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

Diff of /code/trunk/ChangeLog

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

revision 81 by nigel, Sat Feb 24 21:40:59 2007 UTC revision 510 by ph10, Sat Mar 27 17:45:29 2010 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    Version 8.03 26-Mar-2010
5    ------------------------
6    
7    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
8        THEN.
9    
10    
11    Version 8.02 19-Mar-2010
12    ------------------------
13    
14    1.  The Unicode data tables have been updated to Unicode 5.2.0.
15    
16    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
17        configured.
18    
19    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
20        original author of that file, following a query about its status.
21    
22    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
23        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
24    
25    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
26        quantifier applied to a forward-referencing subroutine call, could compile
27        incorrect code or give the error "internal error: previously-checked
28        referenced subpattern not found".
29    
30    6.  Both MS Visual Studio and Symbian OS have problems with initializing
31        variables to point to external functions. For these systems, therefore,
32        pcre_malloc etc. are now initialized to local functions that call the
33        relevant global functions.
34    
35    7.  There were two entries missing in the vectors called coptable and poptable
36        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
37        I've fixed the data, and added a kludgy way of testing at compile time that
38        the lengths are correct (equal to the number of opcodes).
39    
40    8.  Following on from 7, I added a similar kludge to check the length of the
41        eint vector in pcreposix.c.
42    
43    9.  Error texts for pcre_compile() are held as one long string to avoid too
44        much relocation at load time. To find a text, the string is searched,
45        counting zeros. There was no check for running off the end of the string,
46        which could happen if a new error number was added without updating the
47        string.
48    
49    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
50    
51    11. \K was not working if it appeared in an atomic group or in a group that
52        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
53        \K is "not well defined" if used in an assertion. PCRE now accepts it if
54        the assertion is positive, but not if it is negative.
55    
56    12. Change 11 fortuitously reduced the size of the stack frame used in the
57        "match()" function of pcre_exec.c by one pointer. Forthcoming
58        implementation of support for (*MARK) will need an extra pointer on the
59        stack; I have reserved it now, so that the stack frame size does not
60        decrease.
61    
62    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
63        item in branch that calls a recursion is a subroutine call - as in the
64        second branch in the above example - was incorrectly given the compile-
65        time error "recursive call could loop indefinitely" because pcre_compile()
66        was not correctly checking the subroutine for matching a non-empty string.
67    
68    14. The checks for overrunning compiling workspace could trigger after an
69        overrun had occurred. This is a "should never occur" error, but it can be
70        triggered by pathological patterns such as hundreds of nested parentheses.
71        The checks now trigger 100 bytes before the end of the workspace.
72    
73    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
74    
75    
76    Version 8.01 19-Jan-2010
77    ------------------------
78    
79    1.  If a pattern contained a conditional subpattern with only one branch (in
80        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
81        computed the wrong minimum data length (which is of course zero for such
82        subpatterns). This could cause incorrect "no match" results.
83    
84    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
85        the pattern is reset in the first branch, pcre_compile() failed with
86        "internal error: code overflow at offset...". This happened only when
87        the reset was to the original external option setting. (An optimization
88        abstracts leading options settings into an external setting, which was the
89        cause of this.)
90    
91    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
92        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
93        assertion pattern did not match (meaning that the assertion was true), it
94        was incorrectly treated as false if the SKIP had been reached during the
95        matching. This also applied to assertions used as conditions.
96    
97    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
98        assertion subpattern, including such a pattern used as a condition,
99        unpredictable results occurred, instead of the error return
100        PCRE_ERROR_DFA_UITEM.
101    
102    5.  The C++ GlobalReplace function was not working like Perl for the special
103        situation when an empty string is matched. It now does the fancy magic
104        stuff that is necessary.
105    
106    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
107        removed. (These were left over from very, very early versions of PCRE.)
108    
109    7.  Some cosmetic changes to the code to make life easier when compiling it
110        as part of something else:
111    
112        (a) Change DEBUG to PCRE_DEBUG.
113    
114        (b) In pcre_compile(), rename the member of the "branch_chain" structure
115            called "current" as "current_branch", to prevent a collision with the
116            Linux macro when compiled as a kernel module.
117    
118        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
119            prevent a collision with the Linux macro when compiled as a kernel
120            module.
121    
122    8.  In pcre_compile() there are some checks for integer overflows that used to
123        cast potentially large values to (double). This has been changed to that
124        when building, a check for int64_t is made, and if it is found, it is used
125        instead, thus avoiding the use of floating point arithmetic. (There is no
126        other use of FP in PCRE.) If int64_t is not found, the fallback is to
127        double.
128    
129    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
130        2005 (difference between two addresses compared to an unsigned value).
131    
132    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
133        custom one, because of the following reported problem in Windows:
134    
135          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
136              under Win32.
137          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
138              therefore missing the function definition.
139          - The compiler thus generates a "C" signature for the test function.
140          - The linker fails to find the "C" function.
141          - PCRE fails to configure if asked to do so against libbz2.
142    
143    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
144        messages were output:
145    
146          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
147          rerunning libtoolize, to keep the correct libtool macros in-tree.
148          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
149    
150        I have done both of these things.
151    
152    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
153        most of the time, it *can* run out if it is given a pattern that contains a
154        runaway infinite recursion. I updated the discussion in the pcrestack man
155        page.
156    
157    13. Now that we have gone to the x.xx style of version numbers, the minor
158        version may start with zero. Using 08 or 09 is a bad idea because users
159        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
160        interpreted as invalid octal numbers. I've updated the previous comment in
161        configure.ac, and also added a check that gives an error if 08 or 09 are
162        used.
163    
164    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
165        causing partial matching to fail when the end of the subject matched \W
166        in a UTF-8 pattern where \W was quantified with a minimum of 3.
167    
168    15. There were some discrepancies between the declarations in pcre_internal.h
169        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
170        their definitions. The declarations used "const uschar *" and the
171        definitions used USPTR. Even though USPTR is normally defined as "const
172        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
173        reported that: "This difference in casting confuses some C++ compilers, for
174        example, SunCC recognizes above declarations as different functions and
175        generates broken code for hbpcre." I have changed the declarations to use
176        USPTR.
177    
178    16. GNU libtool is named differently on some systems. The autogen.sh script now
179        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
180        (FreeBSD).
181    
182    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
183        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
184        comment: "Figure out how to create a longlong from a string: strtoll and
185        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
186        instance, but it only takes 2 args instead of 3!"
187    
188    18. A subtle bug concerned with back references has been fixed by a change of
189        specification, with a corresponding code fix. A pattern such as
190        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
191        refers, was giving matches when it shouldn't. For example, xa=xaaa would
192        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
193        same bug. Such groups have to be quantified to be useful, or contained
194        inside another quantified group. (If there's no repetition, the reference
195        can never match.) The problem arises because, having left the group and
196        moved on to the rest of the pattern, a later failure that backtracks into
197        the group uses the captured value from the final iteration of the group
198        rather than the correct earlier one. I have fixed this in PCRE by forcing
199        any group that contains a reference to itself to be an atomic group; that
200        is, there cannot be any backtracking into it once it has completed. This is
201        similar to recursive and subroutine calls.
202    
203    
204    Version 8.00 19-Oct-09
205    ----------------------
206    
207    1.  The table for translating pcre_compile() error codes into POSIX error codes
208        was out-of-date, and there was no check on the pcre_compile() error code
209        being within the table. This could lead to an OK return being given in
210        error.
211    
212    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
213        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
214        in a Windows environment.
215    
216    3.  The pcregrep --count option prints the count for each file even when it is
217        zero, as does GNU grep. However, pcregrep was also printing all files when
218        --files-with-matches was added. Now, when both options are given, it prints
219        counts only for those files that have at least one match. (GNU grep just
220        prints the file name in this circumstance, but including the count seems
221        more useful - otherwise, why use --count?) Also ensured that the
222        combination -clh just lists non-zero counts, with no names.
223    
224    4.  The long form of the pcregrep -F option was incorrectly implemented as
225        --fixed_strings instead of --fixed-strings. This is an incompatible change,
226        but it seems right to fix it, and I didn't think it was worth preserving
227        the old behaviour.
228    
229    5.  The command line items --regex=pattern and --regexp=pattern were not
230        recognized by pcregrep, which required --regex pattern or --regexp pattern
231        (with a space rather than an '='). The man page documented the '=' forms,
232        which are compatible with GNU grep; these now work.
233    
234    6.  No libpcreposix.pc file was created for pkg-config; there was just
235        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
236    
237    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
238        when UCP support is not needed, by modifying the Python script that
239        generates it from Unicode data files. This should not matter if the module
240        is correctly used as a library, but I received one complaint about 50K of
241        unwanted data. My guess is that the person linked everything into his
242        program rather than using a library. Anyway, it does no harm.
243    
244    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
245        was a minimum greater than 1 for a wide character in a possessive
246        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
247        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
248        character. Chaos in the form of incorrect output or a compiling loop could
249        result.
250    
251    9.  The restrictions on what a pattern can contain when partial matching is
252        requested for pcre_exec() have been removed. All patterns can now be
253        partially matched by this function. In addition, if there are at least two
254        slots in the offset vector, the offset of the earliest inspected character
255        for the match and the offset of the end of the subject are set in them when
256        PCRE_ERROR_PARTIAL is returned.
257    
258    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
259        synonymous with PCRE_PARTIAL, for backwards compatibility, and
260        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
261        and may be more useful for multi-segment matching.
