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

Diff of /code/trunk/ChangeLog

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

revision 25 by nigel, Sat Feb 24 21:38:45 2007 UTC revision 438 by ph10, Sun Sep 6 20:00:47 2009 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    Version 8.00 ??-???-??
5    ----------------------
6    
7    1.  The table for translating pcre_compile() error codes into POSIX error codes
8        was out-of-date, and there was no check on the pcre_compile() error code
9        being within the table. This could lead to an OK return being given in
10        error.
11    
12    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
13        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
14        in a Windows environment.
15    
16    3.  The pcregrep --count option prints the count for each file even when it is
17        zero, as does GNU grep. However, pcregrep was also printing all files when
18        --files-with-matches was added. Now, when both options are given, it prints
19        counts only for those files that have at least one match. (GNU grep just
20        prints the file name in this circumstance, but including the count seems
21        more useful - otherwise, why use --count?) Also ensured that the
22        combination -clh just lists non-zero counts, with no names.
23    
24    4.  The long form of the pcregrep -F option was incorrectly implemented as
25        --fixed_strings instead of --fixed-strings. This is an incompatible change,
26        but it seems right to fix it, and I didn't think it was worth preserving
27        the old behaviour.
28    
29    5.  The command line items --regex=pattern and --regexp=pattern were not
30        recognized by pcregrep, which required --regex pattern or --regexp pattern
31        (with a space rather than an '='). The man page documented the '=' forms,
32        which are compatible with GNU grep; these now work.
33    
34    6.  No libpcreposix.pc file was created for pkg-config; there was just
35        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
36    
37    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
38        when UCP support is not needed, by modifying the Python script that
39        generates it from Unicode data files. This should not matter if the module
40        is correctly used as a library, but I received one complaint about 50K of
41        unwanted data. My guess is that the person linked everything into his
42        program rather than using a library. Anyway, it does no harm.
43    
44    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
45        was a minimum greater than 1 for a wide character in a possessive
46        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
47        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
48        character. Chaos in the form of incorrect output or a compiling loop could
49        result.
50    
51    9.  The restrictions on what a pattern can contain when partial matching is
52        requested for pcre_exec() have been removed. All patterns can now be
53        partially matched by this function. In addition, if there are at least two
54        slots in the offset vector, the offset of the earliest inspected character
55        for the match and the offset of the end of the subject are set in them when
56        PCRE_ERROR_PARTIAL is returned.
57    
58    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
59        synonymous with PCRE_PARTIAL, for backwards compatibility, and
60        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
61        and may be more useful for multi-segment matching, especially with
62        pcre_exec().
63    
64    11. Partial matching with pcre_exec() is now more intuitive. A partial match
65        used to be given if ever the end of the subject was reached; now it is
66        given only if matching could not proceed because another character was
67        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
68        string "Z", which now yields "no match" instead of "partial match". In the
69        case of pcre_dfa_exec(), "no match" is given if every matching path for the
70        final character ended with (*FAIL).
71    
72    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
73        if the pattern had a "must contain" character that was already found in the
74        earlier partial match, unless partial matching was again requested. For
75        example, with the pattern /dog.(body)?/, the "must contain" character is
76        "g". If the first part-match was for the string "dog", restarting with
77        "sbody" failed. This bug has been fixed.
78    
79    13. The string returned by pcre_dfa_exec() after a partial match has been
80        changed so that it starts at the first inspected character rather than the
81        first character of the match. This makes a difference only if the pattern
82        starts with a lookbehind assertion or \b or \B (\K is not supported by
83        pcre_dfa_exec()). It's an incompatible change, but it makes the two
84        matching functions compatible, and I think it's the right thing to do.
85    
86    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
87        so that the demonstration program is easily available in environments where
88        PCRE has not been installed from source.
89    
90    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
91        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
92        library.
93    
94    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
95        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
96        is not the first non-POSIX option to be added. Clearly some people find
97        these options useful.
98    
99    17. If a caller to the POSIX matching function regexec() passes a non-zero
100        value for nmatch with a NULL value for pmatch, the value of
101        nmatch is forced to zero.
102    
103    18. RunGrepTest did not have a test for the availability of the -u option of
104        the diff command, as RunTest does. It now checks in the same way as
105        RunTest, and also checks for the -b option.
106    
107    19. If an odd number of negated classes containing just a single character
108        interposed, within parentheses, between a forward reference to a named
109        subpattern and the definition of the subpattern, compilation crashed with
110        an internal error, complaining that it could not find the referenced
111        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
112        [The bug was that it was starting one character too far in when skipping
113        over the character class, thus treating the ] as data rather than
114        terminating the class. This meant it could skip too much.]
115    
116    
117    Version 7.9 11-Apr-09
118    ---------------------
119    
120    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
121        (pcretest), all targets were linked against these libraries. This included
122        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
123        libraries. This caused unwanted dependencies to be created. This problem
124        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
125        pcretest is linked with readline.
126    
127    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
128        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
129        moved outside it again, because FALSE and TRUE are already defined in AIX,
130        but BOOL is not.
131    
132    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
133        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
134    
135    4.  The pcregrep documentation said spaces were inserted as well as colons (or
136        hyphens) following file names and line numbers when outputting matching
137        lines. This is not true; no spaces are inserted. I have also clarified the
138        wording for the --colour (or --color) option.
139    
140    5.  In pcregrep, when --colour was used with -o, the list of matching strings
141        was not coloured; this is different to GNU grep, so I have changed it to be
142        the same.
143    
144    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
145        each matching line was coloured. Now it goes on to look for further matches
146        of any of the test patterns, which is the same behaviour as GNU grep.
147    
148    7.  A pattern that could match an empty string could cause pcregrep to loop; it
149        doesn't make sense to accept an empty string match in pcregrep, so I have
150        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
151        seems to be how GNU grep behaves.
