/[pcre]/code/trunk/ChangeLog
ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 55 by nigel, Sat Feb 24 21:39:46 2007 UTC revision 537 by ph10, Tue Jun 8 15:28:26 2010 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    Version 8.10 03-Jun-2010
5    ------------------------
6    
7    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
8        THEN.
9    
10    2.  (*ACCEPT) was not working when inside an atomic group.
11    
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.
15    
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.
19    
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.)
22    
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.
28    
29    7.  Minor change to pcretest.c to avoid a compiler warning.
30    
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).
34    
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.
39    
40    10. Added --line-buffered to pcregrep.
41    
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
46        (#976).
47    
48    12. If a pattern that was studied started with a repeated Unicode property
49        test, for example, \p{Nd}+, there was the theoretical possibility of
50        setting up an incorrect bitmap of starting bytes, but fortunately it could
51        not have actually happened in practice until change 8 above was made (it
52        added property types that matched character-matching opcodes).
53    
54    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
55        possible starting bytes for non-anchored patterns.
56    
57    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
58        \R, and also a number of cases that involve Unicode properties, both
59        explicit and implicit when PCRE_UCP is set.
60    
61    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
62        input, it could crash or give wrong results if characters with values
63        greater than 0xc0 were present in the subject string. (Detail: it assumed
64        UTF-8 input when processing these items.)
65    
66    16. Added a lot of (int) casts to avoid compiler warnings in systems where
67        size_t is 64-bit (#991).
68    
69    17. Added a check for running out of memory when PCRE is compiled with
70        --disable-stack-for-recursion (#990).
71    
72    18. If the last data line in a file for pcretest does not have a newline on
73        the end, a newline was missing in the output.
74    
75    
76    
77    Version 8.02 19-Mar-2010
78    ------------------------
79    
80    1.  The Unicode data tables have been updated to Unicode 5.2.0.
81    
82    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
83        configured.
84    
85    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
86        original author of that file, following a query about its status.
87    
88    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
89        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
90    
91    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
92        quantifier applied to a forward-referencing subroutine call, could compile
93        incorrect code or give the error "internal error: previously-checked
94        referenced subpattern not found".
95    
96    6.  Both MS Visual Studio and Symbian OS have problems with initializing
97        variables to point to external functions. For these systems, therefore,
98        pcre_malloc etc. are now initialized to local functions that call the
99        relevant global functions.
100    
101    7.  There were two entries missing in the vectors called coptable and poptable
102        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
103        I've fixed the data, and added a kludgy way of testing at compile time that
104        the lengths are correct (equal to the number of opcodes).
105    
106    8.  Following on from 7, I added a similar kludge to check the length of the
107        eint vector in pcreposix.c.
108    
109    9.  Error texts for pcre_compile() are held as one long string to avoid too
110        much relocation at load time. To find a text, the string is searched,
111        counting zeros. There was no check for running off the end of the string,
112        which could happen if a new error number was added without updating the
113        string.
114    
115    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
116    
117    11. \K was not working if it appeared in an atomic group or in a group that
118        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
119        \K is "not well defined" if used in an assertion. PCRE now accepts it if
120        the assertion is positive, but not if it is negative.
121    
122    12. Change 11 fortuitously reduced the size of the stack frame used in the
123        "match()" function of pcre_exec.c by one pointer. Forthcoming
124        implementation of support for (*MARK) will need an extra pointer on the
125        stack; I have reserved it now, so that the stack frame size does not
126        decrease.
127    
128    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
129        item in branch that calls a recursion is a subroutine call - as in the
130        second branch in the above example - was incorrectly given the compile-
131        time error "recursive call could loop indefinitely" because pcre_compile()
132        was not correctly checking the subroutine for matching a non-empty string.
133    
134    14. The checks for overrunning compiling workspace could trigger after an
135        overrun had occurred. This is a "should never occur" error, but it can be
136        triggered by pathological patterns such as hundreds of nested parentheses.
137        The checks now trigger 100 bytes before the end of the workspace.
138    
139    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
140    
141    
142    Version 8.01 19-Jan-2010
143    ------------------------
144    
145    1.  If a pattern contained a conditional subpattern with only one branch (in
146        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
147        computed the wrong minimum data length (which is of course zero for such
148        subpatterns). This could cause incorrect "no match" results.
149    
150    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
151        the pattern is reset in the first branch, pcre_compile() failed with
152        "internal error: code overflow at offset...". This happened only when
153        the reset was to the original external option setting. (An optimization
154        abstracts leading options settings into an external setting, which was the
155        cause of this.)
156    
157    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
158        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
159        assertion pattern did not match (meaning that the assertion was true), it
160        was incorrectly treated as false if the SKIP had been reached during the
161        matching. This also applied to assertions used as conditions.
162    
163    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
164        assertion subpattern, including such a pattern used as a condition,
165        unpredictable results occurred, instead of the error return
166        PCRE_ERROR_DFA_UITEM.
167    
168    5.  The C++ GlobalReplace function was not working like Perl for the special
169        situation when an empty string is matched. It now does the fancy magic
170        stuff that is necessary.
171    
172    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
173        removed. (These were left over from very, very early versions of PCRE.)
174    
175    7.  Some cosmetic changes to the code to make life easier when compiling it
176        as part of something else:
177    
178        (a) Change DEBUG to PCRE_DEBUG.
179    
180        (b) In pcre_compile(), rename the member of the "branch_chain" structure
181            called "current" as "current_branch", to prevent a collision with the
182            Linux macro when compiled as a kernel module.
183    
184        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
185            prevent a collision with the Linux macro when compiled as a kernel
186            module.
187    
188    8.  In pcre_compile() there are some checks for integer overflows that used to
189        cast potentially large values to (double). This has been changed to that
190        when building, a check for int64_t is made, and if it is found, it is used
191        instead, thus avoiding the use of floating point arithmetic. (There is no
192        other use of FP in PCRE.) If int64_t is not found, the fallback is to
193        double.
194    
195    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
196        2005 (difference between two addresses compared to an unsigned value).
197    
198    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
199        custom one, because of the following reported problem in Windows:
200    
201          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
202              under Win32.
203          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
204              therefore missing the function definition.
205          - The compiler thus generates a "C" signature for the test function.
206          - The linker fails to find the "C" function.
207          - PCRE fails to configure if asked to do so against libbz2.
208    
209    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
210        messages were output:
211    
212          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
213          rerunning libtoolize, to keep the correct libtool macros in-tree.
214          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
215    
216        I have done both of these things.
217    
218    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
219        most of the time, it *can* run out if it is given a pattern that contains a
220        runaway infinite recursion. I updated the discussion in the pcrestack man
221        page.
222    
223    13. Now that we have gone to the x.xx style of version numbers, the minor
224        version may start with zero. Using 08 or 09 is a bad idea because users
225        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
226        interpreted as invalid octal numbers. I've updated the previous comment in
227        configure.ac, and also added a check that gives an error if 08 or 09 are
228        used.
229    
230    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
231        causing partial matching to fail when the end of the subject matched \W
232        in a UTF-8 pattern where \W was quantified with a minimum of 3.
233    
234    15. There were some discrepancies between the declarations in pcre_internal.h
235        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
236        their definitions. The declarations used "const uschar *" and the
237        definitions used USPTR. Even though USPTR is normally defined as "const
238        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
239        reported that: "This difference in casting confuses some C++ compilers, for
240        example, SunCC recognizes above declarations as different functions and
241        generates broken code for hbpcre." I have changed the declarations to use
242        USPTR.
243    
244    16. GNU libtool is named differently on some systems. The autogen.sh script now
245        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
246        (FreeBSD).
247    
248    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
249        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
250        comment: "Figure out how to create a longlong from a string: strtoll and
251        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
252        instance, but it only takes 2 args instead of 3!"
253    
254    18. A subtle bug concerned with back references has been fixed by a change of
255        specification, with a corresponding code fix. A pattern such as
256        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
257        refers, was giving matches when it shouldn't. For example, xa=xaaa would
258        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
259        same bug. Such groups have to be quantified to be useful, or contained
260        inside another quantified group. (If there's no repetition, the reference
261        can never match.) The problem arises because, having left the group and
262        moved on to the rest of the pattern, a later failure that backtracks into
263        the group uses the captured value from the final iteration of the group
264        rather than the correct earlier one. I have fixed this in PCRE by forcing
265        any group that contains a reference to itself to be an atomic group; that
266        is, there cannot be any backtracking into it once it has completed. This is
267        similar to recursive and subroutine calls.
268    
269    
270    Version 8.00 19-Oct-09
271    ----------------------
272    
273    1.  The table for translating pcre_compile() error codes into POSIX error codes
274        was out-of-date, and there was no check on the pcre_compile() error code
275        being within the table. This could lead to an OK return being given in
276        error.
277    
278    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
279        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
280        in a Windows environment.
281    
282    3.  The pcregrep --count option prints the count for each file even when it is
283        zero, as does GNU grep. However, pcregrep was also printing all files when
284        --files-with-matches was added. Now, when both options are given, it prints
285        counts only for those files that have at least one match. (GNU grep just
286        prints the file name in this circumstance, but including the count seems
287        more useful - otherwise, why use --count?) Also ensured that the
288        combination -clh just lists non-zero counts, with no names.
289    
290    4.  The long form of the pcregrep -F option was incorrectly implemented as
291        --fixed_strings instead of --fixed-strings. This is an incompatible change,
292        but it seems right to fix it, and I didn't think it was worth preserving
293        the old behaviour.
294    
295    5.  The command line items --regex=pattern and --regexp=pattern were not
296        recognized by pcregrep, which required --regex pattern or --regexp pattern
297        (with a space rather than an '='). The man page documented the '=' forms,
298        which are compatible with GNU grep; these now work.
299    
300    6.  No libpcreposix.pc file was created for pkg-config; there was just
301        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
302    
303    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
304        when UCP support is not needed, by modifying the Python script that
305        generates it from Unicode data files. This should not matter if the module
306        is correctly used as a library, but I received one complaint about 50K of
307        unwanted data. My guess is that the person linked everything into his
308        program rather than using a library. Anyway, it does no harm.
309    
310    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
311        was a minimum greater than 1 for a wide character in a possessive
312        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
313        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
314        character. Chaos in the form of incorrect output or a compiling loop could
315        result.
316    
317    9.  The restrictions on what a pattern can contain when partial matching is
318        requested for pcre_exec() have been removed. All patterns can now be
319        partially matched by this function. In addition, if there are at least two
320        slots in the offset vector, the offset of the earliest inspected character
321        for the match and the offset of the end of the subject are set in them when
322        PCRE_ERROR_PARTIAL is returned.
323    
324    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
325        synonymous with PCRE_PARTIAL, for backwards compatibility, and
326        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
327        and may be more useful for multi-segment matching.
328    
329    11. Partial matching with pcre_exec() is now more intuitive. A partial match
330        used to be given if ever the end of the subject was reached; now it is
331        given only if matching could not proceed because another character was
332        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
333        string "Z", which now yields "no match" instead of "partial match". In the
334        case of pcre_dfa_exec(), "no match" is given if every matching path for the
335        final character ended with (*FAIL).
336    
337    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
338        if the pattern had a "must contain" character that was already found in the
339        earlier partial match, unless partial matching was again requested. For
340        example, with the pattern /dog.(body)?/, the "must contain" character is
341        "g". If the first part-match was for the string "dog", restarting with
342        "sbody" failed. This bug has been fixed.
343    
344    13. The string returned by pcre_dfa_exec() after a partial match has been
345        changed so that it starts at the first inspected character rather than the
346        first character of the match. This makes a difference only if the pattern
347        starts with a lookbehind assertion or \b or \B (\K is not supported by
348        pcre_dfa_exec()). It's an incompatible change, but it makes the two
349        matching functions compatible, and I think it's the right thing to do.
350    
351    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
352        so that the demonstration program is easily available in environments where
353        PCRE has not been installed from source.
354    
355    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
356        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
357        library.
358    
359    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
360        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
361        is not the first non-POSIX option to be added. Clearly some people find
362        these options useful.
363    
364    17. If a caller to the POSIX matching function regexec() passes a non-zero
365        value for nmatch with a NULL value for pmatch, the value of
366        nmatch is forced to zero.
367    
368    18. RunGrepTest did not have a test for the availability of the -u option of
369        the diff command, as RunTest does. It now checks in the same way as
370        RunTest, and also checks for the -b option.
371    
372    19. If an odd number of negated classes containing just a single character
373        interposed, within parentheses, between a forward reference to a named
374        subpattern and the definition of the subpattern, compilation crashed with
375        an internal error, complaining that it could not find the referenced
376        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
377        [The bug was that it was starting one character too far in when skipping
378        over the character class, thus treating the ] as data rather than
379        terminating the class. This meant it could skip too much.]
380    
381    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
382        /g option in pcretest when the pattern contains \K, which makes it possible
383        to have an empty string match not at the start, even when the pattern is
384        anchored. Updated pcretest and pcredemo to use this option.
385    
386    21. If the maximum number of capturing subpatterns in a recursion was greater
387        than the maximum at the outer level, the higher number was returned, but
388        with unset values at the outer level. The correct (outer level) value is
389        now given.
390    
391    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
392        PCRE did not set those parentheses (unlike Perl). I have now found a way to
393        make it do so. The string so far is captured, making this feature
394        compatible with Perl.
395    
396    23. The tests have been re-organized, adding tests 11 and 12, to make it
397        possible to check the Perl 5.10 features against Perl 5.10.
398    
399    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
400        pattern matches a fixed length string. PCRE did not allow this; now it
401        does. Neither allows recursion.
402    
403    25. I finally figured out how to implement a request to provide the minimum
404        length of subject string that was needed in order to match a given pattern.
405        (It was back references and recursion that I had previously got hung up
406        on.) This code has now been added to pcre_study(); it finds a lower bound
407        to the length of subject needed. It is not necessarily the greatest lower
408        bound, but using it to avoid searching strings that are too short does give
409        some useful speed-ups. The value is available to calling programs via
410        pcre_fullinfo().
411    
412    26. While implementing 25, I discovered to my embarrassment that pcretest had
413        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
414        study optimizations had never been tested with that matching function.
415        Oops. What is worse, even when it was passed study data, there was a bug in
416        pcre_dfa_exec() that meant it never actually used it. Double oops. There
417        were also very few tests of studied patterns with pcre_dfa_exec().
418    
419    27. If (?| is used to create subpatterns with duplicate numbers, they are now
420        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
421        on the other side of the coin, they are no longer allowed to have different
422        names, because these cannot be distinguished in PCRE, and this has caused
423        confusion. (This is a difference from Perl.)
424    
425    28. When duplicate subpattern names are present (necessarily with different
426        numbers, as required by 27 above), and a test is made by name in a
427        conditional pattern, either for a subpattern having been matched, or for
428        recursion in such a pattern, all the associated numbered subpatterns are
429        tested, and the overall condition is true if the condition is true for any
430        one of them. This is the way Perl works, and is also more like the way
431        testing by number works.