262    
263    11. Partial matching with pcre_exec() is now more intuitive. A partial match
264        used to be given if ever the end of the subject was reached; now it is
265        given only if matching could not proceed because another character was
266        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
267        string "Z", which now yields "no match" instead of "partial match". In the
268        case of pcre_dfa_exec(), "no match" is given if every matching path for the
269        final character ended with (*FAIL).
270    
271    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
272        if the pattern had a "must contain" character that was already found in the
273        earlier partial match, unless partial matching was again requested. For
274        example, with the pattern /dog.(body)?/, the "must contain" character is
275        "g". If the first part-match was for the string "dog", restarting with
276        "sbody" failed. This bug has been fixed.
277    
278    13. The string returned by pcre_dfa_exec() after a partial match has been
279        changed so that it starts at the first inspected character rather than the
280        first character of the match. This makes a difference only if the pattern
281        starts with a lookbehind assertion or \b or \B (\K is not supported by
282        pcre_dfa_exec()). It's an incompatible change, but it makes the two
283        matching functions compatible, and I think it's the right thing to do.
284    
285    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
286        so that the demonstration program is easily available in environments where
287        PCRE has not been installed from source.
288    
289    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
290        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
291        library.
292    
293    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
294        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
295        is not the first non-POSIX option to be added. Clearly some people find
296        these options useful.
297    
298    17. If a caller to the POSIX matching function regexec() passes a non-zero
299        value for nmatch with a NULL value for pmatch, the value of
300        nmatch is forced to zero.
301    
302    18. RunGrepTest did not have a test for the availability of the -u option of
303        the diff command, as RunTest does. It now checks in the same way as
304        RunTest, and also checks for the -b option.
305    
306    19. If an odd number of negated classes containing just a single character
307        interposed, within parentheses, between a forward reference to a named
308        subpattern and the definition of the subpattern, compilation crashed with
309        an internal error, complaining that it could not find the referenced
310        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
311        [The bug was that it was starting one character too far in when skipping
312        over the character class, thus treating the ] as data rather than
313        terminating the class. This meant it could skip too much.]
314    
315    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
316        /g option in pcretest when the pattern contains \K, which makes it possible
317        to have an empty string match not at the start, even when the pattern is
318        anchored. Updated pcretest and pcredemo to use this option.
319    
320    21. If the maximum number of capturing subpatterns in a recursion was greater
321        than the maximum at the outer level, the higher number was returned, but
322        with unset values at the outer level. The correct (outer level) value is
323        now given.
324    
325    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
326        PCRE did not set those parentheses (unlike Perl). I have now found a way to
327        make it do so. The string so far is captured, making this feature
328        compatible with Perl.
329    
330    23. The tests have been re-organized, adding tests 11 and 12, to make it
331        possible to check the Perl 5.10 features against Perl 5.10.
332    
333    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
334        pattern matches a fixed length string. PCRE did not allow this; now it
335        does. Neither allows recursion.
336    
337    25. I finally figured out how to implement a request to provide the minimum
338        length of subject string that was needed in order to match a given pattern.
339        (It was back references and recursion that I had previously got hung up
340        on.) This code has now been added to pcre_study(); it finds a lower bound
341        to the length of subject needed. It is not necessarily the greatest lower
342        bound, but using it to avoid searching strings that are too short does give
343        some useful speed-ups. The value is available to calling programs via
344        pcre_fullinfo().
345    
346    26. While implementing 25, I discovered to my embarrassment that pcretest had
347        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
348        study optimizations had never been tested with that matching function.
349        Oops. What is worse, even when it was passed study data, there was a bug in
350        pcre_dfa_exec() that meant it never actually used it. Double oops. There
351        were also very few tests of studied patterns with pcre_dfa_exec().
352    
353    27. If (?| is used to create subpatterns with duplicate numbers, they are now
354        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
355        on the other side of the coin, they are no longer allowed to have different
356        names, because these cannot be distinguished in PCRE, and this has caused
357        confusion. (This is a difference from Perl.)
358    
359    28. When duplicate subpattern names are present (necessarily with different
360        numbers, as required by 27 above), and a test is made by name in a
361        conditional pattern, either for a subpattern having been matched, or for
362        recursion in such a pattern, all the associated numbered subpatterns are
363        tested, and the overall condition is true if the condition is true for any
364        one of them. This is the way Perl works, and is also more like the way
365        testing by number works.
366    
367    
368    Version 7.9 11-Apr-09
369    ---------------------
370    
371    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
372        (pcretest), all targets were linked against these libraries. This included
373        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
374        libraries. This caused unwanted dependencies to be created. This problem
375        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
376        pcretest is linked with readline.
377    
378    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
379        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
380        moved outside it again, because FALSE and TRUE are already defined in AIX,
381        but BOOL is not.
382    
383    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
384        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
385    
386    4.  The pcregrep documentation said spaces were inserted as well as colons (or
387        hyphens) following file names and line numbers when outputting matching
388        lines. This is not true; no spaces are inserted. I have also clarified the
389        wording for the --colour (or --color) option.
390    
391    5.  In pcregrep, when --colour was used with -o, the list of matching strings
392        was not coloured; this is different to GNU grep, so I have changed it to be
393        the same.
394    
395    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
396        each matching line was coloured. Now it goes on to look for further matches
397        of any of the test patterns, which is the same behaviour as GNU grep.
398    
399    7.  A pattern that could match an empty string could cause pcregrep to loop; it
400        doesn't make sense to accept an empty string match in pcregrep, so I have
401        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
402        seems to be how GNU grep behaves.
403    
404    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
405        start or after a newline", because the conditional assertion was not being
406        correctly handled. The rule now is that both the assertion and what follows
407        in the first alternative must satisfy the test.
408    
409    9.  If auto-callout was enabled in a pattern with a conditional group whose
410        condition was an assertion, PCRE could crash during matching, both with
411        pcre_exec() and pcre_dfa_exec().
412    
413    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
414        used for matching.
415    
416    11. Unicode property support in character classes was not working for
417        characters (bytes) greater than 127 when not in UTF-8 mode.
418    
419    12. Added the -M command line option to pcretest.
420    
421    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
422    
423    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
424    
425    16. Added comments and documentation about mis-use of no_arg in the C++
426        wrapper.
427    
428    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
429        from Martin Jerabek that uses macro names for all relevant character and
430        string constants.
431    
432    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
433        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
434        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
435        these, but not everybody uses configure.
436    
437    19. A conditional group that had only one branch was not being correctly
438        recognized as an item that could match an empty string. This meant that an
439        enclosing group might also not be so recognized, causing infinite looping
440        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
441        with the subject "ab", where knowledge that the repeated group can match
442        nothing is needed in order to break the loop.
443    
444    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
445        exec(), but without supplying a callout function, matching went wrong.
446    
447    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
448        leak if the size of the offset vector was greater than 30. When the vector
449        is smaller, the saved offsets during recursion go onto a local stack
450        vector, but for larger vectors malloc() is used. It was failing to free
451        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
452        error, in fact).
453    
454    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
455        heapframe that is used only when UTF-8 support is enabled. This caused no
456        problem, but was untidy.
457    
458    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
459        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
460        included within another project.
461    
462    24. Steven Van Ingelgem's patches to add more options to the CMake support,
463        slightly modified by me:
464    
465          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
466              not building pcregrep.
467    
468          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
469              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
470    
471    25. Forward references, both numeric and by name, in patterns that made use of
472        duplicate group numbers, could behave incorrectly or give incorrect errors,
473        because when scanning forward to find the reference group, PCRE was not
474        taking into account the duplicate group numbers. A pattern such as
475        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
476    
477    26. Changed a few more instances of "const unsigned char *" to USPTR, making
478        the feature of a custom pointer more persuasive (as requested by a user).
479    
480    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
481        pcretest.c, inside #ifndefs, because it seems they are sometimes already
482        pre-defined.
483    
484    28. Added support for (*UTF8) at the start of a pattern.
485    
486    29. Arrange for flags added by the "release type" setting in CMake to be shown
487        in the configuration summary.
488    
489    
490    Version 7.8 05-Sep-08
491    ---------------------
492    
493    1.  Replaced UCP searching code with optimized version as implemented for Ad
494        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
495        stage table and inline lookup instead of a function, giving speed ups of 2
496        to 5 times on some simple patterns that I tested. Permission was given to
497        distribute the MultiStage2.py script that generates the tables (it's not in
498        the tarball, but is in the Subversion repository).
499    
500    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
501        scripts.
502    
503    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
504        a group with a zero qualifier. The result of the study could be incorrect,
505        or the function might crash, depending on the pattern.
506    
507    4.  Caseless matching was not working for non-ASCII characters in back
508        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
509        It now works when Unicode Property Support is available.
510    
511    5.  In pcretest, an escape such as \x{de} in the data was always generating
512        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
513        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
514        truncation.
515    
516    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
517    
518    7.  Added two (int) casts to pcregrep when printing the difference of two
519        pointers, in case they are 64-bit values.
520    
521    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
522        test 2 if it fails.
523    
524    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
525        and a #define of that name to empty if it is not externally set. This is to
526        allow users of MSVC to set it if necessary.
527    
528    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
529        the convenience functions in the pcre_get.c source file.
530    
531    11. An option change at the start of a pattern that had top-level alternatives
532        could cause overwriting and/or a crash. This command provoked a crash in
533        some environments:
534    
535          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
536    
537        This potential security problem was recorded as CVE-2008-2371.
538    
539    12. For a pattern where the match had to start at the beginning or immediately
540        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
541        pcre_dfa_exec() could read past the end of the passed subject if there was
542        no match. To help with detecting such bugs (e.g. with valgrind), I modified
543        pcretest so that it places the subject at the end of its malloc-ed buffer.