152    
153    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
154        start or after a newline", because the conditional assertion was not being
155        correctly handled. The rule now is that both the assertion and what follows
156        in the first alternative must satisfy the test.
157    
158    9.  If auto-callout was enabled in a pattern with a conditional group whose
159        condition was an assertion, PCRE could crash during matching, both with
160        pcre_exec() and pcre_dfa_exec().
161    
162    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
163        used for matching.
164    
165    11. Unicode property support in character classes was not working for
166        characters (bytes) greater than 127 when not in UTF-8 mode.
167    
168    12. Added the -M command line option to pcretest.
169    
170    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
171    
172    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
173    
174    16. Added comments and documentation about mis-use of no_arg in the C++
175        wrapper.
176    
177    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
178        from Martin Jerabek that uses macro names for all relevant character and
179        string constants.
180    
181    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
182        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
183        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
184        these, but not everybody uses configure.
185    
186    19. A conditional group that had only one branch was not being correctly
187        recognized as an item that could match an empty string. This meant that an
188        enclosing group might also not be so recognized, causing infinite looping
189        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
190        with the subject "ab", where knowledge that the repeated group can match
191        nothing is needed in order to break the loop.
192    
193    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
194        exec(), but without supplying a callout function, matching went wrong.
195    
196    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
197        leak if the size of the offset vector was greater than 30. When the vector
198        is smaller, the saved offsets during recursion go onto a local stack
199        vector, but for larger vectors malloc() is used. It was failing to free
200        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
201        error, in fact).
202    
203    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
204        heapframe that is used only when UTF-8 support is enabled. This caused no
205        problem, but was untidy.
206    
207    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
208        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
209        included within another project.
210    
211    24. Steven Van Ingelgem's patches to add more options to the CMake support,
212        slightly modified by me:
213    
214          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
215              not building pcregrep.
216    
217          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
218              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
219    
220    25. Forward references, both numeric and by name, in patterns that made use of
221        duplicate group numbers, could behave incorrectly or give incorrect errors,
222        because when scanning forward to find the reference group, PCRE was not
223        taking into account the duplicate group numbers. A pattern such as
224        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
225    
226    26. Changed a few more instances of "const unsigned char *" to USPTR, making
227        the feature of a custom pointer more persuasive (as requested by a user).
228    
229    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
230        pcretest.c, inside #ifndefs, because it seems they are sometimes already
231        pre-defined.
232    
233    28. Added support for (*UTF8) at the start of a pattern.
234    
235    29. Arrange for flags added by the "release type" setting in CMake to be shown
236        in the configuration summary.
237    
238    
239    Version 7.8 05-Sep-08
240    ---------------------
241    
242    1.  Replaced UCP searching code with optimized version as implemented for Ad
243        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
244        stage table and inline lookup instead of a function, giving speed ups of 2
245        to 5 times on some simple patterns that I tested. Permission was given to
246        distribute the MultiStage2.py script that generates the tables (it's not in
247        the tarball, but is in the Subversion repository).
248    
249    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
250        scripts.
251    
252    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
253        a group with a zero qualifier. The result of the study could be incorrect,
254        or the function might crash, depending on the pattern.
255    
256    4.  Caseless matching was not working for non-ASCII characters in back
257        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
258        It now works when Unicode Property Support is available.
259    
260    5.  In pcretest, an escape such as \x{de} in the data was always generating
261        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
262        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
263        truncation.
264    
265    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
266    
267    7.  Added two (int) casts to pcregrep when printing the difference of two
268        pointers, in case they are 64-bit values.
269    
270    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
271        test 2 if it fails.
272    
273    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
274        and a #define of that name to empty if it is not externally set. This is to
275        allow users of MSVC to set it if necessary.
276    
277    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
278        the convenience functions in the pcre_get.c source file.
279    
280    11. An option change at the start of a pattern that had top-level alternatives
281        could cause overwriting and/or a crash. This command provoked a crash in
282        some environments:
283    
284          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
285    
286        This potential security problem was recorded as CVE-2008-2371.
287    
288    12. For a pattern where the match had to start at the beginning or immediately
289        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
290        pcre_dfa_exec() could read past the end of the passed subject if there was
291        no match. To help with detecting such bugs (e.g. with valgrind), I modified
292        pcretest so that it places the subject at the end of its malloc-ed buffer.
293    
294    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
295        exec() might read past the end of the data buffer in UTF-8 mode.
296    
297    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
298        the data contained the byte 0x85 as part of a UTF-8 character within its
299        first line. This applied both to normal and DFA matching.
300    
301    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
302        /^[^d]*?$/8 failed to match "abc".
303    
304    16. Added a missing copyright notice to pcrecpp_internal.h.
305    
306    17. Make it more clear in the documentation that values returned from
307        pcre_exec() in ovector are byte offsets, not character counts.
308    
309    18. Tidied a few places to stop certain compilers from issuing warnings.
310    
311    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
312        supplied by Stefan Weber. I made a further small update for 7.8 because
313        there is a change of source arrangements: the pcre_searchfuncs.c module is
314        replaced by pcre_ucd.c.
315    
316    
317    Version 7.7 07-May-08
318    ---------------------
319    
320    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
321        a string to a long long, pretend we don't even have a long long." This is
322        done by checking for the strtoq, strtoll, and _strtoi64 functions.
323    
324    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
325        pre-7.6 versions, which defined a global no_arg variable instead of putting
326        it in the RE class. (See also #8 below.)
327    
328    3.  Remove a line of dead code, identified by coverity and reported by Nuno
329        Lopes.
330    
331    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
332    
333        (1) The include/exclude patterns were being applied to the whole pathnames
334            of files, instead of just to the final components.
335    
336        (2) If there was more than one level of directory, the subdirectories were
337            skipped unless they satisfied the include/exclude conditions. This is
338            inconsistent with GNU grep (and could even be seen as contrary to the
339            pcregrep specification - which I improved to make it absolutely clear).
