well as for a data file.
1733        (l) Added the --colo(u)r option.
1735        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1736            is not present by default.
1738     3. A nasty bug was discovered in the handling of recursive patterns, that is,
1739        items such as (?R) or (?1), when the recursion could match a number of
1740        alternatives. If it matched one of the alternatives, but subsequently,
1741        outside the recursion, there was a failure, the code tried to back up into
1742        the recursion. However, because of the way PCRE is implemented, this is not
1743        possible, and the result was an incorrect result from the match.
1745        In order to prevent this happening, the specification of recursion has
1746        been changed so that all such subpatterns are automatically treated as
1747        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1749     4. I had overlooked the fact that, in some locales, there are characters for
1750        which isalpha() is true but neither isupper() nor islower() are true. In
1751        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1752        and ordfeminine) are like this. This affected the treatment of \w and \W
1753        when they appeared in character classes, but not when they appeared outside
1754        a character class. The bit map for "word" characters is now created
1755        separately from the results of isalnum() instead of just taking it from the
1756        upper, lower, and digit maps. (Plus the underscore character, of course.)
1758     5. The above bug also affected the handling of POSIX character classes such as
1759        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1760        permanent tables. Instead, the bit maps for such a class were previously
1761        created as the appropriate unions of the upper, lower, and digit bitmaps.
1762        Now they are created by subtraction from the [[:word:]] class, which has
1763        its own bitmap.
1765     6. The [[:blank:]] character class matches horizontal, but not vertical space.
1766        It is created by subtracting the vertical space characters (\x09, \x0a,
1767        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1768        subtraction was done in the overall bitmap for a character class, meaning
1769        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1770        be recognized. This bug has been fixed.
1772     7. Patches from the folks at Google:
1774          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1775          real life, but is still worth protecting against".
1777          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1778          regular expressions".
1780          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1781          have it.
1783          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1784          "configure" and the latter not, in order to fix a problem somebody had
1785          with compiling the Arg class on HP-UX.
1787          (e) Improve the error-handling of the C++ wrapper a little bit.
1789          (f) New tests for checking recursion limiting.
1791     8. The pcre_memmove() function, which is used only if the environment does not
1792        have a standard memmove() function (and is therefore rarely compiled),
1793        contained two bugs: (a) use of int instead of size_t, and (b) it was not
1794        returning a result (though PCRE never actually uses the result).
1796     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1797        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1798        returned instead of calling malloc() with an overflowing number that would
1799        most likely cause subsequent chaos.
1801    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1803    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1804        with this option is matched, the nmatch and pmatch options of regexec() are
1805        ignored.
1807    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1808        provided in case anyone wants to the the POSIX interface with UTF-8
1809        strings.
1811    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1812        C++ linking (needed for some HP-UX environments).
1814    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1815        (unused parameter) and in the pcre_printint() function (omitted "default"
1816        switch label when the default is to do nothing).
1818    15. Added some code to make it possible, when PCRE is compiled as a C++
1819        library, to replace subject pointers for pcre_exec() with a smart pointer
1820        class, thus making it possible to process discontinuous strings.
1822    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1823        much the same function. They were added by different people who were trying
1824        to make PCRE easy to compile on non-Unix systems. It has been suggested
1825        that PCRE_EXPORT be abolished now that there is more automatic apparatus
1826        for compiling on Windows systems. I have therefore replaced it with
1827        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1828        defaults to "extern" for C or "extern C" for C++, which works fine on
1829        Unix-like systems. It is now possible to override the value of PCRE_DATA_
1830        SCOPE with something explicit in config.h. In addition:
1832        (a) pcreposix.h still had just "extern" instead of either of these macros;
1833            I have replaced it with PCRE_DATA_SCOPE.
1835        (b) Functions such as _pcre_xclass(), which are internal to the library,
1836            but external in the C sense, all had PCRE_EXPORT in their definitions.
1837            This is apparently wrong for the Windows case, so I have removed it.
1838            (It makes no difference on Unix-like systems.)
1840    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1841        of recursive calls to match(). This is different to MATCH_LIMIT because
1842        that limits the total number of calls to match(), not all of which increase
1843        the depth of recursion. Limiting the recursion depth limits the amount of
1844        stack (or heap if NO_RECURSE is set) that is used. The default can be set
1845        when PCRE is compiled, and changed at run time. A patch from Google adds
1846        this functionality to the C++ interface.
1848    18. Changes to the handling of Unicode character properties:
1850        (a) Updated the table to Unicode 4.1.0.
1852        (b) Recognize characters that are not in the table as "Cn" (undefined).
1854        (c) I revised the way the table is implemented to a much improved format
1855            which includes recognition of ranges. It now supports the ranges that
1856            are defined in UnicodeData.txt, and it also amalgamates other
1857            characters into ranges. This has reduced the number of entries in the
1858            table from around 16,000 to around 3,000, thus reducing its size
1859            considerably. I realized I did not need to use a tree structure after
1860            all - a binary chop search is just as efficient. Having reduced the
1861            number of entries, I extended their size from 6 bytes to 8 bytes to
1862            allow for more data.
1864        (d) Added support for Unicode script names via properties such as \p{Han}.
1866    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1867        matching that character.
1869    20. When matching a repeated Unicode property with a minimum greater than zero,
1870        (for example \pL{2,}), PCRE could look past the end of the subject if it
1871        reached it while seeking the minimum number of characters. This could
1872        happen only if some of the characters were more than one byte long, because
1873        there is a check for at least the minimum number of bytes.
1875    21. Refactored the implementation of \p and \P so as to be more general, to
1876        allow for more different types of property in future. This has changed the
1877        compiled form incompatibly. Anybody with saved compiled patterns that use
1878        \p or \P will have to recompile them.
1880    22. Added "Any" and "L&" to the supported property types.
1882    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1883        but give a compile time error if the value is greater than 0xff.
1885    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1886        accidentally not being installed or uninstalled.