544    
545    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
546        exec() might read past the end of the data buffer in UTF-8 mode.
547    
548    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
549        the data contained the byte 0x85 as part of a UTF-8 character within its
550        first line. This applied both to normal and DFA matching.
551    
552    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
553        /^[^d]*?$/8 failed to match "abc".
554    
555    16. Added a missing copyright notice to pcrecpp_internal.h.
556    
557    17. Make it more clear in the documentation that values returned from
558        pcre_exec() in ovector are byte offsets, not character counts.
559    
560    18. Tidied a few places to stop certain compilers from issuing warnings.
561    
562    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
563        supplied by Stefan Weber. I made a further small update for 7.8 because
564        there is a change of source arrangements: the pcre_searchfuncs.c module is
565        replaced by pcre_ucd.c.
566    
567    
568    Version 7.7 07-May-08
569    ---------------------
570    
571    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
572        a string to a long long, pretend we don't even have a long long." This is
573        done by checking for the strtoq, strtoll, and _strtoi64 functions.
574    
575    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
576        pre-7.6 versions, which defined a global no_arg variable instead of putting
577        it in the RE class. (See also #8 below.)
578    
579    3.  Remove a line of dead code, identified by coverity and reported by Nuno
580        Lopes.
581    
582    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
583    
584        (1) The include/exclude patterns were being applied to the whole pathnames
585            of files, instead of just to the final components.
586    
587        (2) If there was more than one level of directory, the subdirectories were
588            skipped unless they satisfied the include/exclude conditions. This is
589            inconsistent with GNU grep (and could even be seen as contrary to the
590            pcregrep specification - which I improved to make it absolutely clear).
591            The action now is always to scan all levels of directory, and just
592            apply the include/exclude patterns to regular files.
593    
594    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
595        --exclude_dir in the tests to avoid scanning .svn directories.
596    
597    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
598        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
599        doesn't support NULs in patterns.
600    
601    7.  Added some missing "const"s to declarations of static tables in
602        pcre_compile.c and pcre_dfa_exec.c.
603    
604    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
605        caused by fix #2  above. (Subsequently also a second patch to fix the
606        first patch. And a third patch - this was a messy problem.)
607    
608    9.  Applied Craig's patch to remove the use of push_back().
609    
610    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
611        matching function regexec().
612    
613    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
614        which, however, unlike Perl's \g{...}, are subroutine calls, not back
615        references. PCRE supports relative numbers with this syntax (I don't think
616        Oniguruma does).
617    
618    12. Previously, a group with a zero repeat such as (...){0} was completely
619        omitted from the compiled regex. However, this means that if the group
620        was called as a subroutine from elsewhere in the pattern, things went wrong
621        (an internal error was given). Such groups are now left in the compiled
622        pattern, with a new opcode that causes them to be skipped at execution
623        time.
624    
625    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
626        to the way PCRE behaves:
627    
628        (a) A lone ] character is dis-allowed (Perl treats it as data).
629    
630        (b) A back reference to an unmatched subpattern matches an empty string
631            (Perl fails the current match path).
632    
633        (c) A data ] in a character class must be notated as \] because if the
634            first data character in a class is ], it defines an empty class. (In
635            Perl it is not possible to have an empty class.) The empty class []
636            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
637            The negative empty class [^] matches any one character, independently
638            of the DOTALL setting.
639    
640    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
641        non-existent subpattern following a character class starting with ']' and
642        containing () gave an internal compiling error instead of "reference to
643        non-existent subpattern". Fortunately, when the pattern did exist, the
644        compiled code was correct. (When scanning forwards to check for the
645        existencd of the subpattern, it was treating the data ']' as terminating
646        the class, so got the count wrong. When actually compiling, the reference
647        was subsequently set up correctly.)
648    
649    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
650        it was being rejected as not supported by pcre_dfa_exec(), even though
651        other assertions are supported. I have made pcre_dfa_exec() support
652        (*FAIL).
653    
654    16. The implementation of 13c above involved the invention of a new opcode,
655        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
656        cannot be changed at match time, I realized I could make a small
657        improvement to matching performance by compiling OP_ALLANY instead of
658        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
659        on the OP_ANY path.
660    
661    17. Compiling pcretest on Windows with readline support failed without the
662        following two fixes: (1) Make the unistd.h include conditional on
663        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
664    
665    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
666        ncurses library to be included for pcretest when ReadLine support is
667        requested, but also to allow for it to be overridden. This patch came from
668        Daniel Bergström.
669    
670    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
671        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
672        any errors with the current Unicode tables. Thanks to Peter Kankowski for
673        spotting this.
674    
675    
676    Version 7.6 28-Jan-08
677    ---------------------
678    
679    1.  A character class containing a very large number of characters with
680        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
681        overflow.
682    
683    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
684        HAVE_LONG_LONG is not defined.
685    
686    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
687        bring them up to date and include new features. This patch includes:
688    
689        - Fixed PH's badly added libz and libbz2 support.
690        - Fixed a problem with static linking.
691        - Added pcredemo. [But later removed - see 7 below.]
692        - Fixed dftables problem and added an option.
693        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
694            HAVE_LONG_LONG.
695        - Added readline support for pcretest.
696        - Added an listing of the option settings after cmake has run.
697    
698    4.  A user submitted a patch to Makefile that makes it easy to create
699        "pcre.dll" under mingw when using Configure/Make. I added stuff to
700        Makefile.am that cause it to include this special target, without
701        affecting anything else. Note that the same mingw target plus all
702        the other distribution libraries and programs are now supported
703        when configuring with CMake (see 6 below) instead of with
704        Configure/Make.
705    
706    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
707        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
708        exported in the Windows port". It has not yet been confirmed that the patch
709        solves the problem, but it does no harm.
710    
711    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
712        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
713        with CMake, and also correct the comment about stack recursion.
714    
715    7.  Remove the automatic building of pcredemo from the ./configure system and
716        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
717        of a program that users should build themselves after PCRE is installed, so
718        building it automatically is not really right. What is more, it gave
719        trouble in some build environments.
720    
721    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
722    
723    
724    Version 7.5 10-Jan-08
725    ---------------------
726    
727    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
728        values in parens when parsing an RE using the C++ wrapper."
729    
730    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
731        Characters greater than 255 were excluded from the class instead of being
732        included.
733    
734    3.  The same bug as (2) above applied to negated POSIX classes such as
735        [:^space:].
736    
737    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
738        defined or documented. It seems to have been a typo for PCRE_STATIC, so
739        I have changed it.
740    
741    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
742        first named subpattern) and a construct such as (?&a) would reference the
743        first named subpattern whose name started with "a" (in other words, the
744        length check was missing). Both these problems are fixed. "Subpattern name
745        expected" is now given for (?&) (a zero-length name), and this patch also
746        makes it give the same error for \k'' (previously it complained that that
747        was a reference to a non-existent subpattern).
748    
749    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
750        this is right because (?- can be followed by option settings as well as by
751        digits. I have, however, made the messages clearer.
752    
753    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
754        than the number used in the conditional) now cause a compile-time error.
755        This is actually not compatible with Perl, which accepts such patterns, but
756        treats the conditional as always being FALSE (as PCRE used to), but it
757        seems to me that giving a diagnostic is better.
758    
759    8.  Change "alphameric" to the more common word "alphanumeric" in comments
760        and messages.
761    
762    9.  Fix two occurrences of "backslash" in comments that should have been
763        "backspace".
764    
765    10. Remove two redundant lines of code that can never be obeyed (their function
766        was moved elsewhere).
767    
768    11. The program that makes PCRE's Unicode character property table had a bug
769        which caused it to generate incorrect table entries for sequences of
770        characters that have the same character type, but are in different scripts.
771        It amalgamated them into a single range, with the script of the first of
772        them. In other words, some characters were in the wrong script. There were
773        thirteen such cases, affecting characters in the following ranges:
774    
775          U+002b0 - U+002c1
776          U+0060c - U+0060d
777          U+0061e - U+00612
778          U+0064b - U+0065e
779          U+0074d - U+0076d
780          U+01800 - U+01805
781          U+01d00 - U+01d77
782          U+01d9b - U+01dbf
783          U+0200b - U+0200f
784          U+030fc - U+030fe
785          U+03260 - U+0327f
786          U+0fb46 - U+0fbb1
787          U+10450 - U+1049d
788    
789    12. The -o option (show only the matching part of a line) for pcregrep was not
790        compatible with GNU grep in that, if there was more than one match in a
791        line, it showed only the first of them. It now behaves in the same way as
792        GNU grep.
793    
794    13. If the -o and -v options were combined for pcregrep, it printed a blank
795        line for every non-matching line. GNU grep prints nothing, and pcregrep now
796        does the same. The return code can be used to tell if there were any
797        non-matching lines.
798    
799    14. Added --file-offsets and --line-offsets to pcregrep.
800    
801    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
802        infinitely looping recursion. The bug was that positive lookaheads were not
803        being skipped when checking for a possible empty match (negative lookaheads
804        and both kinds of lookbehind were skipped).
805    
806    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
807        inclusion of <windows.h> to before rather than after the definition of
808        INVALID_FILE_ATTRIBUTES (patch from David Byron).
809    
810    17. Specifying a possessive quantifier with a specific limit for a Unicode
811        character property caused pcre_compile() to compile bad code, which led at
812        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
813        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
814        caused the error; without that there was no problem.