432    
433    
434    Version 7.9 11-Apr-09
435    ---------------------
436    
437    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
438        (pcretest), all targets were linked against these libraries. This included
439        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
440        libraries. This caused unwanted dependencies to be created. This problem
441        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
442        pcretest is linked with readline.
443    
444    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
445        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
446        moved outside it again, because FALSE and TRUE are already defined in AIX,
447        but BOOL is not.
448    
449    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
450        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
451    
452    4.  The pcregrep documentation said spaces were inserted as well as colons (or
453        hyphens) following file names and line numbers when outputting matching
454        lines. This is not true; no spaces are inserted. I have also clarified the
455        wording for the --colour (or --color) option.
456    
457    5.  In pcregrep, when --colour was used with -o, the list of matching strings
458        was not coloured; this is different to GNU grep, so I have changed it to be
459        the same.
460    
461    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
462        each matching line was coloured. Now it goes on to look for further matches
463        of any of the test patterns, which is the same behaviour as GNU grep.
464    
465    7.  A pattern that could match an empty string could cause pcregrep to loop; it
466        doesn't make sense to accept an empty string match in pcregrep, so I have
467        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
468        seems to be how GNU grep behaves.
469    
470    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
471        start or after a newline", because the conditional assertion was not being
472        correctly handled. The rule now is that both the assertion and what follows
473        in the first alternative must satisfy the test.
474    
475    9.  If auto-callout was enabled in a pattern with a conditional group whose
476        condition was an assertion, PCRE could crash during matching, both with
477        pcre_exec() and pcre_dfa_exec().
478    
479    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
480        used for matching.
481    
482    11. Unicode property support in character classes was not working for
483        characters (bytes) greater than 127 when not in UTF-8 mode.
484    
485    12. Added the -M command line option to pcretest.
486    
487    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
488    
489    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
490    
491    16. Added comments and documentation about mis-use of no_arg in the C++
492        wrapper.
493    
494    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
495        from Martin Jerabek that uses macro names for all relevant character and
496        string constants.
497    
498    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
499        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
500        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
501        these, but not everybody uses configure.
502    
503    19. A conditional group that had only one branch was not being correctly
504        recognized as an item that could match an empty string. This meant that an
505        enclosing group might also not be so recognized, causing infinite looping
506        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
507        with the subject "ab", where knowledge that the repeated group can match
508        nothing is needed in order to break the loop.
509    
510    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
511        exec(), but without supplying a callout function, matching went wrong.
512    
513    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
514        leak if the size of the offset vector was greater than 30. When the vector
515        is smaller, the saved offsets during recursion go onto a local stack
516        vector, but for larger vectors malloc() is used. It was failing to free
517        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
518        error, in fact).
519    
520    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
521        heapframe that is used only when UTF-8 support is enabled. This caused no
522        problem, but was untidy.
523    
524    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
525        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
526        included within another project.
527    
528    24. Steven Van Ingelgem's patches to add more options to the CMake support,
529        slightly modified by me:
530    
531          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
532              not building pcregrep.
533    
534          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
535              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
536    
537    25. Forward references, both numeric and by name, in patterns that made use of
538        duplicate group numbers, could behave incorrectly or give incorrect errors,
539        because when scanning forward to find the reference group, PCRE was not
540        taking into account the duplicate group numbers. A pattern such as
541        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
542    
543    26. Changed a few more instances of "const unsigned char *" to USPTR, making
544        the feature of a custom pointer more persuasive (as requested by a user).
545    
546    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
547        pcretest.c, inside #ifndefs, because it seems they are sometimes already
548        pre-defined.
549    
550    28. Added support for (*UTF8) at the start of a pattern.
551    
552    29. Arrange for flags added by the "release type" setting in CMake to be shown
553        in the configuration summary.
554    
555    
556    Version 7.8 05-Sep-08
557    ---------------------
558    
559    1.  Replaced UCP searching code with optimized version as implemented for Ad
560        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
561        stage table and inline lookup instead of a function, giving speed ups of 2
562        to 5 times on some simple patterns that I tested. Permission was given to
563        distribute the MultiStage2.py script that generates the tables (it's not in
564        the tarball, but is in the Subversion repository).
565    
566    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
567        scripts.
568    
569    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
570        a group with a zero qualifier. The result of the study could be incorrect,
571        or the function might crash, depending on the pattern.
572    
573    4.  Caseless matching was not working for non-ASCII characters in back
574        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
575        It now works when Unicode Property Support is available.
576    
577    5.  In pcretest, an escape such as \x{de} in the data was always generating
578        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
579        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
580        truncation.
581    
582    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
583    
584    7.  Added two (int) casts to pcregrep when printing the difference of two
585        pointers, in case they are 64-bit values.
586    
587    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
588        test 2 if it fails.
589    
590    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
591        and a #define of that name to empty if it is not externally set. This is to
592        allow users of MSVC to set it if necessary.
593    
594    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
595        the convenience functions in the pcre_get.c source file.
596    
597    11. An option change at the start of a pattern that had top-level alternatives
598        could cause overwriting and/or a crash. This command provoked a crash in
599        some environments:
600    
601          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
602    
603        This potential security problem was recorded as CVE-2008-2371.
604    
605    12. For a pattern where the match had to start at the beginning or immediately
606        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
607        pcre_dfa_exec() could read past the end of the passed subject if there was
608        no match. To help with detecting such bugs (e.g. with valgrind), I modified
609        pcretest so that it places the subject at the end of its malloc-ed buffer.
610    
611    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
612        exec() might read past the end of the data buffer in UTF-8 mode.
613    
614    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
615        the data contained the byte 0x85 as part of a UTF-8 character within its
616        first line. This applied both to normal and DFA matching.
617    
618    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
619        /^[^d]*?$/8 failed to match "abc".
620    
621    16. Added a missing copyright notice to pcrecpp_internal.h.
622    
623    17. Make it more clear in the documentation that values returned from
624        pcre_exec() in ovector are byte offsets, not character counts.
625    
626    18. Tidied a few places to stop certain compilers from issuing warnings.
627    
628    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
629        supplied by Stefan Weber. I made a further small update for 7.8 because
630        there is a change of source arrangements: the pcre_searchfuncs.c module is
631        replaced by pcre_ucd.c.
632    
633    
634    Version 7.7 07-May-08
635    ---------------------
636    
637    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
638        a string to a long long, pretend we don't even have a long long." This is
639        done by checking for the strtoq, strtoll, and _strtoi64 functions.
640    
641    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
642        pre-7.6 versions, which defined a global no_arg variable instead of putting
643        it in the RE class. (See also #8 below.)
644    
645    3.  Remove a line of dead code, identified by coverity and reported by Nuno
646        Lopes.
647    
648    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
649    
650        (1) The include/exclude patterns were being applied to the whole pathnames
651            of files, instead of just to the final components.
652    
653        (2) If there was more than one level of directory, the subdirectories were
654            skipped unless they satisfied the include/exclude conditions. This is
655            inconsistent with GNU grep (and could even be seen as contrary to the
656            pcregrep specification - which I improved to make it absolutely clear).
657            The action now is always to scan all levels of directory, and just
658            apply the include/exclude patterns to regular files.
659    
660    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
661        --exclude_dir in the tests to avoid scanning .svn directories.
662    
663    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
664        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
665        doesn't support NULs in patterns.
666    
667    7.  Added some missing "const"s to declarations of static tables in
668        pcre_compile.c and pcre_dfa_exec.c.
669    
670    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
671        caused by fix #2  above. (Subsequently also a second patch to fix the
672        first patch. And a third patch - this was a messy problem.)
673    
674    9.  Applied Craig's patch to remove the use of push_back().
675    
676    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
677        matching function regexec().
678    
679    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
680        which, however, unlike Perl's \g{...}, are subroutine calls, not back
681        references. PCRE supports relative numbers with this syntax (I don't think
682        Oniguruma does).
683    
684    12. Previously, a group with a zero repeat such as (...){0} was completely
685        omitted from the compiled regex. However, this means that if the group
686        was called as a subroutine from elsewhere in the pattern, things went wrong
687        (an internal error was given). Such groups are now left in the compiled
688        pattern, with a new opcode that causes them to be skipped at execution
689        time.
690    
691    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
692        to the way PCRE behaves:
693    
694        (a) A lone ] character is dis-allowed (Perl treats it as data).
695    
696        (b) A back reference to an unmatched subpattern matches an empty string
697            (Perl fails the current match path).
698    
699        (c) A data ] in a character class must be notated as \] because if the
700            first data character in a class is ], it defines an empty class. (In
701            Perl it is not possible to have an empty class.) The empty class []
702            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
703            The negative empty class [^] matches any one character, independently
704            of the DOTALL setting.
705    
706    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
707        non-existent subpattern following a character class starting with ']' and
708        containing () gave an internal compiling error instead of "reference to
709        non-existent subpattern". Fortunately, when the pattern did exist, the
710        compiled code was correct. (When scanning forwards to check for the
711        existencd of the subpattern, it was treating the data ']' as terminating
712        the class, so got the count wrong. When actually compiling, the reference
713        was subsequently set up correctly.)
714    
715    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
716        it was being rejected as not supported by pcre_dfa_exec(), even though
717        other assertions are supported. I have made pcre_dfa_exec() support
718        (*FAIL).
719    
720    16. The implementation of 13c above involved the invention of a new opcode,
721        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
722        cannot be changed at match time, I realized I could make a small
723        improvement to matching performance by compiling OP_ALLANY instead of
724        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
725        on the OP_ANY path.
726    
727    17. Compiling pcretest on Windows with readline support failed without the
728        following two fixes: (1) Make the unistd.h include conditional on
729        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
730    
731    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
732        ncurses library to be included for pcretest when ReadLine support is
733        requested, but also to allow for it to be overridden. This patch came from
734        Daniel Bergström.
735    
736    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
737        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
738        any errors with the current Unicode tables. Thanks to Peter Kankowski for
739        spotting this.
740    
741    
742    Version 7.6 28-Jan-08
743    ---------------------
744    
745    1.  A character class containing a very large number of characters with
746        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
747        overflow.
748    
749    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
750        HAVE_LONG_LONG is not defined.
751    
752    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
753        bring them up to date and include new features. This patch includes:
754    
755        - Fixed PH's badly added libz and libbz2 support.
756        - Fixed a problem with static linking.
757        - Added pcredemo. [But later removed - see 7 below.]
758        - Fixed dftables problem and added an option.
759        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
760            HAVE_LONG_LONG.
761        - Added readline support for pcretest.
762        - Added an listing of the option settings after cmake has run.
763    
764    4.  A user submitted a patch to Makefile that makes it easy to create
765        "pcre.dll" under mingw when using Configure/Make. I added stuff to
766        Makefile.am that cause it to include this special target, without
767        affecting anything else. Note that the same mingw target plus all
768        the other distribution libraries and programs are now supported
769        when configuring with CMake (see 6 below) instead of with
770        Configure/Make.
771    
772    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
773        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
774        exported in the Windows port". It has not yet been confirmed that the patch
775        solves the problem, but it does no harm.
776    
777    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
778        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
779        with CMake, and also correct the comment about stack recursion.
780    
781    7.  Remove the automatic building of pcredemo from the ./configure system and
782        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
783        of a program that users should build themselves after PCRE is installed, so
784        building it automatically is not really right. What is more, it gave
785        trouble in some build environments.
786    
787    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
788    
789    
790    Version 7.5 10-Jan-08
791    ---------------------
792    
793    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
794        values in parens when parsing an RE using the C++ wrapper."
795    
796    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
797        Characters greater than 255 were excluded from the class instead of being
798        included.
799    
800    3.  The same bug as (2) above applied to negated POSIX classes such as
801        [:^space:].
802    
803    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
804        defined or documented. It seems to have been a typo for PCRE_STATIC, so
805        I have changed it.
806    
807    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
808        first named subpattern) and a construct such as (?&a) would reference the
809        first named subpattern whose name started with "a" (in other words, the
810        length check was missing). Both these problems are fixed. "Subpattern name
811        expected" is now given for (?&) (a zero-length name), and this patch also
812        makes it give the same error for \k'' (previously it complained that that
813        was a reference to a non-existent subpattern).
814    
815    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
816        this is right because (?- can be followed by option settings as well as by
817        digits. I have, however, made the messages clearer.
818    
819    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
820        than the number used in the conditional) now cause a compile-time error.
821        This is actually not compatible with Perl, which accepts such patterns, but
822        treats the conditional as always being FALSE (as PCRE used to), but it
823        seems to me that giving a diagnostic is better.
824    
825    8.  Change "alphameric" to the more common word "alphanumeric" in comments
826        and messages.
827    
828    9.  Fix two occurrences of "backslash" in comments that should have been
829        "backspace".
830    
831    10. Remove two redundant lines of code that can never be obeyed (their function
832        was moved elsewhere).
833    
834    11. The program that makes PCRE's Unicode character property table had a bug
835        which caused it to generate incorrect table entries for sequences of
836        characters that have the same character type, but are in different scripts.
837        It amalgamated them into a single range, with the script of the first of
838        them. In other words, some characters were in the wrong script. There were
839        thirteen such cases, affecting characters in the following ranges:
840    
841          U+002b0 - U+002c1
842          U+0060c - U+0060d
843          U+0061e - U+00612
844          U+0064b - U+0065e
845          U+0074d - U+0076d
846          U+01800 - U+01805
847          U+01d00 - U+01d77
848          U+01d9b - U+01dbf
849          U+0200b - U+0200f
850          U+030fc - U+030fe
851          U+03260 - U+0327f
852          U+0fb46 - U+0fbb1
853          U+10450 - U+1049d
854    
855    12. The -o option (show only the matching part of a line) for pcregrep was not
856        compatible with GNU grep in that, if there was more than one match in a
857        line, it showed only the first of them. It now behaves in the same way as
858        GNU grep.
859    
860    13. If the -o and -v options were combined for pcregrep, it printed a blank
861        line for every non-matching line. GNU grep prints nothing, and pcregrep now
862        does the same. The return code can be used to tell if there were any
863        non-matching lines.
864    
865    14. Added --file-offsets and --line-offsets to pcregrep.
866    
867    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
868        infinitely looping recursion. The bug was that positive lookaheads were not
869        being skipped when checking for a possible empty match (negative lookaheads
870        and both kinds of lookbehind were skipped).
871    
872    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
873        inclusion of <windows.h> to before rather than after the definition of
874        INVALID_FILE_ATTRIBUTES (patch from David Byron).
875    
876    17. Specifying a possessive quantifier with a specific limit for a Unicode
877        character property caused pcre_compile() to compile bad code, which led at
878        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
879        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
880        caused the error; without that there was no problem.
881    
882    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
883    
884    19. Added --enable-pcretest-libreadline.