1888    25. The pcre.h file was built from pcre.h.in, but the only changes that were
1889        made were to insert the current release number. This seemed silly, because
1890        it made things harder for people building PCRE on systems that don't run
1891        "configure". I have turned pcre.h into a distributed file, no longer built
1892        by "configure", with the version identification directly included. There is
1893        no longer a pcre.h.in file.
1895        However, this change necessitated a change to the pcre-config script as
1896        well. It is built from pcre-config.in, and one of the substitutions was the
1897        release number. I have updated configure.ac so that ./configure now finds
1898        the release number by grepping pcre.h.
1900    26. Added the ability to run the tests under valgrind.
1903    Version 6.4 05-Sep-05
1904    ---------------------
1906     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1907        "--" to be printed when multiple files were scanned, even when none of the
1908        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1909        consider it to be a bug, and have restored the previous behaviour.
1911     2. A couple of code tidies to get rid of compiler warnings.
1913     3. The pcretest program used to cheat by referring to symbols in the library
1914        whose names begin with _pcre_. These are internal symbols that are not
1915        really supposed to be visible externally, and in some environments it is
1916        possible to suppress them. The cheating is now confined to including
1917        certain files from the library's source, which is a bit cleaner.
1919     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1920        file's purpose clearer.
1922     5. Reorganized pcre_ucp_findchar().
1925    Version 6.3 15-Aug-05
1926    ---------------------
1928     1. The file libpcre.pc.in did not have general read permission in the tarball.
1930     2. There were some problems when building without C++ support:
1932        (a) If C++ support was not built, "make install" and "make test" still
1933            tried to test it.
1935        (b) There were problems when the value of CXX was explicitly set. Some
1936            changes have been made to try to fix these, and ...
1938        (c) --disable-cpp can now be used to explicitly disable C++ support.
1940        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1941            backslash in a target when C++ was disabled. This confuses some
1942            versions of "make", apparently. Using an intermediate variable solves
1943            this. (Same for CPP_LOBJ.)
1945     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1946        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1947        necessary on certain architectures.
1949     4. Added a setting of -export-symbols-regex to the link command to remove
1950        those symbols that are exported in the C sense, but actually are local
1951        within the library, and not documented. Their names all begin with
1952        "_pcre_". This is not a perfect job, because (a) we have to except some
1953        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1954        available (and never for static libraries). I have made a note to try to
1955        find a way round (a) in the future.
1958    Version 6.2 01-Aug-05
1959    ---------------------
1961     1. There was no test for integer overflow of quantifier values. A construction
1962        such as {1111111111111111} would give undefined results. What is worse, if
1963        a minimum quantifier for a parenthesized subpattern overflowed and became
1964        negative, the calculation of the memory size went wrong. This could have
1965        led to memory overwriting.
1967     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1969     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1970        operating environments where this matters.
1972     4. Applied Giuseppe Maxia's patch to add additional features for controlling
1973        PCRE options from within the C++ wrapper.
1975     5. Named capturing subpatterns were not being correctly counted when a pattern
1976        was compiled. This caused two problems: (a) If there were more than 100
1977        such subpatterns, the calculation of the memory needed for the whole
1978        compiled pattern went wrong, leading to an overflow error. (b) Numerical
1979        back references of the form \12, where the number was greater than 9, were
1980        not recognized as back references, even though there were sufficient
1981        previous subpatterns.
1983     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1984        versions of gcc, e.g. 2.95.4.
1987    Version 6.1 21-Jun-05
1988    ---------------------
1990     1. There was one reference to the variable "posix" in pcretest.c that was not
1991        surrounded by "#if !defined NOPOSIX".
1993     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1994        the cross-check on the old pcre_info() function, for the benefit of the
1995        cut-down version of PCRE that is currently imported into Exim.
1997     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1998        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1999        patterns that start (?i) but I don't think that matters. The use of (?i) is
2000        just an example; this all applies to the other options as well.
2002     4. Since libtool seems to echo the compile commands it is issuing, the output
2003        from "make" can be reduced a bit by putting "@" in front of each libtool
2004        compile command.
2006     5. Patch from the folks at Google for configure.in to be a bit more thorough
2007        in checking for a suitable C++ installation before trying to compile the
2008        C++ stuff. This should fix a reported problem when a compiler was present,
2009        but no suitable headers.
2011     6. The man pages all had just "PCRE" as their title. I have changed them to
2012        be the relevant file name. I have also arranged that these names are
2013        retained in the file doc/pcre.txt, which is a concatenation in text format
2014        of all the man pages except the little individual ones for each function.
2016     7. The NON-UNIX-USE file had not been updated for the different set of source
2017        files that come with release 6. I also added a few comments about the C++
2018        wrapper.
2021    Version 6.0 07-Jun-05
2022    ---------------------
2024     1. Some minor internal re-organization to help with my DFA experiments.
2026     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
2027        didn't matter for the library itself when fully configured, but did matter
2028        when compiling without UCP support, or within Exim, where the ucp files are
2029        not imported.
2031     3. Refactoring of the library code to split up the various functions into
2032        different source modules. The addition of the new DFA matching code (see
2033        below) to a single monolithic source would have made it really too
2034        unwieldy, quite apart from causing all the code to be include in a
2035        statically linked application, when only some functions are used. This is
2036        relevant even without the DFA addition now that patterns can be compiled in
2037        one application and matched in another.
2039        The downside of splitting up is that there have to be some external
2040        functions and data tables that are used internally in different modules of
2041        the library but which are not part of the API. These have all had their
2042        names changed to start with "_pcre_" so that they are unlikely to clash
2043        with other external names.
2045     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
2046        a different (DFA) algorithm. Although it is slower than the original
2047        function, it does have some advantages for certain types of matching
2048        problem.
2050     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
2051        including restarting after a partial match.
2053     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
2054        defined when compiling for Windows was sent to me. I have put it into the
2055        code, though I have no means of testing or verifying it.
2057     7. Added the pcre_refcount() auxiliary function.