340            The action now is always to scan all levels of directory, and just
341            apply the include/exclude patterns to regular files.
342    
343    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
344        --exclude_dir in the tests to avoid scanning .svn directories.
345    
346    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
347        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
348        doesn't support NULs in patterns.
349    
350    7.  Added some missing "const"s to declarations of static tables in
351        pcre_compile.c and pcre_dfa_exec.c.
352    
353    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
354        caused by fix #2  above. (Subsequently also a second patch to fix the
355        first patch. And a third patch - this was a messy problem.)
356    
357    9.  Applied Craig's patch to remove the use of push_back().
358    
359    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
360        matching function regexec().
361    
362    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
363        which, however, unlike Perl's \g{...}, are subroutine calls, not back
364        references. PCRE supports relative numbers with this syntax (I don't think
365        Oniguruma does).
366    
367    12. Previously, a group with a zero repeat such as (...){0} was completely
368        omitted from the compiled regex. However, this means that if the group
369        was called as a subroutine from elsewhere in the pattern, things went wrong
370        (an internal error was given). Such groups are now left in the compiled
371        pattern, with a new opcode that causes them to be skipped at execution
372        time.
373    
374    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
375        to the way PCRE behaves:
376    
377        (a) A lone ] character is dis-allowed (Perl treats it as data).
378    
379        (b) A back reference to an unmatched subpattern matches an empty string
380            (Perl fails the current match path).
381    
382        (c) A data ] in a character class must be notated as \] because if the
383            first data character in a class is ], it defines an empty class. (In
384            Perl it is not possible to have an empty class.) The empty class []
385            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
386            The negative empty class [^] matches any one character, independently
387            of the DOTALL setting.
388    
389    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
390        non-existent subpattern following a character class starting with ']' and
391        containing () gave an internal compiling error instead of "reference to
392        non-existent subpattern". Fortunately, when the pattern did exist, the
393        compiled code was correct. (When scanning forwards to check for the
394        existencd of the subpattern, it was treating the data ']' as terminating
395        the class, so got the count wrong. When actually compiling, the reference
396        was subsequently set up correctly.)
397    
398    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
399        it was being rejected as not supported by pcre_dfa_exec(), even though
400        other assertions are supported. I have made pcre_dfa_exec() support
401        (*FAIL).
402    
403    16. The implementation of 13c above involved the invention of a new opcode,
404        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
405        cannot be changed at match time, I realized I could make a small
406        improvement to matching performance by compiling OP_ALLANY instead of
407        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
408        on the OP_ANY path.
409    
410    17. Compiling pcretest on Windows with readline support failed without the
411        following two fixes: (1) Make the unistd.h include conditional on
412        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
413    
414    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
415        ncurses library to be included for pcretest when ReadLine support is
416        requested, but also to allow for it to be overridden. This patch came from
417        Daniel Bergström.
418    
419    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
420        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
421        any errors with the current Unicode tables. Thanks to Peter Kankowski for
422        spotting this.
423    
424    
425    Version 7.6 28-Jan-08
426    ---------------------
427    
428    1.  A character class containing a very large number of characters with
429        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
430        overflow.
431    
432    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
433        HAVE_LONG_LONG is not defined.
434    
435    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
436        bring them up to date and include new features. This patch includes:
437    
438        - Fixed PH's badly added libz and libbz2 support.
439        - Fixed a problem with static linking.
440        - Added pcredemo. [But later removed - see 7 below.]
441        - Fixed dftables problem and added an option.
442        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
443            HAVE_LONG_LONG.
444        - Added readline support for pcretest.
445        - Added an listing of the option settings after cmake has run.
446    
447    4.  A user submitted a patch to Makefile that makes it easy to create
448        "pcre.dll" under mingw when using Configure/Make. I added stuff to
449        Makefile.am that cause it to include this special target, without
450        affecting anything else. Note that the same mingw target plus all
451        the other distribution libraries and programs are now supported
452        when configuring with CMake (see 6 below) instead of with
453        Configure/Make.
454    
455    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
456        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
457        exported in the Windows port". It has not yet been confirmed that the patch
458        solves the problem, but it does no harm.
459    
460    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
461        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
462        with CMake, and also correct the comment about stack recursion.
463    
464    7.  Remove the automatic building of pcredemo from the ./configure system and
465        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
466        of a program that users should build themselves after PCRE is installed, so
467        building it automatically is not really right. What is more, it gave
468        trouble in some build environments.
469    
470    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
471    
472    
473    Version 7.5 10-Jan-08
474    ---------------------
475    
476    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
477        values in parens when parsing an RE using the C++ wrapper."
478    
479    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
480        Characters greater than 255 were excluded from the class instead of being
481        included.
482    
483    3.  The same bug as (2) above applied to negated POSIX classes such as
484        [:^space:].
485    
486    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
487        defined or documented. It seems to have been a typo for PCRE_STATIC, so
488        I have changed it.
489    
490    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
491        first named subpattern) and a construct such as (?&a) would reference the
492        first named subpattern whose name started with "a" (in other words, the
493        length check was missing). Both these problems are fixed. "Subpattern name
494        expected" is now given for (?&) (a zero-length name), and this patch also
495        makes it give the same error for \k'' (previously it complained that that
496        was a reference to a non-existent subpattern).
497    
498    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
499        this is right because (?- can be followed by option settings as well as by
500        digits. I have, however, made the messages clearer.
501    
502    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
503        than the number used in the conditional) now cause a compile-time error.
504        This is actually not compatible with Perl, which accepts such patterns, but
505        treats the conditional as always being FALSE (as PCRE used to), but it
506        seems to me that giving a diagnostic is better.
507    
508    8.  Change "alphameric" to the more common word "alphanumeric" in comments
509        and messages.
510    
511    9.  Fix two occurrences of "backslash" in comments that should have been
512        "backspace".
513    
514    10. Remove two redundant lines of code that can never be obeyed (their function
515        was moved elsewhere).
