ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 11 by nigel, Sat Feb 24 21:38:17 2007 UTC revision 532 by ph10, Tue Jun 1 16:21:42 2010 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
4    Version 8.10 03 May-2010
5    ------------------------
7    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
8        THEN.
10    2.  (*ACCEPT) was not working when inside an atomic group.
12    3.  Inside a character class, \B is treated as a literal by default, but
13        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
14        causes the error). The code is unchanged, but I tidied the documentation.
16    4.  Inside a character class, PCRE always treated \R and \X as literals,
17        whereas Perl faults them if its -w option is set. I have changed PCRE so
18        that it faults them when PCRE_EXTRA is set.
20    5.  Added support for \N, which always matches any character other than
21        newline. (It is the same as "." when PCRE_DOTALL is not set.)
23    6.  When compiling pcregrep with newer versions of gcc which may have
24        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
25        declared with attribute warn_unused_result" were given. Just casting the
26        result to (void) does not stop the warnings; a more elaborate fudge is
27        needed. I've used a macro to implement this.
29    7.  Minor change to pcretest.c to avoid a compiler warning.
31    8.  Added four artifical Unicode properties to help with an option to make
32        \s etc use properties (see next item). The new properties are: Xan
33        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
35    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
36        use Unicode properties. (*UCP) at the start of a pattern can be used to set
37        this option. Modified pcretest to add /W to test this facility. Added
38        REG_UCP to make it available via the POSIX interface.
40    10. Added --line-buffered to pcregrep.
42    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
43        studied, and the match started with a letter with a code point greater than
44        127 whose first byte was different to the first byte of the other case of
45        the letter, the other case of this starting letter was not recognized.
47    12. pcreposix.c included pcre.h before including pcre_internal.h. This caused a
48        conflict in the definition of PCRE_EXP_DECL. I have removed the include of
49        pcre.h as pcre_internal.h includes pcre.h itself. (This may be a bit of
50        historical tidying that never got done.)
52    13. If a pattern that was studied started with a repeated Unicode property
53        test, for example, \p{Nd}+, there was the theoretical possibility of
54        setting up an incorrect bitmap of starting bytes, but fortunately it could
55        not have actually happened in practice until change 8 above was made (it
56        added property types that matched character-matching opcodes).
58    14. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
59        possible starting bytes for non-anchored patterns.
61    15. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
62        \R, and also a number of cases that involve Unicode properties, both
63        explicit and implicit when PCRE_UCP is set.
65    16. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
66        input, it could crash or give wrong results if characters with values
67        greater than 0xc0 were present in the subject string. (Detail: it assumed
68        UTF-8 input when processing these items.)
70    17. Added a lot of (int) casts to avoid compiler warnings in systems where
71        size_t is 64-bit.
73    18. Added a check for running out of memory when PCRE is compiled with
74        --disable-stack-for-recursion.
78    Version 8.02 19-Mar-2010
79    ------------------------
81    1.  The Unicode data tables have been updated to Unicode 5.2.0.
83    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
84        configured.
86    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
87        original author of that file, following a query about its status.
89    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
90        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
92    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
93        quantifier applied to a forward-referencing subroutine call, could compile
94        incorrect code or give the error "internal error: previously-checked
95        referenced subpattern not found".
97    6.  Both MS Visual Studio and Symbian OS have problems with initializing
98        variables to point to external functions. For these systems, therefore,
99        pcre_malloc etc. are now initialized to local functions that call the
100        relevant global functions.
102    7.  There were two entries missing in the vectors called coptable and poptable
103        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
104        I've fixed the data, and added a kludgy way of testing at compile time that
105        the lengths are correct (equal to the number of opcodes).
107    8.  Following on from 7, I added a similar kludge to check the length of the
108        eint vector in pcreposix.c.
110    9.  Error texts for pcre_compile() are held as one long string to avoid too
111        much relocation at load time. To find a text, the string is searched,
112        counting zeros. There was no check for running off the end of the string,
113        which could happen if a new error number was added without updating the
114        string.
116    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
118    11. \K was not working if it appeared in an atomic group or in a group that
119        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
120        \K is "not well defined" if used in an assertion. PCRE now accepts it if
121        the assertion is positive, but not if it is negative.
123    12. Change 11 fortuitously reduced the size of the stack frame used in the
124        "match()" function of pcre_exec.c by one pointer. Forthcoming
125        implementation of support for (*MARK) will need an extra pointer on the
126        stack; I have reserved it now, so that the stack frame size does not
127        decrease.
129    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
130        item in branch that calls a recursion is a subroutine call - as in the
131        second branch in the above example - was incorrectly given the compile-
132        time error "recursive call could loop indefinitely" because pcre_compile()
133        was not correctly checking the subroutine for matching a non-empty string.
135    14. The checks for overrunning compiling workspace could trigger after an
136        overrun had occurred. This is a "should never occur" error, but it can be
137        triggered by pathological patterns such as hundreds of nested parentheses.
138        The checks now trigger 100 bytes before the end of the workspace.
140    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
143    Version 8.01 19-Jan-2010
144    ------------------------
146    1.  If a pattern contained a conditional subpattern with only one branch (in
147        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
148        computed the wrong minimum data length (which is of course zero for such
149        subpatterns). This could cause incorrect "no match" results.
151    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
152        the pattern is reset in the first branch, pcre_compile() failed with
153        "internal error: code overflow at offset...". This happened only when
154        the reset was to the original external option setting. (An optimization
155        abstracts leading options settings into an external setting, which was the
156        cause of this.)
158    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
159        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
160        assertion pattern did not match (meaning that the assertion was true), it
161        was incorrectly treated as false if the SKIP had been reached during the
162        matching. This also applied to assertions used as conditions.
164    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
165        assertion subpattern, including such a pattern used as a condition,
166        unpredictable results occurred, instead of the error return
169    5.  The C++ GlobalReplace function was not working like Perl for the special
170        situation when an empty string is matched. It now does the fancy magic
171        stuff that is necessary.
173    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
174        removed. (These were left over from very, very early versions of PCRE.)
176    7.  Some cosmetic changes to the code to make life easier when compiling it
177        as part of something else:
179        (a) Change DEBUG to PCRE_DEBUG.
181        (b) In pcre_compile(), rename the member of the "branch_chain" structure
182            called "current" as "current_branch", to prevent a collision with the
183            Linux macro when compiled as a kernel module.
185        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
186            prevent a collision with the Linux macro when compiled as a kernel
187            module.
189    8.  In pcre_compile() there are some checks for integer overflows that used to
190        cast potentially large values to (double). This has been changed to that
191        when building, a check for int64_t is made, and if it is found, it is used
192        instead, thus avoiding the use of floating point arithmetic. (There is no
193        other use of FP in PCRE.) If int64_t is not found, the fallback is to
194        double.
196    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
197        2005 (difference between two addresses compared to an unsigned value).
199    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
200        custom one, because of the following reported problem in Windows:
202          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
203              under Win32.
204          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
205              therefore missing the function definition.
206          - The compiler thus generates a "C" signature for the test function.
207          - The linker fails to find the "C" function.
208          - PCRE fails to configure if asked to do so against libbz2.
210    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
211        messages were output:
213          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
214          rerunning libtoolize, to keep the correct libtool macros in-tree.
215          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
217        I have done both of these things.
219    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
220        most of the time, it *can* run out if it is given a pattern that contains a
221        runaway infinite recursion. I updated the discussion in the pcrestack man
222        page.
224    13. Now that we have gone to the x.xx style of version numbers, the minor
225        version may start with zero. Using 08 or 09 is a bad idea because users
226        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
227        interpreted as invalid octal numbers. I've updated the previous comment in
228        configure.ac, and also added a check that gives an error if 08 or 09 are
229        used.
231    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
232        causing partial matching to fail when the end of the subject matched \W
233        in a UTF-8 pattern where \W was quantified with a minimum of 3.
235    15. There were some discrepancies between the declarations in pcre_internal.h
236        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
237        their definitions. The declarations used "const uschar *" and the
238        definitions used USPTR. Even though USPTR is normally defined as "const
239        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
240        reported that: "This difference in casting confuses some C++ compilers, for
241        example, SunCC recognizes above declarations as different functions and
242        generates broken code for hbpcre." I have changed the declarations to use
243        USPTR.
245    16. GNU libtool is named differently on some systems. The autogen.sh script now
246        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
247        (FreeBSD).
249    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
250        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
251        comment: "Figure out how to create a longlong from a string: strtoll and
252        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
253        instance, but it only takes 2 args instead of 3!"
255    18. A subtle bug concerned with back references has been fixed by a change of
256        specification, with a corresponding code fix. A pattern such as
257        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
258        refers, was giving matches when it shouldn't. For example, xa=xaaa would
259        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
260        same bug. Such groups have to be quantified to be useful, or contained
261        inside another quantified group. (If there's no repetition, the reference
262        can never match.) The problem arises because, having left the group and
263        moved on to the rest of the pattern, a later failure that backtracks into
264        the group uses the captured value from the final iteration of the group
265        rather than the correct earlier one. I have fixed this in PCRE by forcing
266        any group that contains a reference to itself to be an atomic group; that
267        is, there cannot be any backtracking into it once it has completed. This is
268        similar to recursive and subroutine calls.
271    Version 8.00 19-Oct-09
272    ----------------------
274    1.  The table for translating pcre_compile() error codes into POSIX error codes
275        was out-of-date, and there was no check on the pcre_compile() error code
276        being within the table. This could lead to an OK return being given in
277        error.
279    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
280        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
281        in a Windows environment.
283    3.  The pcregrep --count option prints the count for each file even when it is
284        zero, as does GNU grep. However, pcregrep was also printing all files when
285        --files-with-matches was added. Now, when both options are given, it prints
286        counts only for those files that have at least one match. (GNU grep just
287        prints the file name in this circumstance, but including the count seems
288        more useful - otherwise, why use --count?) Also ensured that the
289        combination -clh just lists non-zero counts, with no names.
291    4.  The long form of the pcregrep -F option was incorrectly implemented as
292        --fixed_strings instead of --fixed-strings. This is an incompatible change,
293        but it seems right to fix it, and I didn't think it was worth preserving
294        the old behaviour.
296    5.  The command line items --regex=pattern and --regexp=pattern were not
297        recognized by pcregrep, which required --regex pattern or --regexp pattern
298        (with a space rather than an '='). The man page documented the '=' forms,
299        which are compatible with GNU grep; these now work.
301    6.  No libpcreposix.pc file was created for pkg-config; there was just
302        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
304    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
305        when UCP support is not needed, by modifying the Python script that
306        generates it from Unicode data files. This should not matter if the module
307        is correctly used as a library, but I received one complaint about 50K of
308        unwanted data. My guess is that the person linked everything into his
309        program rather than using a library. Anyway, it does no harm.
311    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
312        was a minimum greater than 1 for a wide character in a possessive
313        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
314        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
315        character. Chaos in the form of incorrect output or a compiling loop could
316        result.
318    9.  The restrictions on what a pattern can contain when partial matching is
319        requested for pcre_exec() have been removed. All patterns can now be
320        partially matched by this function. In addition, if there are at least two
321        slots in the offset vector, the offset of the earliest inspected character
322        for the match and the offset of the end of the subject are set in them when
323        PCRE_ERROR_PARTIAL is returned.
325    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
326        synonymous with PCRE_PARTIAL, for backwards compatibility, and
327        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
328        and may be more useful for multi-segment matching.
330    11. Partial matching with pcre_exec() is now more intuitive. A partial match
331        used to be given if ever the end of the subject was reached; now it is
332        given only if matching could not proceed because another character was
333        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
334        string "Z", which now yields "no match" instead of "partial match". In the
335        case of pcre_dfa_exec(), "no match" is given if every matching path for the
336        final character ended with (*FAIL).
338    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
339        if the pattern had a "must contain" character that was already found in the
340        earlier partial match, unless partial matching was again requested. For
341        example, with the pattern /dog.(body)?/, the "must contain" character is
342        "g". If the first part-match was for the string "dog", restarting with
343        "sbody" failed. This bug has been fixed.
345    13. The string returned by pcre_dfa_exec() after a partial match has been
346        changed so that it starts at the first inspected character rather than the
347        first character of the match. This makes a difference only if the pattern
348        starts with a lookbehind assertion or \b or \B (\K is not supported by
349        pcre_dfa_exec()). It's an incompatible change, but it makes the two
350        matching functions compatible, and I think it's the right thing to do.
352    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
353        so that the demonstration program is easily available in environments where
354        PCRE has not been installed from source.
356    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
357        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
358        library.
360    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
361        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
362        is not the first non-POSIX option to be added. Clearly some people find
363        these options useful.
365    17. If a caller to the POSIX matching function regexec() passes a non-zero
366        value for nmatch with a NULL value for pmatch, the value of
367        nmatch is forced to zero.
369    18. RunGrepTest did not have a test for the availability of the -u option of
370        the diff command, as RunTest does. It now checks in the same way as
371        RunTest, and also checks for the -b option.
373    19. If an odd number of negated classes containing just a single character
374        interposed, within parentheses, between a forward reference to a named
375        subpattern and the definition of the subpattern, compilation crashed with
376        an internal error, complaining that it could not find the referenced
377        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
378        [The bug was that it was starting one character too far in when skipping
379        over the character class, thus treating the ] as data rather than
380        terminating the class. This meant it could skip too much.]
382    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
383        /g option in pcretest when the pattern contains \K, which makes it possible
384        to have an empty string match not at the start, even when the pattern is
385        anchored. Updated pcretest and pcredemo to use this option.
387    21. If the maximum number of capturing subpatterns in a recursion was greater
388        than the maximum at the outer level, the higher number was returned, but
389        with unset values at the outer level. The correct (outer level) value is
390        now given.
392    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
393        PCRE did not set those parentheses (unlike Perl). I have now found a way to
394        make it do so. The string so far is captured, making this feature
395        compatible with Perl.
397    23. The tests have been re-organized, adding tests 11 and 12, to make it
398        possible to check the Perl 5.10 features against Perl 5.10.
400    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
401        pattern matches a fixed length string. PCRE did not allow this; now it
402        does. Neither allows recursion.
404    25. I finally figured out how to implement a request to provide the minimum
405        length of subject string that was needed in order to match a given pattern.
406        (It was back references and recursion that I had previously got hung up
407        on.) This code has now been added to pcre_study(); it finds a lower bound
408        to the length of subject needed. It is not necessarily the greatest lower
409        bound, but using it to avoid searching strings that are too short does give
410        some useful speed-ups. The value is available to calling programs via
411        pcre_fullinfo().
413    26. While implementing 25, I discovered to my embarrassment that pcretest had
414        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
415        study optimizations had never been tested with that matching function.
416        Oops. What is worse, even when it was passed study data, there was a bug in
417        pcre_dfa_exec() that meant it never actually used it. Double oops. There
418        were also very few tests of studied patterns with pcre_dfa_exec().