2059     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
2060        match before or at the first newline in the subject string. In pcretest,
2061        the /f option on a pattern can be used to set this.
2063     9. A repeated \w when used in UTF-8 mode with characters greater than 256
2064        would behave wrongly. This has been present in PCRE since release 4.0.
2066    10. A number of changes to the pcregrep command:
2068        (a) Refactored how -x works; insert ^(...)$ instead of setting
2069            PCRE_ANCHORED and checking the length, in preparation for adding
2070            something similar for -w.
2072        (b) Added the -w (match as a word) option.
2074        (c) Refactored the way lines are read and buffered so as to have more
2075            than one at a time available.
2077        (d) Implemented a pcregrep test script.
2079        (e) Added the -M (multiline match) option. This allows patterns to match
2080            over several lines of the subject. The buffering ensures that at least
2081            8K, or the rest of the document (whichever is the shorter) is available
2082            for matching (and similarly the previous 8K for lookbehind assertions).
2084        (f) Changed the --help output so that it now says
2086              -w, --word-regex(p)
2088            instead of two lines, one with "regex" and the other with "regexp"
2089            because that confused at least one person since the short forms are the
2090            same. (This required a bit of code, as the output is generated
2091            automatically from a table. It wasn't just a text change.)
2093        (g) -- can be used to terminate pcregrep options if the next thing isn't an
2094            option but starts with a hyphen. Could be a pattern or a path name
2095            starting with a hyphen, for instance.
2097        (h) "-" can be given as a file name to represent stdin.
2099        (i) When file names are being printed, "(standard input)" is used for
2100            the standard input, for compatibility with GNU grep. Previously
2101            "<stdin>" was used.
2103        (j) The option --label=xxx can be used to supply a name to be used for
2104            stdin when file names are being printed. There is no short form.
2106        (k) Re-factored the options decoding logic because we are going to add
2107            two more options that take data. Such options can now be given in four
2108            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
2110        (l) Added the -A, -B, and -C options for requesting that lines of context
2111            around matches be printed.
2113        (m) Added the -L option to print the names of files that do not contain
2114            any matching lines, that is, the complement of -l.
2116        (n) The return code is 2 if any file cannot be opened, but pcregrep does
2117            continue to scan other files.
2119        (o) The -s option was incorrectly implemented. For compatibility with other
2120            greps, it now suppresses the error message for a non-existent or non-
2121            accessible file (but not the return code). There is a new option called
2122            -q that suppresses the output of matching lines, which was what -s was
2123            previously doing.
2125        (p) Added --include and --exclude options to specify files for inclusion
2126            and exclusion when recursing.
2128    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
2129        Hopefully, it now does.
2131    12. Missing cast in pcre_study().
2133    13. Added an "uninstall" target to the makefile.
2135    14. Replaced "extern" in the function prototypes in Makefile.in with
2136        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
2137        world, but is set differently for Windows.
2139    15. Added a second compiling function called pcre_compile2(). The only
2140        difference is that it has an extra argument, which is a pointer to an
2141        integer error code. When there is a compile-time failure, this is set
2142        non-zero, in addition to the error test pointer being set to point to an
2143        error message. The new argument may be NULL if no error number is required
2144        (but then you may as well call pcre_compile(), which is now just a
2145        wrapper). This facility is provided because some applications need a
2146        numeric error indication, but it has also enabled me to tidy up the way
2147        compile-time errors are handled in the POSIX wrapper.
2149    16. Added VPATH=.libs to the makefile; this should help when building with one
2150        prefix path and installing with another. (Or so I'm told by someone who
2151        knows more about this stuff than I do.)
2153    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
2154        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
2155        match everything, including newlines. This is not POSIX-compatible, but
2156        somebody wanted the feature. From pcretest it can be activated by using
2157        both the P and the s flags.
2159    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
2161    19. libpcre.pc was being incorrectly installed as executable.
2163    20. A couple of places in pcretest check for end-of-line by looking for '\n';
2164        it now also looks for '\r' so that it will work unmodified on Windows.
2166    21. Added Google's contributed C++ wrapper to the distribution.
2168    22. Added some untidy missing memory free() calls in pcretest, to keep
2169        Electric Fence happy when testing.
2173    Version 5.0 13-Sep-04
2174    ---------------------
2176     1. Internal change: literal characters are no longer packed up into items
2177        containing multiple characters in a single byte-string. Each character
2178        is now matched using a separate opcode. However, there may be more than one
2179        byte in the character in UTF-8 mode.
2181     2. The pcre_callout_block structure has two new fields: pattern_position and
2182        next_item_length. These contain the offset in the pattern to the next match
2183        item, and its length, respectively.
2185     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
2186        insertion of callouts before each pattern item. Added the /C option to
2187        pcretest to make use of this.
2189     4. On the advice of a Windows user, the lines
2191          #if defined(_WIN32) || defined(WIN32)
2192          _setmode( _fileno( stdout ), 0x8000 );
2193          #endif  /* defined(_WIN32) || defined(WIN32) */
2195        have been added to the source of pcretest. This apparently does useful
2196        magic in relation to line terminators.
2198     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
2199        for the benefit of those environments where the "b" makes a difference.
2201     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
2202        to know about it. I have put a hack into configure.in that adds in code
2203        to set GCC=yes if CC=icc. This seems to end up at a point in the
2204        generated configure script that is early enough to affect the setting of
2205        compiler options, which is what is needed, but I have no means of testing
2206        whether it really works. (The user who reported this had patched the
2207        generated configure script, which of course I cannot do.)
2209        LATER: After change 22 below (new libtool files), the configure script
2210        seems to know about icc (and also ecc). Therefore, I have commented out
2211        this hack in configure.in.
2213     7. Added support for pkg-config (2 patches were sent in).