815    
816    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
817    
818    19. Added --enable-pcretest-libreadline.
819    
820    20. In pcrecpp.cc, the variable 'count' was incremented twice in
821        RE::GlobalReplace(). As a result, the number of replacements returned was
822        double what it should be. I removed one of the increments, but Craig sent a
823        later patch that removed the other one (the right fix) and added unit tests
824        that check the return values (which was not done before).
825    
826    21. Several CMake things:
827    
828        (1) Arranged that, when cmake is used on Unix, the libraries end up with
829            the names libpcre and libpcreposix, not just pcre and pcreposix.
830    
831        (2) The above change means that pcretest and pcregrep are now correctly
832            linked with the newly-built libraries, not previously installed ones.
833    
834        (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
835    
836    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
837        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
838        UTF-8 newline character). The key issue is that the pattern starts .*;
839        this means that the match must be either at the beginning, or after a
840        newline. The bug was in the code for advancing after a failed match and
841        checking that the new position followed a newline. It was not taking
842        account of UTF-8 characters correctly.
843    
844    23. PCRE was behaving differently from Perl in the way it recognized POSIX
845        character classes. PCRE was not treating the sequence [:...:] as a
846        character class unless the ... were all letters. Perl, however, seems to
847        allow any characters between [: and :], though of course it rejects as
848        unknown any "names" that contain non-letters, because all the known class
849        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
850        for example, whereas PCRE did not - it did not recognize a POSIX character
851        class. This seemed a bit dangerous, so the code has been changed to be
852        closer to Perl. The behaviour is not identical to Perl, because PCRE will
853        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
854        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
855        Perl does, and where it didn't before.
856    
857    24. Rewrite so as to remove the single use of %n from pcregrep because in some
858        Windows environments %n is disabled by default.
859    
860    
861    Version 7.4 21-Sep-07
862    ---------------------
863    
864    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
865        means that a class such as [\s] counted as "explicit reference to CR or
866        LF". That isn't really right - the whole point of the change was to try to
867        help when there was an actual mention of one of the two characters. So now
868        the change happens only if \r or \n (or a literal CR or LF) character is
869        encountered.
870    
871    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
872        of both had grown to the point where there were only 3 bits left.
873        Fortunately, there was spare space in the data structure, and so I have
874        moved the internal flags into a new 16-bit field to free up more option
875        bits.
876    
877    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
878        but did not set the internal JCHANGED flag - either of these is enough to
879        control the way the "get" function works - but the PCRE_INFO_JCHANGED
880        facility is supposed to tell if (?J) was ever used, so now (?J) at the
881        start sets both bits.
882    
883    4.  Added options (at build time, compile time, exec time) to change \R from
884        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
885    
886    5.  doc/pcresyntax.html was missing from the distribution.
887    
888    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
889        compatibility, even though it is no longer used.
890    
891    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
892        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
893        windows.h file is present (where different names are used). [This was
894        reversed later after testing - see 16 below.]
895    
896    8.  Changed all #include <config.h> to #include "config.h". There were also
897        some further <pcre.h> cases that I changed to "pcre.h".
898    
899    9.  When pcregrep was used with the --colour option, it missed the line ending
900        sequence off the lines that it output.
901    
902    10. It was pointed out to me that arrays of string pointers cause lots of
903        relocations when a shared library is dynamically loaded. A technique of
904        using a single long string with a table of offsets can drastically reduce
905        these. I have refactored PCRE in four places to do this. The result is
906        dramatic:
907    
908          Originally:                          290
909          After changing UCP table:            187
910          After changing error message table:   43
911          After changing table of "verbs"       36
912          After changing table of Posix names   22
913    
914        Thanks to the folks working on Gregex for glib for this insight.
915    
916    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
917        unicode-properties was also set.
918    
919    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
920    
921    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
922        checked only for CRLF.
923    
924    14. Added casts to pcretest.c to avoid compiler warnings.
925    
926    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
927    
928    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
929        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
930        entirely. This removes changes made in 7 above.
931    
932    17. The CMake files have been updated, and there is now more information about
933        building with CMake in the NON-UNIX-USE document.
934    
935    
936    Version 7.3 28-Aug-07
937    ---------------------
938    
939     1. In the rejigging of the build system that eventually resulted in 7.1, the
940        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
941        brackets there is not right, since it causes compilers to look for an
942        installed pcre.h, not the version that is in the source that is being
943        compiled (which of course may be different). I have changed it back to:
944    
945          #include "pcre.h"
946    
947        I have a vague recollection that the change was concerned with compiling in
948        different directories, but in the new build system, that is taken care of
949        by the VPATH setting the Makefile.
950    
951     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
952        when the subject happened to end in the byte 0x85 (e.g. if the last
953        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
954        characters but of course it shouldn't be taken as a newline when it is part
955        of another character. The bug was that, for an unlimited repeat of . in
956        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
957        characters when looking for a newline.
958    
959     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
960    
961     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
962        in debug output.
963    
964     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
965        long printing in the pcrecpp unittest when running under MinGW.
966    
967     6. ESC_K was left out of the EBCDIC table.
968    
969     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
970        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
971        limit also applies to "virtual nesting" when a pattern is recursive, and in
972        this case 1000 isn't so big. I have been able to remove this limit at the
973        expense of backing off one optimization in certain circumstances. Normally,
974        when pcre_exec() would call its internal match() function recursively and
975        immediately return the result unconditionally, it uses a "tail recursion"
976        feature to save stack. However, when a subpattern that can match an empty
977        string has an unlimited repetition quantifier, it no longer makes this
978        optimization. That gives it a stack frame in which to save the data for
979        checking that an empty string has been matched. Previously this was taken
980        from the 1000-entry workspace that had been reserved. So now there is no
981        explicit limit, but more stack is used.
982    
983     8. Applied Daniel's patches to solve problems with the import/export magic
984        syntax that is required for Windows, and which was going wrong for the
985        pcreposix and pcrecpp parts of the library. These were overlooked when this
986        problem was solved for the main library.
987    
988     9. There were some crude static tests to avoid integer overflow when computing
989        the size of patterns that contain repeated groups with explicit upper
990        limits. As the maximum quantifier is 65535, the maximum group length was
991        set at 30,000 so that the product of these two numbers did not overflow a
992        32-bit integer. However, it turns out that people want to use groups that
993        are longer than 30,000 bytes (though not repeat them that many times).
994        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
995        made it possible to implement the integer overflow checks in a much more
996        dynamic way, which I have now done. The artificial limitation on group
997        length has been removed - we now have only the limit on the total length of
998        the compiled pattern, which depends on the LINK_SIZE setting.
999    
1000    10. Fixed a bug in the documentation for get/copy named substring when
1001        duplicate names are permitted. If none of the named substrings are set, the
1002        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1003        empty string.
1004    
1005    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1006        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1007        because the ] is interpreted as the first data character and the
1008        terminating ] is not found. PCRE has been made compatible with Perl in this
1009        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1010        cause memory overwriting.
1011    
1012    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1013        string has been matched (to stop an infinite loop). It was not recognizing
1014        a conditional subpattern that could match an empty string if that
1015        subpattern was within another subpattern. For example, it looped when
1016        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
1017        condition was not nested. This bug has been fixed.
1018    
1019    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1020        past the start of the subject in the presence of bytes with the top bit
1021        set, for example "\x8aBCD".
1022    
1023    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1024        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1025    
1026    14. Optimized (?!) to (*FAIL).
1027    
1028    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1029        This restricts code points to be within the range 0 to 0x10FFFF, excluding
1030        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1031        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1032        does: it's just the validity check that is more restrictive.
1033    
1034    16. Inserted checks for integer overflows during escape sequence (backslash)
1035        processing, and also fixed erroneous offset values for syntax errors during
1036        backslash processing.
1037    
1038    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1039        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1040    
1041    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1042        caused an overrun.
1043    
1044    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1045        something other than just ASCII characters) inside a group that had an
1046        unlimited repeat caused a loop at compile time (while checking to see
1047        whether the group could match an empty string).
1048    
1049    20. Debugging a pattern containing \p or \P could cause a crash. For example,
1050        [\P{Any}] did so. (Error in the code for printing property names.)
1051    
1052    21. An orphan \E inside a character class could cause a crash.
1053    
1054    22. A repeated capturing bracket such as (A)? could cause a wild memory
1055        reference during compilation.
1056    
1057    23. There are several functions in pcre_compile() that scan along a compiled
1058        expression for various reasons (e.g. to see if it's fixed length for look
1059        behind). There were bugs in these functions when a repeated \p or \P was
1060        present in the pattern. These operators have additional parameters compared
1061        with \d, etc, and these were not being taken into account when moving along
1062        the compiled data. Specifically:
1063    
1064        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1065            length.
1066    
1067        (b) An item such as \pL+ within a repeated group could cause crashes or
1068            loops.
1069    
1070        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1071            "reference to non-existent subpattern" error.
1072    
1073        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1074    
1075    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1076        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1077    
1078    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1079    
1080    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1081        character were causing crashes (broken optimization).
1082    
1083    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1084        \p or \P) caused a compile-time loop.
1085    
1086    28. More problems have arisen in unanchored patterns when CRLF is a valid line
1087        break. For example, the unstudied pattern [\r\n]A does not match the string
1088        "\r\nA" because change 7.0/46 below moves the current point on by two
1089        characters after failing to match at the start. However, the pattern \nA
1090        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
1091        the same is true. There doesn't seem any very clean way out of this, but
1092        what I have chosen to do makes the common cases work: PCRE now takes note
1093        of whether there can be an explicit match for \r or \n anywhere in the
1094        pattern, and if so, 7.0/46 no longer applies. As part of this change,
1095        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1096        pattern has explicit CR or LF references.