420    27. If (?| is used to create subpatterns with duplicate numbers, they are now
421        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
422        on the other side of the coin, they are no longer allowed to have different
423        names, because these cannot be distinguished in PCRE, and this has caused
424        confusion. (This is a difference from Perl.)
426    28. When duplicate subpattern names are present (necessarily with different
427        numbers, as required by 27 above), and a test is made by name in a
428        conditional pattern, either for a subpattern having been matched, or for
429        recursion in such a pattern, all the associated numbered subpatterns are
430        tested, and the overall condition is true if the condition is true for any
431        one of them. This is the way Perl works, and is also more like the way
432        testing by number works.
435    Version 7.9 11-Apr-09
436    ---------------------
438    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
439        (pcretest), all targets were linked against these libraries. This included
440        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
441        libraries. This caused unwanted dependencies to be created. This problem
442        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
443        pcretest is linked with readline.
445    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
446        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
447        moved outside it again, because FALSE and TRUE are already defined in AIX,
448        but BOOL is not.
450    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
451        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
453    4.  The pcregrep documentation said spaces were inserted as well as colons (or
454        hyphens) following file names and line numbers when outputting matching
455        lines. This is not true; no spaces are inserted. I have also clarified the
456        wording for the --colour (or --color) option.
458    5.  In pcregrep, when --colour was used with -o, the list of matching strings
459        was not coloured; this is different to GNU grep, so I have changed it to be
460        the same.
462    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
463        each matching line was coloured. Now it goes on to look for further matches
464        of any of the test patterns, which is the same behaviour as GNU grep.
466    7.  A pattern that could match an empty string could cause pcregrep to loop; it
467        doesn't make sense to accept an empty string match in pcregrep, so I have
468        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
469        seems to be how GNU grep behaves.
471    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
472        start or after a newline", because the conditional assertion was not being
473        correctly handled. The rule now is that both the assertion and what follows
474        in the first alternative must satisfy the test.
476    9.  If auto-callout was enabled in a pattern with a conditional group whose
477        condition was an assertion, PCRE could crash during matching, both with
478        pcre_exec() and pcre_dfa_exec().
480    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
481        used for matching.
483    11. Unicode property support in character classes was not working for
484        characters (bytes) greater than 127 when not in UTF-8 mode.
486    12. Added the -M command line option to pcretest.
488    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
490    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
492    16. Added comments and documentation about mis-use of no_arg in the C++
493        wrapper.
495    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
496        from Martin Jerabek that uses macro names for all relevant character and
497        string constants.
499    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
500        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
501        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
502        these, but not everybody uses configure.
504    19. A conditional group that had only one branch was not being correctly
505        recognized as an item that could match an empty string. This meant that an
506        enclosing group might also not be so recognized, causing infinite looping
507        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
508        with the subject "ab", where knowledge that the repeated group can match
509        nothing is needed in order to break the loop.
511    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
512        exec(), but without supplying a callout function, matching went wrong.
514    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
515        leak if the size of the offset vector was greater than 30. When the vector
516        is smaller, the saved offsets during recursion go onto a local stack
517        vector, but for larger vectors malloc() is used. It was failing to free
518        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
519        error, in fact).
521    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
522        heapframe that is used only when UTF-8 support is enabled. This caused no
523        problem, but was untidy.
525    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
526        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
527        included within another project.
529    24. Steven Van Ingelgem's patches to add more options to the CMake support,
530        slightly modified by me:
532          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
533              not building pcregrep.
535          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
536              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
538    25. Forward references, both numeric and by name, in patterns that made use of
539        duplicate group numbers, could behave incorrectly or give incorrect errors,
540        because when scanning forward to find the reference group, PCRE was not
541        taking into account the duplicate group numbers. A pattern such as
542        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
544    26. Changed a few more instances of "const unsigned char *" to USPTR, making
545        the feature of a custom pointer more persuasive (as requested by a user).
547    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
548        pcretest.c, inside #ifndefs, because it seems they are sometimes already
549        pre-defined.
551    28. Added support for (*UTF8) at the start of a pattern.
553    29. Arrange for flags added by the "release type" setting in CMake to be shown
554        in the configuration summary.
557    Version 7.8 05-Sep-08
558    ---------------------
560    1.  Replaced UCP searching code with optimized version as implemented for Ad
561        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
562        stage table and inline lookup instead of a function, giving speed ups of 2
563        to 5 times on some simple patterns that I tested. Permission was given to
564        distribute the MultiStage2.py script that generates the tables (it's not in
565        the tarball, but is in the Subversion repository).
567    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
568        scripts.
570    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
571        a group with a zero qualifier. The result of the study could be incorrect,
572        or the function might crash, depending on the pattern.
574    4.  Caseless matching was not working for non-ASCII characters in back
575        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
576        It now works when Unicode Property Support is available.
578    5.  In pcretest, an escape such as \x{de} in the data was always generating
579        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
580        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
581        truncation.
583    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
585    7.  Added two (int) casts to pcregrep when printing the difference of two
586        pointers, in case they are 64-bit values.
588    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
589        test 2 if it fails.
591    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
592        and a #define of that name to empty if it is not externally set. This is to
593        allow users of MSVC to set it if necessary.
595    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
596        the convenience functions in the pcre_get.c source file.
598    11. An option change at the start of a pattern that had top-level alternatives
599        could cause overwriting and/or a crash. This command provoked a crash in
600        some environments:
602          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
604        This potential security problem was recorded as CVE-2008-2371.
606    12. For a pattern where the match had to start at the beginning or immediately
607        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
608        pcre_dfa_exec() could read past the end of the passed subject if there was
609        no match. To help with detecting such bugs (e.g. with valgrind), I modified
610        pcretest so that it places the subject at the end of its malloc-ed buffer.
612    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
613        exec() might read past the end of the data buffer in UTF-8 mode.
615    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
616        the data contained the byte 0x85 as part of a UTF-8 character within its
617        first line. This applied both to normal and DFA matching.
619    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
620        /^[^d]*?$/8 failed to match "abc".
622    16. Added a missing copyright notice to pcrecpp_internal.h.
624    17. Make it more clear in the documentation that values returned from
625        pcre_exec() in ovector are byte offsets, not character counts.
627    18. Tidied a few places to stop certain compilers from issuing warnings.
629    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
630        supplied by Stefan Weber. I made a further small update for 7.8 because
631        there is a change of source arrangements: the pcre_searchfuncs.c module is
632        replaced by pcre_ucd.c.
635    Version 7.7 07-May-08
636    ---------------------
638    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
639        a string to a long long, pretend we don't even have a long long." This is
640        done by checking for the strtoq, strtoll, and _strtoi64 functions.
642    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
643        pre-7.6 versions, which defined a global no_arg variable instead of putting
644        it in the RE class. (See also #8 below.)
646    3.  Remove a line of dead code, identified by coverity and reported by Nuno
647        Lopes.
649    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
651        (1) The include/exclude patterns were being applied to the whole pathnames
652            of files, instead of just to the final components.
654        (2) If there was more than one level of directory, the subdirectories were
655            skipped unless they satisfied the include/exclude conditions. This is
656            inconsistent with GNU grep (and could even be seen as contrary to the
657            pcregrep specification - which I improved to make it absolutely clear).
658            The action now is always to scan all levels of directory, and just
659            apply the include/exclude patterns to regular files.
661    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
662        --exclude_dir in the tests to avoid scanning .svn directories.
664    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
665        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
666        doesn't support NULs in patterns.
668    7.  Added some missing "const"s to declarations of static tables in
669        pcre_compile.c and pcre_dfa_exec.c.
671    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
672        caused by fix #2  above. (Subsequently also a second patch to fix the
673        first patch. And a third patch - this was a messy problem.)
675    9.  Applied Craig's patch to remove the use of push_back().
677    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
678        matching function regexec().
680    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
681        which, however, unlike Perl's \g{...}, are subroutine calls, not back
682        references. PCRE supports relative numbers with this syntax (I don't think
683        Oniguruma does).
685    12. Previously, a group with a zero repeat such as (...){0} was completely
686        omitted from the compiled regex. However, this means that if the group
687        was called as a subroutine from elsewhere in the pattern, things went wrong
688        (an internal error was given). Such groups are now left in the compiled
689        pattern, with a new opcode that causes them to be skipped at execution
690        time.
692    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
693        to the way PCRE behaves:
695        (a) A lone ] character is dis-allowed (Perl treats it as data).
697        (b) A back reference to an unmatched subpattern matches an empty string
698            (Perl fails the current match path).
700        (c) A data ] in a character class must be notated as \] because if the
701            first data character in a class is ], it defines an empty class. (In
702            Perl it is not possible to have an empty class.) The empty class []
703            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
704            The negative empty class [^] matches any one character, independently
705            of the DOTALL setting.
707    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
708        non-existent subpattern following a character class starting with ']' and
709        containing () gave an internal compiling error instead of "reference to
710        non-existent subpattern". Fortunately, when the pattern did exist, the
711        compiled code was correct. (When scanning forwards to check for the
712        existencd of the subpattern, it was treating the data ']' as terminating
713        the class, so got the count wrong. When actually compiling, the reference
714        was subsequently set up correctly.)
716    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
717        it was being rejected as not supported by pcre_dfa_exec(), even though
718        other assertions are supported. I have made pcre_dfa_exec() support
719        (*FAIL).
721    16. The implementation of 13c above involved the invention of a new opcode,
722        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
723        cannot be changed at match time, I realized I could make a small
724        improvement to matching performance by compiling OP_ALLANY instead of
725        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
726        on the OP_ANY path.
728    17. Compiling pcretest on Windows with readline support failed without the
729        following two fixes: (1) Make the unistd.h include conditional on
730        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
732    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
733        ncurses library to be included for pcretest when ReadLine support is
734        requested, but also to allow for it to be overridden. This patch came from
735        Daniel Bergström.
737    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
738        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
739        any errors with the current Unicode tables. Thanks to Peter Kankowski for
740        spotting this.
743    Version 7.6 28-Jan-08
744    ---------------------
746    1.  A character class containing a very large number of characters with
747        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
748        overflow.
750    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
751        HAVE_LONG_LONG is not defined.
753    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
754        bring them up to date and include new features. This patch includes:
756        - Fixed PH's badly added libz and libbz2 support.
757        - Fixed a problem with static linking.
758        - Added pcredemo. [But later removed - see 7 below.]
759        - Fixed dftables problem and added an option.
760        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
761            HAVE_LONG_LONG.
762        - Added readline support for pcretest.
763        - Added an listing of the option settings after cmake has run.
765    4.  A user submitted a patch to Makefile that makes it easy to create
766        "pcre.dll" under mingw when using Configure/Make. I added stuff to
767        Makefile.am that cause it to include this special target, without
768        affecting anything else. Note that the same mingw target plus all
769        the other distribution libraries and programs are now supported
770        when configuring with CMake (see 6 below) instead of with
771        Configure/Make.
773    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
774        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
775        exported in the Windows port". It has not yet been confirmed that the patch
776        solves the problem, but it does no harm.
778    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
779        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
780        with CMake, and also correct the comment about stack recursion.
782    7.  Remove the automatic building of pcredemo from the ./configure system and
783        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
784        of a program that users should build themselves after PCRE is installed, so
785        building it automatically is not really right. What is more, it gave
786        trouble in some build environments.
788    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
791    Version 7.5 10-Jan-08
792    ---------------------
794    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
795        values in parens when parsing an RE using the C++ wrapper."
797    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
798        Characters greater than 255 were excluded from the class instead of being
799        included.
801    3.  The same bug as (2) above applied to negated POSIX classes such as
802        [:^space:].
804    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
805        defined or documented. It seems to have been a typo for PCRE_STATIC, so
806        I have changed it.
808    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
809        first named subpattern) and a construct such as (?&a) would reference the
810        first named subpattern whose name started with "a" (in other words, the
811        length check was missing). Both these problems are fixed. "Subpattern name
812        expected" is now given for (?&) (a zero-length name), and this patch also
813        makes it give the same error for \k'' (previously it complained that that
814        was a reference to a non-existent subpattern).
816    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
817        this is right because (?- can be followed by option settings as well as by
818        digits. I have, however, made the messages clearer.
820    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
821        than the number used in the conditional) now cause a compile-time error.
822        This is actually not compatible with Perl, which accepts such patterns, but
823        treats the conditional as always being FALSE (as PCRE used to), but it
824        seems to me that giving a diagnostic is better.
826    8.  Change "alphameric" to the more common word "alphanumeric" in comments
827        and messages.
829    9.  Fix two occurrences of "backslash" in comments that should have been
830        "backspace".
832    10. Remove two redundant lines of code that can never be obeyed (their function
833        was moved elsewhere).
835    11. The program that makes PCRE's Unicode character property table had a bug
836        which caused it to generate incorrect table entries for sequences of
837        characters that have the same character type, but are in different scripts.
838        It amalgamated them into a single range, with the script of the first of
839        them. In other words, some characters were in the wrong script. There were
840        thirteen such cases, affecting characters in the following ranges:
842          U+002b0 - U+002c1
843          U+0060c - U+0060d
844          U+0061e - U+00612
845          U+0064b - U+0065e
846          U+0074d - U+0076d
847          U+01800 - U+01805
848          U+01d00 - U+01d77
849          U+01d9b - U+01dbf
850          U+0200b - U+0200f
851          U+030fc - U+030fe
852          U+03260 - U+0327f
853          U+0fb46 - U+0fbb1
854          U+10450 - U+1049d
856    12. The -o option (show only the matching part of a line) for pcregrep was not
857        compatible with GNU grep in that, if there was more than one match in a
858        line, it showed only the first of them. It now behaves in the same way as
859        GNU grep.
861    13. If the -o and -v options were combined for pcregrep, it printed a blank
862        line for every non-matching line. GNU grep prints nothing, and pcregrep now
863        does the same. The return code can be used to tell if there were any
864        non-matching lines.
866    14. Added --file-offsets and --line-offsets to pcregrep.
868    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
869        infinitely looping recursion. The bug was that positive lookaheads were not
870        being skipped when checking for a possible empty match (negative lookaheads
871        and both kinds of lookbehind were skipped).
873    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
874        inclusion of <windows.h> to before rather than after the definition of
875        INVALID_FILE_ATTRIBUTES (patch from David Byron).
877    17. Specifying a possessive quantifier with a specific limit for a Unicode
878        character property caused pcre_compile() to compile bad code, which led at
879        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
880        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
881        caused the error; without that there was no problem.
883    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
885    19. Added --enable-pcretest-libreadline.
887    20. In pcrecpp.cc, the variable 'count' was incremented twice in
888        RE::GlobalReplace(). As a result, the number of replacements returned was
889        double what it should be. I removed one of the increments, but Craig sent a
890        later patch that removed the other one (the right fix) and added unit tests
891        that check the return values (which was not done before).