2215     8. Negated POSIX character classes that used a combination of internal tables
2216        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2217        [[:^ascii]]. Typically, they would match almost any characters. The other
2218        POSIX classes were not broken in this way.
2220     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2221        to find the match, as PCRE was deluded into thinking that the match had to
2222        start at the start point or following a newline. The same bug applied to
2223        patterns with negative forward assertions or any backward assertions
2224        preceding ".*" at the start, unless the pattern required a fixed first
2225        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2227    10. In UTF-8 mode, when moving forwards in the subject after a failed match
2228        starting at the last subject character, bytes beyond the end of the subject
2229        string were read.
2231    11. Renamed the variable "class" as "classbits" to make life easier for C++
2232        users. (Previously there was a macro definition, but it apparently wasn't
2233        enough.)
2235    12. Added the new field "tables" to the extra data so that tables can be passed
2236        in at exec time, or the internal tables can be re-selected. This allows
2237        a compiled regex to be saved and re-used at a later time by a different
2238        program that might have everything at different addresses.
2240    13. Modified the pcre-config script so that, when run on Solaris, it shows a
2241        -R library as well as a -L library.
2243    14. The debugging options of pcretest (-d on the command line or D on a
2244        pattern) showed incorrect output for anything following an extended class
2245        that contained multibyte characters and which was followed by a quantifier.
2247    15. Added optional support for general category Unicode character properties
2248        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2249        support. It adds about 90K to the size of the library. The meanings of the
2250        inbuilt class escapes such as \d and \s have NOT been changed.
2252    16. Updated pcredemo.c to include calls to free() to release the memory for the
2253        compiled pattern.
2255    17. The generated file chartables.c was being created in the source directory
2256        instead of in the building directory. This caused the build to fail if the
2257        source directory was different from the building directory, and was
2258        read-only.
2260    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2261        file. No doubt somebody will tell me if they don't make sense... Also added
2262        Dan Mooney's comments about building on OpenVMS.
2264    19. Added support for partial matching via the PCRE_PARTIAL option for
2265        pcre_exec() and the \P data escape in pcretest.
2267    20. Extended pcretest with 3 new pattern features:
2269        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
2270              write the compiled pattern to the file whose name is "rest-of-line".
2271              This is a straight binary dump of the data, with the saved pointer to
2272              the character tables forced to be NULL. The study data, if any, is
2273              written too. After writing, pcretest reads a new pattern.
2275        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2276              compiled pattern from the given file. There must not be any
2277              occurrences of "<" in the file name (pretty unlikely); if there are,
2278              pcretest will instead treat the initial "<" as a pattern delimiter.
2279              After reading in the pattern, pcretest goes on to read data lines as
2280              usual.
2282        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2283              and 16-bit fields in a compiled pattern, to simulate a pattern that
2284              was compiled on a host of opposite endianness.
2286    21. The pcre-exec() function can now cope with patterns that were compiled on
2287        hosts of opposite endianness, with this restriction:
2289          As for any compiled expression that is saved and used later, the tables
2290          pointer field cannot be preserved; the extra_data field in the arguments
2291          to pcre_exec() should be used to pass in a tables address if a value
2292          other than the default internal tables were used at compile time.
2294    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2295        now diagnosed as an error. Previously, most of the time, a negative number
2296        would have been treated as zero, but if in addition "ovector" was passed as
2297        NULL, a crash could occur.
2299    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2300        new versions from the libtool 1.5 distribution (the last one is a copy of
2301        a file called libtool.m4). This seems to have fixed the need to patch
2302        "configure" to support Darwin 1.3 (which I used to do). However, I still
2303        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2304        workstation).
2306    24. Changed the PCRE licence to be the more standard "BSD" licence.
2309    Version 4.5 01-Dec-03
2310    ---------------------
2312     1. There has been some re-arrangement of the code for the match() function so
2313        that it can be compiled in a version that does not call itself recursively.
2314        Instead, it keeps those local variables that need separate instances for
2315        each "recursion" in a frame on the heap, and gets/frees frames whenever it
2316        needs to "recurse". Keeping track of where control must go is done by means
2317        of setjmp/longjmp. The whole thing is implemented by a set of macros that
2318        hide most of the details from the main code, and operates only if
2319        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2320        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2321        operating.
2323        To make it easier for callers to provide specially tailored get/free
2324        functions for this usage, two new functions, pcre_stack_malloc, and
2325        pcre_stack_free, are used. They are always called in strict stacking order,
2326        and the size of block requested is always the same.
2328        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2329        PCRE has been compiled to use the stack or the heap for recursion. The
2330        -C option of pcretest uses this to show which version is compiled.
2332        A new data escape \S, is added to pcretest; it causes the amounts of store
2333        obtained and freed by both kinds of malloc/free at match time to be added
2334        to the output.
2336     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2337        what's available on my current Linux desktop machine.
2339     3. When matching a UTF-8 string, the test for a valid string at the start has
2340        been extended. If start_offset is not zero, PCRE now checks that it points
2341        to a byte that is the start of a UTF-8 character. If not, it returns
2342        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2343        this is necessary because there may be backward assertions in the pattern.
2344        When matching the same subject several times, it may save resources to use
2345        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2347     4. The code for checking the validity of UTF-8 strings has been tightened so
2348        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2349        containing "overlong sequences".
2351     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2352        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2353        should have been "&", but it just so happened that all the cases this let
2354        through by mistake were picked up later in the function.
2356     6. I had used a variable called "isblank" - this is a C99 function, causing
2357        some compilers to warn. To avoid this, I renamed it (as "blankclass").
2359     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2360        prompting; (b) run "./pcretest /dev/null" at the start of the test script
2361        so the version is shown; (c) stop "make test" echoing "./RunTest".
2363     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2365     9. The prototype for memmove() for systems that don't have it was using
2366        size_t, but the inclusion of the header that defines size_t was later. I've
2367        moved the #includes for the C headers earlier to avoid this.
2369    10. Added some adjustments to the code to make it easier to compiler on certain
2370        special systems:
2372          (a) Some "const" qualifiers were missing.
2373          (b) Added the macro EXPORT before all exported functions; by default this
2374              is defined to be empty.