1097    
1098    29. Added (*CR) etc for changing newline setting at start of pattern.
1099    
1100    
1101    Version 7.2 19-Jun-07
1102    ---------------------
1103    
1104     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
1105        which is apparently normally available under Windows.
1106    
1107     2. Re-jig the pcregrep tests with different newline settings in an attempt
1108        to make them independent of the local environment's newline setting.
1109    
1110     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
1111    
1112     4. Some of the "internals" tests were previously cut out when the link size
1113        was not 2, because the output contained actual offsets. The recent new
1114        "Z" feature of pcretest means that these can be cut out, making the tests
1115        usable with all link sizes.
1116    
1117     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
1118        stack recursion. This gives a massive performance boost under BSD, but just
1119        a small improvement under Linux. However, it saves one field in the frame
1120        in all cases.
1121    
1122     6. Added more features from the forthcoming Perl 5.10:
1123    
1124        (a) (?-n) (where n is a string of digits) is a relative subroutine or
1125            recursion call. It refers to the nth most recently opened parentheses.
1126    
1127        (b) (?+n) is also a relative subroutine call; it refers to the nth next
1128            to be opened parentheses.
1129    
1130        (c) Conditions that refer to capturing parentheses can be specified
1131            relatively, for example, (?(-2)... or (?(+3)...
1132    
1133        (d) \K resets the start of the current match so that everything before
1134            is not part of it.
1135    
1136        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1137    
1138        (f) \g{name} is another synonym - part of Perl 5.10's unification of
1139            reference syntax.
1140    
1141        (g) (?| introduces a group in which the numbering of parentheses in each
1142            alternative starts with the same number.
1143    
1144        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
1145    
1146     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
1147        PCRE_INFO_JCHANGED.
1148    
1149     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
1150        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
1151        for detecting groups that can match an empty string.
1152    
1153     9. A pattern with a very large number of alternatives (more than several
1154        hundred) was running out of internal workspace during the pre-compile
1155        phase, where pcre_compile() figures out how much memory will be needed. A
1156        bit of new cunning has reduced the workspace needed for groups with
1157        alternatives. The 1000-alternative test pattern now uses 12 bytes of
1158        workspace instead of running out of the 4096 that are available.
1159    
1160    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
1161    
1162    11. Applied patch from Google to remove an optimization that didn't quite work.
1163        The report of the bug said:
1164    
1165          pcrecpp::RE("a*").FullMatch("aaa") matches, while
1166          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
1167          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
1168    
1169    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
1170        it matched the wrong number of bytes.
1171    
1172    
1173    Version 7.1 24-Apr-07
1174    ---------------------
1175    
1176     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
1177        that is more "standard", making use of automake and other Autotools. There
1178        is some re-arrangement of the files and adjustment of comments consequent
1179        on this.
1180    
1181     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
1182        for recursive directory scanning broke on some systems because the files
1183        are not scanned in any specific order and on different systems the order
1184        was different. A call to "sort" has been inserted into RunGrepTest for the
1185        approprate test as a short-term fix. In the longer term there may be an
1186        alternative.
1187    
1188     3. I had an email from Eric Raymond about problems translating some of PCRE's
1189        man pages to HTML (despite the fact that I distribute HTML pages, some
1190        people do their own conversions for various reasons). The problems
1191        concerned the use of low-level troff macros .br and .in. I have therefore
1192        removed all such uses from the man pages (some were redundant, some could
1193        be replaced by .nf/.fi pairs). The 132html script that I use to generate
1194        HTML has been updated to handle .nf/.fi and to complain if it encounters
1195        .br or .in.
1196    
1197     4. Updated comments in configure.ac that get placed in config.h.in and also
1198        arranged for config.h to be included in the distribution, with the name
1199        config.h.generic, for the benefit of those who have to compile without
1200        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
1201    
1202     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
1203        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
1204        makevp.bat for the current PCRE, using the additional files
1205        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
1206    
1207     6. A Windows user reported a minor discrepancy with test 2, which turned out
1208        to be caused by a trailing space on an input line that had got lost in his
1209        copy. The trailing space was an accident, so I've just removed it.
1210    
1211     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
1212        that is needed.
1213    
1214     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
1215        as "const" (a) because they are and (b) because it helps the PHP
1216        maintainers who have recently made a script to detect big data structures
1217        in the php code that should be moved to the .rodata section. I remembered
1218        to update Builducptable as well, so it won't revert if ucptable.h is ever
1219        re-created.
1220    
1221     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
1222        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
1223        order to be able to cut out the UTF-8 tables in the latter when UTF-8
1224        support is not required. This saves 1.5-2K of code, which is important in
1225        some applications.
1226    
1227        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
1228        so as not to refer to the tables, even though these functions will never be
1229        called when UTF-8 support is disabled. Otherwise there are problems with a
1230        shared library.
1231    
1232    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
1233    
1234        (a) It was defining its arguments as char * instead of void *.
1235    
1236        (b) It was assuming that all moves were upwards in memory; this was true
1237            a long time ago when I wrote it, but is no longer the case.
1238    
1239        The emulated memove() is provided for those environments that have neither
1240        memmove() nor bcopy(). I didn't think anyone used it these days, but that
1241        is clearly not the case, as these two bugs were recently reported.
1242    
1243    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
1244        and Detrail to create the HTML documentation, the .txt form of the man
1245        pages, and it removes trailing spaces from listed files. It also creates
1246        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
1247        case, it wraps all the #defines with #ifndefs. This script should be run
1248        before "make dist".
1249    
1250    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1251        with Unicode property support.
1252    
1253        (a) For a maximizing quantifier, if the two different cases of the
1254            character were of different lengths in their UTF-8 codings (there are
1255            some cases like this - I found 11), and the matching function had to
1256            back up over a mixture of the two cases, it incorrectly assumed they
1257            were both the same length.
1258    
1259        (b) When PCRE was configured to use the heap rather than the stack for
1260            recursion during matching, it was not correctly preserving the data for
1261            the other case of a UTF-8 character when checking ahead for a match
1262            while processing a minimizing repeat. If the check also involved
1263            matching a wide character, but failed, corruption could cause an
1264            erroneous result when trying to check for a repeat of the original
1265            character.
1266    
1267    13. Some tidying changes to the testing mechanism:
1268    
1269        (a) The RunTest script now detects the internal link size and whether there
1270            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1271            values substituted by "configure". (The RunGrepTest script already did
1272            this for UTF-8.) The configure.ac script no longer substitutes the
1273            relevant variables.
1274    
1275        (b) The debugging options /B and /D in pcretest show the compiled bytecode
1276            with length and offset values. This means that the output is different
1277            for different internal link sizes. Test 2 is skipped for link sizes
1278            other than 2 because of this, bypassing the problem. Unfortunately,
1279            there was also a test in test 3 (the locale tests) that used /B and
1280            failed for link sizes other than 2. Rather than cut the whole test out,
1281            I have added a new /Z option to pcretest that replaces the length and
1282            offset values with spaces. This is now used to make test 3 independent
1283            of link size. (Test 2 will be tidied up later.)
1284    
1285    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1286        segmentation fault instead of returning the appropriate error message.
1287    
1288    15. In multiline mode when the newline sequence was set to "any", the pattern
1289        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1290        This doesn't seem right; it now treats the CRLF combination as the line
1291        ending, and so does not match in that case. It's only a pattern such as ^$
1292        that would hit this one: something like ^ABC$ would have failed after \r
1293        and then tried again after \r\n.
1294    
1295    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1296        in an attempt to make files that differ only in their line terminators
1297        compare equal. This works on Linux.
1298    
1299    17. Under certain error circumstances pcregrep might try to free random memory
1300        as it exited. This is now fixed, thanks to valgrind.
1301    
1302    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1303        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1304        was because its rules for how to advance for /g after matching an empty
1305        string at the end of a line did not allow for this case. They now check for
1306        it specially.
1307    
1308    20. pcretest is supposed to handle patterns and data of any length, by
1309        extending its buffers when necessary. It was getting this wrong when the
1310        buffer for a data line had to be extended.
1311    
1312    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1313        CRLF as a newline sequence.
1314    
1315    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1316        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1317        I have nevertheless tidied it up.
1318    
1319    23. Added some casts to kill warnings from HP-UX ia64 compiler.
1320    
1321    24. Added a man page for pcre-config.
1322    
1323    
1324    Version 7.0 19-Dec-06
1325    ---------------------
1326    
1327     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1328        moving to gcc 4.1.1.
1329    
1330     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1331        sys/time.h, which is documented as needed for this function. It doesn't
1332        seem to matter on Linux, but it showed up on some releases of OS X.
1333    
1334     3. It seems that there are systems where bytes whose values are greater than
1335        127 match isprint() in the "C" locale. The "C" locale should be the
1336        default when a C program starts up. In most systems, only ASCII printing
1337        characters match isprint(). This difference caused the output from pcretest
1338        to vary, making some of the tests fail. I have changed pcretest so that:
1339    
1340        (a) When it is outputting text in the compiled version of a pattern, bytes
1341            other than 32-126 are always shown as hex escapes.
1342    
1343        (b) When it is outputting text that is a matched part of a subject string,
1344            it does the same, unless a different locale has been set for the match
1345            (using the /L modifier). In this case, it uses isprint() to decide.