516    
517    11. The program that makes PCRE's Unicode character property table had a bug
518        which caused it to generate incorrect table entries for sequences of
519        characters that have the same character type, but are in different scripts.
520        It amalgamated them into a single range, with the script of the first of
521        them. In other words, some characters were in the wrong script. There were
522        thirteen such cases, affecting characters in the following ranges:
523    
524          U+002b0 - U+002c1
525          U+0060c - U+0060d
526          U+0061e - U+00612
527          U+0064b - U+0065e
528          U+0074d - U+0076d
529          U+01800 - U+01805
530          U+01d00 - U+01d77
531          U+01d9b - U+01dbf
532          U+0200b - U+0200f
533          U+030fc - U+030fe
534          U+03260 - U+0327f
535          U+0fb46 - U+0fbb1
536          U+10450 - U+1049d
537    
538    12. The -o option (show only the matching part of a line) for pcregrep was not
539        compatible with GNU grep in that, if there was more than one match in a
540        line, it showed only the first of them. It now behaves in the same way as
541        GNU grep.
542    
543    13. If the -o and -v options were combined for pcregrep, it printed a blank
544        line for every non-matching line. GNU grep prints nothing, and pcregrep now
545        does the same. The return code can be used to tell if there were any
546        non-matching lines.
547    
548    14. Added --file-offsets and --line-offsets to pcregrep.
549    
550    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
551        infinitely looping recursion. The bug was that positive lookaheads were not
552        being skipped when checking for a possible empty match (negative lookaheads
553        and both kinds of lookbehind were skipped).
554    
555    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
556        inclusion of <windows.h> to before rather than after the definition of
557        INVALID_FILE_ATTRIBUTES (patch from David Byron).
558    
559    17. Specifying a possessive quantifier with a specific limit for a Unicode
560        character property caused pcre_compile() to compile bad code, which led at
561        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
562        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
563        caused the error; without that there was no problem.
564    
565    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
566    
567    19. Added --enable-pcretest-libreadline.
568    
569    20. In pcrecpp.cc, the variable 'count' was incremented twice in
570        RE::GlobalReplace(). As a result, the number of replacements returned was
571        double what it should be. I removed one of the increments, but Craig sent a
572        later patch that removed the other one (the right fix) and added unit tests
573        that check the return values (which was not done before).
574    
575    21. Several CMake things:
576    
577        (1) Arranged that, when cmake is used on Unix, the libraries end up with
578            the names libpcre and libpcreposix, not just pcre and pcreposix.
579    
580        (2) The above change means that pcretest and pcregrep are now correctly
581            linked with the newly-built libraries, not previously installed ones.
582    
583        (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
584    
585    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
586        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
587        UTF-8 newline character). The key issue is that the pattern starts .*;
588        this means that the match must be either at the beginning, or after a
589        newline. The bug was in the code for advancing after a failed match and
590        checking that the new position followed a newline. It was not taking
591        account of UTF-8 characters correctly.
592    
593    23. PCRE was behaving differently from Perl in the way it recognized POSIX
594        character classes. PCRE was not treating the sequence [:...:] as a
595        character class unless the ... were all letters. Perl, however, seems to
596        allow any characters between [: and :], though of course it rejects as
597        unknown any "names" that contain non-letters, because all the known class
598        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
599        for example, whereas PCRE did not - it did not recognize a POSIX character
600        class. This seemed a bit dangerous, so the code has been changed to be
601        closer to Perl. The behaviour is not identical to Perl, because PCRE will
602        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
603        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
604        Perl does, and where it didn't before.
605    
606    24. Rewrite so as to remove the single use of %n from pcregrep because in some
607        Windows environments %n is disabled by default.
608    
609    
610    Version 7.4 21-Sep-07
611    ---------------------
612    
613    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
614        means that a class such as [\s] counted as "explicit reference to CR or
615        LF". That isn't really right - the whole point of the change was to try to
616        help when there was an actual mention of one of the two characters. So now
617        the change happens only if \r or \n (or a literal CR or LF) character is
618        encountered.
619    
620    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
621        of both had grown to the point where there were only 3 bits left.
622        Fortunately, there was spare space in the data structure, and so I have
623        moved the internal flags into a new 16-bit field to free up more option
624        bits.
625    
626    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
627        but did not set the internal JCHANGED flag - either of these is enough to
628        control the way the "get" function works - but the PCRE_INFO_JCHANGED
629        facility is supposed to tell if (?J) was ever used, so now (?J) at the
630        start sets both bits.
631    
632    4.  Added options (at build time, compile time, exec time) to change \R from
633        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
634    
635    5.  doc/pcresyntax.html was missing from the distribution.
636    
637    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
638        compatibility, even though it is no longer used.
639    
640    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
641        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
642        windows.h file is present (where different names are used). [This was
643        reversed later after testing - see 16 below.]
644    
645    8.  Changed all #include <config.h> to #include "config.h". There were also
646        some further <pcre.h> cases that I changed to "pcre.h".
647    
648    9.  When pcregrep was used with the --colour option, it missed the line ending
649        sequence off the lines that it output.
650    
651    10. It was pointed out to me that arrays of string pointers cause lots of
652        relocations when a shared library is dynamically loaded. A technique of
653        using a single long string with a table of offsets can drastically reduce
654        these. I have refactored PCRE in four places to do this. The result is
655        dramatic:
656    
657          Originally:                          290
658          After changing UCP table:            187
659          After changing error message table:   43
660          After changing table of "verbs"       36
661          After changing table of Posix names   22
662    
663        Thanks to the folks working on Gregex for glib for this insight.
664    
665    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
666        unicode-properties was also set.
667    
668    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
669    
670    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
671        checked only for CRLF.
672    
673    14. Added casts to pcretest.c to avoid compiler warnings.
674    
675    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
676    
677    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
678        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
679        entirely. This removes changes made in 7 above.