2375          (c) Changed the dftables auxiliary program (that builds chartables.c) so
2376              that it reads its output file name as an argument instead of writing
2377              to the standard output and assuming this can be redirected.
2379    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2380        class containing characters with values greater than 255, PCRE compilation
2381        went into a loop.
2383    12. A recursive reference to a subpattern that was within another subpattern
2384        that had a minimum quantifier of zero caused PCRE to crash. For example,
2385        (x(y(?2))z)? provoked this bug with a subject that got as far as the
2386        recursion. If the recursively-called subpattern itself had a zero repeat,
2387        that was OK.
2389    13. In pcretest, the buffer for reading a data line was set at 30K, but the
2390        buffer into which it was copied (for escape processing) was still set at
2391        1024, so long lines caused crashes.
2393    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2394        "internal error: code overflow...". This applied to any character class
2395        that was followed by a possessive quantifier.
2397    15. Modified the Makefile to add libpcre.la as a prerequisite for
2398        libpcreposix.la because I was told this is needed for a parallel build to
2399        work.
2401    16. If a pattern that contained .* following optional items at the start was
2402        studied, the wrong optimizing data was generated, leading to matching
2403        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2404        matching string must start with a or b or c. The correct conclusion for
2405        this pattern is that a match can start with any character.
2408    Version 4.4 13-Aug-03
2409    ---------------------
2411     1. In UTF-8 mode, a character class containing characters with values between
2412        127 and 255 was not handled correctly if the compiled pattern was studied.
2413        In fixing this, I have also improved the studying algorithm for such
2414        classes (slightly).
2416     2. Three internal functions had redundant arguments passed to them. Removal
2417        might give a very teeny performance improvement.
2419     3. Documentation bug: the value of the capture_top field in a callout is *one
2420        more than* the number of the hightest numbered captured substring.
2422     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2423        in incorrectly linking with a previously installed version. They now link
2424        explicitly with libpcre.la.
2426     5. configure.in no longer needs to recognize Cygwin specially.
2428     6. A problem in pcre.in for Windows platforms is fixed.
2430     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2431        pcretest, it used to include the size of the study block as part of its
2432        output. Unfortunately, the structure contains a field that has a different
2433        size on different hardware architectures. This meant that the tests that
2434        showed this size failed. As the block is currently always of a fixed size,
2435        this information isn't actually particularly useful in pcretest output, so
2436        I have just removed it.
2438     8. Three pre-processor statements accidentally did not start in column 1.
2439        Sadly, there are *still* compilers around that complain, even though
2440        standard C has not required this for well over a decade. Sigh.
2442     9. In pcretest, the code for checking callouts passed small integers in the
2443        callout_data field, which is a void * field. However, some picky compilers
2444        complained about the casts involved for this on 64-bit systems. Now
2445        pcretest passes the address of the small integer instead, which should get
2446        rid of the warnings.
2448    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2449        both compile and run time, and gives an error if an invalid UTF-8 sequence
2450        is found. There is a option for disabling this check in cases where the
2451        string is known to be correct and/or the maximum performance is wanted.
2453    11. In response to a bug report, I changed one line in Makefile.in from
2455            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2456        to
2457            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2459        to look similar to other lines, but I have no way of telling whether this
2460        is the right thing to do, as I do not use Windows. No doubt I'll get told
2461        if it's wrong...
2464    Version 4.3 21-May-03
2465    ---------------------
2467    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2468       Makefile.
2470    2. Some refactoring to improve the quality of the code:
2472       (i)   The utf8_table... variables are now declared "const".
2474       (ii)  The code for \cx, which used the "case flipping" table to upper case
2475             lower case letters, now just substracts 32. This is ASCII-specific,
2476             but the whole concept of \cx is ASCII-specific, so it seems
2477             reasonable.
2479       (iii) PCRE was using its character types table to recognize decimal and
2480             hexadecimal digits in the pattern. This is silly, because it handles
2481             only 0-9, a-f, and A-F, but the character types table is locale-
2482             specific, which means strange things might happen. A private
2483             table is now used for this - though it costs 256 bytes, a table is
2484             much faster than multiple explicit tests. Of course, the standard
2485             character types table is still used for matching digits in subject
2486             strings against \d.
2488       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2489             ending in _t are). So I've renamed it as ESC_tee.
2491    3. The first argument for regexec() in the POSIX wrapper should have been
2492       defined as "const".
2494    4. Changed pcretest to use malloc() for its buffers so that they can be
2495       Electric Fenced for debugging.
2497    5. There were several places in the code where, in UTF-8 mode, PCRE would try
2498       to read one or more bytes before the start of the subject string. Often this
2499       had no effect on PCRE's behaviour, but in some circumstances it could
2500       provoke a segmentation fault.
2502    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2503       to try to read one or more bytes before the start of the subject string.
2505    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
2506       UTF-8 support could misbehave in various ways if the subject string
2507       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
2508       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
2509       back over UTF-8 characters.)
2512    Version 4.2 14-Apr-03
2513    ---------------------
2515    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
2517    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
2518         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
2519         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
2520         [WIN_PREFIX]: new variable, "cyg" for Cygwin
2521         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
2522           and BUILD_EXEEXT
2523         Note: automatic setting of the BUILD variables is not yet working
2524         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
2525           compile-time but not at link-time
2526         [LINK]: use for linking executables only
2527         make different versions for Windows and non-Windows
2528         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
2529           libraries
2530         [LINK_FOR_BUILD]: new variable
2531         [OBJEXT]: use throughout
2532         [EXEEXT]: use throughout
2533         <winshared>: new target
2534         <wininstall>: new target
2535         <dftables.o>: use native compiler
2536         <dftables>: use native linker
2537         <install>: handle Windows platform correctly
2538         <clean>: ditto
2539         <check>: ditto
2540         copy DLL to top builddir before testing
2542       As part of these changes, -no-undefined was removed again. This was reported
2543       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
2544       in any case.