1346    
1347     4. Fixed a major bug that caused incorrect computation of the amount of memory
1348        required for a compiled pattern when options that changed within the
1349        pattern affected the logic of the preliminary scan that determines the
1350        length. The relevant options are -x, and -i in UTF-8 mode. The result was
1351        that the computed length was too small. The symptoms of this bug were
1352        either the PCRE error "internal error: code overflow" from pcre_compile(),
1353        or a glibc crash with a message such as "pcretest: free(): invalid next
1354        size (fast)". Examples of patterns that provoked this bug (shown in
1355        pcretest format) are:
1356    
1357          /(?-x: )/x
1358          /(?x)(?-x: \s*#\s*)/
1359          /((?i)[\x{c0}])/8
1360          /(?i:[\x{c0}])/8
1361    
1362        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1363        is now done differently.
1364    
1365     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1366        wrapper classes; (b) implement a new function in the C++ scanner that is
1367        more efficient than the old way of doing things because it avoids levels of
1368        recursion in the regex matching; (c) add a paragraph to the documentation
1369        for the FullMatch() function.
1370    
1371     6. The escape sequence \n was being treated as whatever was defined as
1372        "newline". Not only was this contrary to the documentation, which states
1373        that \n is character 10 (hex 0A), but it also went horribly wrong when
1374        "newline" was defined as CRLF. This has been fixed.
1375    
1376     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1377        was being set to -1 for the "end of line" case (supposedly a value that no
1378        character can have). Though this value is never used (the check for end of
1379        line is "zero bytes in current character"), it caused compiler complaints.
1380        I've changed it to 0xffffffff.
1381    
1382     8. In pcre_version.c, the version string was being built by a sequence of
1383        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1384        string (as it is for production releases) called a macro with an empty
1385        argument. The C standard says the result of this is undefined. The gcc
1386        compiler treats it as an empty string (which was what was wanted) but it is
1387        reported that Visual C gives an error. The source has been hacked around to
1388        avoid this problem.
1389    
1390     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1391        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1392        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1393        of them did).
1394    
1395    10. Originally, pcretest opened its input and output without "b"; then I was
1396        told that "b" was needed in some environments, so it was added for release
1397        5.0 to both the input and output. (It makes no difference on Unix-like
1398        systems.) Later I was told that it is wrong for the input on Windows. I've
1399        now abstracted the modes into two macros, to make it easier to fiddle with
1400        them, and removed "b" from the input mode under Windows.
1401    
1402    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1403    
1404    12. Added -help and --help to pcretest as an official way of being reminded
1405        of the options.
1406    
1407    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1408        and pcrecpp.cc because they annoy compilers at high warning levels.
1409    
1410    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1411    
1412    15. Fixed an occurrence of == in configure.ac that should have been = (shell
1413        scripts are not C programs :-) and which was not noticed because it works
1414        on Linux.
1415    
1416    16. pcretest is supposed to handle any length of pattern and data line (as one
1417        line or as a continued sequence of lines) by extending its input buffer if
1418        necessary. This feature was broken for very long pattern lines, leading to
1419        a string of junk being passed to pcre_compile() if the pattern was longer
1420        than about 50K.
1421    
1422    17. I have done a major re-factoring of the way pcre_compile() computes the
1423        amount of memory needed for a compiled pattern. Previously, there was code
1424        that made a preliminary scan of the pattern in order to do this. That was
1425        OK when PCRE was new, but as the facilities have expanded, it has become
1426        harder and harder to keep it in step with the real compile phase, and there
1427        have been a number of bugs (see for example, 4 above). I have now found a
1428        cunning way of running the real compile function in a "fake" mode that
1429        enables it to compute how much memory it would need, while actually only
1430        ever using a few hundred bytes of working memory and without too many
1431        tests of the mode. This should make future maintenance and development
1432        easier. A side effect of this work is that the limit of 200 on the nesting
1433        depth of parentheses has been removed (though this was never a serious
1434        limitation, I suspect). However, there is a downside: pcre_compile() now
1435        runs more slowly than before (30% or more, depending on the pattern). I
1436        hope this isn't a big issue. There is no effect on runtime performance.
1437    
1438    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1439        newline (only possible for the last line of a file) and it was a
1440        pattern that set a locale (followed by /Lsomething), pcretest crashed.
1441    
1442    19. Added additional timing features to pcretest. (1) The -tm option now times
1443        matching only, not compiling. (2) Both -t and -tm can be followed, as a
1444        separate command line item, by a number that specifies the number of
1445        repeats to use when timing. The default is 50000; this gives better
1446        precision, but takes uncomfortably long for very large patterns.
1447    
1448    20. Extended pcre_study() to be more clever in cases where a branch of a
1449        subpattern has no definite first character. For example, (a*|b*)[cd] would
1450        previously give no result from pcre_study(). Now it recognizes that the
1451        first character must be a, b, c, or d.
1452    
1453    21. There was an incorrect error "recursive call could loop indefinitely" if
1454        a subpattern (or the entire pattern) that was being tested for matching an
1455        empty string contained only one non-empty item after a nested subpattern.
1456        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1457        incorrectly, because the \d was being skipped in the check.
1458    
1459    22. The pcretest program now has a new pattern option /B and a command line
1460        option -b, which is equivalent to adding /B to every pattern. This causes
1461        it to show the compiled bytecode, without the additional information that
1462        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1463        is the same as /B/I).
1464    
1465    23. A new optimization is now able automatically to treat some sequences such
1466        as a*b as a*+b. More specifically, if something simple (such as a character
1467        or a simple class like \d) has an unlimited quantifier, and is followed by
1468        something that cannot possibly match the quantified thing, the quantifier
1469        is automatically "possessified".
1470    
1471    24. A recursive reference to a subpattern whose number was greater than 39
1472        went wrong under certain circumstances in UTF-8 mode. This bug could also
1473        have affected the operation of pcre_study().
1474    
1475    25. Realized that a little bit of performance could be had by replacing
1476        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1477    
1478    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1479    
1480    27. Possessive quantifiers such as a++ were previously implemented by turning
1481        them into atomic groups such as ($>a+). Now they have their own opcodes,
1482        which improves performance. This includes the automatically created ones
1483        from 23 above.
1484    
1485    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1486        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1487        the first matched character to be a colon. This applied both to named and
1488        numbered groups.
1489    
1490    29. The ucpinternal.h header file was missing its idempotency #ifdef.
1491    
1492    30. I was sent a "project" file called libpcre.a.dev which I understand makes
1493        building PCRE on Windows easier, so I have included it in the distribution.
1494    
1495    31. There is now a check in pcretest against a ridiculously large number being
1496        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1497        loop, the loop is abandoned.
1498    
1499    32. Forward references to subpatterns in conditions such as (?(2)...) where
1500        subpattern 2 is defined later cause pcre_compile() to search forwards in
1501        the pattern for the relevant set of parentheses. This search went wrong
1502        when there were unescaped parentheses in a character class, parentheses
1503        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1504    
1505    33. "Subroutine" calls and backreferences were previously restricted to
1506        referencing subpatterns earlier in the regex. This restriction has now
1507        been removed.
1508    
1509    34. Added a number of extra features that are going to be in Perl 5.10. On the
1510        whole, these are just syntactic alternatives for features that PCRE had
1511        previously implemented using the Python syntax or my own invention. The
1512        other formats are all retained for compatibility.
1513    
1514        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1515            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1516            also .NET compatible.
1517    
1518        (b) A recursion or subroutine call to a named group can now be defined as
1519            (?&name) as well as (?P>name).
1520    
1521        (c) A backreference to a named group can now be defined as \k<name> or
1522            \k'name' as well as (?P=name). The new forms, as well as being in Perl
1523            5.10, are also .NET compatible.
1524    
1525        (d) A conditional reference to a named group can now use the syntax
1526            (?(<name>) or (?('name') as well as (?(name).
1527    
1528        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1529            groups (named and numbered) that are never evaluated inline, but can be
1530            called as "subroutines" from elsewhere. In effect, the DEFINE condition
1531            is always false. There may be only one alternative in such a group.
1532    
1533        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1534            as the simple (?(R). The condition is true only if the most recent
1535            recursion is that of the given number or name. It does not search out
1536            through the entire recursion stack.
1537    
1538        (g) The escape \gN or \g{N} has been added, where N is a positive or
1539            negative number, specifying an absolute or relative reference.
1540    
1541    35. Tidied to get rid of some further signed/unsigned compiler warnings and
1542        some "unreachable code" warnings.
1543    
1544    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1545        things, this adds five new scripts.
1546    
1547    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1548        There were also incompatibilities regarding the handling of \Q..\E inside
1549        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1550        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1551    
1552    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1553        matches an empty string, and forcibly breaks the loop. There were bugs in
1554        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
1555        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
1556        separate and independent bugs (that affected different cases) have been
1557        fixed.
1558    
1559    39. Refactored the code to abolish the use of different opcodes for small
1560        capturing bracket numbers. This is a tidy that I avoided doing when I
1561        removed the limit on the number of capturing brackets for 3.5 back in 2001.
1562        The new approach is not only tidier, it makes it possible to reduce the
1563        memory needed to fix the previous bug (38).
1564    
1565    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1566        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1567        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1568        mode.
1569    
1570    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1571        report.
1572    
1573    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1574        copy construction and assignment in the C++ wrapper.
1575    
1576    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1577        couple of bugs that could have given wrong results in the "--newline=crlf"
1578        case.