893    21. Several CMake things:
895        (1) Arranged that, when cmake is used on Unix, the libraries end up with
896            the names libpcre and libpcreposix, not just pcre and pcreposix.
898        (2) The above change means that pcretest and pcregrep are now correctly
899            linked with the newly-built libraries, not previously installed ones.
903    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
904        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
905        UTF-8 newline character). The key issue is that the pattern starts .*;
906        this means that the match must be either at the beginning, or after a
907        newline. The bug was in the code for advancing after a failed match and
908        checking that the new position followed a newline. It was not taking
909        account of UTF-8 characters correctly.
911    23. PCRE was behaving differently from Perl in the way it recognized POSIX
912        character classes. PCRE was not treating the sequence [:...:] as a
913        character class unless the ... were all letters. Perl, however, seems to
914        allow any characters between [: and :], though of course it rejects as
915        unknown any "names" that contain non-letters, because all the known class
916        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
917        for example, whereas PCRE did not - it did not recognize a POSIX character
918        class. This seemed a bit dangerous, so the code has been changed to be
919        closer to Perl. The behaviour is not identical to Perl, because PCRE will
920        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
921        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
922        Perl does, and where it didn't before.
924    24. Rewrite so as to remove the single use of %n from pcregrep because in some
925        Windows environments %n is disabled by default.
928    Version 7.4 21-Sep-07
929    ---------------------
931    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
932        means that a class such as [\s] counted as "explicit reference to CR or
933        LF". That isn't really right - the whole point of the change was to try to
934        help when there was an actual mention of one of the two characters. So now
935        the change happens only if \r or \n (or a literal CR or LF) character is
936        encountered.
938    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
939        of both had grown to the point where there were only 3 bits left.
940        Fortunately, there was spare space in the data structure, and so I have
941        moved the internal flags into a new 16-bit field to free up more option
942        bits.
944    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
945        but did not set the internal JCHANGED flag - either of these is enough to
946        control the way the "get" function works - but the PCRE_INFO_JCHANGED
947        facility is supposed to tell if (?J) was ever used, so now (?J) at the
948        start sets both bits.
950    4.  Added options (at build time, compile time, exec time) to change \R from
951        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
953    5.  doc/pcresyntax.html was missing from the distribution.
955    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
956        compatibility, even though it is no longer used.
958    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
959        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
960        windows.h file is present (where different names are used). [This was
961        reversed later after testing - see 16 below.]
963    8.  Changed all #include <config.h> to #include "config.h". There were also
964        some further <pcre.h> cases that I changed to "pcre.h".
966    9.  When pcregrep was used with the --colour option, it missed the line ending
967        sequence off the lines that it output.
969    10. It was pointed out to me that arrays of string pointers cause lots of
970        relocations when a shared library is dynamically loaded. A technique of
971        using a single long string with a table of offsets can drastically reduce
972        these. I have refactored PCRE in four places to do this. The result is
973        dramatic:
975          Originally:                          290
976          After changing UCP table:            187
977          After changing error message table:   43
978          After changing table of "verbs"       36
979          After changing table of Posix names   22
981        Thanks to the folks working on Gregex for glib for this insight.
983    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
984        unicode-properties was also set.
986    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
988    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
989        checked only for CRLF.
991    14. Added casts to pcretest.c to avoid compiler warnings.
993    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
995    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
996        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
997        entirely. This removes changes made in 7 above.
999    17. The CMake files have been updated, and there is now more information about
1000        building with CMake in the NON-UNIX-USE document.
1003    Version 7.3 28-Aug-07
1004    ---------------------
1006     1. In the rejigging of the build system that eventually resulted in 7.1, the
1007        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
1008        brackets there is not right, since it causes compilers to look for an
1009        installed pcre.h, not the version that is in the source that is being
1010        compiled (which of course may be different). I have changed it back to:
1012          #include "pcre.h"
1014        I have a vague recollection that the change was concerned with compiling in
1015        different directories, but in the new build system, that is taken care of
1016        by the VPATH setting the Makefile.
1018     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
1019        when the subject happened to end in the byte 0x85 (e.g. if the last
1020        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
1021        characters but of course it shouldn't be taken as a newline when it is part
1022        of another character. The bug was that, for an unlimited repeat of . in
1023        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
1024        characters when looking for a newline.
1026     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
1028     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
1029        in debug output.
1031     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
1032        long printing in the pcrecpp unittest when running under MinGW.
1034     6. ESC_K was left out of the EBCDIC table.
1036     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
1037        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
1038        limit also applies to "virtual nesting" when a pattern is recursive, and in
1039        this case 1000 isn't so big. I have been able to remove this limit at the
1040        expense of backing off one optimization in certain circumstances. Normally,
1041        when pcre_exec() would call its internal match() function recursively and
1042        immediately return the result unconditionally, it uses a "tail recursion"
1043        feature to save stack. However, when a subpattern that can match an empty
1044        string has an unlimited repetition quantifier, it no longer makes this
1045        optimization. That gives it a stack frame in which to save the data for
1046        checking that an empty string has been matched. Previously this was taken
1047        from the 1000-entry workspace that had been reserved. So now there is no
1048        explicit limit, but more stack is used.
1050     8. Applied Daniel's patches to solve problems with the import/export magic
1051        syntax that is required for Windows, and which was going wrong for the
1052        pcreposix and pcrecpp parts of the library. These were overlooked when this
1053        problem was solved for the main library.
1055     9. There were some crude static tests to avoid integer overflow when computing
1056        the size of patterns that contain repeated groups with explicit upper
1057        limits. As the maximum quantifier is 65535, the maximum group length was
1058        set at 30,000 so that the product of these two numbers did not overflow a
1059        32-bit integer. However, it turns out that people want to use groups that
1060        are longer than 30,000 bytes (though not repeat them that many times).
1061        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1062        made it possible to implement the integer overflow checks in a much more
1063        dynamic way, which I have now done. The artificial limitation on group
1064        length has been removed - we now have only the limit on the total length of
1065        the compiled pattern, which depends on the LINK_SIZE setting.
1067    10. Fixed a bug in the documentation for get/copy named substring when
1068        duplicate names are permitted. If none of the named substrings are set, the
1069        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1070        empty string.
1072    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1073        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1074        because the ] is interpreted as the first data character and the
1075        terminating ] is not found. PCRE has been made compatible with Perl in this
1076        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1077        cause memory overwriting.
1079    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1080        string has been matched (to stop an infinite loop). It was not recognizing
1081        a conditional subpattern that could match an empty string if that
1082        subpattern was within another subpattern. For example, it looped when
1083        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
1084        condition was not nested. This bug has been fixed.
1086    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1087        past the start of the subject in the presence of bytes with the top bit
1088        set, for example "\x8aBCD".
1090    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1091        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1093    14. Optimized (?!) to (*FAIL).
1095    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1096        This restricts code points to be within the range 0 to 0x10FFFF, excluding
1097        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1098        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1099        does: it's just the validity check that is more restrictive.
1101    16. Inserted checks for integer overflows during escape sequence (backslash)
1102        processing, and also fixed erroneous offset values for syntax errors during
1103        backslash processing.
1105    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1106        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1108    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1109        caused an overrun.
1111    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1112        something other than just ASCII characters) inside a group that had an
1113        unlimited repeat caused a loop at compile time (while checking to see
1114        whether the group could match an empty string).
1116    20. Debugging a pattern containing \p or \P could cause a crash. For example,
1117        [\P{Any}] did so. (Error in the code for printing property names.)
1119    21. An orphan \E inside a character class could cause a crash.
1121    22. A repeated capturing bracket such as (A)? could cause a wild memory
1122        reference during compilation.
1124    23. There are several functions in pcre_compile() that scan along a compiled
1125        expression for various reasons (e.g. to see if it's fixed length for look
1126        behind). There were bugs in these functions when a repeated \p or \P was
1127        present in the pattern. These operators have additional parameters compared
1128        with \d, etc, and these were not being taken into account when moving along
1129        the compiled data. Specifically:
1131        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1132            length.
1134        (b) An item such as \pL+ within a repeated group could cause crashes or
1135            loops.
1137        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1138            "reference to non-existent subpattern" error.
1140        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1142    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1143        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1145    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1147    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1148        character were causing crashes (broken optimization).
1150    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1151        \p or \P) caused a compile-time loop.
1153    28. More problems have arisen in unanchored patterns when CRLF is a valid line
1154        break. For example, the unstudied pattern [\r\n]A does not match the string
1155        "\r\nA" because change 7.0/46 below moves the current point on by two
1156        characters after failing to match at the start. However, the pattern \nA
1157        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
1158        the same is true. There doesn't seem any very clean way out of this, but
1159        what I have chosen to do makes the common cases work: PCRE now takes note
1160        of whether there can be an explicit match for \r or \n anywhere in the
1161        pattern, and if so, 7.0/46 no longer applies. As part of this change,
1162        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1163        pattern has explicit CR or LF references.
1165    29. Added (*CR) etc for changing newline setting at start of pattern.
1168    Version 7.2 19-Jun-07
1169    ---------------------
1171     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
1172        which is apparently normally available under Windows.
1174     2. Re-jig the pcregrep tests with different newline settings in an attempt
1175        to make them independent of the local environment's newline setting.
1177     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
1179     4. Some of the "internals" tests were previously cut out when the link size
1180        was not 2, because the output contained actual offsets. The recent new
1181        "Z" feature of pcretest means that these can be cut out, making the tests
1182        usable with all link sizes.
1184     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
1185        stack recursion. This gives a massive performance boost under BSD, but just
1186        a small improvement under Linux. However, it saves one field in the frame
1187        in all cases.
1189     6. Added more features from the forthcoming Perl 5.10:
1191        (a) (?-n) (where n is a string of digits) is a relative subroutine or
1192            recursion call. It refers to the nth most recently opened parentheses.
1194        (b) (?+n) is also a relative subroutine call; it refers to the nth next
1195            to be opened parentheses.
1197        (c) Conditions that refer to capturing parentheses can be specified
1198            relatively, for example, (?(-2)... or (?(+3)...
1200        (d) \K resets the start of the current match so that everything before
1201            is not part of it.
1203        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1205        (f) \g{name} is another synonym - part of Perl 5.10's unification of
1206            reference syntax.
1208        (g) (?| introduces a group in which the numbering of parentheses in each
1209            alternative starts with the same number.
1211        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
1213     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
1216     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
1217        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
1218        for detecting groups that can match an empty string.
1220     9. A pattern with a very large number of alternatives (more than several
1221        hundred) was running out of internal workspace during the pre-compile
1222        phase, where pcre_compile() figures out how much memory will be needed. A
1223        bit of new cunning has reduced the workspace needed for groups with
1224        alternatives. The 1000-alternative test pattern now uses 12 bytes of
1225        workspace instead of running out of the 4096 that are available.
1227    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
1229    11. Applied patch from Google to remove an optimization that didn't quite work.
1230        The report of the bug said:
1232          pcrecpp::RE("a*").FullMatch("aaa") matches, while
1233          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
1234          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
1236    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
1237        it matched the wrong number of bytes.
1240    Version 7.1 24-Apr-07
1241    ---------------------
1243     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
1244        that is more "standard", making use of automake and other Autotools. There
1245        is some re-arrangement of the files and adjustment of comments consequent
1246        on this.
1248     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
1249        for recursive directory scanning broke on some systems because the files
1250        are not scanned in any specific order and on different systems the order
1251        was different. A call to "sort" has been inserted into RunGrepTest for the
1252        approprate test as a short-term fix. In the longer term there may be an
1253        alternative.
1255     3. I had an email from Eric Raymond about problems translating some of PCRE's
1256        man pages to HTML (despite the fact that I distribute HTML pages, some
1257        people do their own conversions for various reasons). The problems
1258        concerned the use of low-level troff macros .br and .in. I have therefore
1259        removed all such uses from the man pages (some were redundant, some could
1260        be replaced by .nf/.fi pairs). The 132html script that I use to generate
1261        HTML has been updated to handle .nf/.fi and to complain if it encounters
1262        .br or .in.
1264     4. Updated comments in configure.ac that get placed in config.h.in and also
1265        arranged for config.h to be included in the distribution, with the name
1266        config.h.generic, for the benefit of those who have to compile without
1267        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
1269     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
1270        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
1271        makevp.bat for the current PCRE, using the additional files
1272        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
1274     6. A Windows user reported a minor discrepancy with test 2, which turned out
1275        to be caused by a trailing space on an input line that had got lost in his
1276        copy. The trailing space was an accident, so I've just removed it.
1278     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
1279        that is needed.
1281     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
1282        as "const" (a) because they are and (b) because it helps the PHP
1283        maintainers who have recently made a script to detect big data structures
1284        in the php code that should be moved to the .rodata section. I remembered
1285        to update Builducptable as well, so it won't revert if ucptable.h is ever
1286        re-created.
1288     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
1289        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
1290        order to be able to cut out the UTF-8 tables in the latter when UTF-8
1291        support is not required. This saves 1.5-2K of code, which is important in
1292        some applications.
1294        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
1295        so as not to refer to the tables, even though these functions will never be
1296        called when UTF-8 support is disabled. Otherwise there are problems with a
1297        shared library.
1299    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
1301        (a) It was defining its arguments as char * instead of void *.
1303        (b) It was assuming that all moves were upwards in memory; this was true
1304            a long time ago when I wrote it, but is no longer the case.
1306        The emulated memove() is provided for those environments that have neither
1307        memmove() nor bcopy(). I didn't think anyone used it these days, but that
1308        is clearly not the case, as these two bugs were recently reported.
1310    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
1311        and Detrail to create the HTML documentation, the .txt form of the man
1312        pages, and it removes trailing spaces from listed files. It also creates
1313        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
1314        case, it wraps all the #defines with #ifndefs. This script should be run
1315        before "make dist".
1317    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1318        with Unicode property support.
1320        (a) For a maximizing quantifier, if the two different cases of the
1321            character were of different lengths in their UTF-8 codings (there are
1322            some cases like this - I found 11), and the matching function had to
1323            back up over a mixture of the two cases, it incorrectly assumed they
1324            were both the same length.
1326        (b) When PCRE was configured to use the heap rather than the stack for
1327            recursion during matching, it was not correctly preserving the data for
1328            the other case of a UTF-8 character when checking ahead for a match
1329            while processing a minimizing repeat. If the check also involved
1330            matching a wide character, but failed, corruption could cause an
1331            erroneous result when trying to check for a repeat of the original
1332            character.
1334    13. Some tidying changes to the testing mechanism:
1336        (a) The RunTest script now detects the internal link size and whether there
1337            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1338            values substituted by "configure". (The RunGrepTest script already did
1339            this for UTF-8.) The configure.ac script no longer substitutes the
1340            relevant variables.