2546    3. Some tidies to get rid of compiler warnings:
2548       . In the match_data structure, match_limit was an unsigned long int, whereas
2549         match_call_count was an int. I've made them both unsigned long ints.
2551       . In pcretest the fact that a const uschar * doesn't automatically cast to
2552         a void * provoked a warning.
2554       . Turning on some more compiler warnings threw up some "shadow" variables
2555         and a few more missing casts.
2557    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
2558       option, a class that contained a single character with a value between 128
2559       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
2561    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
2562       option, a class that contained several characters, but with at least one
2563       whose value was between 128 and 255 caused PCRE to crash.
2566    Version 4.1 12-Mar-03
2567    ---------------------
2569    1. Compiling with gcc -pedantic found a couple of places where casts were
2570    needed, and a string in dftables.c that was longer than standard compilers are
2571    required to support.
2573    2. Compiling with Sun's compiler found a few more places where the code could
2574    be tidied up in order to avoid warnings.
2576    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
2577    first of these names is deprecated in the latest Autoconf in favour of the name
2578    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
2579    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2580    analogy I have changed it to CFLAGS_FOR_BUILD.
2582    4. Added -no-undefined to the linking command in the Makefile, because this is
2583    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2584    linking step for the pcreposix library.
2586    5. PCRE was failing to diagnose the case of two named groups with the same
2587    name.
2589    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2590    literal character that is needed in the subject for a match, and scans along to
2591    ensure that it is present before embarking on the full matching process. This
2592    saves time in cases of nested unlimited repeats that are never going to match.
2593    Problem: the scan can take a lot of time if the subject is very long (e.g.
2594    megabytes), thus penalizing straightforward matches. It is now done only if the
2595    amount of subject to be scanned is less than 1000 bytes.
2597    7. A lesser problem with the same optimization is that it was recording the
2598    first character of an anchored pattern as "needed", thus provoking a search
2599    right along the subject, even when the first match of the pattern was going to
2600    fail. The "needed" character is now not set for anchored patterns, unless it
2601    follows something in the pattern that is of non-fixed length. Thus, it still
2602    fulfils its original purpose of finding quick non-matches in cases of nested
2603    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
2606    Version 4.0 17-Feb-03
2607    ---------------------
2609    1. If a comment in an extended regex that started immediately after a meta-item
2610    extended to the end of string, PCRE compiled incorrect data. This could lead to
2611    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
2613    2. Moved to autoconf 2.53 and libtool 1.4.2.
2615    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
2616    the special perltest8 script is no longer needed - all the tests can be run
2617    from a single perltest script.
2619    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
2620    by \s. It has now been removed in PCRE. This means it isn't recognized as
2621    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
2622    class [:space:] *does* include VT, thereby creating a mess.
2624    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
2625    space and tab.
2627    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
2628    its new features into the main test script, reducing the number of scripts.
2630    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
2631    were backward compatible, and made the (?i) apply to the whole pattern, as if
2632    /i were given. Now it behaves more logically, and applies the option setting
2633    only to what follows. PCRE has been changed to follow suit. However, if it
2634    finds options settings right at the start of the pattern, it extracts them into
2635    the global options, as before. Thus, they show up in the info data.
2637    8. Added support for the \Q...\E escape sequence. Characters in between are
2638    treated as literals. This is slightly different from Perl in that $ and @ are
2639    also handled as literals inside the quotes. In Perl, they will cause variable
2640    interpolation. Note the following examples:
2642        Pattern            PCRE matches      Perl matches
2644        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
2645        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
2646        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
2648    For compatibility with Perl, \Q...\E sequences are recognized inside character
2649    classes as well as outside them.
2651    9. Re-organized 3 code statements in pcretest to avoid "overflow in
2652    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2653    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2654    signed/unsigned warnings.
2656    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2657    option for pcretest, so I've replaced it by a simple function that does just
2658    that job.
2660    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2661    "pcregrep -".
2663    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2664    Java package. This provides some syntactic sugar for simple cases of what my
2665    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2666    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2667    item, you can use this simplified notation. Note that only makes sense with
2668    greedy quantifiers. Consequently, the use of the possessive quantifier forces
2669    greediness, whatever the setting of the PCRE_UNGREEDY option.
2671    13. A change of greediness default within a pattern was not taking effect at
2672    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2673    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2674    was abstracted outside.
2676    14. PCRE now supports the \G assertion. It is true when the current matching
2677    position is at the start point of the match. This differs from \A when the
2678    starting offset is non-zero. Used with the /g option of pcretest (or similar
2679    code), it works in the same way as it does for Perl's /g option. If all
2680    alternatives of a regex begin with \G, the expression is anchored to the start
2681    match position, and the "anchored" flag is set in the compiled expression.
2683    15. Some bugs concerning the handling of certain option changes within patterns
2684    have been fixed. These applied to options other than (?ims). For example,
2685    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2686    been the other way round. Some of this was related to change 7 above.
2688    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2689    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2690    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2691    POSIX classes only within a class (e.g. /[[:alpha:]]/).
2693    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2694    mode. Unlike ".", it always matches newline, whatever the setting of
2695    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2696    assertions. Perl allows it, but it doesn't (in general) work because it can't
2697    calculate the length of the lookbehind. At least, that's the case for Perl
2698    5.8.0 - I've been told they are going to document that it doesn't work in
2699    future.
2701    18. Added an error diagnosis for escapes that PCRE does not support: these are
2702    \L, \l, \N, \P, \p, \U, \u, and \X.
2704    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2705    reading past the end of the pattern in cases such as /[abcd/.
2707    20. PCRE was getting more memory than necessary for patterns with classes that
2708    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2710    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2711    compiling PCRE for use with Virtual Pascal.
2713    22. Small fix to the Makefile to make it work properly if the build is done
2714    outside the source tree.
2716    23. Added a new extension: a condition to go with recursion. If a conditional
2717    subpattern starts with (?(R) the "true" branch is used if recursion has
2718    happened, whereas the "false" branch is used only at the top level.