885    
886    20. In pcrecpp.cc, the variable 'count' was incremented twice in
887        RE::GlobalReplace(). As a result, the number of replacements returned was
888        double what it should be. I removed one of the increments, but Craig sent a
889        later patch that removed the other one (the right fix) and added unit tests
890        that check the return values (which was not done before).
891    
892    21. Several CMake things:
893    
894        (1) Arranged that, when cmake is used on Unix, the libraries end up with
895            the names libpcre and libpcreposix, not just pcre and pcreposix.
896    
897        (2) The above change means that pcretest and pcregrep are now correctly
898            linked with the newly-built libraries, not previously installed ones.
899    
900        (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
901    
902    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
903        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
904        UTF-8 newline character). The key issue is that the pattern starts .*;
905        this means that the match must be either at the beginning, or after a
906        newline. The bug was in the code for advancing after a failed match and
907        checking that the new position followed a newline. It was not taking
908        account of UTF-8 characters correctly.
909    
910    23. PCRE was behaving differently from Perl in the way it recognized POSIX
911        character classes. PCRE was not treating the sequence [:...:] as a
912        character class unless the ... were all letters. Perl, however, seems to
913        allow any characters between [: and :], though of course it rejects as
914        unknown any "names" that contain non-letters, because all the known class
915        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
916        for example, whereas PCRE did not - it did not recognize a POSIX character
917        class. This seemed a bit dangerous, so the code has been changed to be
918        closer to Perl. The behaviour is not identical to Perl, because PCRE will
919        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
920        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
921        Perl does, and where it didn't before.
922    
923    24. Rewrite so as to remove the single use of %n from pcregrep because in some
924        Windows environments %n is disabled by default.
925    
926    
927    Version 7.4 21-Sep-07
928    ---------------------
929    
930    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
931        means that a class such as [\s] counted as "explicit reference to CR or
932        LF". That isn't really right - the whole point of the change was to try to
933        help when there was an actual mention of one of the two characters. So now
934        the change happens only if \r or \n (or a literal CR or LF) character is
935        encountered.
936    
937    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
938        of both had grown to the point where there were only 3 bits left.
939        Fortunately, there was spare space in the data structure, and so I have
940        moved the internal flags into a new 16-bit field to free up more option
941        bits.
942    
943    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
944        but did not set the internal JCHANGED flag - either of these is enough to
945        control the way the "get" function works - but the PCRE_INFO_JCHANGED
946        facility is supposed to tell if (?J) was ever used, so now (?J) at the
947        start sets both bits.
948    
949    4.  Added options (at build time, compile time, exec time) to change \R from
950        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
951    
952    5.  doc/pcresyntax.html was missing from the distribution.
953    
954    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
955        compatibility, even though it is no longer used.
956    
957    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
958        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
959        windows.h file is present (where different names are used). [This was
960        reversed later after testing - see 16 below.]
961    
962    8.  Changed all #include <config.h> to #include "config.h". There were also
963        some further <pcre.h> cases that I changed to "pcre.h".
964    
965    9.  When pcregrep was used with the --colour option, it missed the line ending
966        sequence off the lines that it output.
967    
968    10. It was pointed out to me that arrays of string pointers cause lots of
969        relocations when a shared library is dynamically loaded. A technique of
970        using a single long string with a table of offsets can drastically reduce
971        these. I have refactored PCRE in four places to do this. The result is
972        dramatic:
973    
974          Originally:                          290
975          After changing UCP table:            187
976          After changing error message table:   43
977          After changing table of "verbs"       36
978          After changing table of Posix names   22
979    
980        Thanks to the folks working on Gregex for glib for this insight.
981    
982    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
983        unicode-properties was also set.
984    
985    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
986    
987    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
988        checked only for CRLF.
989    
990    14. Added casts to pcretest.c to avoid compiler warnings.
991    
992    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
993    
994    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
995        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
996        entirely. This removes changes made in 7 above.
997    
998    17. The CMake files have been updated, and there is now more information about
999        building with CMake in the NON-UNIX-USE document.
1000    
1001    
1002    Version 7.3 28-Aug-07
1003    ---------------------
1004    
1005     1. In the rejigging of the build system that eventually resulted in 7.1, the
1006        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
1007        brackets there is not right, since it causes compilers to look for an
1008        installed pcre.h, not the version that is in the source that is being
1009        compiled (which of course may be different). I have changed it back to:
1010    
1011          #include "pcre.h"
1012    
1013        I have a vague recollection that the change was concerned with compiling in
1014        different directories, but in the new build system, that is taken care of
1015        by the VPATH setting the Makefile.
1016    
1017     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
1018        when the subject happened to end in the byte 0x85 (e.g. if the last
1019        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
1020        characters but of course it shouldn't be taken as a newline when it is part
1021        of another character. The bug was that, for an unlimited repeat of . in
1022        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
1023        characters when looking for a newline.
1024    
1025     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
1026    
1027     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
1028        in debug output.
1029    
1030     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
1031        long printing in the pcrecpp unittest when running under MinGW.
1032    
1033     6. ESC_K was left out of the EBCDIC table.
1034    
1035     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
1036        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
1037        limit also applies to "virtual nesting" when a pattern is recursive, and in
1038        this case 1000 isn't so big. I have been able to remove this limit at the
1039        expense of backing off one optimization in certain circumstances. Normally,
1040        when pcre_exec() would call its internal match() function recursively and
1041        immediately return the result unconditionally, it uses a "tail recursion"
1042        feature to save stack. However, when a subpattern that can match an empty
1043        string has an unlimited repetition quantifier, it no longer makes this
1044        optimization. That gives it a stack frame in which to save the data for
1045        checking that an empty string has been matched. Previously this was taken
1046        from the 1000-entry workspace that had been reserved. So now there is no
1047        explicit limit, but more stack is used.
1048    
1049     8. Applied Daniel's patches to solve problems with the import/export magic
1050        syntax that is required for Windows, and which was going wrong for the
1051        pcreposix and pcrecpp parts of the library. These were overlooked when this
1052        problem was solved for the main library.
1053    
1054     9. There were some crude static tests to avoid integer overflow when computing
1055        the size of patterns that contain repeated groups with explicit upper
1056        limits. As the maximum quantifier is 65535, the maximum group length was
1057        set at 30,000 so that the product of these two numbers did not overflow a
1058        32-bit integer. However, it turns out that people want to use groups that
1059        are longer than 30,000 bytes (though not repeat them that many times).
1060        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1061        made it possible to implement the integer overflow checks in a much more
1062        dynamic way, which I have now done. The artificial limitation on group
1063        length has been removed - we now have only the limit on the total length of
1064        the compiled pattern, which depends on the LINK_SIZE setting.
1065    
1066    10. Fixed a bug in the documentation for get/copy named substring when
1067        duplicate names are permitted. If none of the named substrings are set, the
1068        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1069        empty string.
1070    
1071    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1072        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1073        because the ] is interpreted as the first data character and the
1074        terminating ] is not found. PCRE has been made compatible with Perl in this
1075        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1076        cause memory overwriting.
1077    
1078    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1079        string has been matched (to stop an infinite loop). It was not recognizing
1080        a conditional subpattern that could match an empty string if that
1081        subpattern was within another subpattern. For example, it looped when
1082        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
1083        condition was not nested. This bug has been fixed.
1084    
1085    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1086        past the start of the subject in the presence of bytes with the top bit
1087        set, for example "\x8aBCD".
1088    
1089    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1090        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1091    
1092    14. Optimized (?!) to (*FAIL).
1093    
1094    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1095        This restricts code points to be within the range 0 to 0x10FFFF, excluding
1096        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1097        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1098        does: it's just the validity check that is more restrictive.
1099    
1100    16. Inserted checks for integer overflows during escape sequence (backslash)
1101        processing, and also fixed erroneous offset values for syntax errors during
1102        backslash processing.
1103    
1104    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1105        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1106    
1107    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1108        caused an overrun.
1109    
1110    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1111        something other than just ASCII characters) inside a group that had an
1112        unlimited repeat caused a loop at compile time (while checking to see
1113        whether the group could match an empty string).
1114    
1115    20. Debugging a pattern containing \p or \P could cause a crash. For example,
1116        [\P{Any}] did so. (Error in the code for printing property names.)
1117    
1118    21. An orphan \E inside a character class could cause a crash.
1119    
1120    22. A repeated capturing bracket such as (A)? could cause a wild memory
1121        reference during compilation.
1122    
1123    23. There are several functions in pcre_compile() that scan along a compiled
1124        expression for various reasons (e.g. to see if it's fixed length for look
1125        behind). There were bugs in these functions when a repeated \p or \P was
1126        present in the pattern. These operators have additional parameters compared
1127        with \d, etc, and these were not being taken into account when moving along
1128        the compiled data. Specifically:
1129    
1130        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1131            length.
1132    
1133        (b) An item such as \pL+ within a repeated group could cause crashes or
1134            loops.
1135    
1136        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1137            "reference to non-existent subpattern" error.
1138    
1139        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1140    
1141    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1142        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1143    
1144    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1145    
1146    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1147        character were causing crashes (broken optimization).
1148    
1149    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1150        \p or \P) caused a compile-time loop.
1151    
1152    28. More problems have arisen in unanchored patterns when CRLF is a valid line
1153        break. For example, the unstudied pattern [\r\n]A does not match the string
1154        "\r\nA" because change 7.0/46 below moves the current point on by two
1155        characters after failing to match at the start. However, the pattern \nA
1156        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
1157        the same is true. There doesn't seem any very clean way out of this, but
1158        what I have chosen to do makes the common cases work: PCRE now takes note
1159        of whether there can be an explicit match for \r or \n anywhere in the
1160        pattern, and if so, 7.0/46 no longer applies. As part of this change,
1161        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1162        pattern has explicit CR or LF references.
1163    
1164    29. Added (*CR) etc for changing newline setting at start of pattern.
1165    
1166    
1167    Version 7.2 19-Jun-07
1168    ---------------------
1169    
1170     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
1171        which is apparently normally available under Windows.
1172    
1173     2. Re-jig the pcregrep tests with different newline settings in an attempt
1174        to make them independent of the local environment's newline setting.
1175    
1176     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
1177    
1178     4. Some of the "internals" tests were previously cut out when the link size
1179        was not 2, because the output contained actual offsets. The recent new
1180        "Z" feature of pcretest means that these can be cut out, making the tests
1181        usable with all link sizes.
1182    
1183     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
1184        stack recursion. This gives a massive performance boost under BSD, but just
1185        a small improvement under Linux. However, it saves one field in the frame
1186        in all cases.
1187    
1188     6. Added more features from the forthcoming Perl 5.10:
1189    
1190        (a) (?-n) (where n is a string of digits) is a relative subroutine or
1191            recursion call. It refers to the nth most recently opened parentheses.
1192    
1193        (b) (?+n) is also a relative subroutine call; it refers to the nth next
1194            to be opened parentheses.
1195    
1196        (c) Conditions that refer to capturing parentheses can be specified
1197            relatively, for example, (?(-2)... or (?(+3)...
1198    
1199        (d) \K resets the start of the current match so that everything before
1200            is not part of it.
1201    
1202        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1203    
1204        (f) \g{name} is another synonym - part of Perl 5.10's unification of
1205            reference syntax.
1206    
1207        (g) (?| introduces a group in which the numbering of parentheses in each
1208            alternative starts with the same number.
1209    
1210        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
1211    
1212     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
1213        PCRE_INFO_JCHANGED.
1214    
1215     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
1216        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
1217        for detecting groups that can match an empty string.
1218    
1219     9. A pattern with a very large number of alternatives (more than several
1220        hundred) was running out of internal workspace during the pre-compile
1221        phase, where pcre_compile() figures out how much memory will be needed. A
1222        bit of new cunning has reduced the workspace needed for groups with
1223        alternatives. The 1000-alternative test pattern now uses 12 bytes of
1224        workspace instead of running out of the 4096 that are available.
1225    
1226    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
1227    
1228    11. Applied patch from Google to remove an optimization that didn't quite work.
1229        The report of the bug said:
1230    
1231          pcrecpp::RE("a*").FullMatch("aaa") matches, while
1232          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
1233          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
1234    
1235    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
1236        it matched the wrong number of bytes.
1237    
1238    
1239    Version 7.1 24-Apr-07
1240    ---------------------
1241    
1242     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
1243        that is more "standard", making use of automake and other Autotools. There
1244        is some re-arrangement of the files and adjustment of comments consequent
1245        on this.
1246    
1247     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
1248        for recursive directory scanning broke on some systems because the files
1249        are not scanned in any specific order and on different systems the order
1250        was different. A call to "sort" has been inserted into RunGrepTest for the
1251        approprate test as a short-term fix. In the longer term there may be an
1252        alternative.
1253    
1254     3. I had an email from Eric Raymond about problems translating some of PCRE's
1255        man pages to HTML (despite the fact that I distribute HTML pages, some
1256        people do their own conversions for various reasons). The problems
1257        concerned the use of low-level troff macros .br and .in. I have therefore
1258        removed all such uses from the man pages (some were redundant, some could
1259        be replaced by .nf/.fi pairs). The 132html script that I use to generate
1260        HTML has been updated to handle .nf/.fi and to complain if it encounters
1261        .br or .in.
1262    
1263     4. Updated comments in configure.ac that get placed in config.h.in and also
1264        arranged for config.h to be included in the distribution, with the name
1265        config.h.generic, for the benefit of those who have to compile without
1266        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
1267    
1268     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
1269        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
1270        makevp.bat for the current PCRE, using the additional files
1271        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
1272    
1273     6. A Windows user reported a minor discrepancy with test 2, which turned out
1274        to be caused by a trailing space on an input line that had got lost in his
1275        copy. The trailing space was an accident, so I've just removed it.
1276    
1277     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
1278        that is needed.
1279    
1280     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
1281        as "const" (a) because they are and (b) because it helps the PHP
1282        maintainers who have recently made a script to detect big data structures
1283        in the php code that should be moved to the .rodata section. I remembered
1284        to update Builducptable as well, so it won't revert if ucptable.h is ever
1285        re-created.
1286    
1287     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
1288        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
1289        order to be able to cut out the UTF-8 tables in the latter when UTF-8
1290        support is not required. This saves 1.5-2K of code, which is important in
1291        some applications.
1292    
1293        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
1294        so as not to refer to the tables, even though these functions will never be
1295        called when UTF-8 support is disabled. Otherwise there are problems with a
1296        shared library.
1297    
1298    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
1299    
1300        (a) It was defining its arguments as char * instead of void *.
1301    
1302        (b) It was assuming that all moves were upwards in memory; this was true
1303            a long time ago when I wrote it, but is no longer the case.
1304    
1305        The emulated memove() is provided for those environments that have neither
1306        memmove() nor bcopy(). I didn't think anyone used it these days, but that
1307        is clearly not the case, as these two bugs were recently reported.