1342        (b) The debugging options /B and /D in pcretest show the compiled bytecode
1343            with length and offset values. This means that the output is different
1344            for different internal link sizes. Test 2 is skipped for link sizes
1345            other than 2 because of this, bypassing the problem. Unfortunately,
1346            there was also a test in test 3 (the locale tests) that used /B and
1347            failed for link sizes other than 2. Rather than cut the whole test out,
1348            I have added a new /Z option to pcretest that replaces the length and
1349            offset values with spaces. This is now used to make test 3 independent
1350            of link size. (Test 2 will be tidied up later.)
1352    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1353        segmentation fault instead of returning the appropriate error message.
1355    15. In multiline mode when the newline sequence was set to "any", the pattern
1356        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1357        This doesn't seem right; it now treats the CRLF combination as the line
1358        ending, and so does not match in that case. It's only a pattern such as ^$
1359        that would hit this one: something like ^ABC$ would have failed after \r
1360        and then tried again after \r\n.
1362    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1363        in an attempt to make files that differ only in their line terminators
1364        compare equal. This works on Linux.
1366    17. Under certain error circumstances pcregrep might try to free random memory
1367        as it exited. This is now fixed, thanks to valgrind.
1369    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1370        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1371        was because its rules for how to advance for /g after matching an empty
1372        string at the end of a line did not allow for this case. They now check for
1373        it specially.
1375    20. pcretest is supposed to handle patterns and data of any length, by
1376        extending its buffers when necessary. It was getting this wrong when the
1377        buffer for a data line had to be extended.
1379    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1380        CRLF as a newline sequence.
1382    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1383        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1384        I have nevertheless tidied it up.
1386    23. Added some casts to kill warnings from HP-UX ia64 compiler.
1388    24. Added a man page for pcre-config.
1391    Version 7.0 19-Dec-06
1392    ---------------------
1394     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1395        moving to gcc 4.1.1.
1397     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1398        sys/time.h, which is documented as needed for this function. It doesn't
1399        seem to matter on Linux, but it showed up on some releases of OS X.
1401     3. It seems that there are systems where bytes whose values are greater than
1402        127 match isprint() in the "C" locale. The "C" locale should be the
1403        default when a C program starts up. In most systems, only ASCII printing
1404        characters match isprint(). This difference caused the output from pcretest
1405        to vary, making some of the tests fail. I have changed pcretest so that:
1407        (a) When it is outputting text in the compiled version of a pattern, bytes
1408            other than 32-126 are always shown as hex escapes.
1410        (b) When it is outputting text that is a matched part of a subject string,
1411            it does the same, unless a different locale has been set for the match
1412            (using the /L modifier). In this case, it uses isprint() to decide.
1414     4. Fixed a major bug that caused incorrect computation of the amount of memory
1415        required for a compiled pattern when options that changed within the
1416        pattern affected the logic of the preliminary scan that determines the
1417        length. The relevant options are -x, and -i in UTF-8 mode. The result was
1418        that the computed length was too small. The symptoms of this bug were
1419        either the PCRE error "internal error: code overflow" from pcre_compile(),
1420        or a glibc crash with a message such as "pcretest: free(): invalid next
1421        size (fast)". Examples of patterns that provoked this bug (shown in
1422        pcretest format) are:
1424          /(?-x: )/x
1425          /(?x)(?-x: \s*#\s*)/
1426          /((?i)[\x{c0}])/8
1427          /(?i:[\x{c0}])/8
1429        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1430        is now done differently.
1432     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1433        wrapper classes; (b) implement a new function in the C++ scanner that is
1434        more efficient than the old way of doing things because it avoids levels of
1435        recursion in the regex matching; (c) add a paragraph to the documentation
1436        for the FullMatch() function.
1438     6. The escape sequence \n was being treated as whatever was defined as
1439        "newline". Not only was this contrary to the documentation, which states
1440        that \n is character 10 (hex 0A), but it also went horribly wrong when
1441        "newline" was defined as CRLF. This has been fixed.
1443     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1444        was being set to -1 for the "end of line" case (supposedly a value that no
1445        character can have). Though this value is never used (the check for end of
1446        line is "zero bytes in current character"), it caused compiler complaints.
1447        I've changed it to 0xffffffff.
1449     8. In pcre_version.c, the version string was being built by a sequence of
1450        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1451        string (as it is for production releases) called a macro with an empty
1452        argument. The C standard says the result of this is undefined. The gcc
1453        compiler treats it as an empty string (which was what was wanted) but it is
1454        reported that Visual C gives an error. The source has been hacked around to
1455        avoid this problem.
1457     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1458        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1459        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1460        of them did).
1462    10. Originally, pcretest opened its input and output without "b"; then I was
1463        told that "b" was needed in some environments, so it was added for release
1464        5.0 to both the input and output. (It makes no difference on Unix-like
1465        systems.) Later I was told that it is wrong for the input on Windows. I've
1466        now abstracted the modes into two macros, to make it easier to fiddle with
1467        them, and removed "b" from the input mode under Windows.
1469    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1471    12. Added -help and --help to pcretest as an official way of being reminded
1472        of the options.
1474    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1475        and pcrecpp.cc because they annoy compilers at high warning levels.
1477    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1479    15. Fixed an occurrence of == in configure.ac that should have been = (shell
1480        scripts are not C programs :-) and which was not noticed because it works
1481        on Linux.
1483    16. pcretest is supposed to handle any length of pattern and data line (as one
1484        line or as a continued sequence of lines) by extending its input buffer if
1485        necessary. This feature was broken for very long pattern lines, leading to
1486        a string of junk being passed to pcre_compile() if the pattern was longer
1487        than about 50K.
1489    17. I have done a major re-factoring of the way pcre_compile() computes the
1490        amount of memory needed for a compiled pattern. Previously, there was code
1491        that made a preliminary scan of the pattern in order to do this. That was
1492        OK when PCRE was new, but as the facilities have expanded, it has become
1493        harder and harder to keep it in step with the real compile phase, and there
1494        have been a number of bugs (see for example, 4 above). I have now found a
1495        cunning way of running the real compile function in a "fake" mode that
1496        enables it to compute how much memory it would need, while actually only
1497        ever using a few hundred bytes of working memory and without too many
1498        tests of the mode. This should make future maintenance and development
1499        easier. A side effect of this work is that the limit of 200 on the nesting
1500        depth of parentheses has been removed (though this was never a serious
1501        limitation, I suspect). However, there is a downside: pcre_compile() now
1502        runs more slowly than before (30% or more, depending on the pattern). I
1503        hope this isn't a big issue. There is no effect on runtime performance.
1505    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1506        newline (only possible for the last line of a file) and it was a
1507        pattern that set a locale (followed by /Lsomething), pcretest crashed.
1509    19. Added additional timing features to pcretest. (1) The -tm option now times
1510        matching only, not compiling. (2) Both -t and -tm can be followed, as a
1511        separate command line item, by a number that specifies the number of
1512        repeats to use when timing. The default is 50000; this gives better
1513        precision, but takes uncomfortably long for very large patterns.
1515    20. Extended pcre_study() to be more clever in cases where a branch of a
1516        subpattern has no definite first character. For example, (a*|b*)[cd] would
1517        previously give no result from pcre_study(). Now it recognizes that the
1518        first character must be a, b, c, or d.
1520    21. There was an incorrect error "recursive call could loop indefinitely" if
1521        a subpattern (or the entire pattern) that was being tested for matching an
1522        empty string contained only one non-empty item after a nested subpattern.
1523        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1524        incorrectly, because the \d was being skipped in the check.
1526    22. The pcretest program now has a new pattern option /B and a command line
1527        option -b, which is equivalent to adding /B to every pattern. This causes
1528        it to show the compiled bytecode, without the additional information that
1529        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1530        is the same as /B/I).
1532    23. A new optimization is now able automatically to treat some sequences such
1533        as a*b as a*+b. More specifically, if something simple (such as a character
1534        or a simple class like \d) has an unlimited quantifier, and is followed by
1535        something that cannot possibly match the quantified thing, the quantifier
1536        is automatically "possessified".
1538    24. A recursive reference to a subpattern whose number was greater than 39
1539        went wrong under certain circumstances in UTF-8 mode. This bug could also
1540        have affected the operation of pcre_study().
1542    25. Realized that a little bit of performance could be had by replacing
1543        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1545    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1547    27. Possessive quantifiers such as a++ were previously implemented by turning
1548        them into atomic groups such as ($>a+). Now they have their own opcodes,
1549        which improves performance. This includes the automatically created ones
1550        from 23 above.
1552    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1553        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1554        the first matched character to be a colon. This applied both to named and
1555        numbered groups.
1557    29. The ucpinternal.h header file was missing its idempotency #ifdef.
1559    30. I was sent a "project" file called libpcre.a.dev which I understand makes
1560        building PCRE on Windows easier, so I have included it in the distribution.
1562    31. There is now a check in pcretest against a ridiculously large number being
1563        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1564        loop, the loop is abandoned.
1566    32. Forward references to subpatterns in conditions such as (?(2)...) where
1567        subpattern 2 is defined later cause pcre_compile() to search forwards in
1568        the pattern for the relevant set of parentheses. This search went wrong
1569        when there were unescaped parentheses in a character class, parentheses
1570        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1572    33. "Subroutine" calls and backreferences were previously restricted to
1573        referencing subpatterns earlier in the regex. This restriction has now
1574        been removed.
1576    34. Added a number of extra features that are going to be in Perl 5.10. On the
1577        whole, these are just syntactic alternatives for features that PCRE had
1578        previously implemented using the Python syntax or my own invention. The
1579        other formats are all retained for compatibility.
1581        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1582            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1583            also .NET compatible.
1585        (b) A recursion or subroutine call to a named group can now be defined as
1586            (?&name) as well as (?P>name).
1588        (c) A backreference to a named group can now be defined as \k<name> or
1589            \k'name' as well as (?P=name). The new forms, as well as being in Perl
1590            5.10, are also .NET compatible.
1592        (d) A conditional reference to a named group can now use the syntax
1593            (?(<name>) or (?('name') as well as (?(name).
1595        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1596            groups (named and numbered) that are never evaluated inline, but can be
1597            called as "subroutines" from elsewhere. In effect, the DEFINE condition
1598            is always false. There may be only one alternative in such a group.
1600        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1601            as the simple (?(R). The condition is true only if the most recent
1602            recursion is that of the given number or name. It does not search out
1603            through the entire recursion stack.
1605        (g) The escape \gN or \g{N} has been added, where N is a positive or
1606            negative number, specifying an absolute or relative reference.
1608    35. Tidied to get rid of some further signed/unsigned compiler warnings and
1609        some "unreachable code" warnings.
1611    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1612        things, this adds five new scripts.
1614    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1615        There were also incompatibilities regarding the handling of \Q..\E inside
1616        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1617        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1619    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1620        matches an empty string, and forcibly breaks the loop. There were bugs in
1621        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
1622        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
1623        separate and independent bugs (that affected different cases) have been
1624        fixed.
1626    39. Refactored the code to abolish the use of different opcodes for small
1627        capturing bracket numbers. This is a tidy that I avoided doing when I
1628        removed the limit on the number of capturing brackets for 3.5 back in 2001.
1629        The new approach is not only tidier, it makes it possible to reduce the
1630        memory needed to fix the previous bug (38).
1632    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1633        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1634        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1635        mode.
1637    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1638        report.
1640    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1641        copy construction and assignment in the C++ wrapper.
1643    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1644        couple of bugs that could have given wrong results in the "--newline=crlf"
1645        case.
1647    44. Added a number of casts and did some reorganization of signed/unsigned int
1648        variables following suggestions from Dair Grant. Also renamed the variable
1649        "this" as "item" because it is a C++ keyword.
1651    45. Arranged for dftables to add
1653          #include "pcre_internal.h"
1655        to pcre_chartables.c because without it, gcc 4.x may remove the array
1656        definition from the final binary if PCRE is built into a static library and
1657        dead code stripping is activated.
1659    46. For an unanchored pattern, if a match attempt fails at the start of a
1660        newline sequence, and the newline setting is CRLF or ANY, and the next two
1661        characters are CRLF, advance by two characters instead of one.
1664    Version 6.7 04-Jul-06
1665    ---------------------
1667     1. In order to handle tests when input lines are enormously long, pcretest has
1668        been re-factored so that it automatically extends its buffers when
1669        necessary. The code is crude, but this _is_ just a test program. The
1670        default size has been increased from 32K to 50K.
1672     2. The code in pcre_study() was using the value of the re argument before
1673        testing it for NULL. (Of course, in any sensible call of the function, it
1674        won't be NULL.)
1676     3. The memmove() emulation function in pcre_internal.h, which is used on
1677        systems that lack both memmove() and bcopy() - that is, hardly ever -
1678        was missing a "static" storage class specifier.
1680     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1681        containing an extended class (one that cannot be represented by a bitmap
1682        because it contains high-valued characters or Unicode property items, e.g.
1683        [\pZ]). Almost always one would set UTF-8 mode when processing such a
1684        pattern, but PCRE should not loop if you do not (it no longer does).
1685        [Detail: two cases were found: (a) a repeated subpattern containing an
1686        extended class; (b) a recursive reference to a subpattern that followed a
1687        previous extended class. It wasn't skipping over the extended class
1688        correctly when UTF-8 mode was not set.]
1690     5. A negated single-character class was not being recognized as fixed-length
1691        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1692        compile error "lookbehind assertion is not fixed length".
1694     6. The RunPerlTest auxiliary script was showing an unexpected difference
1695        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1696        write a Perl script that can interpret lines of an input file either as
1697        byte characters or as UTF-8, which is what "perltest" was being required to
1698        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1699        can't do is switch easily at run time between having the "use utf8;" pragma
1700        or not. In the end, I fudged it by using the RunPerlTest script to insert
1701        "use utf8;" explicitly for the UTF-8 tests.
1703     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1704        the end of the subject string, contrary to the documentation and to what
1705        Perl does. This was true of both matching functions. Now it matches only at
1706        the start of the subject and immediately after *internal* newlines.
1708     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1709        a pointer to an int instead of a pointer to an unsigned long int. This
1710        caused problems on 64-bit systems.
1712     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1713        instance of the 'standard' template library not being so standard".
1715    10. There was no check on the number of named subpatterns nor the maximum
1716        length of a subpattern name. The product of these values is used to compute
1717        the size of the memory block for a compiled pattern. By supplying a very
1718        long subpattern name and a large number of named subpatterns, the size
1719        computation could be caused to overflow. This is now prevented by limiting
1720        the length of names to 32 characters, and the number of named subpatterns
1721        to 10,000.
1723    11. Subpatterns that are repeated with specific counts have to be replicated in
1724        the compiled pattern. The size of memory for this was computed from the
1725        length of the subpattern and the repeat count. The latter is limited to
1726        65535, but there was no limit on the former, meaning that integer overflow
1727        could in principle occur. The compiled length of a repeated subpattern is
1728        now limited to 30,000 bytes in order to prevent this.
1730    12. Added the optional facility to have named substrings with the same name.
1732    13. Added the ability to use a named substring as a condition, using the
1733        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1734        are numbers (not recommended). Forward references are permitted.