2720    24. When there was a very long string of literal characters (over 255 bytes
2721    without UTF support, over 250 bytes with UTF support), the computation of how
2722    much memory was required could be incorrect, leading to segfaults or other
2723    strange effects.
2725    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2726    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2727    there was a subsequent back reference to those brackets. This meant that, for
2728    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2729    possible to check for precisely this case. All we can do is abandon the
2730    optimization if .* occurs inside capturing brackets when there are any back
2731    references whatsoever. (See below for a better fix that came later.)
2733    26. The handling of the optimization for finding the first character of a
2734    non-anchored pattern, and for finding a character that is required later in the
2735    match were failing in some cases. This didn't break the matching; it just
2736    failed to optimize when it could. The way this is done has been re-implemented.
2738    27. Fixed typo in error message for invalid (?R item (it said "(?p").
2740    28. Added a new feature that provides some of the functionality that Perl
2741    provides with (?{...}). The facility is termed a "callout". The way it is done
2742    in PCRE is for the caller to provide an optional function, by setting
2743    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2744    global variable. By default it is unset, which disables all calling out. To get
2745    the function called, the regex must include (?C) at appropriate points. This
2746    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2747    This provides a means of identifying different callout points. When PCRE
2748    reaches such a point in the regex, if pcre_callout has been set, the external
2749    function is called. It is provided with data in a structure called
2750    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2751    matching continues; if it returns a non-zero value, the match at the current
2752    point fails. However, backtracking will occur if possible. [This was changed
2753    later and other features added - see item 49 below.]
2755    29. pcretest is upgraded to test the callout functionality. It provides a
2756    callout function that displays information. By default, it shows the start of
2757    the match and the current position in the text. There are some new data escapes
2758    to vary what happens:
2760        \C+         in addition, show current contents of captured substrings
2761        \C-         do not supply a callout function
2762        \C!n        return 1 when callout number n is reached
2763        \C!n!m      return 1 when callout number n is reached for the mth time
2765    30. If pcregrep was called with the -l option and just a single file name, it
2766    output "<stdin>" if a match was found, instead of the file name.
2768    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2769    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2770    pcre_exec(). This saves a malloc/free per call. The default value of
2771    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2772    when configuring.
2774    32. The default maximum size of a compiled pattern is 64K. There have been a
2775    few cases of people hitting this limit. The code now uses macros to handle the
2776    storing of links as offsets within the compiled pattern. It defaults to 2-byte
2777    links, but this can be changed to 3 or 4 bytes by --with-link-size when
2778    configuring. Tests 2 and 5 work only with 2-byte links because they output
2779    debugging information about compiled patterns.
2781    33. Internal code re-arrangements:
2783    (a) Moved the debugging function for printing out a compiled regex into
2784        its own source file (printint.c) and used #include to pull it into
2785        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2786        separate copies.
2788    (b) Defined the list of op-code names for debugging as a macro in
2789        internal.h so that it is next to the definition of the opcodes.
2791    (c) Defined a table of op-code lengths for simpler skipping along compiled
2792        code. This is again a macro in internal.h so that it is next to the
2793        definition of the opcodes.
2795    34. Added support for recursive calls to individual subpatterns, along the
2796    lines of Robin Houston's patch (but implemented somewhat differently).
2798    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2799    allow it to read and process whole directories in Win32. This code was
2800    contributed by Lionel Fourquaux; it has not been tested by me.
2802    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2803    used to name a group. Names consist of alphanumerics and underscores, and must
2804    be unique. Back references use the syntax (?P=name) and recursive calls use
2805    (?P>name) which is a PCRE extension to the Python extension. Groups still have
2806    numbers. The function pcre_fullinfo() can be used after compilation to extract
2807    a name/number map. There are three relevant calls:
2809      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
2810      PCRE_INFO_NAMECOUNT            yields the number of entries
2811      PCRE_INFO_NAMETABLE            yields a pointer to the map.
2813    The map is a vector of fixed-size entries. The size of each entry depends on
2814    the length of the longest name used. The first two bytes of each entry are the
2815    group number, most significant byte first. There follows the corresponding
2816    name, zero terminated. The names are in alphabetical order.
2818    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2819    case instead of 255. Making it the same both with and without UTF-8 support
2820    means that the same test output works with both.
2822    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2823    calling malloc() with a zero argument.
2825    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2826    optimization. I've improved things by keeping a bitmap of backreferences with
2827    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2828    fact referenced, the optimization can be applied. It is unlikely that a
2829    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2830    the match to follow \n) will appear inside brackets with a number greater than
2831    31, but if it does, any back reference > 31 suppresses the optimization.
2833    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2834    of disabling numbered capturing parentheses. Any opening parenthesis that is
2835    not followed by ? behaves as if it were followed by ?: but named parentheses
2836    can still be used for capturing (and they will acquire numbers in the usual
2837    way).
2839    41. Redesigned the return codes from the match() function into yes/no/error so
2840    that errors can be passed back from deep inside the nested calls. A malloc
2841    failure while inside a recursive subpattern call now causes the
2842    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2844    42. It is now possible to set a limit on the number of times the match()
2845    function is called in a call to pcre_exec(). This facility makes it possible to
2846    limit the amount of recursion and backtracking, though not in a directly
2847    obvious way, because the match() function is used in a number of different
2848    circumstances. The count starts from zero for each position in the subject
2849    string (for non-anchored patterns). The default limit is, for compatibility, a
2850    large number, namely 10 000 000. You can change this in two ways:
2852    (a) When configuring PCRE before making, you can use --with-match-limit=n
2853        to set a default value for the compiled library.
2855    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2856        a different value is set. See 45 below.
2858    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2860    43. Added a new function pcre_config(int, void *) to enable run-time extraction
2861    of things that can be changed at compile time. The first argument specifies
2862    what is wanted and the second points to where the information is to be placed.