680    
681    17. The CMake files have been updated, and there is now more information about
682        building with CMake in the NON-UNIX-USE document.
683    
684    
685    Version 7.3 28-Aug-07
686    ---------------------
687    
688     1. In the rejigging of the build system that eventually resulted in 7.1, the
689        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
690        brackets there is not right, since it causes compilers to look for an
691        installed pcre.h, not the version that is in the source that is being
692        compiled (which of course may be different). I have changed it back to:
693    
694          #include "pcre.h"
695    
696        I have a vague recollection that the change was concerned with compiling in
697        different directories, but in the new build system, that is taken care of
698        by the VPATH setting the Makefile.
699    
700     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
701        when the subject happened to end in the byte 0x85 (e.g. if the last
702        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
703        characters but of course it shouldn't be taken as a newline when it is part
704        of another character. The bug was that, for an unlimited repeat of . in
705        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
706        characters when looking for a newline.
707    
708     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
709    
710     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
711        in debug output.
712    
713     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
714        long printing in the pcrecpp unittest when running under MinGW.
715    
716     6. ESC_K was left out of the EBCDIC table.
717    
718     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
719        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
720        limit also applies to "virtual nesting" when a pattern is recursive, and in
721        this case 1000 isn't so big. I have been able to remove this limit at the
722        expense of backing off one optimization in certain circumstances. Normally,
723        when pcre_exec() would call its internal match() function recursively and
724        immediately return the result unconditionally, it uses a "tail recursion"
725        feature to save stack. However, when a subpattern that can match an empty
726        string has an unlimited repetition quantifier, it no longer makes this
727        optimization. That gives it a stack frame in which to save the data for
728        checking that an empty string has been matched. Previously this was taken
729        from the 1000-entry workspace that had been reserved. So now there is no
730        explicit limit, but more stack is used.
731    
732     8. Applied Daniel's patches to solve problems with the import/export magic
733        syntax that is required for Windows, and which was going wrong for the
734        pcreposix and pcrecpp parts of the library. These were overlooked when this
735        problem was solved for the main library.
736    
737     9. There were some crude static tests to avoid integer overflow when computing
738        the size of patterns that contain repeated groups with explicit upper
739        limits. As the maximum quantifier is 65535, the maximum group length was
740        set at 30,000 so that the product of these two numbers did not overflow a
741        32-bit integer. However, it turns out that people want to use groups that
742        are longer than 30,000 bytes (though not repeat them that many times).
743        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
744        made it possible to implement the integer overflow checks in a much more
745        dynamic way, which I have now done. The artificial limitation on group
746        length has been removed - we now have only the limit on the total length of
747        the compiled pattern, which depends on the LINK_SIZE setting.
748    
749    10. Fixed a bug in the documentation for get/copy named substring when
750        duplicate names are permitted. If none of the named substrings are set, the
751        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
752        empty string.
753    
754    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
755        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
756        because the ] is interpreted as the first data character and the
757        terminating ] is not found. PCRE has been made compatible with Perl in this
758        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
759        cause memory overwriting.
760    
761    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
762        string has been matched (to stop an infinite loop). It was not recognizing
763        a conditional subpattern that could match an empty string if that
764        subpattern was within another subpattern. For example, it looped when
765        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
766        condition was not nested. This bug has been fixed.
767    
768    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
769        past the start of the subject in the presence of bytes with the top bit
770        set, for example "\x8aBCD".
771    
772    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
773        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
774    
775    14. Optimized (?!) to (*FAIL).
776    
777    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
778        This restricts code points to be within the range 0 to 0x10FFFF, excluding
779        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
780        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
781        does: it's just the validity check that is more restrictive.
782    
783    16. Inserted checks for integer overflows during escape sequence (backslash)
784        processing, and also fixed erroneous offset values for syntax errors during
785        backslash processing.
786    
787    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
788        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
789    
790    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
791        caused an overrun.
792    
793    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
794        something other than just ASCII characters) inside a group that had an
795        unlimited repeat caused a loop at compile time (while checking to see
796        whether the group could match an empty string).
797    
798    20. Debugging a pattern containing \p or \P could cause a crash. For example,
799        [\P{Any}] did so. (Error in the code for printing property names.)
800    
801    21. An orphan \E inside a character class could cause a crash.
802    
803    22. A repeated capturing bracket such as (A)? could cause a wild memory
804        reference during compilation.
805    
806    23. There are several functions in pcre_compile() that scan along a compiled
807        expression for various reasons (e.g. to see if it's fixed length for look
808        behind). There were bugs in these functions when a repeated \p or \P was
809        present in the pattern. These operators have additional parameters compared
810        with \d, etc, and these were not being taken into account when moving along
811        the compiled data. Specifically:
812    
813        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
814            length.
815    
816        (b) An item such as \pL+ within a repeated group could cause crashes or
817            loops.
818    
819        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
820            "reference to non-existent subpattern" error.
821    
822        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
823    
824    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
825        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
826    
827    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
828    
829    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
830        character were causing crashes (broken optimization).
831    
832    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
833        \p or \P) caused a compile-time loop.
834    
835    28. More problems have arisen in unanchored patterns when CRLF is a valid line
836        break. For example, the unstudied pattern [\r\n]A does not match the string
837        "\r\nA" because change 7.0/46 below moves the current point on by two
838        characters after failing to match at the start. However, the pattern \nA
839        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
840        the same is true. There doesn't seem any very clean way out of this, but
841        what I have chosen to do makes the common cases work: PCRE now takes note
842        of whether there can be an explicit match for \r or \n anywhere in the
843        pattern, and if so, 7.0/46 no longer applies. As part of this change,
844        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
845        pattern has explicit CR or LF references.
846    
847    29. Added (*CR) etc for changing newline setting at start of pattern.
848    
849    
850    Version 7.2 19-Jun-07
851    ---------------------
852    
853     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
854        which is apparently normally available under Windows.
