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

Legend:
Removed from v.81  
changed lines
  Added in v.476

  ViewVC Help
Powered by ViewVC 1.1.5