2863    The current list of available information is:
2865      PCRE_CONFIG_UTF8
2867    The output is an integer that is set to one if UTF-8 support is available;
2868    otherwise it is set to zero.
2872    The output is an integer that it set to the value of the code that is used for
2873    newline. It is either LF (10) or CR (13).
2877    The output is an integer that contains the number of bytes used for internal
2878    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2882    The output is an integer that contains the threshold above which the POSIX
2883    interface uses malloc() for output vectors. See item 31 above.
2887    The output is an unsigned integer that contains the default limit of the number
2888    of match() calls in a pcre_exec() execution. See 42 above.
2890    44. pcretest has been upgraded by the addition of the -C option. This causes it
2891    to extract all the available output from the new pcre_config() function, and to
2892    output it. The program then exits immediately.
2894    45. A need has arisen to pass over additional data with calls to pcre_exec() in
2895    order to support additional features. One way would have been to define
2896    pcre_exec2() (for example) with extra arguments, but this would not have been
2897    extensible, and would also have required all calls to the original function to
2898    be mapped to the new one. Instead, I have chosen to extend the mechanism that
2899    is used for passing in "extra" data from pcre_study().
2901    The pcre_extra structure is now exposed and defined in pcre.h. It currently
2902    contains the following fields:
2904      flags         a bitmap indicating which of the following fields are set
2905      study_data    opaque data from pcre_study()
2906      match_limit   a way of specifying a limit on match() calls for a specific
2907                      call to pcre_exec()
2908      callout_data  data for callouts (see 49 below)
2910    The flag bits are also defined in pcre.h, and are
2916    The pcre_study() function now returns one of these new pcre_extra blocks, with
2917    the actual study data pointed to by the study_data field, and the
2918    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2919    before. That is, this change is entirely upwards-compatible and requires no
2920    change to existing code.
2922    If you want to pass in additional data to pcre_exec(), you can either place it
2923    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2924    block.
2926    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2927    data string contains the escape sequence \M, pcretest calls pcre_exec() several
2928    times with different match limits, until it finds the minimum value needed for
2929    pcre_exec() to complete. The value is then output. This can be instructive; for
2930    most simple matches the number is quite small, but for pathological cases it
2931    gets very large very quickly.
2933    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2934    returns the size of the data block pointed to by the study_data field in a
2935    pcre_extra block, that is, the value that was passed as the argument to
2936    pcre_malloc() when PCRE was getting memory in which to place the information
2937    created by pcre_study(). The fourth argument should point to a size_t variable.
2938    pcretest has been extended so that this information is shown after a successful
2939    pcre_study() call when information about the compiled regex is being displayed.
2941    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2942    because what follows is always an absolute path. (Later: it turns out that this
2943    is more than cosmetic for MinGW, because it doesn't like empty path
2944    components.)
2946    49. Some changes have been made to the callout feature (see 28 above):
2948    (i)  A callout function now has three choices for what it returns:
2950           0  =>  success, carry on matching
2951         > 0  =>  failure at this point, but backtrack if possible
2952         < 0  =>  serious error, return this value from pcre_exec()
2954         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2955         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2956         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2957         use by callout functions. It will never be used by PCRE itself.
2959    (ii) The pcre_extra structure (see 45 above) has a void * field called
2960         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2961         pcre_callout_block structure has a field of the same name. The contents of
2962         the field passed in the pcre_extra structure are passed to the callout
2963         function in the corresponding field in the callout block. This makes it
2964         easier to use the same callout-containing regex from multiple threads. For
2965         testing, the pcretest program has a new data escape
2967           \C*n        pass the number n (may be negative) as callout_data
2969         If the callout function in pcretest receives a non-zero value as
2970         callout_data, it returns that value.
2972    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2973    there were some redundant $(CFLAGS) in commands that are now specified as
2974    $(LINK), which already includes $(CFLAGS).
2976    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2977    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2978    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2979    one-byte characters throughout. Note that case-insensitive matching applies
2980    only to characters whose values are less than 256. PCRE doesn't support the
2981    notion of cases for higher-valued characters.
2983    (i)   A character class whose characters are all within 0-255 is handled as
2984          a bit map, and the map is inverted for negative classes. Previously, a
2985          character > 255 always failed to match such a class; however it should
2986          match if the class was a negative one (e.g. [^ab]). This has been fixed.
2988    (ii)  A negated character class with a single character < 255 is coded as
2989          "not this character" (OP_NOT). This wasn't working properly when the test
2990          character was multibyte, either singly or repeated.
2992    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2993          mode, for example: \x{100}{2,3}.
2995    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2996          singly or repeated) now correctly test multibyte characters. However,
2997          PCRE doesn't recognize any characters with values greater than 255 as
2998          digits, spaces, or word characters. Such characters always match \D, \S,
2999          and \W, and never match \d, \s, or \w.
3001    (v)   Classes may now contain characters and character ranges with values
3002          greater than 255. For example: [ab\x{100}-\x{400}].
3004    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
3005          PCRE in UTF-8 mode.
3007    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
3008    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
3009    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
3010    value.)
3012    53. The single man page has become too large. I have therefore split it up into
3013    a number of separate man pages. These also give rise to individual HTML pages;
3014    these are now put in a separate directory, and there is an index.html page that
3015    lists them all. Some hyperlinking between the pages has been installed.
3017    54. Added convenience functions for handling named capturing parentheses.
3019    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
3020    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
3021    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
3022    are faulted.
3024    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
3025    calling configure. These values are used when compiling the dftables.c program
3026    which is run to generate the source of the default character tables. They
3027    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
3028    you will need to set these values.
3030    57. Updated the building process for Windows DLL, as provided by Fred Cox.
3033    Version 3.9 02-Jan-02
3034  ---------------------  ---------------------
3036  1. A bit of extraneous text had somehow crept into the pcregrep documentation.  1. A bit of extraneous text had somehow crept into the pcregrep documentation.

Removed from v.61  
changed lines
  Added in v.484

  ViewVC Help
Powered by ViewVC 1.1.5