1579    
1580    44. Added a number of casts and did some reorganization of signed/unsigned int
1581        variables following suggestions from Dair Grant. Also renamed the variable
1582        "this" as "item" because it is a C++ keyword.
1583    
1584    45. Arranged for dftables to add
1585    
1586          #include "pcre_internal.h"
1587    
1588        to pcre_chartables.c because without it, gcc 4.x may remove the array
1589        definition from the final binary if PCRE is built into a static library and
1590        dead code stripping is activated.
1591    
1592    46. For an unanchored pattern, if a match attempt fails at the start of a
1593        newline sequence, and the newline setting is CRLF or ANY, and the next two
1594        characters are CRLF, advance by two characters instead of one.
1595    
1596    
1597    Version 6.7 04-Jul-06
1598    ---------------------
1599    
1600     1. In order to handle tests when input lines are enormously long, pcretest has
1601        been re-factored so that it automatically extends its buffers when
1602        necessary. The code is crude, but this _is_ just a test program. The
1603        default size has been increased from 32K to 50K.
1604    
1605     2. The code in pcre_study() was using the value of the re argument before
1606        testing it for NULL. (Of course, in any sensible call of the function, it
1607        won't be NULL.)
1608    
1609     3. The memmove() emulation function in pcre_internal.h, which is used on
1610        systems that lack both memmove() and bcopy() - that is, hardly ever -
1611        was missing a "static" storage class specifier.
1612    
1613     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1614        containing an extended class (one that cannot be represented by a bitmap
1615        because it contains high-valued characters or Unicode property items, e.g.
1616        [\pZ]). Almost always one would set UTF-8 mode when processing such a
1617        pattern, but PCRE should not loop if you do not (it no longer does).
1618        [Detail: two cases were found: (a) a repeated subpattern containing an
1619        extended class; (b) a recursive reference to a subpattern that followed a
1620        previous extended class. It wasn't skipping over the extended class
1621        correctly when UTF-8 mode was not set.]
1622    
1623     5. A negated single-character class was not being recognized as fixed-length
1624        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1625        compile error "lookbehind assertion is not fixed length".
1626    
1627     6. The RunPerlTest auxiliary script was showing an unexpected difference
1628        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1629        write a Perl script that can interpret lines of an input file either as
1630        byte characters or as UTF-8, which is what "perltest" was being required to
1631        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1632        can't do is switch easily at run time between having the "use utf8;" pragma
1633        or not. In the end, I fudged it by using the RunPerlTest script to insert
1634        "use utf8;" explicitly for the UTF-8 tests.
1635    
1636     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1637        the end of the subject string, contrary to the documentation and to what
1638        Perl does. This was true of both matching functions. Now it matches only at
1639        the start of the subject and immediately after *internal* newlines.
1640    
1641     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1642        a pointer to an int instead of a pointer to an unsigned long int. This
1643        caused problems on 64-bit systems.
1644    
1645     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1646        instance of the 'standard' template library not being so standard".
1647    
1648    10. There was no check on the number of named subpatterns nor the maximum
1649        length of a subpattern name. The product of these values is used to compute
1650        the size of the memory block for a compiled pattern. By supplying a very
1651        long subpattern name and a large number of named subpatterns, the size
1652        computation could be caused to overflow. This is now prevented by limiting
1653        the length of names to 32 characters, and the number of named subpatterns
1654        to 10,000.
1655    
1656    11. Subpatterns that are repeated with specific counts have to be replicated in
1657        the compiled pattern. The size of memory for this was computed from the
1658        length of the subpattern and the repeat count. The latter is limited to
1659        65535, but there was no limit on the former, meaning that integer overflow
1660        could in principle occur. The compiled length of a repeated subpattern is
1661        now limited to 30,000 bytes in order to prevent this.
1662    
1663    12. Added the optional facility to have named substrings with the same name.
1664    
1665    13. Added the ability to use a named substring as a condition, using the
1666        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1667        are numbers (not recommended). Forward references are permitted.
1668    
1669    14. Added forward references in named backreferences (if you see what I mean).
1670    
1671    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1672        pattern could run off the end of the subject. For example, the pattern
1673        "(?s)(.{1,5})"8 did this with the subject "ab".
1674    
1675    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1676        PCRE_CASELESS was set when matching characters that were quantified with ?
1677        or *.
1678    
1679    17. A character class other than a single negated character that had a minimum
1680        but no maximum quantifier - for example [ab]{6,} - was not handled
1681        correctly by pce_dfa_exec(). It would match only one character.
1682    
1683    18. A valid (though odd) pattern that looked like a POSIX character
1684        class but used an invalid character after [ (for example [[,abc,]]) caused
1685        pcre_compile() to give the error "Failed: internal error: code overflow" or
1686        in some cases to crash with a glibc free() error. This could even happen if
1687        the pattern terminated after [[ but there just happened to be a sequence of
1688        letters, a binary zero, and a closing ] in the memory that followed.
1689    
1690    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1691        over the years. Originally (before any Unicode support), just the bottom 8
1692        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1693        output from "man perlunicode" includes this:
1694    
1695          The regular expression compiler produces polymorphic opcodes.  That
1696          is, the pattern adapts to the data and automatically switches to
1697          the Unicode character scheme when presented with Unicode data--or
1698          instead uses a traditional byte scheme when presented with byte
1699          data.
1700    
1701        Sadly, a wide octal escape does not cause a switch, and in a string with
1702        no other multibyte characters, these octal escapes are treated as before.
1703        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
1704        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1705        Unicode string.
1706    
1707        I have not perpetrated such confusion in PCRE. Up till now, it took just
1708        the bottom 8 bits, as in old Perl. I have now made octal escapes with
1709        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1710        translate to the appropriate multibyte character.
1711    
1712    29. Applied some refactoring to reduce the number of warnings from Microsoft
1713        and Borland compilers. This has included removing the fudge introduced
1714        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1715        a warning about an unused variable.
1716    
1717    21. PCRE has not included VT (character 0x0b) in the set of whitespace
1718        characters since release 4.0, because Perl (from release 5.004) does not.
1719        [Or at least, is documented not to: some releases seem to be in conflict
1720        with the documentation.] However, when a pattern was studied with
1721        pcre_study() and all its branches started with \s, PCRE still included VT
1722        as a possible starting character. Of course, this did no harm; it just
1723        caused an unnecessary match attempt.
1724    
1725    22. Removed a now-redundant internal flag bit that recorded the fact that case
1726        dependency changed within the pattern. This was once needed for "required
1727        byte" processing, but is no longer used. This recovers a now-scarce options
1728        bit. Also moved the least significant internal flag bit to the most-
1729        significant bit of the word, which was not previously used (hangover from
1730        the days when it was an int rather than a uint) to free up another bit for
1731        the future.
1732    
1733    23. Added support for CRLF line endings as well as CR and LF. As well as the
1734        default being selectable at build time, it can now be changed at runtime
1735        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1736        specify that it is scanning data with non-default line endings.
1737    
1738    24. Changed the definition of CXXLINK to make it agree with the definition of
1739        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1740    
1741    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1742        recursions. This makes a big different to stack usage for some patterns.
1743    
1744    26. If a subpattern containing a named recursion or subroutine reference such
1745        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1746        the space required for the compiled pattern went wrong and gave too small a
1747        value. Depending on the environment, this could lead to "Failed: internal
1748        error: code overflow at offset 49" or "glibc detected double free or
1749        corruption" errors.
1750    
1751    27. Applied patches from Google (a) to support the new newline modes and (b) to
1752        advance over multibyte UTF-8 characters in GlobalReplace.
1753    
1754    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1755        difference for some implementation of PCRE in some Windows version.
1756    
1757    29. Added some extra testing facilities to pcretest:
1758    
1759        \q<number>   in a data line sets the "match limit" value
1760        \Q<number>   in a data line sets the "match recursion limt" value
1761        -S <number>  sets the stack size, where <number> is in megabytes
1762    
1763        The -S option isn't available for Windows.
1764    
1765    
1766    Version 6.6 06-Feb-06
1767    ---------------------
1768    
1769     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1770        in pcreposix.h. I have copied the definition from pcre.h.
1771    
1772     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1773        because pcre.h is no longer a built file.
1774    
1775     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1776        not normally included in the compiled code.
1777    
1778    
1779    Version 6.5 01-Feb-06
1780    ---------------------
1781    
1782     1. When using the partial match feature with pcre_dfa_exec(), it was not
1783        anchoring the second and subsequent partial matches at the new starting
1784        point. This could lead to incorrect results. For example, with the pattern
1785        /1234/, partially matching against "123" and then "a4" gave a match.
1786    
1787     2. Changes to pcregrep:
1788    
1789        (a) All non-match returns from pcre_exec() were being treated as failures
1790            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1791            error message is output. Some extra information is given for the
1792            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
1793            probably the only errors that are likely to be caused by users (by
1794            specifying a regex that has nested indefinite repeats, for instance).
1795            If there are more than 20 of these errors, pcregrep is abandoned.
1796    
1797        (b) A binary zero was treated as data while matching, but terminated the
1798            output line if it was written out. This has been fixed: binary zeroes
1799            are now no different to any other data bytes.
1800    
1801        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1802            used to set a locale for matching. The --locale=xxxx long option has
1803            been added (no short equivalent) to specify a locale explicitly on the
1804            pcregrep command, overriding the environment variables.
1805    
1806        (d) When -B was used with -n, some line numbers in the output were one less
1807            than they should have been.
1808    
1809        (e) Added the -o (--only-matching) option.