1736    14. Added forward references in named backreferences (if you see what I mean).
1738    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1739        pattern could run off the end of the subject. For example, the pattern
1740        "(?s)(.{1,5})"8 did this with the subject "ab".
1742    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1743        PCRE_CASELESS was set when matching characters that were quantified with ?
1744        or *.
1746    17. A character class other than a single negated character that had a minimum
1747        but no maximum quantifier - for example [ab]{6,} - was not handled
1748        correctly by pce_dfa_exec(). It would match only one character.
1750    18. A valid (though odd) pattern that looked like a POSIX character
1751        class but used an invalid character after [ (for example [[,abc,]]) caused
1752        pcre_compile() to give the error "Failed: internal error: code overflow" or
1753        in some cases to crash with a glibc free() error. This could even happen if
1754        the pattern terminated after [[ but there just happened to be a sequence of
1755        letters, a binary zero, and a closing ] in the memory that followed.
1757    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1758        over the years. Originally (before any Unicode support), just the bottom 8
1759        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1760        output from "man perlunicode" includes this:
1762          The regular expression compiler produces polymorphic opcodes.  That
1763          is, the pattern adapts to the data and automatically switches to
1764          the Unicode character scheme when presented with Unicode data--or
1765          instead uses a traditional byte scheme when presented with byte
1766          data.
1768        Sadly, a wide octal escape does not cause a switch, and in a string with
1769        no other multibyte characters, these octal escapes are treated as before.
1770        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
1771        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1772        Unicode string.
1774        I have not perpetrated such confusion in PCRE. Up till now, it took just
1775        the bottom 8 bits, as in old Perl. I have now made octal escapes with
1776        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1777        translate to the appropriate multibyte character.
1779    29. Applied some refactoring to reduce the number of warnings from Microsoft
1780        and Borland compilers. This has included removing the fudge introduced
1781        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1782        a warning about an unused variable.
1784    21. PCRE has not included VT (character 0x0b) in the set of whitespace
1785        characters since release 4.0, because Perl (from release 5.004) does not.
1786        [Or at least, is documented not to: some releases seem to be in conflict
1787        with the documentation.] However, when a pattern was studied with
1788        pcre_study() and all its branches started with \s, PCRE still included VT
1789        as a possible starting character. Of course, this did no harm; it just
1790        caused an unnecessary match attempt.
1792    22. Removed a now-redundant internal flag bit that recorded the fact that case
1793        dependency changed within the pattern. This was once needed for "required
1794        byte" processing, but is no longer used. This recovers a now-scarce options
1795        bit. Also moved the least significant internal flag bit to the most-
1796        significant bit of the word, which was not previously used (hangover from
1797        the days when it was an int rather than a uint) to free up another bit for
1798        the future.
1800    23. Added support for CRLF line endings as well as CR and LF. As well as the
1801        default being selectable at build time, it can now be changed at runtime
1802        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1803        specify that it is scanning data with non-default line endings.
1805    24. Changed the definition of CXXLINK to make it agree with the definition of
1806        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1808    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1809        recursions. This makes a big different to stack usage for some patterns.
1811    26. If a subpattern containing a named recursion or subroutine reference such
1812        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1813        the space required for the compiled pattern went wrong and gave too small a
1814        value. Depending on the environment, this could lead to "Failed: internal
1815        error: code overflow at offset 49" or "glibc detected double free or
1816        corruption" errors.
1818    27. Applied patches from Google (a) to support the new newline modes and (b) to
1819        advance over multibyte UTF-8 characters in GlobalReplace.
1821    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1822        difference for some implementation of PCRE in some Windows version.
1824    29. Added some extra testing facilities to pcretest:
1826        \q<number>   in a data line sets the "match limit" value
1827        \Q<number>   in a data line sets the "match recursion limt" value
1828        -S <number>  sets the stack size, where <number> is in megabytes
1830        The -S option isn't available for Windows.
1833    Version 6.6 06-Feb-06
1834    ---------------------
1836     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1837        in pcreposix.h. I have copied the definition from pcre.h.
1839     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1840        because pcre.h is no longer a built file.
1842     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1843        not normally included in the compiled code.
1846    Version 6.5 01-Feb-06
1847    ---------------------
1849     1. When using the partial match feature with pcre_dfa_exec(), it was not
1850        anchoring the second and subsequent partial matches at the new starting
1851        point. This could lead to incorrect results. For example, with the pattern
1852        /1234/, partially matching against "123" and then "a4" gave a match.
1854     2. Changes to pcregrep:
1856        (a) All non-match returns from pcre_exec() were being treated as failures
1857            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1858            error message is output. Some extra information is given for the
1859            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
1860            probably the only errors that are likely to be caused by users (by
1861            specifying a regex that has nested indefinite repeats, for instance).
1862            If there are more than 20 of these errors, pcregrep is abandoned.
1864        (b) A binary zero was treated as data while matching, but terminated the
1865            output line if it was written out. This has been fixed: binary zeroes
1866            are now no different to any other data bytes.
1868        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1869            used to set a locale for matching. The --locale=xxxx long option has
1870            been added (no short equivalent) to specify a locale explicitly on the
1871            pcregrep command, overriding the environment variables.
1873        (d) When -B was used with -n, some line numbers in the output were one less
1874            than they should have been.
1876        (e) Added the -o (--only-matching) option.
1878        (f) If -A or -C was used with -c (count only), some lines of context were
1879            accidentally printed for the final match.
1881        (g) Added the -H (--with-filename) option.
1883        (h) The combination of options -rh failed to suppress file names for files
1884            that were found from directory arguments.
1886        (i) Added the -D (--devices) and -d (--directories) options.
1888        (j) Added the -F (--fixed-strings) option.
1890        (k) Allow "-" to be used as a file name for -f as well as for a data file.
1892        (l) Added the --colo(u)r option.
1894        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1895            is not present by default.
1897     3. A nasty bug was discovered in the handling of recursive patterns, that is,
1898        items such as (?R) or (?1), when the recursion could match a number of
1899        alternatives. If it matched one of the alternatives, but subsequently,
1900        outside the recursion, there was a failure, the code tried to back up into
1901        the recursion. However, because of the way PCRE is implemented, this is not
1902        possible, and the result was an incorrect result from the match.
1904        In order to prevent this happening, the specification of recursion has
1905        been changed so that all such subpatterns are automatically treated as
1906        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1908     4. I had overlooked the fact that, in some locales, there are characters for
1909        which isalpha() is true but neither isupper() nor islower() are true. In
1910        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1911        and ordfeminine) are like this. This affected the treatment of \w and \W
1912        when they appeared in character classes, but not when they appeared outside
1913        a character class. The bit map for "word" characters is now created
1914        separately from the results of isalnum() instead of just taking it from the
1915        upper, lower, and digit maps. (Plus the underscore character, of course.)
1917     5. The above bug also affected the handling of POSIX character classes such as
1918        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1919        permanent tables. Instead, the bit maps for such a class were previously
1920        created as the appropriate unions of the upper, lower, and digit bitmaps.
1921        Now they are created by subtraction from the [[:word:]] class, which has
1922        its own bitmap.
1924     6. The [[:blank:]] character class matches horizontal, but not vertical space.
1925        It is created by subtracting the vertical space characters (\x09, \x0a,
1926        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1927        subtraction was done in the overall bitmap for a character class, meaning
1928        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1929        be recognized. This bug has been fixed.
1931     7. Patches from the folks at Google:
1933          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1934          real life, but is still worth protecting against".
1936          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1937          regular expressions".
1939          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1940          have it.
1942          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1943          "configure" and the latter not, in order to fix a problem somebody had
1944          with compiling the Arg class on HP-UX.
1946          (e) Improve the error-handling of the C++ wrapper a little bit.
1948          (f) New tests for checking recursion limiting.
1950     8. The pcre_memmove() function, which is used only if the environment does not
1951        have a standard memmove() function (and is therefore rarely compiled),
1952        contained two bugs: (a) use of int instead of size_t, and (b) it was not
1953        returning a result (though PCRE never actually uses the result).
1955     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1956        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1957        returned instead of calling malloc() with an overflowing number that would
1958        most likely cause subsequent chaos.
1960    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1962    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1963        with this option is matched, the nmatch and pmatch options of regexec() are
1964        ignored.
1966    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1967        provided in case anyone wants to the the POSIX interface with UTF-8
1968        strings.
1970    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1971        C++ linking (needed for some HP-UX environments).
1973    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1974        (unused parameter) and in the pcre_printint() function (omitted "default"
1975        switch label when the default is to do nothing).
1977    15. Added some code to make it possible, when PCRE is compiled as a C++
1978        library, to replace subject pointers for pcre_exec() with a smart pointer
1979        class, thus making it possible to process discontinuous strings.
1981    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1982        much the same function. They were added by different people who were trying
1983        to make PCRE easy to compile on non-Unix systems. It has been suggested
1984        that PCRE_EXPORT be abolished now that there is more automatic apparatus
1985        for compiling on Windows systems. I have therefore replaced it with
1986        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1987        defaults to "extern" for C or "extern C" for C++, which works fine on
1988        Unix-like systems. It is now possible to override the value of PCRE_DATA_
1989        SCOPE with something explicit in config.h. In addition:
1991        (a) pcreposix.h still had just "extern" instead of either of these macros;
1992            I have replaced it with PCRE_DATA_SCOPE.
1994        (b) Functions such as _pcre_xclass(), which are internal to the library,
1995            but external in the C sense, all had PCRE_EXPORT in their definitions.
1996            This is apparently wrong for the Windows case, so I have removed it.
1997            (It makes no difference on Unix-like systems.)
1999    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
2000        of recursive calls to match(). This is different to MATCH_LIMIT because
2001        that limits the total number of calls to match(), not all of which increase
2002        the depth of recursion. Limiting the recursion depth limits the amount of
2003        stack (or heap if NO_RECURSE is set) that is used. The default can be set
2004        when PCRE is compiled, and changed at run time. A patch from Google adds
2005        this functionality to the C++ interface.
2007    18. Changes to the handling of Unicode character properties:
2009        (a) Updated the table to Unicode 4.1.0.
2011        (b) Recognize characters that are not in the table as "Cn" (undefined).
2013        (c) I revised the way the table is implemented to a much improved format
2014            which includes recognition of ranges. It now supports the ranges that
2015            are defined in UnicodeData.txt, and it also amalgamates other
2016            characters into ranges. This has reduced the number of entries in the
2017            table from around 16,000 to around 3,000, thus reducing its size
2018            considerably. I realized I did not need to use a tree structure after
2019            all - a binary chop search is just as efficient. Having reduced the
2020            number of entries, I extended their size from 6 bytes to 8 bytes to
2021            allow for more data.
2023        (d) Added support for Unicode script names via properties such as \p{Han}.
2025    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
2026        matching that character.
2028    20. When matching a repeated Unicode property with a minimum greater than zero,
2029        (for example \pL{2,}), PCRE could look past the end of the subject if it
2030        reached it while seeking the minimum number of characters. This could
2031        happen only if some of the characters were more than one byte long, because
2032        there is a check for at least the minimum number of bytes.
2034    21. Refactored the implementation of \p and \P so as to be more general, to
2035        allow for more different types of property in future. This has changed the
2036        compiled form incompatibly. Anybody with saved compiled patterns that use
2037        \p or \P will have to recompile them.
2039    22. Added "Any" and "L&" to the supported property types.
2041    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
2042        but give a compile time error if the value is greater than 0xff.
2044    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
2045        accidentally not being installed or uninstalled.
2047    25. The pcre.h file was built from pcre.h.in, but the only changes that were
2048        made were to insert the current release number. This seemed silly, because
2049        it made things harder for people building PCRE on systems that don't run
2050        "configure". I have turned pcre.h into a distributed file, no longer built
2051        by "configure", with the version identification directly included. There is
2052        no longer a pcre.h.in file.
2054        However, this change necessitated a change to the pcre-config script as
2055        well. It is built from pcre-config.in, and one of the substitutions was the
2056        release number. I have updated configure.ac so that ./configure now finds
2057        the release number by grepping pcre.h.
2059    26. Added the ability to run the tests under valgrind.
2062    Version 6.4 05-Sep-05
2063    ---------------------
2065     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
2066        "--" to be printed when multiple files were scanned, even when none of the
2067        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
2068        consider it to be a bug, and have restored the previous behaviour.
2070     2. A couple of code tidies to get rid of compiler warnings.
2072     3. The pcretest program used to cheat by referring to symbols in the library
2073        whose names begin with _pcre_. These are internal symbols that are not
2074        really supposed to be visible externally, and in some environments it is
2075        possible to suppress them. The cheating is now confined to including
2076        certain files from the library's source, which is a bit cleaner.
2078     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
2079        file's purpose clearer.
2081     5. Reorganized pcre_ucp_findchar().
2084    Version 6.3 15-Aug-05
2085    ---------------------
2087     1. The file libpcre.pc.in did not have general read permission in the tarball.
2089     2. There were some problems when building without C++ support:
2091        (a) If C++ support was not built, "make install" and "make test" still
2092            tried to test it.
2094        (b) There were problems when the value of CXX was explicitly set. Some
2095            changes have been made to try to fix these, and ...
2097        (c) --disable-cpp can now be used to explicitly disable C++ support.
2099        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
2100            backslash in a target when C++ was disabled. This confuses some
2101            versions of "make", apparently. Using an intermediate variable solves
2102            this. (Same for CPP_LOBJ.)
2104     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
2105        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
2106        necessary on certain architectures.
2108     4. Added a setting of -export-symbols-regex to the link command to remove
2109        those symbols that are exported in the C sense, but actually are local
2110        within the library, and not documented. Their names all begin with
2111        "_pcre_". This is not a perfect job, because (a) we have to except some
2112        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
2113        available (and never for static libraries). I have made a note to try to
2114        find a way round (a) in the future.
2117    Version 6.2 01-Aug-05
2118    ---------------------
2120     1. There was no test for integer overflow of quantifier values. A construction
2121        such as {1111111111111111} would give undefined results. What is worse, if
2122        a minimum quantifier for a parenthesized subpattern overflowed and became
2123        negative, the calculation of the memory size went wrong. This could have
2124        led to memory overwriting.
2126     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
2128     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
2129        operating environments where this matters.
2131     4. Applied Giuseppe Maxia's patch to add additional features for controlling
2132        PCRE options from within the C++ wrapper.
2134     5. Named capturing subpatterns were not being correctly counted when a pattern
2135        was compiled. This caused two problems: (a) If there were more than 100
2136        such subpatterns, the calculation of the memory needed for the whole
2137        compiled pattern went wrong, leading to an overflow error. (b) Numerical
2138        back references of the form \12, where the number was greater than 9, were
2139        not recognized as back references, even though there were sufficient
2140        previous subpatterns.
2142     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
2143        versions of gcc, e.g. 2.95.4.