1308    
1309    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
1310        and Detrail to create the HTML documentation, the .txt form of the man
1311        pages, and it removes trailing spaces from listed files. It also creates
1312        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
1313        case, it wraps all the #defines with #ifndefs. This script should be run
1314        before "make dist".
1315    
1316    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1317        with Unicode property support.
1318    
1319        (a) For a maximizing quantifier, if the two different cases of the
1320            character were of different lengths in their UTF-8 codings (there are
1321            some cases like this - I found 11), and the matching function had to
1322            back up over a mixture of the two cases, it incorrectly assumed they
1323            were both the same length.
1324    
1325        (b) When PCRE was configured to use the heap rather than the stack for
1326            recursion during matching, it was not correctly preserving the data for
1327            the other case of a UTF-8 character when checking ahead for a match
1328            while processing a minimizing repeat. If the check also involved
1329            matching a wide character, but failed, corruption could cause an
1330            erroneous result when trying to check for a repeat of the original
1331            character.
1332    
1333    13. Some tidying changes to the testing mechanism:
1334    
1335        (a) The RunTest script now detects the internal link size and whether there
1336            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1337            values substituted by "configure". (The RunGrepTest script already did
1338            this for UTF-8.) The configure.ac script no longer substitutes the
1339            relevant variables.
1340    
1341        (b) The debugging options /B and /D in pcretest show the compiled bytecode
1342            with length and offset values. This means that the output is different
1343            for different internal link sizes. Test 2 is skipped for link sizes
1344            other than 2 because of this, bypassing the problem. Unfortunately,
1345            there was also a test in test 3 (the locale tests) that used /B and
1346            failed for link sizes other than 2. Rather than cut the whole test out,
1347            I have added a new /Z option to pcretest that replaces the length and
1348            offset values with spaces. This is now used to make test 3 independent
1349            of link size. (Test 2 will be tidied up later.)
1350    
1351    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1352        segmentation fault instead of returning the appropriate error message.
1353    
1354    15. In multiline mode when the newline sequence was set to "any", the pattern
1355        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1356        This doesn't seem right; it now treats the CRLF combination as the line
1357        ending, and so does not match in that case. It's only a pattern such as ^$
1358        that would hit this one: something like ^ABC$ would have failed after \r
1359        and then tried again after \r\n.
1360    
1361    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1362        in an attempt to make files that differ only in their line terminators
1363        compare equal. This works on Linux.
1364    
1365    17. Under certain error circumstances pcregrep might try to free random memory
1366        as it exited. This is now fixed, thanks to valgrind.
1367    
1368    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1369        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1370        was because its rules for how to advance for /g after matching an empty
1371        string at the end of a line did not allow for this case. They now check for
1372        it specially.
1373    
1374    20. pcretest is supposed to handle patterns and data of any length, by
1375        extending its buffers when necessary. It was getting this wrong when the
1376        buffer for a data line had to be extended.
1377    
1378    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1379        CRLF as a newline sequence.
1380    
1381    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1382        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1383        I have nevertheless tidied it up.
1384    
1385    23. Added some casts to kill warnings from HP-UX ia64 compiler.
1386    
1387    24. Added a man page for pcre-config.
1388    
1389    
1390    Version 7.0 19-Dec-06
1391    ---------------------
1392    
1393     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1394        moving to gcc 4.1.1.
1395    
1396     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1397        sys/time.h, which is documented as needed for this function. It doesn't
1398        seem to matter on Linux, but it showed up on some releases of OS X.
1399    
1400     3. It seems that there are systems where bytes whose values are greater than
1401        127 match isprint() in the "C" locale. The "C" locale should be the
1402        default when a C program starts up. In most systems, only ASCII printing
1403        characters match isprint(). This difference caused the output from pcretest
1404        to vary, making some of the tests fail. I have changed pcretest so that:
1405    
1406        (a) When it is outputting text in the compiled version of a pattern, bytes
1407            other than 32-126 are always shown as hex escapes.
1408    
1409        (b) When it is outputting text that is a matched part of a subject string,
1410            it does the same, unless a different locale has been set for the match
1411            (using the /L modifier). In this case, it uses isprint() to decide.
1412    
1413     4. Fixed a major bug that caused incorrect computation of the amount of memory
1414        required for a compiled pattern when options that changed within the
1415        pattern affected the logic of the preliminary scan that determines the
1416        length. The relevant options are -x, and -i in UTF-8 mode. The result was
1417        that the computed length was too small. The symptoms of this bug were
1418        either the PCRE error "internal error: code overflow" from pcre_compile(),
1419        or a glibc crash with a message such as "pcretest: free(): invalid next
1420        size (fast)". Examples of patterns that provoked this bug (shown in
1421        pcretest format) are:
1422    
1423          /(?-x: )/x
1424          /(?x)(?-x: \s*#\s*)/
1425          /((?i)[\x{c0}])/8
1426          /(?i:[\x{c0}])/8
1427    
1428        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1429        is now done differently.
1430    
1431     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1432        wrapper classes; (b) implement a new function in the C++ scanner that is
1433        more efficient than the old way of doing things because it avoids levels of
1434        recursion in the regex matching; (c) add a paragraph to the documentation
1435        for the FullMatch() function.
1436    
1437     6. The escape sequence \n was being treated as whatever was defined as
1438        "newline". Not only was this contrary to the documentation, which states
1439        that \n is character 10 (hex 0A), but it also went horribly wrong when
1440        "newline" was defined as CRLF. This has been fixed.
1441    
1442     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1443        was being set to -1 for the "end of line" case (supposedly a value that no
1444        character can have). Though this value is never used (the check for end of
1445        line is "zero bytes in current character"), it caused compiler complaints.
1446        I've changed it to 0xffffffff.
1447    
1448     8. In pcre_version.c, the version string was being built by a sequence of
1449        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1450        string (as it is for production releases) called a macro with an empty
1451        argument. The C standard says the result of this is undefined. The gcc
1452        compiler treats it as an empty string (which was what was wanted) but it is
1453        reported that Visual C gives an error. The source has been hacked around to
1454        avoid this problem.
1455    
1456     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1457        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1458        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1459        of them did).
1460    
1461    10. Originally, pcretest opened its input and output without "b"; then I was
1462        told that "b" was needed in some environments, so it was added for release
1463        5.0 to both the input and output. (It makes no difference on Unix-like
1464        systems.) Later I was told that it is wrong for the input on Windows. I've
1465        now abstracted the modes into two macros, to make it easier to fiddle with
1466        them, and removed "b" from the input mode under Windows.
1467    
1468    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1469    
1470    12. Added -help and --help to pcretest as an official way of being reminded
1471        of the options.
1472    
1473    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1474        and pcrecpp.cc because they annoy compilers at high warning levels.
1475    
1476    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1477    
1478    15. Fixed an occurrence of == in configure.ac that should have been = (shell
1479        scripts are not C programs :-) and which was not noticed because it works
1480        on Linux.
1481    
1482    16. pcretest is supposed to handle any length of pattern and data line (as one
1483        line or as a continued sequence of lines) by extending its input buffer if
1484        necessary. This feature was broken for very long pattern lines, leading to
1485        a string of junk being passed to pcre_compile() if the pattern was longer
1486        than about 50K.
1487    
1488    17. I have done a major re-factoring of the way pcre_compile() computes the
1489        amount of memory needed for a compiled pattern. Previously, there was code
1490        that made a preliminary scan of the pattern in order to do this. That was
1491        OK when PCRE was new, but as the facilities have expanded, it has become
1492        harder and harder to keep it in step with the real compile phase, and there
1493        have been a number of bugs (see for example, 4 above). I have now found a
1494        cunning way of running the real compile function in a "fake" mode that
1495        enables it to compute how much memory it would need, while actually only
1496        ever using a few hundred bytes of working memory and without too many
1497        tests of the mode. This should make future maintenance and development
1498        easier. A side effect of this work is that the limit of 200 on the nesting
1499        depth of parentheses has been removed (though this was never a serious
1500        limitation, I suspect). However, there is a downside: pcre_compile() now
1501        runs more slowly than before (30% or more, depending on the pattern). I
1502        hope this isn't a big issue. There is no effect on runtime performance.
1503    
1504    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1505        newline (only possible for the last line of a file) and it was a
1506        pattern that set a locale (followed by /Lsomething), pcretest crashed.
1507    
1508    19. Added additional timing features to pcretest. (1) The -tm option now times
1509        matching only, not compiling. (2) Both -t and -tm can be followed, as a
1510        separate command line item, by a number that specifies the number of
1511        repeats to use when timing. The default is 50000; this gives better
1512        precision, but takes uncomfortably long for very large patterns.
1513    
1514    20. Extended pcre_study() to be more clever in cases where a branch of a
1515        subpattern has no definite first character. For example, (a*|b*)[cd] would
1516        previously give no result from pcre_study(). Now it recognizes that the
1517        first character must be a, b, c, or d.
1518    
1519    21. There was an incorrect error "recursive call could loop indefinitely" if
1520        a subpattern (or the entire pattern) that was being tested for matching an
1521        empty string contained only one non-empty item after a nested subpattern.
1522        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1523        incorrectly, because the \d was being skipped in the check.
1524    
1525    22. The pcretest program now has a new pattern option /B and a command line
1526        option -b, which is equivalent to adding /B to every pattern. This causes
1527        it to show the compiled bytecode, without the additional information that
1528        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1529        is the same as /B/I).
1530    
1531    23. A new optimization is now able automatically to treat some sequences such
1532        as a*b as a*+b. More specifically, if something simple (such as a character
1533        or a simple class like \d) has an unlimited quantifier, and is followed by
1534        something that cannot possibly match the quantified thing, the quantifier
1535        is automatically "possessified".
1536    
1537    24. A recursive reference to a subpattern whose number was greater than 39
1538        went wrong under certain circumstances in UTF-8 mode. This bug could also
1539        have affected the operation of pcre_study().
1540    
1541    25. Realized that a little bit of performance could be had by replacing
1542        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1543    
1544    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1545    
1546    27. Possessive quantifiers such as a++ were previously implemented by turning
1547        them into atomic groups such as ($>a+). Now they have their own opcodes,
1548        which improves performance. This includes the automatically created ones
1549        from 23 above.
1550    
1551    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1552        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1553        the first matched character to be a colon. This applied both to named and
1554        numbered groups.
1555    
1556    29. The ucpinternal.h header file was missing its idempotency #ifdef.
1557    
1558    30. I was sent a "project" file called libpcre.a.dev which I understand makes
1559        building PCRE on Windows easier, so I have included it in the distribution.
1560    
1561    31. There is now a check in pcretest against a ridiculously large number being
1562        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1563        loop, the loop is abandoned.
1564    
1565    32. Forward references to subpatterns in conditions such as (?(2)...) where
1566        subpattern 2 is defined later cause pcre_compile() to search forwards in
1567        the pattern for the relevant set of parentheses. This search went wrong
1568        when there were unescaped parentheses in a character class, parentheses
1569        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1570    
1571    33. "Subroutine" calls and backreferences were previously restricted to
1572        referencing subpatterns earlier in the regex. This restriction has now
1573        been removed.
1574    
1575    34. Added a number of extra features that are going to be in Perl 5.10. On the
1576        whole, these are just syntactic alternatives for features that PCRE had
1577        previously implemented using the Python syntax or my own invention. The
1578        other formats are all retained for compatibility.
1579    
1580        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1581            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1582            also .NET compatible.
1583    
1584        (b) A recursion or subroutine call to a named group can now be defined as
1585            (?&name) as well as (?P>name).
1586    
1587        (c) A backreference to a named group can now be defined as \k<name> or
1588            \k'name' as well as (?P=name). The new forms, as well as being in Perl
1589            5.10, are also .NET compatible.
1590    
1591        (d) A conditional reference to a named group can now use the syntax
1592            (?(<name>) or (?('name') as well as (?(name).
1593    
1594        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1595            groups (named and numbered) that are never evaluated inline, but can be
1596            called as "subroutines" from elsewhere. In effect, the DEFINE condition
1597            is always false. There may be only one alternative in such a group.
1598    
1599        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1600            as the simple (?(R). The condition is true only if the most recent
1601            recursion is that of the given number or name. It does not search out
1602            through the entire recursion stack.
1603    
1604        (g) The escape \gN or \g{N} has been added, where N is a positive or
1605            negative number, specifying an absolute or relative reference.
1606    
1607    35. Tidied to get rid of some further signed/unsigned compiler warnings and
1608        some "unreachable code" warnings.
1609    
1610    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1611        things, this adds five new scripts.
1612    
1613    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1614        There were also incompatibilities regarding the handling of \Q..\E inside
1615        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1616        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1617    
1618    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1619        matches an empty string, and forcibly breaks the loop. There were bugs in
1620        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
1621        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
1622        separate and independent bugs (that affected different cases) have been
1623        fixed.
1624    
1625    39. Refactored the code to abolish the use of different opcodes for small
1626        capturing bracket numbers. This is a tidy that I avoided doing when I
1627        removed the limit on the number of capturing brackets for 3.5 back in 2001.
1628        The new approach is not only tidier, it makes it possible to reduce the
1629        memory needed to fix the previous bug (38).
1630    
1631    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1632        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1633        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1634        mode.
1635    
1636    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1637        report.
1638    
1639    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1640        copy construction and assignment in the C++ wrapper.
1641    
1642    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1643        couple of bugs that could have given wrong results in the "--newline=crlf"
1644        case.
1645    
1646    44. Added a number of casts and did some reorganization of signed/unsigned int
1647        variables following suggestions from Dair Grant. Also renamed the variable
1648        "this" as "item" because it is a C++ keyword.
1649    
1650    45. Arranged for dftables to add
1651    
1652          #include "pcre_internal.h"
1653    
1654        to pcre_chartables.c because without it, gcc 4.x may remove the array
1655        definition from the final binary if PCRE is built into a static library and
1656        dead code stripping is activated.
1657    
1658    46. For an unanchored pattern, if a match attempt fails at the start of a
1659        newline sequence, and the newline setting is CRLF or ANY, and the next two
1660        characters are CRLF, advance by two characters instead of one.
1661    
1662    
1663    Version 6.7 04-Jul-06
1664    ---------------------
1665    
1666     1. In order to handle tests when input lines are enormously long, pcretest has
1667        been re-factored so that it automatically extends its buffers when
1668        necessary. The code is crude, but this _is_ just a test program. The
1669        default size has been increased from 32K to 50K.
1670    
1671     2. The code in pcre_study() was using the value of the re argument before
1672        testing it for NULL. (Of course, in any sensible call of the function, it
1673        won't be NULL.)
1674    
1675     3. The memmove() emulation function in pcre_internal.h, which is used on
1676        systems that lack both memmove() and bcopy() - that is, hardly ever -
1677        was missing a "static" storage class specifier.
1678    
1679     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1680        containing an extended class (one that cannot be represented by a bitmap
1681        because it contains high-valued characters or Unicode property items, e.g.