1810    
1811        (f) If -A or -C was used with -c (count only), some lines of context were
1812            accidentally printed for the final match.
1813    
1814        (g) Added the -H (--with-filename) option.
1815    
1816        (h) The combination of options -rh failed to suppress file names for files
1817            that were found from directory arguments.
1818    
1819        (i) Added the -D (--devices) and -d (--directories) options.
1820    
1821        (j) Added the -F (--fixed-strings) option.
1822    
1823        (k) Allow "-" to be used as a file name for -f as well as for a data file.
1824    
1825        (l) Added the --colo(u)r option.
1826    
1827        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1828            is not present by default.
1829    
1830     3. A nasty bug was discovered in the handling of recursive patterns, that is,
1831        items such as (?R) or (?1), when the recursion could match a number of
1832        alternatives. If it matched one of the alternatives, but subsequently,
1833        outside the recursion, there was a failure, the code tried to back up into
1834        the recursion. However, because of the way PCRE is implemented, this is not
1835        possible, and the result was an incorrect result from the match.
1836    
1837        In order to prevent this happening, the specification of recursion has
1838        been changed so that all such subpatterns are automatically treated as
1839        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1840    
1841     4. I had overlooked the fact that, in some locales, there are characters for
1842        which isalpha() is true but neither isupper() nor islower() are true. In
1843        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1844        and ordfeminine) are like this. This affected the treatment of \w and \W
1845        when they appeared in character classes, but not when they appeared outside
1846        a character class. The bit map for "word" characters is now created
1847        separately from the results of isalnum() instead of just taking it from the
1848        upper, lower, and digit maps. (Plus the underscore character, of course.)
1849    
1850     5. The above bug also affected the handling of POSIX character classes such as
1851        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1852        permanent tables. Instead, the bit maps for such a class were previously
1853        created as the appropriate unions of the upper, lower, and digit bitmaps.
1854        Now they are created by subtraction from the [[:word:]] class, which has
1855        its own bitmap.
1856    
1857     6. The [[:blank:]] character class matches horizontal, but not vertical space.
1858        It is created by subtracting the vertical space characters (\x09, \x0a,
1859        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1860        subtraction was done in the overall bitmap for a character class, meaning
1861        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1862        be recognized. This bug has been fixed.
1863    
1864     7. Patches from the folks at Google:
1865    
1866          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1867          real life, but is still worth protecting against".
1868    
1869          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1870          regular expressions".
1871    
1872          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1873          have it.
1874    
1875          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1876          "configure" and the latter not, in order to fix a problem somebody had
1877          with compiling the Arg class on HP-UX.
1878    
1879          (e) Improve the error-handling of the C++ wrapper a little bit.
1880    
1881          (f) New tests for checking recursion limiting.
1882    
1883     8. The pcre_memmove() function, which is used only if the environment does not
1884        have a standard memmove() function (and is therefore rarely compiled),
1885        contained two bugs: (a) use of int instead of size_t, and (b) it was not
1886        returning a result (though PCRE never actually uses the result).
1887    
1888     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1889        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1890        returned instead of calling malloc() with an overflowing number that would
1891        most likely cause subsequent chaos.
1892    
1893    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1894    
1895    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1896        with this option is matched, the nmatch and pmatch options of regexec() are
1897        ignored.
1898    
1899    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1900        provided in case anyone wants to the the POSIX interface with UTF-8
1901        strings.
1902    
1903    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1904        C++ linking (needed for some HP-UX environments).
1905    
1906    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1907        (unused parameter) and in the pcre_printint() function (omitted "default"
1908        switch label when the default is to do nothing).
1909    
1910    15. Added some code to make it possible, when PCRE is compiled as a C++
1911        library, to replace subject pointers for pcre_exec() with a smart pointer
1912        class, thus making it possible to process discontinuous strings.
1913    
1914    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1915        much the same function. They were added by different people who were trying
1916        to make PCRE easy to compile on non-Unix systems. It has been suggested
1917        that PCRE_EXPORT be abolished now that there is more automatic apparatus
1918        for compiling on Windows systems. I have therefore replaced it with
1919        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1920        defaults to "extern" for C or "extern C" for C++, which works fine on
1921        Unix-like systems. It is now possible to override the value of PCRE_DATA_
1922        SCOPE with something explicit in config.h. In addition:
1923    
1924        (a) pcreposix.h still had just "extern" instead of either of these macros;
1925            I have replaced it with PCRE_DATA_SCOPE.
1926    
1927        (b) Functions such as _pcre_xclass(), which are internal to the library,
1928            but external in the C sense, all had PCRE_EXPORT in their definitions.
1929            This is apparently wrong for the Windows case, so I have removed it.
1930            (It makes no difference on Unix-like systems.)
1931    
1932    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1933        of recursive calls to match(). This is different to MATCH_LIMIT because
1934        that limits the total number of calls to match(), not all of which increase
1935        the depth of recursion. Limiting the recursion depth limits the amount of
1936        stack (or heap if NO_RECURSE is set) that is used. The default can be set
1937        when PCRE is compiled, and changed at run time. A patch from Google adds
1938        this functionality to the C++ interface.
1939    
1940    18. Changes to the handling of Unicode character properties:
1941    
1942        (a) Updated the table to Unicode 4.1.0.
1943    
1944        (b) Recognize characters that are not in the table as "Cn" (undefined).
1945    
1946        (c) I revised the way the table is implemented to a much improved format
1947            which includes recognition of ranges. It now supports the ranges that
1948            are defined in UnicodeData.txt, and it also amalgamates other
1949            characters into ranges. This has reduced the number of entries in the
1950            table from around 16,000 to around 3,000, thus reducing its size
1951            considerably. I realized I did not need to use a tree structure after
1952            all - a binary chop search is just as efficient. Having reduced the
1953            number of entries, I extended their size from 6 bytes to 8 bytes to
1954            allow for more data.
1955    
1956        (d) Added support for Unicode script names via properties such as \p{Han}.
1957    
1958    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1959        matching that character.
1960    
1961    20. When matching a repeated Unicode property with a minimum greater than zero,
1962        (for example \pL{2,}), PCRE could look past the end of the subject if it
1963        reached it while seeking the minimum number of characters. This could
1964        happen only if some of the characters were more than one byte long, because
1965        there is a check for at least the minimum number of bytes.
1966    
1967    21. Refactored the implementation of \p and \P so as to be more general, to
1968        allow for more different types of property in future. This has changed the
1969        compiled form incompatibly. Anybody with saved compiled patterns that use
1970        \p or \P will have to recompile them.
1971    
1972    22. Added "Any" and "L&" to the supported property types.
1973    
1974    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1975        but give a compile time error if the value is greater than 0xff.
1976    
1977    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1978        accidentally not being installed or uninstalled.
1979    
1980    25. The pcre.h file was built from pcre.h.in, but the only changes that were
1981        made were to insert the current release number. This seemed silly, because
1982        it made things harder for people building PCRE on systems that don't run
1983        "configure". I have turned pcre.h into a distributed file, no longer built
1984        by "configure", with the version identification directly included. There is
1985        no longer a pcre.h.in file.
1986    
1987        However, this change necessitated a change to the pcre-config script as
1988        well. It is built from pcre-config.in, and one of the substitutions was the
1989        release number. I have updated configure.ac so that ./configure now finds
1990        the release number by grepping pcre.h.
1991    
1992    26. Added the ability to run the tests under valgrind.
1993    
1994    
1995    Version 6.4 05-Sep-05
1996    ---------------------
1997    
1998     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1999        "--" to be printed when multiple files were scanned, even when none of the
2000        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
2001        consider it to be a bug, and have restored the previous behaviour.
2002    
2003     2. A couple of code tidies to get rid of compiler warnings.
2004    
2005     3. The pcretest program used to cheat by referring to symbols in the library
2006        whose names begin with _pcre_. These are internal symbols that are not
2007        really supposed to be visible externally, and in some environments it is
2008        possible to suppress them. The cheating is now confined to including
2009        certain files from the library's source, which is a bit cleaner.
2010    
2011     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
2012        file's purpose clearer.
2013    
2014     5. Reorganized pcre_ucp_findchar().
2015    
2016    
2017    Version 6.3 15-Aug-05
2018    ---------------------
2019    
2020     1. The file libpcre.pc.in did not have general read permission in the tarball.
2021    
2022     2. There were some problems when building without C++ support:
2023    
2024        (a) If C++ support was not built, "make install" and "make test" still
2025            tried to test it.
2026    
2027        (b) There were problems when the value of CXX was explicitly set. Some
2028            changes have been made to try to fix these, and ...
2029    
2030        (c) --disable-cpp can now be used to explicitly disable C++ support.
2031    
2032        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
2033            backslash in a target when C++ was disabled. This confuses some
2034            versions of "make", apparently. Using an intermediate variable solves
2035            this. (Same for CPP_LOBJ.)
2036    
2037     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
2038        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
2039        necessary on certain architectures.
2040    
2041     4. Added a setting of -export-symbols-regex to the link command to remove
2042        those symbols that are exported in the C sense, but actually are local
2043        within the library, and not documented. Their names all begin with
2044        "_pcre_". This is not a perfect job, because (a) we have to except some
2045        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
2046        available (and never for static libraries). I have made a note to try to
2047        find a way round (a) in the future.
2048    
2049    
2050  Version 6.2 01-Aug-05  Version 6.2 01-Aug-05
2051  ---------------------  ---------------------
2052    

Legend:
Removed from v.81  
changed lines
  Added in v.510

  ViewVC Help
Powered by ViewVC 1.1.5