2146    Version 6.1 21-Jun-05
2147    ---------------------
2149     1. There was one reference to the variable "posix" in pcretest.c that was not
2150        surrounded by "#if !defined NOPOSIX".
2152     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
2153        the cross-check on the old pcre_info() function, for the benefit of the
2154        cut-down version of PCRE that is currently imported into Exim.
2156     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
2157        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
2158        patterns that start (?i) but I don't think that matters. The use of (?i) is
2159        just an example; this all applies to the other options as well.
2161     4. Since libtool seems to echo the compile commands it is issuing, the output
2162        from "make" can be reduced a bit by putting "@" in front of each libtool
2163        compile command.
2165     5. Patch from the folks at Google for configure.in to be a bit more thorough
2166        in checking for a suitable C++ installation before trying to compile the
2167        C++ stuff. This should fix a reported problem when a compiler was present,
2168        but no suitable headers.
2170     6. The man pages all had just "PCRE" as their title. I have changed them to
2171        be the relevant file name. I have also arranged that these names are
2172        retained in the file doc/pcre.txt, which is a concatenation in text format
2173        of all the man pages except the little individual ones for each function.
2175     7. The NON-UNIX-USE file had not been updated for the different set of source
2176        files that come with release 6. I also added a few comments about the C++
2177        wrapper.
2180    Version 6.0 07-Jun-05
2181    ---------------------
2183     1. Some minor internal re-organization to help with my DFA experiments.
2185     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
2186        didn't matter for the library itself when fully configured, but did matter
2187        when compiling without UCP support, or within Exim, where the ucp files are
2188        not imported.
2190     3. Refactoring of the library code to split up the various functions into
2191        different source modules. The addition of the new DFA matching code (see
2192        below) to a single monolithic source would have made it really too
2193        unwieldy, quite apart from causing all the code to be include in a
2194        statically linked application, when only some functions are used. This is
2195        relevant even without the DFA addition now that patterns can be compiled in
2196        one application and matched in another.
2198        The downside of splitting up is that there have to be some external
2199        functions and data tables that are used internally in different modules of
2200        the library but which are not part of the API. These have all had their
2201        names changed to start with "_pcre_" so that they are unlikely to clash
2202        with other external names.
2204     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
2205        a different (DFA) algorithm. Although it is slower than the original
2206        function, it does have some advantages for certain types of matching
2207        problem.
2209     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
2210        including restarting after a partial match.
2212     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
2213        defined when compiling for Windows was sent to me. I have put it into the
2214        code, though I have no means of testing or verifying it.
2216     7. Added the pcre_refcount() auxiliary function.
2218     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
2219        match before or at the first newline in the subject string. In pcretest,
2220        the /f option on a pattern can be used to set this.
2222     9. A repeated \w when used in UTF-8 mode with characters greater than 256
2223        would behave wrongly. This has been present in PCRE since release 4.0.
2225    10. A number of changes to the pcregrep command:
2227        (a) Refactored how -x works; insert ^(...)$ instead of setting
2228            PCRE_ANCHORED and checking the length, in preparation for adding
2229            something similar for -w.
2231        (b) Added the -w (match as a word) option.
2233        (c) Refactored the way lines are read and buffered so as to have more
2234            than one at a time available.
2236        (d) Implemented a pcregrep test script.
2238        (e) Added the -M (multiline match) option. This allows patterns to match
2239            over several lines of the subject. The buffering ensures that at least
2240            8K, or the rest of the document (whichever is the shorter) is available
2241            for matching (and similarly the previous 8K for lookbehind assertions).
2243        (f) Changed the --help output so that it now says
2245              -w, --word-regex(p)
2247            instead of two lines, one with "regex" and the other with "regexp"
2248            because that confused at least one person since the short forms are the
2249            same. (This required a bit of code, as the output is generated
2250            automatically from a table. It wasn't just a text change.)
2252        (g) -- can be used to terminate pcregrep options if the next thing isn't an
2253            option but starts with a hyphen. Could be a pattern or a path name
2254            starting with a hyphen, for instance.
2256        (h) "-" can be given as a file name to represent stdin.
2258        (i) When file names are being printed, "(standard input)" is used for
2259            the standard input, for compatibility with GNU grep. Previously
2260            "<stdin>" was used.
2262        (j) The option --label=xxx can be used to supply a name to be used for
2263            stdin when file names are being printed. There is no short form.
2265        (k) Re-factored the options decoding logic because we are going to add
2266            two more options that take data. Such options can now be given in four
2267            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
2269        (l) Added the -A, -B, and -C options for requesting that lines of context
2270            around matches be printed.
2272        (m) Added the -L option to print the names of files that do not contain
2273            any matching lines, that is, the complement of -l.
2275        (n) The return code is 2 if any file cannot be opened, but pcregrep does
2276            continue to scan other files.
2278        (o) The -s option was incorrectly implemented. For compatibility with other
2279            greps, it now suppresses the error message for a non-existent or non-
2280            accessible file (but not the return code). There is a new option called
2281            -q that suppresses the output of matching lines, which was what -s was
2282            previously doing.
2284        (p) Added --include and --exclude options to specify files for inclusion
2285            and exclusion when recursing.
2287    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
2288        Hopefully, it now does.
2290    12. Missing cast in pcre_study().
2292    13. Added an "uninstall" target to the makefile.
2294    14. Replaced "extern" in the function prototypes in Makefile.in with
2295        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
2296        world, but is set differently for Windows.
2298    15. Added a second compiling function called pcre_compile2(). The only
2299        difference is that it has an extra argument, which is a pointer to an
2300        integer error code. When there is a compile-time failure, this is set
2301        non-zero, in addition to the error test pointer being set to point to an
2302        error message. The new argument may be NULL if no error number is required
2303        (but then you may as well call pcre_compile(), which is now just a
2304        wrapper). This facility is provided because some applications need a
2305        numeric error indication, but it has also enabled me to tidy up the way
2306        compile-time errors are handled in the POSIX wrapper.
2308    16. Added VPATH=.libs to the makefile; this should help when building with one
2309        prefix path and installing with another. (Or so I'm told by someone who
2310        knows more about this stuff than I do.)
2312    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
2313        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
2314        match everything, including newlines. This is not POSIX-compatible, but
2315        somebody wanted the feature. From pcretest it can be activated by using
2316        both the P and the s flags.
2318    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
2320    19. libpcre.pc was being incorrectly installed as executable.
2322    20. A couple of places in pcretest check for end-of-line by looking for '\n';
2323        it now also looks for '\r' so that it will work unmodified on Windows.
2325    21. Added Google's contributed C++ wrapper to the distribution.
2327    22. Added some untidy missing memory free() calls in pcretest, to keep
2328        Electric Fence happy when testing.
2332    Version 5.0 13-Sep-04
2333    ---------------------
2335     1. Internal change: literal characters are no longer packed up into items
2336        containing multiple characters in a single byte-string. Each character
2337        is now matched using a separate opcode. However, there may be more than one
2338        byte in the character in UTF-8 mode.
2340     2. The pcre_callout_block structure has two new fields: pattern_position and
2341        next_item_length. These contain the offset in the pattern to the next match
2342        item, and its length, respectively.
2344     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
2345        insertion of callouts before each pattern item. Added the /C option to
2346        pcretest to make use of this.
2348     4. On the advice of a Windows user, the lines
2350          #if defined(_WIN32) || defined(WIN32)
2351          _setmode( _fileno( stdout ), 0x8000 );
2352          #endif  /* defined(_WIN32) || defined(WIN32) */
2354        have been added to the source of pcretest. This apparently does useful
2355        magic in relation to line terminators.
2357     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
2358        for the benefit of those environments where the "b" makes a difference.
2360     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
2361        to know about it. I have put a hack into configure.in that adds in code
2362        to set GCC=yes if CC=icc. This seems to end up at a point in the
2363        generated configure script that is early enough to affect the setting of
2364        compiler options, which is what is needed, but I have no means of testing
2365        whether it really works. (The user who reported this had patched the
2366        generated configure script, which of course I cannot do.)
2368        LATER: After change 22 below (new libtool files), the configure script
2369        seems to know about icc (and also ecc). Therefore, I have commented out
2370        this hack in configure.in.
2372     7. Added support for pkg-config (2 patches were sent in).
2374     8. Negated POSIX character classes that used a combination of internal tables
2375        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2376        [[:^ascii]]. Typically, they would match almost any characters. The other
2377        POSIX classes were not broken in this way.
2379     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2380        to find the match, as PCRE was deluded into thinking that the match had to
2381        start at the start point or following a newline. The same bug applied to
2382        patterns with negative forward assertions or any backward assertions
2383        preceding ".*" at the start, unless the pattern required a fixed first
2384        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2386    10. In UTF-8 mode, when moving forwards in the subject after a failed match
2387        starting at the last subject character, bytes beyond the end of the subject
2388        string were read.
2390    11. Renamed the variable "class" as "classbits" to make life easier for C++
2391        users. (Previously there was a macro definition, but it apparently wasn't
2392        enough.)
2394    12. Added the new field "tables" to the extra data so that tables can be passed
2395        in at exec time, or the internal tables can be re-selected. This allows
2396        a compiled regex to be saved and re-used at a later time by a different
2397        program that might have everything at different addresses.
2399    13. Modified the pcre-config script so that, when run on Solaris, it shows a
2400        -R library as well as a -L library.
2402    14. The debugging options of pcretest (-d on the command line or D on a
2403        pattern) showed incorrect output for anything following an extended class
2404        that contained multibyte characters and which was followed by a quantifier.
2406    15. Added optional support for general category Unicode character properties
2407        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2408        support. It adds about 90K to the size of the library. The meanings of the
2409        inbuilt class escapes such as \d and \s have NOT been changed.
2411    16. Updated pcredemo.c to include calls to free() to release the memory for the
2412        compiled pattern.
2414    17. The generated file chartables.c was being created in the source directory
2415        instead of in the building directory. This caused the build to fail if the
2416        source directory was different from the building directory, and was
2417        read-only.
2419    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2420        file. No doubt somebody will tell me if they don't make sense... Also added
2421        Dan Mooney's comments about building on OpenVMS.
2423    19. Added support for partial matching via the PCRE_PARTIAL option for
2424        pcre_exec() and the \P data escape in pcretest.
2426    20. Extended pcretest with 3 new pattern features:
2428        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
2429              write the compiled pattern to the file whose name is "rest-of-line".
2430              This is a straight binary dump of the data, with the saved pointer to
2431              the character tables forced to be NULL. The study data, if any, is
2432              written too. After writing, pcretest reads a new pattern.
2434        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2435              compiled pattern from the given file. There must not be any
2436              occurrences of "<" in the file name (pretty unlikely); if there are,
2437              pcretest will instead treat the initial "<" as a pattern delimiter.
2438              After reading in the pattern, pcretest goes on to read data lines as
2439              usual.
2441        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2442              and 16-bit fields in a compiled pattern, to simulate a pattern that
2443              was compiled on a host of opposite endianness.
2445    21. The pcre-exec() function can now cope with patterns that were compiled on
2446        hosts of opposite endianness, with this restriction:
2448          As for any compiled expression that is saved and used later, the tables
2449          pointer field cannot be preserved; the extra_data field in the arguments
2450          to pcre_exec() should be used to pass in a tables address if a value
2451          other than the default internal tables were used at compile time.
2453    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2454        now diagnosed as an error. Previously, most of the time, a negative number
2455        would have been treated as zero, but if in addition "ovector" was passed as
2456        NULL, a crash could occur.
2458    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2459        new versions from the libtool 1.5 distribution (the last one is a copy of
2460        a file called libtool.m4). This seems to have fixed the need to patch
2461        "configure" to support Darwin 1.3 (which I used to do). However, I still
2462        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2463        workstation).
2465    24. Changed the PCRE licence to be the more standard "BSD" licence.
2468    Version 4.5 01-Dec-03
2469    ---------------------
2471     1. There has been some re-arrangement of the code for the match() function so
2472        that it can be compiled in a version that does not call itself recursively.
2473        Instead, it keeps those local variables that need separate instances for
2474        each "recursion" in a frame on the heap, and gets/frees frames whenever it
2475        needs to "recurse". Keeping track of where control must go is done by means
2476        of setjmp/longjmp. The whole thing is implemented by a set of macros that
2477        hide most of the details from the main code, and operates only if
2478        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2479        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2480        operating.
2482        To make it easier for callers to provide specially tailored get/free
2483        functions for this usage, two new functions, pcre_stack_malloc, and
2484        pcre_stack_free, are used. They are always called in strict stacking order,
2485        and the size of block requested is always the same.
2487        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2488        PCRE has been compiled to use the stack or the heap for recursion. The
2489        -C option of pcretest uses this to show which version is compiled.
2491        A new data escape \S, is added to pcretest; it causes the amounts of store
2492        obtained and freed by both kinds of malloc/free at match time to be added
2493        to the output.
2495     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2496        what's available on my current Linux desktop machine.
2498     3. When matching a UTF-8 string, the test for a valid string at the start has
2499        been extended. If start_offset is not zero, PCRE now checks that it points
2500        to a byte that is the start of a UTF-8 character. If not, it returns
2501        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2502        this is necessary because there may be backward assertions in the pattern.
2503        When matching the same subject several times, it may save resources to use
2504        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2506     4. The code for checking the validity of UTF-8 strings has been tightened so
2507        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2508        containing "overlong sequences".
2510     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2511        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2512        should have been "&", but it just so happened that all the cases this let
2513        through by mistake were picked up later in the function.
2515     6. I had used a variable called "isblank" - this is a C99 function, causing
2516        some compilers to warn. To avoid this, I renamed it (as "blankclass").
2518     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2519        prompting; (b) run "./pcretest /dev/null" at the start of the test script
2520        so the version is shown; (c) stop "make test" echoing "./RunTest".
2522     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2524     9. The prototype for memmove() for systems that don't have it was using
2525        size_t, but the inclusion of the header that defines size_t was later. I've
2526        moved the #includes for the C headers earlier to avoid this.
2528    10. Added some adjustments to the code to make it easier to compiler on certain
2529        special systems:
2531          (a) Some "const" qualifiers were missing.
2532          (b) Added the macro EXPORT before all exported functions; by default this
2533              is defined to be empty.
2534          (c) Changed the dftables auxiliary program (that builds chartables.c) so
2535              that it reads its output file name as an argument instead of writing
2536              to the standard output and assuming this can be redirected.
2538    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2539        class containing characters with values greater than 255, PCRE compilation
2540        went into a loop.
2542    12. A recursive reference to a subpattern that was within another subpattern
2543        that had a minimum quantifier of zero caused PCRE to crash. For example,
2544        (x(y(?2))z)? provoked this bug with a subject that got as far as the
2545        recursion. If the recursively-called subpattern itself had a zero repeat,
2546        that was OK.
2548    13. In pcretest, the buffer for reading a data line was set at 30K, but the
2549        buffer into which it was copied (for escape processing) was still set at
2550        1024, so long lines caused crashes.