1682        [\pZ]). Almost always one would set UTF-8 mode when processing such a
1683        pattern, but PCRE should not loop if you do not (it no longer does).
1684        [Detail: two cases were found: (a) a repeated subpattern containing an
1685        extended class; (b) a recursive reference to a subpattern that followed a
1686        previous extended class. It wasn't skipping over the extended class
1687        correctly when UTF-8 mode was not set.]
1688    
1689     5. A negated single-character class was not being recognized as fixed-length
1690        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1691        compile error "lookbehind assertion is not fixed length".
1692    
1693     6. The RunPerlTest auxiliary script was showing an unexpected difference
1694        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1695        write a Perl script that can interpret lines of an input file either as
1696        byte characters or as UTF-8, which is what "perltest" was being required to
1697        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1698        can't do is switch easily at run time between having the "use utf8;" pragma
1699        or not. In the end, I fudged it by using the RunPerlTest script to insert
1700        "use utf8;" explicitly for the UTF-8 tests.
1701    
1702     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1703        the end of the subject string, contrary to the documentation and to what
1704        Perl does. This was true of both matching functions. Now it matches only at
1705        the start of the subject and immediately after *internal* newlines.
1706    
1707     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1708        a pointer to an int instead of a pointer to an unsigned long int. This
1709        caused problems on 64-bit systems.
1710    
1711     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1712        instance of the 'standard' template library not being so standard".
1713    
1714    10. There was no check on the number of named subpatterns nor the maximum
1715        length of a subpattern name. The product of these values is used to compute
1716        the size of the memory block for a compiled pattern. By supplying a very
1717        long subpattern name and a large number of named subpatterns, the size
1718        computation could be caused to overflow. This is now prevented by limiting
1719        the length of names to 32 characters, and the number of named subpatterns
1720        to 10,000.
1721    
1722    11. Subpatterns that are repeated with specific counts have to be replicated in
1723        the compiled pattern. The size of memory for this was computed from the
1724        length of the subpattern and the repeat count. The latter is limited to
1725        65535, but there was no limit on the former, meaning that integer overflow
1726        could in principle occur. The compiled length of a repeated subpattern is
1727        now limited to 30,000 bytes in order to prevent this.
1728    
1729    12. Added the optional facility to have named substrings with the same name.
1730    
1731    13. Added the ability to use a named substring as a condition, using the
1732        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1733        are numbers (not recommended). Forward references are permitted.
1734    
1735    14. Added forward references in named backreferences (if you see what I mean).
1736    
1737    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1738        pattern could run off the end of the subject. For example, the pattern
1739        "(?s)(.{1,5})"8 did this with the subject "ab".
1740    
1741    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1742        PCRE_CASELESS was set when matching characters that were quantified with ?
1743        or *.
1744    
1745    17. A character class other than a single negated character that had a minimum
1746        but no maximum quantifier - for example [ab]{6,} - was not handled
1747        correctly by pce_dfa_exec(). It would match only one character.
1748    
1749    18. A valid (though odd) pattern that looked like a POSIX character
1750        class but used an invalid character after [ (for example [[,abc,]]) caused
1751        pcre_compile() to give the error "Failed: internal error: code overflow" or
1752        in some cases to crash with a glibc free() error. This could even happen if
1753        the pattern terminated after [[ but there just happened to be a sequence of
1754        letters, a binary zero, and a closing ] in the memory that followed.
1755    
1756    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1757        over the years. Originally (before any Unicode support), just the bottom 8
1758        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1759        output from "man perlunicode" includes this:
1760    
1761          The regular expression compiler produces polymorphic opcodes.  That
1762          is, the pattern adapts to the data and automatically switches to
1763          the Unicode character scheme when presented with Unicode data--or
1764          instead uses a traditional byte scheme when presented with byte
1765          data.
1766    
1767        Sadly, a wide octal escape does not cause a switch, and in a string with
1768        no other multibyte characters, these octal escapes are treated as before.
1769        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
1770        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1771        Unicode string.
1772    
1773        I have not perpetrated such confusion in PCRE. Up till now, it took just
1774        the bottom 8 bits, as in old Perl. I have now made octal escapes with
1775        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1776        translate to the appropriate multibyte character.
1777    
1778    29. Applied some refactoring to reduce the number of warnings from Microsoft
1779        and Borland compilers. This has included removing the fudge introduced
1780        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1781        a warning about an unused variable.
1782    
1783    21. PCRE has not included VT (character 0x0b) in the set of whitespace
1784        characters since release 4.0, because Perl (from release 5.004) does not.
1785        [Or at least, is documented not to: some releases seem to be in conflict
1786        with the documentation.] However, when a pattern was studied with
1787        pcre_study() and all its branches started with \s, PCRE still included VT
1788        as a possible starting character. Of course, this did no harm; it just
1789        caused an unnecessary match attempt.
1790    
1791    22. Removed a now-redundant internal flag bit that recorded the fact that case
1792        dependency changed within the pattern. This was once needed for "required
1793        byte" processing, but is no longer used. This recovers a now-scarce options
1794        bit. Also moved the least significant internal flag bit to the most-
1795        significant bit of the word, which was not previously used (hangover from
1796        the days when it was an int rather than a uint) to free up another bit for
1797        the future.
1798    
1799    23. Added support for CRLF line endings as well as CR and LF. As well as the
1800        default being selectable at build time, it can now be changed at runtime
1801        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1802        specify that it is scanning data with non-default line endings.
1803    
1804    24. Changed the definition of CXXLINK to make it agree with the definition of
1805        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1806    
1807    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1808        recursions. This makes a big different to stack usage for some patterns.
1809    
1810    26. If a subpattern containing a named recursion or subroutine reference such
1811        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1812        the space required for the compiled pattern went wrong and gave too small a
1813        value. Depending on the environment, this could lead to "Failed: internal
1814        error: code overflow at offset 49" or "glibc detected double free or
1815        corruption" errors.
1816    
1817    27. Applied patches from Google (a) to support the new newline modes and (b) to
1818        advance over multibyte UTF-8 characters in GlobalReplace.
1819    
1820    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1821        difference for some implementation of PCRE in some Windows version.
1822    
1823    29. Added some extra testing facilities to pcretest:
1824    
1825        \q<number>   in a data line sets the "match limit" value
1826        \Q<number>   in a data line sets the "match recursion limt" value
1827        -S <number>  sets the stack size, where <number> is in megabytes
1828    
1829        The -S option isn't available for Windows.
1830    
1831    
1832    Version 6.6 06-Feb-06
1833    ---------------------
1834    
1835     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1836        in pcreposix.h. I have copied the definition from pcre.h.
1837    
1838     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1839        because pcre.h is no longer a built file.
1840    
1841     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1842        not normally included in the compiled code.
1843    
1844    
1845    Version 6.5 01-Feb-06
1846    ---------------------
1847    
1848     1. When using the partial match feature with pcre_dfa_exec(), it was not
1849        anchoring the second and subsequent partial matches at the new starting
1850        point. This could lead to incorrect results. For example, with the pattern
1851        /1234/, partially matching against "123" and then "a4" gave a match.
1852    
1853     2. Changes to pcregrep:
1854    
1855        (a) All non-match returns from pcre_exec() were being treated as failures
1856            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1857            error message is output. Some extra information is given for the
1858            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
1859            probably the only errors that are likely to be caused by users (by
1860            specifying a regex that has nested indefinite repeats, for instance).
1861            If there are more than 20 of these errors, pcregrep is abandoned.
1862    
1863        (b) A binary zero was treated as data while matching, but terminated the
1864            output line if it was written out. This has been fixed: binary zeroes
1865            are now no different to any other data bytes.
1866    
1867        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1868            used to set a locale for matching. The --locale=xxxx long option has
1869            been added (no short equivalent) to specify a locale explicitly on the
1870            pcregrep command, overriding the environment variables.
1871    
1872        (d) When -B was used with -n, some line numbers in the output were one less
1873            than they should have been.
1874    
1875        (e) Added the -o (--only-matching) option.
1876    
1877        (f) If -A or -C was used with -c (count only), some lines of context were
1878            accidentally printed for the final match.
1879    
1880        (g) Added the -H (--with-filename) option.
1881    
1882        (h) The combination of options -rh failed to suppress file names for files
1883            that were found from directory arguments.
1884    
1885        (i) Added the -D (--devices) and -d (--directories) options.
1886    
1887        (j) Added the -F (--fixed-strings) option.
1888    
1889        (k) Allow "-" to be used as a file name for -f as well as for a data file.
1890    
1891        (l) Added the --colo(u)r option.
1892    
1893        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1894            is not present by default.
1895    
1896     3. A nasty bug was discovered in the handling of recursive patterns, that is,
1897        items such as (?R) or (?1), when the recursion could match a number of
1898        alternatives. If it matched one of the alternatives, but subsequently,
1899        outside the recursion, there was a failure, the code tried to back up into
1900        the recursion. However, because of the way PCRE is implemented, this is not
1901        possible, and the result was an incorrect result from the match.
1902    
1903        In order to prevent this happening, the specification of recursion has
1904        been changed so that all such subpatterns are automatically treated as
1905        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1906    
1907     4. I had overlooked the fact that, in some locales, there are characters for
1908        which isalpha() is true but neither isupper() nor islower() are true. In
1909        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1910        and ordfeminine) are like this. This affected the treatment of \w and \W
1911        when they appeared in character classes, but not when they appeared outside
1912        a character class. The bit map for "word" characters is now created
1913        separately from the results of isalnum() instead of just taking it from the
1914        upper, lower, and digit maps. (Plus the underscore character, of course.)
1915    
1916     5. The above bug also affected the handling of POSIX character classes such as
1917        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1918        permanent tables. Instead, the bit maps for such a class were previously
1919        created as the appropriate unions of the upper, lower, and digit bitmaps.
1920        Now they are created by subtraction from the [[:word:]] class, which has
1921        its own bitmap.
1922    
1923     6. The [[:blank:]] character class matches horizontal, but not vertical space.
1924        It is created by subtracting the vertical space characters (\x09, \x0a,
1925        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1926        subtraction was done in the overall bitmap for a character class, meaning
1927        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1928        be recognized. This bug has been fixed.
1929    
1930     7. Patches from the folks at Google:
1931    
1932          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1933          real life, but is still worth protecting against".
1934    
1935          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1936          regular expressions".
1937    
1938          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1939          have it.
1940    
1941          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1942          "configure" and the latter not, in order to fix a problem somebody had
1943          with compiling the Arg class on HP-UX.
1944    
1945          (e) Improve the error-handling of the C++ wrapper a little bit.
1946    
1947          (f) New tests for checking recursion limiting.
1948    
1949     8. The pcre_memmove() function, which is used only if the environment does not
1950        have a standard memmove() function (and is therefore rarely compiled),
1951        contained two bugs: (a) use of int instead of size_t, and (b) it was not
1952        returning a result (though PCRE never actually uses the result).
1953    
1954     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1955        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1956        returned instead of calling malloc() with an overflowing number that would
1957        most likely cause subsequent chaos.
1958    
1959    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1960    
1961    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1962        with this option is matched, the nmatch and pmatch options of regexec() are
1963        ignored.
1964    
1965    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1966        provided in case anyone wants to the the POSIX interface with UTF-8
1967        strings.
1968    
1969    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1970        C++ linking (needed for some HP-UX environments).
1971    
1972    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1973        (unused parameter) and in the pcre_printint() function (omitted "default"
1974        switch label when the default is to do nothing).
1975    
1976    15. Added some code to make it possible, when PCRE is compiled as a C++
1977        library, to replace subject pointers for pcre_exec() with a smart pointer
1978        class, thus making it possible to process discontinuous strings.
1979    
1980    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1981        much the same function. They were added by different people who were trying
1982        to make PCRE easy to compile on non-Unix systems. It has been suggested
1983        that PCRE_EXPORT be abolished now that there is more automatic apparatus
1984        for compiling on Windows systems. I have therefore replaced it with
1985        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1986        defaults to "extern" for C or "extern C" for C++, which works fine on
1987        Unix-like systems. It is now possible to override the value of PCRE_DATA_
1988        SCOPE with something explicit in config.h. In addition:
1989    
1990        (a) pcreposix.h still had just "extern" instead of either of these macros;
1991            I have replaced it with PCRE_DATA_SCOPE.
1992    
1993        (b) Functions such as _pcre_xclass(), which are internal to the library,
1994            but external in the C sense, all had PCRE_EXPORT in their definitions.
1995            This is apparently wrong for the Windows case, so I have removed it.
1996            (It makes no difference on Unix-like systems.)
1997    
1998    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1999        of recursive calls to match(). This is different to MATCH_LIMIT because
2000        that limits the total number of calls to match(), not all of which increase
2001        the depth of recursion. Limiting the recursion depth limits the amount of
2002        stack (or heap if NO_RECURSE is set) that is used. The default can be set
2003        when PCRE is compiled, and changed at run time. A patch from Google adds
2004        this functionality to the C++ interface.
2005    
2006    18. Changes to the handling of Unicode character properties:
2007    
2008        (a) Updated the table to Unicode 4.1.0.
2009    
2010        (b) Recognize characters that are not in the table as "Cn" (undefined).
2011    
2012        (c) I revised the way the table is implemented to a much improved format
2013            which includes recognition of ranges. It now supports the ranges that
2014            are defined in UnicodeData.txt, and it also amalgamates other
2015            characters into ranges. This has reduced the number of entries in the
2016            table from around 16,000 to around 3,000, thus reducing its size
2017            considerably. I realized I did not need to use a tree structure after
2018            all - a binary chop search is just as efficient. Having reduced the
2019            number of entries, I extended their size from 6 bytes to 8 bytes to
2020            allow for more data.
2021    
2022        (d) Added support for Unicode script names via properties such as \p{Han}.
2023    
2024    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
2025        matching that character.
2026    
2027    20. When matching a repeated Unicode property with a minimum greater than zero,
2028        (for example \pL{2,}), PCRE could look past the end of the subject if it
2029        reached it while seeking the minimum number of characters. This could
2030        happen only if some of the characters were more than one byte long, because
2031        there is a check for at least the minimum number of bytes.
2032    
2033    21. Refactored the implementation of \p and \P so as to be more general, to
2034        allow for more different types of property in future. This has changed the
2035        compiled form incompatibly. Anybody with saved compiled patterns that use
2036        \p or \P will have to recompile them.
2037    
2038    22. Added "Any" and "L&" to the supported property types.
2039    
2040    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
2041        but give a compile time error if the value is greater than 0xff.
2042    
2043    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
2044        accidentally not being installed or uninstalled.
2045    
2046    25. The pcre.h file was built from pcre.h.in, but the only changes that were
2047        made were to insert the current release number. This seemed silly, because
2048        it made things harder for people building PCRE on systems that don't run
2049        "configure". I have turned pcre.h into a distributed file, no longer built
2050        by "configure", with the version identification directly included. There is
2051        no longer a pcre.h.in file.