855    
856     2. Re-jig the pcregrep tests with different newline settings in an attempt
857        to make them independent of the local environment's newline setting.
858    
859     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
860    
861     4. Some of the "internals" tests were previously cut out when the link size
862        was not 2, because the output contained actual offsets. The recent new
863        "Z" feature of pcretest means that these can be cut out, making the tests
864        usable with all link sizes.
865    
866     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
867        stack recursion. This gives a massive performance boost under BSD, but just
868        a small improvement under Linux. However, it saves one field in the frame
869        in all cases.
870    
871     6. Added more features from the forthcoming Perl 5.10:
872    
873        (a) (?-n) (where n is a string of digits) is a relative subroutine or
874            recursion call. It refers to the nth most recently opened parentheses.
875    
876        (b) (?+n) is also a relative subroutine call; it refers to the nth next
877            to be opened parentheses.
878    
879        (c) Conditions that refer to capturing parentheses can be specified
880            relatively, for example, (?(-2)... or (?(+3)...
881    
882        (d) \K resets the start of the current match so that everything before
883            is not part of it.
884    
885        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
886    
887        (f) \g{name} is another synonym - part of Perl 5.10's unification of
888            reference syntax.
889    
890        (g) (?| introduces a group in which the numbering of parentheses in each
891            alternative starts with the same number.
892    
893        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
894    
895     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
896        PCRE_INFO_JCHANGED.
897    
898     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
899        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
900        for detecting groups that can match an empty string.
901    
902     9. A pattern with a very large number of alternatives (more than several
903        hundred) was running out of internal workspace during the pre-compile
904        phase, where pcre_compile() figures out how much memory will be needed. A
905        bit of new cunning has reduced the workspace needed for groups with
906        alternatives. The 1000-alternative test pattern now uses 12 bytes of
907        workspace instead of running out of the 4096 that are available.
908    
909    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
910    
911    11. Applied patch from Google to remove an optimization that didn't quite work.
912        The report of the bug said:
913    
914          pcrecpp::RE("a*").FullMatch("aaa") matches, while
915          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
916          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
917    
918    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
919        it matched the wrong number of bytes.
920    
921    
922    Version 7.1 24-Apr-07
923    ---------------------
924    
925     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
926        that is more "standard", making use of automake and other Autotools. There
927        is some re-arrangement of the files and adjustment of comments consequent
928        on this.
929    
930     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
931        for recursive directory scanning broke on some systems because the files
932        are not scanned in any specific order and on different systems the order
933        was different. A call to "sort" has been inserted into RunGrepTest for the
934        approprate test as a short-term fix. In the longer term there may be an
935        alternative.
936    
937     3. I had an email from Eric Raymond about problems translating some of PCRE's
938        man pages to HTML (despite the fact that I distribute HTML pages, some
939        people do their own conversions for various reasons). The problems
940        concerned the use of low-level troff macros .br and .in. I have therefore
941        removed all such uses from the man pages (some were redundant, some could
942        be replaced by .nf/.fi pairs). The 132html script that I use to generate
943        HTML has been updated to handle .nf/.fi and to complain if it encounters
944        .br or .in.
945    
946     4. Updated comments in configure.ac that get placed in config.h.in and also
947        arranged for config.h to be included in the distribution, with the name
948        config.h.generic, for the benefit of those who have to compile without
949        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
950    
951     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
952        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
953        makevp.bat for the current PCRE, using the additional files
954        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
955    
956     6. A Windows user reported a minor discrepancy with test 2, which turned out
957        to be caused by a trailing space on an input line that had got lost in his
958        copy. The trailing space was an accident, so I've just removed it.
959    
960     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
961        that is needed.
962    
963     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
964        as "const" (a) because they are and (b) because it helps the PHP
965        maintainers who have recently made a script to detect big data structures
966        in the php code that should be moved to the .rodata section. I remembered
967        to update Builducptable as well, so it won't revert if ucptable.h is ever
968        re-created.
969    
970     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
971        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
972        order to be able to cut out the UTF-8 tables in the latter when UTF-8
973        support is not required. This saves 1.5-2K of code, which is important in
974        some applications.
975    
976        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
977        so as not to refer to the tables, even though these functions will never be
978        called when UTF-8 support is disabled. Otherwise there are problems with a
979        shared library.
980    
981    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
982    
983        (a) It was defining its arguments as char * instead of void *.
984    
985        (b) It was assuming that all moves were upwards in memory; this was true
986            a long time ago when I wrote it, but is no longer the case.
987    
988        The emulated memove() is provided for those environments that have neither
989        memmove() nor bcopy(). I didn't think anyone used it these days, but that
990        is clearly not the case, as these two bugs were recently reported.
991    
992    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
993        and Detrail to create the HTML documentation, the .txt form of the man
994        pages, and it removes trailing spaces from listed files. It also creates
995        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
996        case, it wraps all the #defines with #ifndefs. This script should be run
997        before "make dist".
998    
999    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1000        with Unicode property support.
1001    
1002        (a) For a maximizing quantifier, if the two different cases of the
1003            character were of different lengths in their UTF-8 codings (there are
1004            some cases like this - I found 11), and the matching function had to
1005            back up over a mixture of the two cases, it incorrectly assumed they
1006            were both the same length.
1007    
1008        (b) When PCRE was configured to use the heap rather than the stack for
1009            recursion during matching, it was not correctly preserving the data for
1010            the other case of a UTF-8 character when checking ahead for a match
1011            while processing a minimizing repeat. If the check also involved
1012            matching a wide character, but failed, corruption could cause an
1013            erroneous result when trying to check for a repeat of the original
1014            character.
1015    
1016    13. Some tidying changes to the testing mechanism:
1017    
1018        (a) The RunTest script now detects the internal link size and whether there
1019            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1020            values substituted by "configure". (The RunGrepTest script already did
1021            this for UTF-8.) The configure.ac script no longer substitutes the
1022            relevant variables.