2552    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2553        "internal error: code overflow...". This applied to any character class
2554        that was followed by a possessive quantifier.
2556    15. Modified the Makefile to add libpcre.la as a prerequisite for
2557        libpcreposix.la because I was told this is needed for a parallel build to
2558        work.
2560    16. If a pattern that contained .* following optional items at the start was
2561        studied, the wrong optimizing data was generated, leading to matching
2562        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2563        matching string must start with a or b or c. The correct conclusion for
2564        this pattern is that a match can start with any character.
2567    Version 4.4 13-Aug-03
2568    ---------------------
2570     1. In UTF-8 mode, a character class containing characters with values between
2571        127 and 255 was not handled correctly if the compiled pattern was studied.
2572        In fixing this, I have also improved the studying algorithm for such
2573        classes (slightly).
2575     2. Three internal functions had redundant arguments passed to them. Removal
2576        might give a very teeny performance improvement.
2578     3. Documentation bug: the value of the capture_top field in a callout is *one
2579        more than* the number of the hightest numbered captured substring.
2581     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2582        in incorrectly linking with a previously installed version. They now link
2583        explicitly with libpcre.la.
2585     5. configure.in no longer needs to recognize Cygwin specially.
2587     6. A problem in pcre.in for Windows platforms is fixed.
2589     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2590        pcretest, it used to include the size of the study block as part of its
2591        output. Unfortunately, the structure contains a field that has a different
2592        size on different hardware architectures. This meant that the tests that
2593        showed this size failed. As the block is currently always of a fixed size,
2594        this information isn't actually particularly useful in pcretest output, so
2595        I have just removed it.
2597     8. Three pre-processor statements accidentally did not start in column 1.
2598        Sadly, there are *still* compilers around that complain, even though
2599        standard C has not required this for well over a decade. Sigh.
2601     9. In pcretest, the code for checking callouts passed small integers in the
2602        callout_data field, which is a void * field. However, some picky compilers
2603        complained about the casts involved for this on 64-bit systems. Now
2604        pcretest passes the address of the small integer instead, which should get
2605        rid of the warnings.
2607    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2608        both compile and run time, and gives an error if an invalid UTF-8 sequence
2609        is found. There is a option for disabling this check in cases where the
2610        string is known to be correct and/or the maximum performance is wanted.
2612    11. In response to a bug report, I changed one line in Makefile.in from
2614            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2615        to
2616            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2618        to look similar to other lines, but I have no way of telling whether this
2619        is the right thing to do, as I do not use Windows. No doubt I'll get told
2620        if it's wrong...
2623    Version 4.3 21-May-03
2624    ---------------------
2626    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2627       Makefile.
2629    2. Some refactoring to improve the quality of the code:
2631       (i)   The utf8_table... variables are now declared "const".
2633       (ii)  The code for \cx, which used the "case flipping" table to upper case
2634             lower case letters, now just substracts 32. This is ASCII-specific,
2635             but the whole concept of \cx is ASCII-specific, so it seems
2636             reasonable.
2638       (iii) PCRE was using its character types table to recognize decimal and
2639             hexadecimal digits in the pattern. This is silly, because it handles
2640             only 0-9, a-f, and A-F, but the character types table is locale-
2641             specific, which means strange things might happen. A private
2642             table is now used for this - though it costs 256 bytes, a table is
2643             much faster than multiple explicit tests. Of course, the standard
2644             character types table is still used for matching digits in subject
2645             strings against \d.
2647       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2648             ending in _t are). So I've renamed it as ESC_tee.
2650    3. The first argument for regexec() in the POSIX wrapper should have been
2651       defined as "const".
2653    4. Changed pcretest to use malloc() for its buffers so that they can be
2654       Electric Fenced for debugging.
2656    5. There were several places in the code where, in UTF-8 mode, PCRE would try
2657       to read one or more bytes before the start of the subject string. Often this
2658       had no effect on PCRE's behaviour, but in some circumstances it could
2659       provoke a segmentation fault.
2661    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2662       to try to read one or more bytes before the start of the subject string.
2664    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
2665       UTF-8 support could misbehave in various ways if the subject string
2666       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
2667       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
2668       back over UTF-8 characters.)
2671    Version 4.2 14-Apr-03
2672    ---------------------
2674    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
2676    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
2677         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
2678         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
2679         [WIN_PREFIX]: new variable, "cyg" for Cygwin
2680         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
2681           and BUILD_EXEEXT
2682         Note: automatic setting of the BUILD variables is not yet working
2683         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
2684           compile-time but not at link-time
2685         [LINK]: use for linking executables only
2686         make different versions for Windows and non-Windows
2687         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
2688           libraries
2689         [LINK_FOR_BUILD]: new variable
2690         [OBJEXT]: use throughout
2691         [EXEEXT]: use throughout
2692         <winshared>: new target
2693         <wininstall>: new target
2694         <dftables.o>: use native compiler
2695         <dftables>: use native linker
2696         <install>: handle Windows platform correctly
2697         <clean>: ditto
2698         <check>: ditto
2699         copy DLL to top builddir before testing
2701       As part of these changes, -no-undefined was removed again. This was reported
2702       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
2703       in any case.
2705    3. Some tidies to get rid of compiler warnings:
2707       . In the match_data structure, match_limit was an unsigned long int, whereas
2708         match_call_count was an int. I've made them both unsigned long ints.
2710       . In pcretest the fact that a const uschar * doesn't automatically cast to
2711         a void * provoked a warning.
2713       . Turning on some more compiler warnings threw up some "shadow" variables
2714         and a few more missing casts.
2716    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
2717       option, a class that contained a single character with a value between 128
2718       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
2720    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
2721       option, a class that contained several characters, but with at least one
2722       whose value was between 128 and 255 caused PCRE to crash.
2725    Version 4.1 12-Mar-03
2726    ---------------------
2728    1. Compiling with gcc -pedantic found a couple of places where casts were
2729    needed, and a string in dftables.c that was longer than standard compilers are
2730    required to support.
2732    2. Compiling with Sun's compiler found a few more places where the code could
2733    be tidied up in order to avoid warnings.
2735    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
2736    first of these names is deprecated in the latest Autoconf in favour of the name
2737    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
2738    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2739    analogy I have changed it to CFLAGS_FOR_BUILD.
2741    4. Added -no-undefined to the linking command in the Makefile, because this is
2742    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2743    linking step for the pcreposix library.
2745    5. PCRE was failing to diagnose the case of two named groups with the same
2746    name.
2748    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2749    literal character that is needed in the subject for a match, and scans along to
2750    ensure that it is present before embarking on the full matching process. This
2751    saves time in cases of nested unlimited repeats that are never going to match.
2752    Problem: the scan can take a lot of time if the subject is very long (e.g.
2753    megabytes), thus penalizing straightforward matches. It is now done only if the
2754    amount of subject to be scanned is less than 1000 bytes.
2756    7. A lesser problem with the same optimization is that it was recording the
2757    first character of an anchored pattern as "needed", thus provoking a search
2758    right along the subject, even when the first match of the pattern was going to
2759    fail. The "needed" character is now not set for anchored patterns, unless it
2760    follows something in the pattern that is of non-fixed length. Thus, it still
2761    fulfils its original purpose of finding quick non-matches in cases of nested
2762    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
2765    Version 4.0 17-Feb-03
2766    ---------------------
2768    1. If a comment in an extended regex that started immediately after a meta-item
2769    extended to the end of string, PCRE compiled incorrect data. This could lead to
2770    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
2772    2. Moved to autoconf 2.53 and libtool 1.4.2.
2774    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
2775    the special perltest8 script is no longer needed - all the tests can be run
2776    from a single perltest script.
2778    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
2779    by \s. It has now been removed in PCRE. This means it isn't recognized as
2780    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
2781    class [:space:] *does* include VT, thereby creating a mess.
2783    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
2784    space and tab.
2786    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
2787    its new features into the main test script, reducing the number of scripts.
2789    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
2790    were backward compatible, and made the (?i) apply to the whole pattern, as if
2791    /i were given. Now it behaves more logically, and applies the option setting
2792    only to what follows. PCRE has been changed to follow suit. However, if it
2793    finds options settings right at the start of the pattern, it extracts them into
2794    the global options, as before. Thus, they show up in the info data.
2796    8. Added support for the \Q...\E escape sequence. Characters in between are
2797    treated as literals. This is slightly different from Perl in that $ and @ are
2798    also handled as literals inside the quotes. In Perl, they will cause variable
2799    interpolation. Note the following examples:
2801        Pattern            PCRE matches      Perl matches
2803        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
2804        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
2805        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
2807    For compatibility with Perl, \Q...\E sequences are recognized inside character
2808    classes as well as outside them.
2810    9. Re-organized 3 code statements in pcretest to avoid "overflow in
2811    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2812    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2813    signed/unsigned warnings.
2815    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2816    option for pcretest, so I've replaced it by a simple function that does just
2817    that job.
2819    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2820    "pcregrep -".
2822    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2823    Java package. This provides some syntactic sugar for simple cases of what my
2824    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2825    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2826    item, you can use this simplified notation. Note that only makes sense with
2827    greedy quantifiers. Consequently, the use of the possessive quantifier forces
2828    greediness, whatever the setting of the PCRE_UNGREEDY option.
2830    13. A change of greediness default within a pattern was not taking effect at
2831    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2832    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2833    was abstracted outside.
2835    14. PCRE now supports the \G assertion. It is true when the current matching
2836    position is at the start point of the match. This differs from \A when the
2837    starting offset is non-zero. Used with the /g option of pcretest (or similar
2838    code), it works in the same way as it does for Perl's /g option. If all
2839    alternatives of a regex begin with \G, the expression is anchored to the start
2840    match position, and the "anchored" flag is set in the compiled expression.
2842    15. Some bugs concerning the handling of certain option changes within patterns
2843    have been fixed. These applied to options other than (?ims). For example,
2844    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2845    been the other way round. Some of this was related to change 7 above.
2847    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2848    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2849    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2850    POSIX classes only within a class (e.g. /[[:alpha:]]/).
2852    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2853    mode. Unlike ".", it always matches newline, whatever the setting of
2854    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2855    assertions. Perl allows it, but it doesn't (in general) work because it can't
2856    calculate the length of the lookbehind. At least, that's the case for Perl
2857    5.8.0 - I've been told they are going to document that it doesn't work in
2858    future.
2860    18. Added an error diagnosis for escapes that PCRE does not support: these are
2861    \L, \l, \N, \P, \p, \U, \u, and \X.
2863    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2864    reading past the end of the pattern in cases such as /[abcd/.
2866    20. PCRE was getting more memory than necessary for patterns with classes that
2867    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2869    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2870    compiling PCRE for use with Virtual Pascal.
2872    22. Small fix to the Makefile to make it work properly if the build is done
2873    outside the source tree.
2875    23. Added a new extension: a condition to go with recursion. If a conditional
2876    subpattern starts with (?(R) the "true" branch is used if recursion has
2877    happened, whereas the "false" branch is used only at the top level.
2879    24. When there was a very long string of literal characters (over 255 bytes
2880    without UTF support, over 250 bytes with UTF support), the computation of how
2881    much memory was required could be incorrect, leading to segfaults or other
2882    strange effects.
2884    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2885    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2886    there was a subsequent back reference to those brackets. This meant that, for
2887    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2888    possible to check for precisely this case. All we can do is abandon the
2889    optimization if .* occurs inside capturing brackets when there are any back
2890    references whatsoever. (See below for a better fix that came later.)
2892    26. The handling of the optimization for finding the first character of a
2893    non-anchored pattern, and for finding a character that is required later in the
2894    match were failing in some cases. This didn't break the matching; it just
2895    failed to optimize when it could. The way this is done has been re-implemented.
2897    27. Fixed typo in error message for invalid (?R item (it said "(?p").
2899    28. Added a new feature that provides some of the functionality that Perl
2900    provides with (?{...}). The facility is termed a "callout". The way it is done
2901    in PCRE is for the caller to provide an optional function, by setting
2902    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2903    global variable. By default it is unset, which disables all calling out. To get
2904    the function called, the regex must include (?C) at appropriate points. This
2905    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2906    This provides a means of identifying different callout points. When PCRE
2907    reaches such a point in the regex, if pcre_callout has been set, the external
2908    function is called. It is provided with data in a structure called
2909    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2910    matching continues; if it returns a non-zero value, the match at the current
2911    point fails. However, backtracking will occur if possible. [This was changed
2912    later and other features added - see item 49 below.]
2914    29. pcretest is upgraded to test the callout functionality. It provides a
2915    callout function that displays information. By default, it shows the start of
2916    the match and the current position in the text. There are some new data escapes
2917    to vary what happens:
2919        \C+         in addition, show current contents of captured substrings
2920        \C-         do not supply a callout function
2921        \C!n        return 1 when callout number n is reached
2922        \C!n!m      return 1 when callout number n is reached for the mth time
2924    30. If pcregrep was called with the -l option and just a single file name, it
2925    output "<stdin>" if a match was found, instead of the file name.
2927    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2928    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2929    pcre_exec(). This saves a malloc/free per call. The default value of
2930    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2931    when configuring.
2933    32. The default maximum size of a compiled pattern is 64K. There have been a
2934    few cases of people hitting this limit. The code now uses macros to handle the
2935    storing of links as offsets within the compiled pattern. It defaults to 2-byte
2936    links, but this can be changed to 3 or 4 bytes by --with-link-size when
2937    configuring. Tests 2 and 5 work only with 2-byte links because they output
2938    debugging information about compiled patterns.
2940    33. Internal code re-arrangements:
2942    (a) Moved the debugging function for printing out a compiled regex into
2943        its own source file (printint.c) and used #include to pull it into
2944        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2945        separate copies.
2947    (b) Defined the list of op-code names for debugging as a macro in
2948        internal.h so that it is next to the definition of the opcodes.
2950    (c) Defined a table of op-code lengths for simpler skipping along compiled
2951        code. This is again a macro in internal.h so that it is next to the
2952        definition of the opcodes.
2954    34. Added support for recursive calls to individual subpatterns, along the
2955    lines of Robin Houston's patch (but implemented somewhat differently).
2957    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2958    allow it to read and process whole directories in Win32. This code was
2959    contributed by Lionel Fourquaux; it has not been tested by me.
2961    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2962    used to name a group. Names consist of alphanumerics and underscores, and must
2963    be unique. Back references use the syntax (?P=name) and recursive calls use
2964    (?P>name) which is a PCRE extension to the Python extension. Groups still have
2965    numbers. The function pcre_fullinfo() can be used after compilation to extract
2966    a name/number map. There are three relevant calls:
2968      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
2969      PCRE_INFO_NAMECOUNT            yields the number of entries
2970      PCRE_INFO_NAMETABLE            yields a pointer to the map.
2972    The map is a vector of fixed-size entries. The size of each entry depends on
2973    the length of the longest name used. The first two bytes of each entry are the
2974    group number, most significant byte first. There follows the corresponding
2975    name, zero terminated. The names are in alphabetical order.
