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

Diff of /code/trunk/ChangeLog

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

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24        was out-of-date, and there was no check on the pcre_compile() error code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

Legend:
Removed from v.53  
changed lines
  Added in v.472

  ViewVC Help
Powered by ViewVC 1.1.5