1023    
1024        (b) The debugging options /B and /D in pcretest show the compiled bytecode
1025            with length and offset values. This means that the output is different
1026            for different internal link sizes. Test 2 is skipped for link sizes
1027            other than 2 because of this, bypassing the problem. Unfortunately,
1028            there was also a test in test 3 (the locale tests) that used /B and
1029            failed for link sizes other than 2. Rather than cut the whole test out,
1030            I have added a new /Z option to pcretest that replaces the length and
1031            offset values with spaces. This is now used to make test 3 independent
1032            of link size. (Test 2 will be tidied up later.)
1033    
1034    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1035        segmentation fault instead of returning the appropriate error message.
1036    
1037    15. In multiline mode when the newline sequence was set to "any", the pattern
1038        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1039        This doesn't seem right; it now treats the CRLF combination as the line
1040        ending, and so does not match in that case. It's only a pattern such as ^$
1041        that would hit this one: something like ^ABC$ would have failed after \r
1042        and then tried again after \r\n.
1043    
1044    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1045        in an attempt to make files that differ only in their line terminators
1046        compare equal. This works on Linux.
1047    
1048    17. Under certain error circumstances pcregrep might try to free random memory
1049        as it exited. This is now fixed, thanks to valgrind.
1050    
1051    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1052        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1053        was because its rules for how to advance for /g after matching an empty
1054        string at the end of a line did not allow for this case. They now check for
1055        it specially.
1056    
1057    20. pcretest is supposed to handle patterns and data of any length, by
1058        extending its buffers when necessary. It was getting this wrong when the
1059        buffer for a data line had to be extended.
1060    
1061    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1062        CRLF as a newline sequence.
1063    
1064    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1065        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1066        I have nevertheless tidied it up.
1067    
1068    23. Added some casts to kill warnings from HP-UX ia64 compiler.
1069    
1070    24. Added a man page for pcre-config.
1071    
1072    
1073    Version 7.0 19-Dec-06
1074    ---------------------
1075    
1076     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1077        moving to gcc 4.1.1.
1078    
1079     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1080        sys/time.h, which is documented as needed for this function. It doesn't
1081        seem to matter on Linux, but it showed up on some releases of OS X.
1082    
1083     3. It seems that there are systems where bytes whose values are greater than
1084        127 match isprint() in the "C" locale. The "C" locale should be the
1085        default when a C program starts up. In most systems, only ASCII printing
1086        characters match isprint(). This difference caused the output from pcretest
1087        to vary, making some of the tests fail. I have changed pcretest so that:
1088    
1089        (a) When it is outputting text in the compiled version of a pattern, bytes
1090            other than 32-126 are always shown as hex escapes.
1091    
1092        (b) When it is outputting text that is a matched part of a subject string,
1093            it does the same, unless a different locale has been set for the match
1094            (using the /L modifier). In this case, it uses isprint() to decide.
1095    
1096     4. Fixed a major bug that caused incorrect computation of the amount of memory
1097        required for a compiled pattern when options that changed within the
1098        pattern affected the logic of the preliminary scan that determines the
1099        length. The relevant options are -x, and -i in UTF-8 mode. The result was
1100        that the computed length was too small. The symptoms of this bug were
1101        either the PCRE error "internal error: code overflow" from pcre_compile(),
1102        or a glibc crash with a message such as "pcretest: free(): invalid next
1103        size (fast)". Examples of patterns that provoked this bug (shown in
1104        pcretest format) are:
1105    
1106          /(?-x: )/x
1107          /(?x)(?-x: \s*#\s*)/
1108          /((?i)[\x{c0}])/8
1109          /(?i:[\x{c0}])/8
1110    
1111        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1112        is now done differently.
1113    
1114     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1115        wrapper classes; (b) implement a new function in the C++ scanner that is
1116        more efficient than the old way of doing things because it avoids levels of
1117        recursion in the regex matching; (c) add a paragraph to the documentation
1118        for the FullMatch() function.
1119    
1120     6. The escape sequence \n was being treated as whatever was defined as
1121        "newline". Not only was this contrary to the documentation, which states
1122        that \n is character 10 (hex 0A), but it also went horribly wrong when
1123        "newline" was defined as CRLF. This has been fixed.
1124    
1125     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1126        was being set to -1 for the "end of line" case (supposedly a value that no
1127        character can have). Though this value is never used (the check for end of
1128        line is "zero bytes in current character"), it caused compiler complaints.
1129        I've changed it to 0xffffffff.
1130    
1131     8. In pcre_version.c, the version string was being built by a sequence of
1132        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1133        string (as it is for production releases) called a macro with an empty
1134        argument. The C standard says the result of this is undefined. The gcc
1135        compiler treats it as an empty string (which was what was wanted) but it is
1136        reported that Visual C gives an error. The source has been hacked around to
1137        avoid this problem.
1138    
1139     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1140        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1141        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1142        of them did).
1143    
1144    10. Originally, pcretest opened its input and output without "b"; then I was
1145        told that "b" was needed in some environments, so it was added for release
1146        5.0 to both the input and output. (It makes no difference on Unix-like
1147        systems.) Later I was told that it is wrong for the input on Windows. I've
1148        now abstracted the modes into two macros, to make it easier to fiddle with
1149        them, and removed "b" from the input mode under Windows.
1150    
1151    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1152    
1153    12. Added -help and --help to pcretest as an official way of being reminded
1154        of the options.
1155    
1156    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1157        and pcrecpp.cc because they annoy compilers at high warning levels.
1158    
1159    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1160    
1161    15. Fixed an occurrence of == in configure.ac that should have been = (shell
1162        scripts are not C programs :-) and which was not noticed because it works
1163        on Linux.