2052    
2053        However, this change necessitated a change to the pcre-config script as
2054        well. It is built from pcre-config.in, and one of the substitutions was the
2055        release number. I have updated configure.ac so that ./configure now finds
2056        the release number by grepping pcre.h.
2057    
2058    26. Added the ability to run the tests under valgrind.
2059    
2060    
2061    Version 6.4 05-Sep-05
2062    ---------------------
2063    
2064     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
2065        "--" to be printed when multiple files were scanned, even when none of the
2066        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
2067        consider it to be a bug, and have restored the previous behaviour.
2068    
2069     2. A couple of code tidies to get rid of compiler warnings.
2070    
2071     3. The pcretest program used to cheat by referring to symbols in the library
2072        whose names begin with _pcre_. These are internal symbols that are not
2073        really supposed to be visible externally, and in some environments it is
2074        possible to suppress them. The cheating is now confined to including
2075        certain files from the library's source, which is a bit cleaner.
2076    
2077     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
2078        file's purpose clearer.
2079    
2080     5. Reorganized pcre_ucp_findchar().
2081    
2082    
2083    Version 6.3 15-Aug-05
2084    ---------------------
2085    
2086     1. The file libpcre.pc.in did not have general read permission in the tarball.
2087    
2088     2. There were some problems when building without C++ support:
2089    
2090        (a) If C++ support was not built, "make install" and "make test" still
2091            tried to test it.
2092    
2093        (b) There were problems when the value of CXX was explicitly set. Some
2094            changes have been made to try to fix these, and ...
2095    
2096        (c) --disable-cpp can now be used to explicitly disable C++ support.
2097    
2098        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
2099            backslash in a target when C++ was disabled. This confuses some
2100            versions of "make", apparently. Using an intermediate variable solves
2101            this. (Same for CPP_LOBJ.)
2102    
2103     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
2104        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
2105        necessary on certain architectures.
2106    
2107     4. Added a setting of -export-symbols-regex to the link command to remove
2108        those symbols that are exported in the C sense, but actually are local
2109        within the library, and not documented. Their names all begin with
2110        "_pcre_". This is not a perfect job, because (a) we have to except some
2111        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
2112        available (and never for static libraries). I have made a note to try to
2113        find a way round (a) in the future.
2114    
2115    
2116    Version 6.2 01-Aug-05
2117    ---------------------
2118    
2119     1. There was no test for integer overflow of quantifier values. A construction
2120        such as {1111111111111111} would give undefined results. What is worse, if
2121        a minimum quantifier for a parenthesized subpattern overflowed and became
2122        negative, the calculation of the memory size went wrong. This could have
2123        led to memory overwriting.
2124    
2125     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
2126    
2127     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
2128        operating environments where this matters.
2129    
2130     4. Applied Giuseppe Maxia's patch to add additional features for controlling
2131        PCRE options from within the C++ wrapper.
2132    
2133     5. Named capturing subpatterns were not being correctly counted when a pattern
2134        was compiled. This caused two problems: (a) If there were more than 100
2135        such subpatterns, the calculation of the memory needed for the whole
2136        compiled pattern went wrong, leading to an overflow error. (b) Numerical
2137        back references of the form \12, where the number was greater than 9, were
2138        not recognized as back references, even though there were sufficient
2139        previous subpatterns.
2140    
2141     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
2142        versions of gcc, e.g. 2.95.4.
2143    
2144    
2145    Version 6.1 21-Jun-05
2146    ---------------------
2147    
2148     1. There was one reference to the variable "posix" in pcretest.c that was not
2149        surrounded by "#if !defined NOPOSIX".
2150    
2151     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
2152        the cross-check on the old pcre_info() function, for the benefit of the
2153        cut-down version of PCRE that is currently imported into Exim.
2154    
2155     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
2156        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
2157        patterns that start (?i) but I don't think that matters. The use of (?i) is
2158        just an example; this all applies to the other options as well.
2159    
2160     4. Since libtool seems to echo the compile commands it is issuing, the output
2161        from "make" can be reduced a bit by putting "@" in front of each libtool
2162        compile command.
2163    
2164     5. Patch from the folks at Google for configure.in to be a bit more thorough
2165        in checking for a suitable C++ installation before trying to compile the
2166        C++ stuff. This should fix a reported problem when a compiler was present,
2167        but no suitable headers.
2168    
2169     6. The man pages all had just "PCRE" as their title. I have changed them to
2170        be the relevant file name. I have also arranged that these names are
2171        retained in the file doc/pcre.txt, which is a concatenation in text format
2172        of all the man pages except the little individual ones for each function.
2173    
2174     7. The NON-UNIX-USE file had not been updated for the different set of source
2175        files that come with release 6. I also added a few comments about the C++
2176        wrapper.
2177    
2178    
2179    Version 6.0 07-Jun-05
2180    ---------------------
2181    
2182     1. Some minor internal re-organization to help with my DFA experiments.
2183    
2184     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
2185        didn't matter for the library itself when fully configured, but did matter
2186        when compiling without UCP support, or within Exim, where the ucp files are
2187        not imported.
2188    
2189     3. Refactoring of the library code to split up the various functions into
2190        different source modules. The addition of the new DFA matching code (see
2191        below) to a single monolithic source would have made it really too
2192        unwieldy, quite apart from causing all the code to be include in a
2193        statically linked application, when only some functions are used. This is
2194        relevant even without the DFA addition now that patterns can be compiled in
2195        one application and matched in another.
2196    
2197        The downside of splitting up is that there have to be some external
2198        functions and data tables that are used internally in different modules of
2199        the library but which are not part of the API. These have all had their
2200        names changed to start with "_pcre_" so that they are unlikely to clash
2201        with other external names.
2202    
2203     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
2204        a different (DFA) algorithm. Although it is slower than the original
2205        function, it does have some advantages for certain types of matching
2206        problem.
2207    
2208     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
2209        including restarting after a partial match.
2210    
2211     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
2212        defined when compiling for Windows was sent to me. I have put it into the
2213        code, though I have no means of testing or verifying it.
2214    
2215     7. Added the pcre_refcount() auxiliary function.
2216    
2217     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
2218        match before or at the first newline in the subject string. In pcretest,
2219        the /f option on a pattern can be used to set this.
2220    
2221     9. A repeated \w when used in UTF-8 mode with characters greater than 256
2222        would behave wrongly. This has been present in PCRE since release 4.0.
2223    
2224    10. A number of changes to the pcregrep command:
2225    
2226        (a) Refactored how -x works; insert ^(...)$ instead of setting
2227            PCRE_ANCHORED and checking the length, in preparation for adding
2228            something similar for -w.
2229    
2230        (b) Added the -w (match as a word) option.
2231    
2232        (c) Refactored the way lines are read and buffered so as to have more
2233            than one at a time available.
2234    
2235        (d) Implemented a pcregrep test script.
2236    
2237        (e) Added the -M (multiline match) option. This allows patterns to match
2238            over several lines of the subject. The buffering ensures that at least
2239            8K, or the rest of the document (whichever is the shorter) is available
2240            for matching (and similarly the previous 8K for lookbehind assertions).
2241    
2242        (f) Changed the --help output so that it now says
2243    
2244              -w, --word-regex(p)
2245    
2246            instead of two lines, one with "regex" and the other with "regexp"
2247            because that confused at least one person since the short forms are the
2248            same. (This required a bit of code, as the output is generated
2249            automatically from a table. It wasn't just a text change.)
2250    
2251        (g) -- can be used to terminate pcregrep options if the next thing isn't an
2252            option but starts with a hyphen. Could be a pattern or a path name
2253            starting with a hyphen, for instance.
2254    
2255        (h) "-" can be given as a file name to represent stdin.
2256    
2257        (i) When file names are being printed, "(standard input)" is used for
2258            the standard input, for compatibility with GNU grep. Previously
2259            "<stdin>" was used.
2260    
2261        (j) The option --label=xxx can be used to supply a name to be used for
2262            stdin when file names are being printed. There is no short form.
2263    
2264        (k) Re-factored the options decoding logic because we are going to add
2265            two more options that take data. Such options can now be given in four
2266            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
2267    
2268        (l) Added the -A, -B, and -C options for requesting that lines of context
2269            around matches be printed.
2270    
2271        (m) Added the -L option to print the names of files that do not contain
2272            any matching lines, that is, the complement of -l.
2273    
2274        (n) The return code is 2 if any file cannot be opened, but pcregrep does
2275            continue to scan other files.
2276    
2277        (o) The -s option was incorrectly implemented. For compatibility with other
2278            greps, it now suppresses the error message for a non-existent or non-
2279            accessible file (but not the return code). There is a new option called
2280            -q that suppresses the output of matching lines, which was what -s was
2281            previously doing.
2282    
2283        (p) Added --include and --exclude options to specify files for inclusion
2284            and exclusion when recursing.
2285    
2286    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
2287        Hopefully, it now does.
2288    
2289    12. Missing cast in pcre_study().
2290    
2291    13. Added an "uninstall" target to the makefile.
2292    
2293    14. Replaced "extern" in the function prototypes in Makefile.in with
2294        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
2295        world, but is set differently for Windows.
2296    
2297    15. Added a second compiling function called pcre_compile2(). The only
2298        difference is that it has an extra argument, which is a pointer to an
2299        integer error code. When there is a compile-time failure, this is set
2300        non-zero, in addition to the error test pointer being set to point to an
2301        error message. The new argument may be NULL if no error number is required
2302        (but then you may as well call pcre_compile(), which is now just a
2303        wrapper). This facility is provided because some applications need a
2304        numeric error indication, but it has also enabled me to tidy up the way
2305        compile-time errors are handled in the POSIX wrapper.
2306    
2307    16. Added VPATH=.libs to the makefile; this should help when building with one
2308        prefix path and installing with another. (Or so I'm told by someone who
2309        knows more about this stuff than I do.)
2310    
2311    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
2312        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
2313        match everything, including newlines. This is not POSIX-compatible, but
2314        somebody wanted the feature. From pcretest it can be activated by using
2315        both the P and the s flags.
2316    
2317    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
2318    
2319    19. libpcre.pc was being incorrectly installed as executable.
2320    
2321    20. A couple of places in pcretest check for end-of-line by looking for '\n';
2322        it now also looks for '\r' so that it will work unmodified on Windows.
2323    
2324    21. Added Google's contributed C++ wrapper to the distribution.
2325    
2326    22. Added some untidy missing memory free() calls in pcretest, to keep
2327        Electric Fence happy when testing.
2328    
2329    
2330    
2331    Version 5.0 13-Sep-04
2332    ---------------------
2333    
2334     1. Internal change: literal characters are no longer packed up into items
2335        containing multiple characters in a single byte-string. Each character
2336        is now matched using a separate opcode. However, there may be more than one
2337        byte in the character in UTF-8 mode.
2338    
2339     2. The pcre_callout_block structure has two new fields: pattern_position and
2340        next_item_length. These contain the offset in the pattern to the next match
2341        item, and its length, respectively.
2342    
2343     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
2344        insertion of callouts before each pattern item. Added the /C option to
2345        pcretest to make use of this.
2346    
2347     4. On the advice of a Windows user, the lines
2348    
2349          #if defined(_WIN32) || defined(WIN32)
2350          _setmode( _fileno( stdout ), 0x8000 );
2351          #endif  /* defined(_WIN32) || defined(WIN32) */
2352    
2353        have been added to the source of pcretest. This apparently does useful
2354        magic in relation to line terminators.
2355    
2356     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
2357        for the benefit of those environments where the "b" makes a difference.
2358    
2359     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
2360        to know about it. I have put a hack into configure.in that adds in code
2361        to set GCC=yes if CC=icc. This seems to end up at a point in the
2362        generated configure script that is early enough to affect the setting of
2363        compiler options, which is what is needed, but I have no means of testing
2364        whether it really works. (The user who reported this had patched the
2365        generated configure script, which of course I cannot do.)
2366    
2367        LATER: After change 22 below (new libtool files), the configure script
2368        seems to know about icc (and also ecc). Therefore, I have commented out
2369        this hack in configure.in.
2370    
2371     7. Added support for pkg-config (2 patches were sent in).
2372    
2373     8. Negated POSIX character classes that used a combination of internal tables
2374        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2375        [[:^ascii]]. Typically, they would match almost any characters. The other
2376        POSIX classes were not broken in this way.
2377    
2378     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2379        to find the match, as PCRE was deluded into thinking that the match had to
2380        start at the start point or following a newline. The same bug applied to
2381        patterns with negative forward assertions or any backward assertions
2382        preceding ".*" at the start, unless the pattern required a fixed first
2383        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2384    
2385    10. In UTF-8 mode, when moving forwards in the subject after a failed match
2386        starting at the last subject character, bytes beyond the end of the subject
2387        string were read.
2388    
2389    11. Renamed the variable "class" as "classbits" to make life easier for C++
2390        users. (Previously there was a macro definition, but it apparently wasn't
2391        enough.)
2392    
2393    12. Added the new field "tables" to the extra data so that tables can be passed
2394        in at exec time, or the internal tables can be re-selected. This allows
2395        a compiled regex to be saved and re-used at a later time by a different
2396        program that might have everything at different addresses.
2397    
2398    13. Modified the pcre-config script so that, when run on Solaris, it shows a
2399        -R library as well as a -L library.
2400    
2401    14. The debugging options of pcretest (-d on the command line or D on a
2402        pattern) showed incorrect output for anything following an extended class
2403        that contained multibyte characters and which was followed by a quantifier.
2404    
2405    15. Added optional support for general category Unicode character properties
2406        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2407        support. It adds about 90K to the size of the library. The meanings of the
2408        inbuilt class escapes such as \d and \s have NOT been changed.
2409    
2410    16. Updated pcredemo.c to include calls to free() to release the memory for the
2411        compiled pattern.
2412    
2413    17. The generated file chartables.c was being created in the source directory
2414        instead of in the building directory. This caused the build to fail if the
2415        source directory was different from the building directory, and was
2416        read-only.
2417    
2418    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2419        file. No doubt somebody will tell me if they don't make sense... Also added
2420        Dan Mooney's comments about building on OpenVMS.
2421    
2422    19. Added support for partial matching via the PCRE_PARTIAL option for
2423        pcre_exec() and the \P data escape in pcretest.
2424    
2425    20. Extended pcretest with 3 new pattern features:
2426    
2427        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
2428              write the compiled pattern to the file whose name is "rest-of-line".
2429              This is a straight binary dump of the data, with the saved pointer to
2430              the character tables forced to be NULL. The study data, if any, is
2431              written too. After writing, pcretest reads a new pattern.
2432    
2433        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2434              compiled pattern from the given file. There must not be any
2435              occurrences of "<" in the file name (pretty unlikely); if there are,
2436              pcretest will instead treat the initial "<" as a pattern delimiter.
2437              After reading in the pattern, pcretest goes on to read data lines as
2438              usual.