2977    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2978    case instead of 255. Making it the same both with and without UTF-8 support
2979    means that the same test output works with both.
2981    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2982    calling malloc() with a zero argument.
2984    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2985    optimization. I've improved things by keeping a bitmap of backreferences with
2986    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2987    fact referenced, the optimization can be applied. It is unlikely that a
2988    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2989    the match to follow \n) will appear inside brackets with a number greater than
2990    31, but if it does, any back reference > 31 suppresses the optimization.
2992    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2993    of disabling numbered capturing parentheses. Any opening parenthesis that is
2994    not followed by ? behaves as if it were followed by ?: but named parentheses
2995    can still be used for capturing (and they will acquire numbers in the usual
2996    way).
2998    41. Redesigned the return codes from the match() function into yes/no/error so
2999    that errors can be passed back from deep inside the nested calls. A malloc
3000    failure while inside a recursive subpattern call now causes the
3001    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
3003    42. It is now possible to set a limit on the number of times the match()
3004    function is called in a call to pcre_exec(). This facility makes it possible to
3005    limit the amount of recursion and backtracking, though not in a directly
3006    obvious way, because the match() function is used in a number of different
3007    circumstances. The count starts from zero for each position in the subject
3008    string (for non-anchored patterns). The default limit is, for compatibility, a
3009    large number, namely 10 000 000. You can change this in two ways:
3011    (a) When configuring PCRE before making, you can use --with-match-limit=n
3012        to set a default value for the compiled library.
3014    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
3015        a different value is set. See 45 below.
3017    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
3019    43. Added a new function pcre_config(int, void *) to enable run-time extraction
3020    of things that can be changed at compile time. The first argument specifies
3021    what is wanted and the second points to where the information is to be placed.
3022    The current list of available information is:
3024      PCRE_CONFIG_UTF8
3026    The output is an integer that is set to one if UTF-8 support is available;
3027    otherwise it is set to zero.
3031    The output is an integer that it set to the value of the code that is used for
3032    newline. It is either LF (10) or CR (13).
3036    The output is an integer that contains the number of bytes used for internal
3037    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
3041    The output is an integer that contains the threshold above which the POSIX
3042    interface uses malloc() for output vectors. See item 31 above.
3046    The output is an unsigned integer that contains the default limit of the number
3047    of match() calls in a pcre_exec() execution. See 42 above.
3049    44. pcretest has been upgraded by the addition of the -C option. This causes it
3050    to extract all the available output from the new pcre_config() function, and to
3051    output it. The program then exits immediately.
3053    45. A need has arisen to pass over additional data with calls to pcre_exec() in
3054    order to support additional features. One way would have been to define
3055    pcre_exec2() (for example) with extra arguments, but this would not have been
3056    extensible, and would also have required all calls to the original function to
3057    be mapped to the new one. Instead, I have chosen to extend the mechanism that
3058    is used for passing in "extra" data from pcre_study().
3060    The pcre_extra structure is now exposed and defined in pcre.h. It currently
3061    contains the following fields:
3063      flags         a bitmap indicating which of the following fields are set
3064      study_data    opaque data from pcre_study()
3065      match_limit   a way of specifying a limit on match() calls for a specific
3066                      call to pcre_exec()
3067      callout_data  data for callouts (see 49 below)
3069    The flag bits are also defined in pcre.h, and are
3075    The pcre_study() function now returns one of these new pcre_extra blocks, with
3076    the actual study data pointed to by the study_data field, and the
3077    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
3078    before. That is, this change is entirely upwards-compatible and requires no
3079    change to existing code.
3081    If you want to pass in additional data to pcre_exec(), you can either place it
3082    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
3083    block.
3085    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
3086    data string contains the escape sequence \M, pcretest calls pcre_exec() several
3087    times with different match limits, until it finds the minimum value needed for
3088    pcre_exec() to complete. The value is then output. This can be instructive; for
3089    most simple matches the number is quite small, but for pathological cases it
3090    gets very large very quickly.
3092    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
3093    returns the size of the data block pointed to by the study_data field in a
3094    pcre_extra block, that is, the value that was passed as the argument to
3095    pcre_malloc() when PCRE was getting memory in which to place the information
3096    created by pcre_study(). The fourth argument should point to a size_t variable.
3097    pcretest has been extended so that this information is shown after a successful
3098    pcre_study() call when information about the compiled regex is being displayed.
3100    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
3101    because what follows is always an absolute path. (Later: it turns out that this
3102    is more than cosmetic for MinGW, because it doesn't like empty path
3103    components.)
3105    49. Some changes have been made to the callout feature (see 28 above):
3107    (i)  A callout function now has three choices for what it returns:
3109           0  =>  success, carry on matching
3110         > 0  =>  failure at this point, but backtrack if possible
3111         < 0  =>  serious error, return this value from pcre_exec()
3113         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
3114         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
3115         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
3116         use by callout functions. It will never be used by PCRE itself.
3118    (ii) The pcre_extra structure (see 45 above) has a void * field called
3119         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
3120         pcre_callout_block structure has a field of the same name. The contents of
3121         the field passed in the pcre_extra structure are passed to the callout
3122         function in the corresponding field in the callout block. This makes it
3123         easier to use the same callout-containing regex from multiple threads. For
3124         testing, the pcretest program has a new data escape
3126           \C*n        pass the number n (may be negative) as callout_data
3128         If the callout function in pcretest receives a non-zero value as
3129         callout_data, it returns that value.
3131    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
3132    there were some redundant $(CFLAGS) in commands that are now specified as
3133    $(LINK), which already includes $(CFLAGS).
3135    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
3136    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
3137    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
3138    one-byte characters throughout. Note that case-insensitive matching applies
3139    only to characters whose values are less than 256. PCRE doesn't support the
3140    notion of cases for higher-valued characters.
3142    (i)   A character class whose characters are all within 0-255 is handled as
3143          a bit map, and the map is inverted for negative classes. Previously, a
3144          character > 255 always failed to match such a class; however it should
3145          match if the class was a negative one (e.g. [^ab]). This has been fixed.
3147    (ii)  A negated character class with a single character < 255 is coded as
3148          "not this character" (OP_NOT). This wasn't working properly when the test
3149          character was multibyte, either singly or repeated.
3151    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
3152          mode, for example: \x{100}{2,3}.
3154    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
3155          singly or repeated) now correctly test multibyte characters. However,
3156          PCRE doesn't recognize any characters with values greater than 255 as
3157          digits, spaces, or word characters. Such characters always match \D, \S,
3158          and \W, and never match \d, \s, or \w.
3160    (v)   Classes may now contain characters and character ranges with values
3161          greater than 255. For example: [ab\x{100}-\x{400}].
3163    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
3164          PCRE in UTF-8 mode.
3166    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
3167    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
3168    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
3169    value.)
3171    53. The single man page has become too large. I have therefore split it up into
3172    a number of separate man pages. These also give rise to individual HTML pages;
3173    these are now put in a separate directory, and there is an index.html page that
3174    lists them all. Some hyperlinking between the pages has been installed.
3176    54. Added convenience functions for handling named capturing parentheses.
3178    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
3179    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
3180    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
3181    are faulted.
3183    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
3184    calling configure. These values are used when compiling the dftables.c program
3185    which is run to generate the source of the default character tables. They
3186    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
3187    you will need to set these values.
3189    57. Updated the building process for Windows DLL, as provided by Fred Cox.
3192    Version 3.9 02-Jan-02
3193    ---------------------
3195    1. A bit of extraneous text had somehow crept into the pcregrep documentation.
3197    2. If --disable-static was given, the building process failed when trying to
3198    build pcretest and pcregrep. (For some reason it was using libtool to compile
3199    them, which is not right, as they aren't part of the library.)
3202    Version 3.8 18-Dec-01
3203    ---------------------
3205    1. The experimental UTF-8 code was completely screwed up. It was packing the
3206    bytes in the wrong order. How dumb can you get?
3209    Version 3.7 29-Oct-01
3210    ---------------------
3212    1. In updating pcretest to check change 1 of version 3.6, I screwed up.
3213    This caused pcretest, when used on the test data, to segfault. Unfortunately,
3214    this didn't happen under Solaris 8, where I normally test things.
3216    2. The Makefile had to be changed to make it work on BSD systems, where 'make'
3217    doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
3218    isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
3219    this fix an hour or so after the initial 3.7 release.)
3222    Version 3.6 23-Oct-01
3223    ---------------------
3225    1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
3226    offsets passed as NULL with zero offset count.
3228    2. The config.guess and config.sub files had not been updated when I moved to
3229    the latest autoconf.
3232    Version 3.5 15-Aug-01
3233    ---------------------
3235    1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
3236    had been forgotten.
3238    2. By using declared but undefined structures, we can avoid using "void"
3239    definitions in pcre.h while keeping the internal definitions of the structures
3240    private.
3242    3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
3243    user point of view, this means that both static and shared libraries are built
3244    by default, but this can be individually controlled. More of the work of
3245    handling this static/shared cases is now inside libtool instead of PCRE's make
3246    file.
3248    4. The pcretest utility is now installed along with pcregrep because it is
3249    useful for users (to test regexs) and by doing this, it automatically gets
3250    relinked by libtool. The documentation has been turned into a man page, so
3251    there are now .1, .txt, and .html versions in /doc.
3253    5. Upgrades to pcregrep:
3254       (i)   Added long-form option names like gnu grep.
3255       (ii)  Added --help to list all options with an explanatory phrase.
3256       (iii) Added -r, --recursive to recurse into sub-directories.
3257       (iv)  Added -f, --file to read patterns from a file.
3259    6. pcre_exec() was referring to its "code" argument before testing that
3260    argument for NULL (and giving an error if it was NULL).
3262    7. Upgraded Makefile.in to allow for compiling in a different directory from
3263    the source directory.
3265    8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
3266    options bits, the pointer it was passed was to an int instead of to an unsigned
3267    long int. This mattered only on 64-bit systems.
3269    9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
3270    generated) instead of pcre.in, which it its source. Also made the same change
3271    in several of the .c files.
3273    10. A new release of gcc defines printf() as a macro, which broke pcretest
3274    because it had an ifdef in the middle of a string argument for printf(). Fixed
3275    by using separate calls to printf().
3277    11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
3278    script, to force use of CR or LF instead of \n in the source. On non-Unix
3279    systems, the value can be set in config.h.
3281    12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
3282    absolute limit. Changed the text of the error message to make this clear, and
3283    likewise updated the man page.
3285    13. The limit of 99 on the number of capturing subpatterns has been removed.
3286    The new limit is 65535, which I hope will not be a "real" limit.
3289    Version 3.4 22-Aug-00
3290    ---------------------
3292    1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
3294    2. Diagnose condition (?(0) as an error instead of crashing on matching.
3297    Version 3.3 01-Aug-00
3298    ---------------------
3300    1. If an octal character was given, but the value was greater than \377, it
3301    was not getting masked to the least significant bits, as documented. This could
3302    lead to crashes in some systems.
3304    2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
3305    the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
3307    3. Added the functions pcre_free_substring() and pcre_free_substring_list().
3308    These just pass their arguments on to (pcre_free)(), but they are provided
3309    because some uses of PCRE bind it to non-C systems that can call its functions,
3310    but cannot call free() or pcre_free() directly.
3312    4. Add "make test" as a synonym for "make check". Corrected some comments in
3313    the Makefile.
3315    5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
3316    Makefile.
3318    6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
3319    command called pgrep for grepping around the active processes.
3321    7. Added the beginnings of support for UTF-8 character strings.
3323    8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
3324    RANLIB to ./ltconfig so that they are used by libtool. I think these are all
3325    the relevant ones. (AR is not passed because ./ltconfig does its own figuring
3326    out for the ar command.)
3329    Version 3.2 12-May-00
3330    ---------------------
3332    This is purely a bug fixing release.
3334    1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
3335    of ZA. This was just one example of several cases that could provoke this bug,
3336    which was introduced by change 9 of version 2.00. The code for breaking
3337    infinite loops after an iteration that matches an empty string was't working
3338    correctly.
3340    2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
3341    when matched against abbab (for example). After matching an empty string, it
3342    wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
3343    caused it to match further down the string than it should.
3345    3. The code contained an inclusion of sys/types.h. It isn't clear why this
3346    was there because it doesn't seem to be needed, and it causes trouble on some
3347    systems, as it is not a Standard C header. It has been removed.
3349    4. Made 4 silly changes to the source to avoid stupid compiler warnings that
3350    were reported on the Macintosh. The changes were from
3352      while ((c = *(++ptr)) != 0 && c != '\n');
3353    to
3354      while ((c = *(++ptr)) != 0 && c != '\n') ;
3356    Totally extraordinary, but if that's what it takes...
3358    5. PCRE is being used in one environment where neither memmove() nor bcopy() is
3359    available. Added HAVE_BCOPY and an autoconf test for it; if neither
3360    HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
3361    assumes the way PCRE uses memmove() (always moving upwards).
3363    6. PCRE is being used in one environment where strchr() is not available. There
3364    was only one use in pcre.c, and writing it out to avoid strchr() probably gives
3365    faster code anyway.
3368    Version 3.1 09-Feb-00
3369    ---------------------
3371    The only change in this release is the fixing of some bugs in Makefile.in for
3372    the "install" target:
3374    (1) It was failing to install pcreposix.h.
3376    (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
3379    Version 3.0 01-Feb-00
3380    ---------------------
3382    1. Add support for the /+ modifier to perltest (to output $` like it does in
3383    pcretest).
3385    2. Add support for the /g modifier to perltest.
3387    3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
3388    matches null strings.
3390    4. Fix perltest so that it doesn't do unwanted things when fed an empty
3391    pattern. Perl treats empty patterns specially - it reuses the most recent
3392    pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
3393    effect.
3395    5. The POSIX interface was broken in that it was just handing over the POSIX
3396    captured string vector to pcre_exec(), but (since release 2.00) PCRE has
3397    required a bigger vector, with some working space on the end. This means that
3398    the POSIX wrapper now has to get and free some memory, and copy the results.
3400    6. Added some simple autoconf support, placing the test data and the
3401    documentation in separate directories, re-organizing some of the
3402    information files, and making it build pcre-config (a GNU standard). Also added
3403    libtool support for building PCRE as a shared library, which is now the
3404    default.
3406    7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
3407    09 are not valid octal constants. Single digits will be used for minor values
3408    less than 10.
3410    8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
3411    existing programs that set these in the POSIX interface can use PCRE without
3412    modification.
3414    9. Added a new function, pcre_fullinfo() with an extensible interface. It can
3415    return all that pcre_info() returns, plus additional data. The pcre_info()
3416    function is retained for compatibility, but is considered to be obsolete.
3418    10. Added experimental recursion feature (?R) to handle one common case that
3419 &nb