2439    
2440        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2441              and 16-bit fields in a compiled pattern, to simulate a pattern that
2442              was compiled on a host of opposite endianness.
2443    
2444    21. The pcre-exec() function can now cope with patterns that were compiled on
2445        hosts of opposite endianness, with this restriction:
2446    
2447          As for any compiled expression that is saved and used later, the tables
2448          pointer field cannot be preserved; the extra_data field in the arguments
2449          to pcre_exec() should be used to pass in a tables address if a value
2450          other than the default internal tables were used at compile time.
2451    
2452    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2453        now diagnosed as an error. Previously, most of the time, a negative number
2454        would have been treated as zero, but if in addition "ovector" was passed as
2455        NULL, a crash could occur.
2456    
2457    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2458        new versions from the libtool 1.5 distribution (the last one is a copy of
2459        a file called libtool.m4). This seems to have fixed the need to patch
2460        "configure" to support Darwin 1.3 (which I used to do). However, I still
2461        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2462        workstation).
2463    
2464    24. Changed the PCRE licence to be the more standard "BSD" licence.
2465    
2466    
2467    Version 4.5 01-Dec-03
2468    ---------------------
2469    
2470     1. There has been some re-arrangement of the code for the match() function so
2471        that it can be compiled in a version that does not call itself recursively.
2472        Instead, it keeps those local variables that need separate instances for
2473        each "recursion" in a frame on the heap, and gets/frees frames whenever it
2474        needs to "recurse". Keeping track of where control must go is done by means
2475        of setjmp/longjmp. The whole thing is implemented by a set of macros that
2476        hide most of the details from the main code, and operates only if
2477        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2478        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2479        operating.
2480    
2481        To make it easier for callers to provide specially tailored get/free
2482        functions for this usage, two new functions, pcre_stack_malloc, and
2483        pcre_stack_free, are used. They are always called in strict stacking order,
2484        and the size of block requested is always the same.
2485    
2486        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2487        PCRE has been compiled to use the stack or the heap for recursion. The
2488        -C option of pcretest uses this to show which version is compiled.
2489    
2490        A new data escape \S, is added to pcretest; it causes the amounts of store
2491        obtained and freed by both kinds of malloc/free at match time to be added
2492        to the output.
2493    
2494     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2495        what's available on my current Linux desktop machine.
2496    
2497     3. When matching a UTF-8 string, the test for a valid string at the start has
2498        been extended. If start_offset is not zero, PCRE now checks that it points
2499        to a byte that is the start of a UTF-8 character. If not, it returns
2500        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2501        this is necessary because there may be backward assertions in the pattern.
2502        When matching the same subject several times, it may save resources to use
2503        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2504    
2505     4. The code for checking the validity of UTF-8 strings has been tightened so
2506        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2507        containing "overlong sequences".
2508    
2509     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2510        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2511        should have been "&", but it just so happened that all the cases this let
2512        through by mistake were picked up later in the function.
2513    
2514     6. I had used a variable called "isblank" - this is a C99 function, causing
2515        some compilers to warn. To avoid this, I renamed it (as "blankclass").
2516    
2517     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2518        prompting; (b) run "./pcretest /dev/null" at the start of the test script
2519        so the version is shown; (c) stop "make test" echoing "./RunTest".
2520    
2521     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2522    
2523     9. The prototype for memmove() for systems that don't have it was using
2524        size_t, but the inclusion of the header that defines size_t was later. I've
2525        moved the #includes for the C headers earlier to avoid this.
2526    
2527    10. Added some adjustments to the code to make it easier to compiler on certain
2528        special systems:
2529    
2530          (a) Some "const" qualifiers were missing.
2531          (b) Added the macro EXPORT before all exported functions; by default this
2532              is defined to be empty.
2533          (c) Changed the dftables auxiliary program (that builds chartables.c) so
2534              that it reads its output file name as an argument instead of writing
2535              to the standard output and assuming this can be redirected.
2536    
2537    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2538        class containing characters with values greater than 255, PCRE compilation
2539        went into a loop.
2540    
2541    12. A recursive reference to a subpattern that was within another subpattern
2542        that had a minimum quantifier of zero caused PCRE to crash. For example,
2543        (x(y(?2))z)? provoked this bug with a subject that got as far as the
2544        recursion. If the recursively-called subpattern itself had a zero repeat,
2545        that was OK.
2546    
2547    13. In pcretest, the buffer for reading a data line was set at 30K, but the
2548        buffer into which it was copied (for escape processing) was still set at
2549        1024, so long lines caused crashes.
2550    
2551    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2552        "internal error: code overflow...". This applied to any character class
2553        that was followed by a possessive quantifier.
2554    
2555    15. Modified the Makefile to add libpcre.la as a prerequisite for
2556        libpcreposix.la because I was told this is needed for a parallel build to
2557        work.
2558    
2559    16. If a pattern that contained .* following optional items at the start was
2560        studied, the wrong optimizing data was generated, leading to matching
2561        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2562        matching string must start with a or b or c. The correct conclusion for
2563        this pattern is that a match can start with any character.
2564    
2565    
2566    Version 4.4 13-Aug-03
2567    ---------------------
2568    
2569     1. In UTF-8 mode, a character class containing characters with values between
2570        127 and 255 was not handled correctly if the compiled pattern was studied.
2571        In fixing this, I have also improved the studying algorithm for such
2572        classes (slightly).
2573    
2574     2. Three internal functions had redundant arguments passed to them. Removal
2575        might give a very teeny performance improvement.
2576    
2577     3. Documentation bug: the value of the capture_top field in a callout is *one
2578        more than* the number of the hightest numbered captured substring.
2579    
2580     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2581        in incorrectly linking with a previously installed version. They now link
2582        explicitly with libpcre.la.
2583    
2584     5. configure.in no longer needs to recognize Cygwin specially.
2585    
2586     6. A problem in pcre.in for Windows platforms is fixed.
2587    
2588     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2589        pcretest, it used to include the size of the study block as part of its
2590        output. Unfortunately, the structure contains a field that has a different
2591        size on different hardware architectures. This meant that the tests that
2592        showed this size failed. As the block is currently always of a fixed size,
2593        this information isn't actually particularly useful in pcretest output, so
2594        I have just removed it.
2595    
2596     8. Three pre-processor statements accidentally did not start in column 1.
2597        Sadly, there are *still* compilers around that complain, even though
2598        standard C has not required this for well over a decade. Sigh.
2599    
2600     9. In pcretest, the code for checking callouts passed small integers in the
2601        callout_data field, which is a void * field. However, some picky compilers
2602        complained about the casts involved for this on 64-bit systems. Now
2603        pcretest passes the address of the small integer instead, which should get
2604        rid of the warnings.
2605    
2606    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2607        both compile and run time, and gives an error if an invalid UTF-8 sequence
2608        is found. There is a option for disabling this check in cases where the
2609        string is known to be correct and/or the maximum performance is wanted.
2610    
2611    11. In response to a bug report, I changed one line in Makefile.in from
2612    
2613            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2614        to
2615            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2616    
2617        to look similar to other lines, but I have no way of telling whether this
2618        is the right thing to do, as I do not use Windows. No doubt I'll get told
2619        if it's wrong...
2620    
2621    
2622    Version 4.3 21-May-03
2623    ---------------------
2624    
2625    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2626       Makefile.
2627    
2628    2. Some refactoring to improve the quality of the code:
2629    
2630       (i)   The utf8_table... variables are now declared "const".
2631    
2632       (ii)  The code for \cx, which used the "case flipping" table to upper case
2633             lower case letters, now just substracts 32. This is ASCII-specific,
2634             but the whole concept of \cx is ASCII-specific, so it seems
2635             reasonable.
2636    
2637       (iii) PCRE was using its character types table to recognize decimal and
2638             hexadecimal digits in the pattern. This is silly, because it handles
2639             only 0-9, a-f, and A-F, but the character types table is locale-
2640             specific, which means strange things might happen. A private
2641             table is now used for this - though it costs 256 bytes, a table is
2642             much faster than multiple explicit tests. Of course, the standard
2643             character types table is still used for matching digits in subject
2644             strings against \d.
2645    
2646       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2647             ending in _t are). So I've renamed it as ESC_tee.
2648    
2649    3. The first argument for regexec() in the POSIX wrapper should have been
2650       defined as "const".
2651    
2652    4. Changed pcretest to use malloc() for its buffers so that they can be
2653       Electric Fenced for debugging.
2654    
2655    5. There were several places in the code where, in UTF-8 mode, PCRE would try
2656       to read one or more bytes before the start of the subject string. Often this
2657       had no effect on PCRE's behaviour, but in some circumstances it could
2658       provoke a segmentation fault.
2659    
2660    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2661       to try to read one or more bytes before the start of the subject string.
2662    
2663    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
2664       UTF-8 support could misbehave in various ways if the subject string
2665       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
2666       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
2667       back over UTF-8 characters.)
2668    
2669    
2670    Version 4.2 14-Apr-03
2671    ---------------------
2672    
2673    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
2674    
2675    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
2676         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
2677         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
2678         [WIN_PREFIX]: new variable, "cyg" for Cygwin
2679         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
2680           and BUILD_EXEEXT
2681         Note: automatic setting of the BUILD variables is not yet working
2682         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
2683           compile-time but not at link-time
2684         [LINK]: use for linking executables only
2685         make different versions for Windows and non-Windows
2686         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
2687           libraries
2688         [LINK_FOR_BUILD]: new variable
2689         [OBJEXT]: use throughout
2690         [EXEEXT]: use throughout
2691         <winshared>: new target
2692         <wininstall>: new target
2693         <dftables.o>: use native compiler
2694         <dftables>: use native linker
2695         <install>: handle Windows platform correctly
2696         <clean>: ditto
2697         <check>: ditto
2698         copy DLL to top builddir before testing
2699    
2700       As part of these changes, -no-undefined was removed again. This was reported
2701       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
2702       in any case.
2703    
2704    3. Some tidies to get rid of compiler warnings:
2705    
2706       . In the match_data structure, match_limit was an unsigned long int, whereas
2707         match_call_count was an int. I've made them both unsigned long ints.
2708    
2709       . In pcretest the fact that a const uschar * doesn't automatically cast to
2710         a void * provoked a warning.
2711    
2712       . Turning on some more compiler warnings threw up some "shadow" variables
2713         and a few more missing casts.
2714    
2715    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
2716       option, a class that contained a single character with a value between 128
2717       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
2718    
2719    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
2720       option, a class that contained several characters, but with at least one
2721       whose value was between 128 and 255 caused PCRE to crash.
2722    
2723    
2724    Version 4.1 12-Mar-03
2725    ---------------------
2726    
2727    1. Compiling with gcc -pedantic found a couple of places where casts were
2728    needed, and a string in dftables.c that was longer than standard compilers are
2729    required to support.
2730    
2731    2. Compiling with Sun's compiler found a few more places where the code could
2732    be tidied up in order to avoid warnings.
2733    
2734    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
2735    first of these names is deprecated in the latest Autoconf in favour of the name
2736    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
2737    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2738    analogy I have changed it to CFLAGS_FOR_BUILD.
2739    
2740    4. Added -no-undefined to the linking command in the Makefile, because this is
2741    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2742    linking step for the pcreposix library.
2743    
2744    5. PCRE was failing to diagnose the case of two named groups with the same
2745    name.
2746    
2747    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2748    literal character that is needed in the subject for a match, and scans along to
2749    ensure that it is present before embarking on the full matching process. This
2750    saves time in cases of nested unlimited repeats that are never going to match.
2751    Problem: the scan can take a lot of time if the subject is very long (e.g.
2752    megabytes), thus penalizing straightforward matches. It is now done only if the
2753    amount of subject to be scanned is less than 1000 bytes.
2754    
2755    7. A lesser problem with the same optimization is that it was recording the
2756    first character of an anchored pattern as "needed", thus provoking a search
2757    right along the subject, even when the first match of the pattern was going to
2758    fail. The "needed" character is now not set for anchored patterns, unless it
2759    follows something in the pattern that is of non-fixed length. Thus, it still
2760    fulfils its original purpose of finding quick non-matches in cases of nested
2761    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
2762    
2763    
2764    Version 4.0 17-Feb-03
2765    ---------------------
2766    
2767    1. If a comment in an extended regex that started immediately after a meta-item
2768    extended to the end of string, PCRE compiled incorrect data. This could lead to
2769    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
2770    
2771    2. Moved to autoconf 2.53 and libtool 1.4.2.
2772    
2773    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
2774    the special perltest8 script is no longer needed - all the tests can be run
2775    from a single perltest script.
2776    
2777    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
2778    by \s. It has now been removed in PCRE. This means it isn't recognized as
2779    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
2780    class [:space:] *does* include VT, thereby creating a mess.
2781    
2782    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
2783    space and tab.
2784    
2785    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
2786    its new features into the main test script, reducing the number of scripts.
2787    
2788    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
2789    were backward compatible, and made the (?i) apply to the whole pattern, as if
2790    /i were given. Now it behaves more logically, and applies the option setting
2791    only to what follows. PCRE has been changed to follow suit. However, if it
2792    finds options settings right at the start of the pattern, it extracts them into
2793    the global options, as before. Thus, they show up in the info data.
2794    
2795    8. Added support for the \Q...\E escape sequence. Characters in between are
2796    treated as literals. This is slightly different from Perl in that $ and @ are
2797    also handled as literals inside the quotes. In Perl, they will cause variable
2798    interpolation. Note the following examples:
2799    
2800        Pattern            PCRE matches      Perl matches
2801    
2802        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
2803        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
2804        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
2805    
2806    For compatibility with Perl, \Q...\E sequences are recognized inside character
2807    classes as well as outside them.
2808    
2809    9. Re-organized 3 code statements in pcretest to avoid "overflow in
2810    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2811    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2812    signed/unsigned warnings.
2813    
2814    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2815    option for pcretest, so I've replaced it by a simple function that does just
2816    that job.
2817    
2818    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2819    "pcregrep -".
2820    
2821    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2822    Java package. This provides some syntactic sugar for simple cases of what my
2823    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2824    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2825    item, you can use this simplified notation. Note that only makes sense with
2826    greedy quantifiers. Consequently, the use of the possessive quantifier forces
2827    greediness, whatever the setting of the PCRE_UNGREEDY option.
2828    
2829    13. A change of greediness default within a pattern was not taking effect at
2830    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2831    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2832    was abstracted outside.
2833    
2834    14. PCRE now supports the \G assertion. It is true when the current matching
2835    position is at the start point of the match. This differs from \A when the
2836    starting offset is non-zero. Used with the /g option of pcretest (or similar
2837    code), it works in the same way as it does for Perl's /g option. If all
2838    alternatives of a regex begin with \G, the expression is anchored to the start
2839    match position, and the "anchored" flag is set in the compiled expression.
2840    
2841    15. Some bugs concerning the handling of certain option changes within patterns
2842    have been fixed. These applied to options other than (?ims). For example,
2843    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2844    been the other way round. Some of this was related to change 7 above.
2845    
2846    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2847    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2848    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2849    POSIX classes only within a class (e.g. /[[:alpha:]]/).
2850    
2851    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2852    mode. Unlike ".", it always matches newline, whatever the setting of
2853    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2854    assertions. Perl allows it, but it doesn't (in general) work because it can't
2855    calculate the length of the lookbehind. At least, that's the case for Perl
2856    5.8.0 - I've been told they are going to document that it doesn't work in
2857    future.
2858    
2859    18. Added an error diagnosis for escapes that PCRE does not support: these are
2860    \L, \l, \N, \P, \p, \U, \u, and \X.
2861    
2862    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2863    reading past the end of the pattern in cases such as /[abcd/.
2864    
2865    20. PCRE was getting more memory than necessary for patterns with classes that
2866    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2867    
2868    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2869    compiling PCRE for use with Virtual Pascal.
2870    
2871    22. Small fix to the Makefile to make it work properly if the build is done
2872    outside the source tree.
2873    
2874    23. Added a new extension: a condition to go with recursion. If a conditional
2875    subpattern starts with (?(R) the "true" branch is used if recursion has
2876    happened, whereas the "false" branch is used only at the top level.
2877    
2878    24. When there was a very long string of literal characters (over 255 bytes
2879    without UTF support, over 250 bytes with UTF support), the computation of how
2880    much memory was required could be incorrect, leading to segfaults or other
2881    strange effects.
2882    
2883    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2884    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2885    there was a subsequent back reference to those brackets. This meant that, for
2886    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2887    possible to check for precisely this case. All we can do is abandon the
2888    optimization if .* occurs inside capturing brackets when there are any back
2889    references whatsoever. (See below for a better fix that came later.)
2890    
2891    26. The handling of the optimization for finding the first character of a
2892    non-anchored pattern, and for finding a character that is required later in the
2893    match were failing in some cases. This didn't break the matching; it just
2894    failed to optimize when it could. The way this is done has been re-implemented.
2895    
2896    27. Fixed typo in error message for invalid (?R item (it said "(?p").
2897    
2898    28. Added a new feature that provides some of the functionality that Perl
2899    provides with (?{...}). The facility is termed a "callout". The way it is done
2900    in PCRE is for the caller to provide an optional function, by setting
2901    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2902    global variable. By default it is unset, which disables all calling out. To get
2903    the function called, the regex must include (?C) at appropriate points. This
2904    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2905    This provides a means of identifying different callout points. When PCRE
2906    reaches such a point in the regex, if pcre_callout has been set, the external
2907    function is called. It is provided with data in a structure called
2908    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2909    matching continues; if it returns a non-zero value, the match at the current
2910    point fails. However, backtracking will occur if possible. [This was changed
2911    later and other features added - see item 49 below.]
2912    
2913    29. pcretest is upgraded to test the callout functionality. It provides a
2914    callout function that displays information. By default, it shows the start of
2915    the match and the current position in the text. There are some new data escapes
2916    to vary what happens:
2917    
2918        \C+         in addition, show current contents of captured substrings
2919        \C-         do not supply a callout function
2920        \C!n        return 1 when callout number n is reached
2921        \C!n!m      return 1 when callout number n is reached for the mth time
2922    
2923    30. If pcregrep was called with the -l option and just a single file name, it
2924    output "<stdin>" if a match was found, instead of the file name.
2925    
2926    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2927    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2928    pcre_exec(). This saves a malloc/free per call. The default value of
2929    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2930    when configuring.
2931    
2932    32. The default maximum size of a compiled pattern is 64K. There have been a
2933    few cases of people hitting this limit. The code now uses macros to handle the
2934    storing of links as offsets within the compiled pattern. It defaults to 2-byte
2935    links, but this can be changed to 3 or 4 bytes by --with-link-size when
2936    configuring. Tests 2 and 5 work only with 2-byte links because they output
2937    debugging information about compiled patterns.
2938    
2939    33. Internal code re-arrangements:
2940    
2941    (a) Moved the debugging function for printing out a compiled regex into
2942        its own source file (printint.c) and used #include to pull it into
2943        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2944        separate copies.
2945    
2946    (b) Defined the list of op-code names for debugging as a macro in
2947        internal.h so that it is next to the definition of the opcodes.
2948    
2949    (c) Defined a table of op-code lengths for simpler skipping along compiled
2950        code. This is again a macro in internal.h so that it is next to the
2951        definition of the opcodes.
2952    
2953    34. Added support for recursive calls to individual subpatterns, along the
2954    lines of Robin Houston's patch (but implemented somewhat differently).
2955    
2956    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2957    allow it to read and process whole directories in Win32. This code was
2958    contributed by Lionel Fourquaux; it has not been tested by me.
2959    
2960    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2961    used to name a group. Names consist of alphanumerics and underscores, and must
2962    be unique. Back references use the syntax (?P=name) and recursive calls use
2963    (?P>name) which is a PCRE extension to the Python extension. Groups still have
2964    numbers. The function pcre_fullinfo() can be used after compilation to extract
2965    a name/number map. There are three relevant calls:
2966    
2967      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
2968      PCRE_INFO_NAMECOUNT            yields the number of entries
2969      PCRE_INFO_NAMETABLE            yields a pointer to the map.
2970    
2971    The map is a vector of fixed-size entries. The size of each entry depends on
2972    the length of the longest name used. The first two bytes of each entry are the
2973    group number, most significant byte first. There follows the corresponding
2974    name, zero terminated. The names are in alphabetical order.
2975    
2976    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2977    case instead of 255. Making it the same both with and without UTF-8 support
2978    means that the same test output works with both.
2979    
2980    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2981    calling malloc() with a zero argument.
2982    
2983    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2984    optimization. I've improved things by keeping a bitmap of backreferences with
2985    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2986    fact referenced, the optimization can be applied. It is unlikely that a
2987    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2988    the match to follow \n) will appear inside brackets with a number greater than
2989    31, but if it does, any back reference > 31 suppresses the optimization.
2990    
2991    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2992    of disabling numbered capturing parentheses. Any opening parenthesis that is
2993    not followed by ? behaves as if it were followed by ?: but named parentheses
2994    can still be used for capturing (and they will acquire numbers in the usual
2995    way).
2996    
2997    41. Redesigned the return codes from the match() function into yes/no/error so
2998    that errors can be passed back from deep inside the nested calls. A malloc
2999    failure while inside a recursive subpattern call now causes the
3000    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
3001    
3002    42. It is now possible to set a limit on the number of times the match()
3003    function is called in a call to pcre_exec(). This facility makes it possible to
3004    limit the amount of recursion and backtracking, though not in a directly
3005    obvious way, because the match() function is used in a number of different
3006    circumstances. The count starts from zero for each position in the subject
3007    string (for non-anchored patterns). The default limit is, for compatibility, a
3008    large number, namely 10 000 000. You can change this in two ways:
3009    
3010    (a) When configuring PCRE before making, you can use --with-match-limit=n
3011        to set a default value for the compiled library.
3012    
3013    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
3014        a different value is set. See 45 below.
3015    
3016    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
3017    
3018    43. Added a new function pcre_config(int, void *) to enable run-time extraction
3019    of things that can be changed at compile time. The first argument specifies
3020    what is wanted and the second points to where the information is to be placed.
3021    The current list of available information is:
3022    
3023      PCRE_CONFIG_UTF8
3024    
3025    The output is an integer that is set to one if UTF-8 support is available;
3026    otherwise it is set to zero.
3027    
3028      PCRE_CONFIG_NEWLINE
3029    
3030    The output is an integer that it set to the value of the code that is used for
3031    newline. It is either LF (10) or CR (13).
3032    
3033      PCRE_CONFIG_LINK_SIZE
3034    
3035    The output is an integer that contains the number of bytes used for internal
3036    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
3037    
3038      PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
3039    
3040    The output is an integer that contains the threshold above which the POSIX
3041    interface uses malloc() for output vectors. See item 31 above.
3042    
3043      PCRE_CONFIG_MATCH_LIMIT
3044    
3045    The output is an unsigned integer that contains the default limit of the number
3046    of match() calls in a pcre_exec() execution. See 42 above.
3047    
3048    44. pcretest has been upgraded by the addition of the -C option. This causes it
3049    to extract all the available output from the new pcre_config() function, and to
3050    output it. The program then exits immediately.
3051    
3052    45. A need has arisen to pass over additional data with calls to pcre_exec() in
3053    order to support additional features. One way would have been to define
3054    pcre_exec2() (for example) with extra arguments, but this would not have been
3055    extensible, and would also have required all calls to the original function to
3056    be mapped to the new one. Instead, I have chosen to extend the mechanism that
3057    is used for passing in "extra" data from pcre_study().
3058    
3059    The pcre_extra structure is now exposed and defined in pcre.h. It currently
3060    contains the following fields:
3061    
3062      flags         a bitmap indicating which of the following fields are set
3063      study_data    opaque data from pcre_study()
3064      match_limit   a way of specifying a limit on match() calls for a specific
3065                      call to pcre_exec()
3066      callout_data  data for callouts (see 49 below)
3067    
3068    The flag bits are also defined in pcre.h, and are
3069    
3070      PCRE_EXTRA_STUDY_DATA
3071      PCRE_EXTRA_MATCH_LIMIT
3072      PCRE_EXTRA_CALLOUT_DATA
3073    
3074    The pcre_study() function now returns one of these new pcre_extra blocks, with
3075    the actual study data pointed to by the study_data field, and the
3076    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
3077    before. That is, this change is entirely upwards-compatible and requires no
3078    change to existing code.
3079    
3080    If you want to pass in additional data to pcre_exec(), you can either place it
3081    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
3082    block.
3083    
3084    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
3085    data string contains the escape sequence \M, pcretest calls pcre_exec() several
3086    times with different match limits, until it finds the minimum value needed for
3087    pcre_exec() to complete. The value is then output. This can be instructive; for
3088    most simple matches the number is quite small, but for pathological cases it
3089    gets very large very quickly.
3090    
3091    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
3092    returns the size of the data block pointed to by the study_data field in a
3093    pcre_extra block, that is, the value that was passed as the argument to
3094    pcre_malloc() when PCRE was getting memory in which to place the information
3095    created by pcre_study(). The fourth argument should point to a size_t variable.
3096    pcretest has been extended so that this information is shown after a successful
3097    pcre_study() call when information about the compiled regex is being displayed.
3098    
3099    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
3100    because what follows is always an absolute path. (Later: it turns out that this
3101    is more than cosmetic for MinGW, because it doesn't like empty path
3102    components.)
3103    
3104    49. Some changes have been made to the callout feature (see 28 above):
3105    
3106    (i)  A callout function now has three choices for what it returns:
3107    
3108           0  =>  success, carry on matching
3109         > 0  =>  failure at this point, but backtrack if possible
3110         < 0  =>  serious error, return this value from pcre_exec()
3111    
3112         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
3113         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
3114         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
3115         use by callout functions. It will never be used by PCRE itself.
3116    
3117    (ii) The pcre_extra structure (see 45 above) has a void * field called
3118         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
3119         pcre_callout_block structure has a field of the same name. The contents of
3120         the field passed in the pcre_extra structure are passed to the callout
3121         function in the corresponding field in the callout block. This makes it
3122         easier to use the same callout-containing regex from multiple threads. For
3123         testing, the pcretest program has a new data escape
3124    
3125           \C*n        pass the number n (may be negative) as callout_data
3126    
3127         If the callout function in pcretest receives a non-zero value as
3128         callout_data, it returns that value.
3129    
3130    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
3131    there were some redundant $(CFLAGS) in commands that are now specified as
3132    $(LINK), which already includes $(CFLAGS).
3133    
3134    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
3135    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
3136    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
3137    one-byte characters throughout. Note that case-insensitive matching applies
3138    only to characters whose values are less than 256. PCRE doesn't support the
3139    notion of cases for higher-valued characters.
3140    
3141    (i)   A character class whose characters are all within 0-255 is handled as
3142          a bit map, and the map is inverted for negative classes. Previously, a
3143          character > 255 always failed to match such a class; however it should
3144          match if the class was a negative one (e.g. [^ab]). This has been fixed.
3145    
3146    (ii)  A negated character class with a single character < 255 is coded as
3147          "not this character" (OP_NOT). This wasn't working properly when the test
3148          character was multibyte, either singly or repeated.
3149    
3150    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
3151          mode, for example: \x{100}{2,3}.
3152    
3153    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
3154          singly or repeated) now correctly test multibyte characters. However,
3155          PCRE doesn't recognize any characters with values greater than 255 as
3156          digits, spaces, or word characters. Such characters always match \D, \S,
3157          and \W, and never match \d, \s, or \w.
3158    
3159    (v)   Classes may now contain characters and character ranges with values
3160          greater than 255. For example: [ab\x{100}-\x{400}].
3161    
3162    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
3163          PCRE in UTF-8 mode.
3164    
3165    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
3166    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
3167    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
3168    value.)
3169    
3170    53. The single man page has become too large. I have therefore split it up into
3171    a number of separate man pages. These also give rise to individual HTML pages;
3172    these are now put in a separate directory, and there is an index.html page that
3173    lists them all. Some hyperlinking between the pages has been installed.
3174    
3175    54. Added convenience functions for handling named capturing parentheses.
3176    
3177    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
3178    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
3179    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
3180    are faulted.
3181    
3182    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
3183    calling configure. These values are used when compiling the dftables.c program
3184    which is run to generate the source of the default character tables. They
3185    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
3186    you will need to set these values.
3187    
3188    57. Updated the building process for Windows DLL, as provided by Fred Cox.
3189    
3190    
3191    Version 3.9 02-Jan-02
3192    ---------------------
3193    
3194    1. A bit of extraneous text had somehow crept into the pcregrep documentation.
3195    
3196    2. If --disable-static was given, the building process failed when trying to
3197    build pcretest and pcregrep. (For some reason it was using libtool to compile
3198    them, which is not right, as they aren't part of the library.)
3199    
3200    
3201    Version 3.8 18-Dec-01
3202    ---------------------
3203    
3204    1. The experimental UTF-8 code was completely screwed up. It was packing the
3205    bytes in the wrong order. How dumb can you get?
3206    
3207    
3208    Version 3.7 29-Oct-01
3209    ---------------------
3210    
3211    1. In updating pcretest to check change 1 of version 3.6, I screwed up.
3212    This caused pcretest, when used on the test data, to segfault. Unfortunately,
3213    this didn't happen under Solaris 8, where I normally test things.
3214    
3215    2. The Makefile had to be changed to make it work on BSD systems, where 'make'
3216    doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
3217    isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
3218    this fix an hour or so after the initial 3.7 release.)
3219    
3220    
3221  Version 3.6 23-Oct-01  Version 3.6 23-Oct-01
3222  ---------------------  ---------------------
3223    

Legend:
Removed from v.55  
changed lines
  Added in v.537

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12