1164    
1165    16. pcretest is supposed to handle any length of pattern and data line (as one
1166        line or as a continued sequence of lines) by extending its input buffer if
1167        necessary. This feature was broken for very long pattern lines, leading to
1168        a string of junk being passed to pcre_compile() if the pattern was longer
1169        than about 50K.
1170    
1171    17. I have done a major re-factoring of the way pcre_compile() computes the
1172        amount of memory needed for a compiled pattern. Previously, there was code
1173        that made a preliminary scan of the pattern in order to do this. That was
1174        OK when PCRE was new, but as the facilities have expanded, it has become
1175        harder and harder to keep it in step with the real compile phase, and there
1176        have been a number of bugs (see for example, 4 above). I have now found a
1177        cunning way of running the real compile function in a "fake" mode that
1178        enables it to compute how much memory it would need, while actually only
1179        ever using a few hundred bytes of working memory and without too many
1180        tests of the mode. This should make future maintenance and development
1181        easier. A side effect of this work is that the limit of 200 on the nesting
1182        depth of parentheses has been removed (though this was never a serious
1183        limitation, I suspect). However, there is a downside: pcre_compile() now
1184        runs more slowly than before (30% or more, depending on the pattern). I
1185        hope this isn't a big issue. There is no effect on runtime performance.
1186    
1187    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1188        newline (only possible for the last line of a file) and it was a
1189        pattern that set a locale (followed by /Lsomething), pcretest crashed.
1190    
1191    19. Added additional timing features to pcretest. (1) The -tm option now times
1192        matching only, not compiling. (2) Both -t and -tm can be followed, as a
1193        separate command line item, by a number that specifies the number of
1194        repeats to use when timing. The default is 50000; this gives better
1195        precision, but takes uncomfortably long for very large patterns.
1196    
1197    20. Extended pcre_study() to be more clever in cases where a branch of a
1198        subpattern has no definite first character. For example, (a*|b*)[cd] would
1199        previously give no result from pcre_study(). Now it recognizes that the
1200        first character must be a, b, c, or d.
1201    
1202    21. There was an incorrect error "recursive call could loop indefinitely" if
1203        a subpattern (or the entire pattern) that was being tested for matching an
1204        empty string contained only one non-empty item after a nested subpattern.
1205        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1206        incorrectly, because the \d was being skipped in the check.
1207    
1208    22. The pcretest program now has a new pattern option /B and a command line
1209        option -b, which is equivalent to adding /B to every pattern. This causes
1210        it to show the compiled bytecode, without the additional information that
1211        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1212        is the same as /B/I).
1213    
1214    23. A new optimization is now able automatically to treat some sequences such
1215        as a*b as a*+b. More specifically, if something simple (such as a character
1216        or a simple class like \d) has an unlimited quantifier, and is followed by
1217        something that cannot possibly match the quantified thing, the quantifier
1218        is automatically "possessified".
1219    
1220    24. A recursive reference to a subpattern whose number was greater than 39
1221        went wrong under certain circumstances in UTF-8 mode. This bug could also
1222        have affected the operation of pcre_study().
1223    
1224    25. Realized that a little bit of performance could be had by replacing
1225        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1226    
1227    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1228    
1229    27. Possessive quantifiers such as a++ were previously implemented by turning
1230        them into atomic groups such as ($>a+). Now they have their own opcodes,
1231        which improves performance. This includes the automatically created ones
1232        from 23 above.
1233    
1234    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1235        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1236        the first matched character to be a colon. This applied both to named and
1237        numbered groups.
1238    
1239    29. The ucpinternal.h header file was missing its idempotency #ifdef.
1240    
1241    30. I was sent a "project" file called libpcre.a.dev which I understand makes
1242        building PCRE on Windows easier, so I have included it in the distribution.
1243    
1244    31. There is now a check in pcretest against a ridiculously large number being
1245        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1246        loop, the loop is abandoned.
1247    
1248    32. Forward references to subpatterns in conditions such as (?(2)...) where
1249        subpattern 2 is defined later cause pcre_compile() to search forwards in
1250        the pattern for the relevant set of parentheses. This search went wrong
1251        when there were unescaped parentheses in a character class, parentheses
1252        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1253    
1254    33. "Subroutine" calls and backreferences were previously restricted to
1255        referencing subpatterns earlier in the regex. This restriction has now
1256        been removed.
1257    
1258    34. Added a number of extra features that are going to be in Perl 5.10. On the
1259        whole, these are just syntactic alternatives for features that PCRE had
1260        previously implemented using the Python syntax or my own invention. The
1261        other formats are all retained for compatibility.
1262    
1263        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1264            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1265            also .NET compatible.
1266    
1267        (b) A recursion or subroutine call to a named group can now be defined as
1268            (?&name) as well as (?P>name).
1269    
1270        (c) A backreference to a named group can now be defined as \k<name> or
1271            \k'name' as well as (?P=name). The new forms, as well as being in Perl
1272            5.10, are also .NET compatible.
1273    
1274        (d) A conditional reference to a named group can now use the syntax
1275            (?(<name>) or (?('name') as well as (?(name).
1276    
1277        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1278            groups (named and numbered) that are never evaluated inline, but can be
1279            called as "subroutines" from elsewhere. In effect, the DEFINE condition
1280            is always false. There may be only one alternative in such a group.
1281    
1282        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1283            as the simple (?(R). The condition is true only if the most recent
1284            recursion is that of the given number or name. It does not search out
1285            through the entire recursion stack.
1286    
1287        (g) The escape \gN or \g{N} has been added, where N is a positive or
1288            negative number, specifying an absolute or relative reference.
1289    
1290    35. Tidied to get rid of some further signed/unsigned compiler warnings and
1291        some "unreachable code" warnings.
1292    
1293    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1294        things, this adds five new scripts.
1295    
1296    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1297        There were also incompatibilities regarding the handling of \Q..\E inside
1298        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1299        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1300    
1301    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1302        matches an empty string, and forcibly breaks the loop. There were bugs in
1303        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
1304        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
1305        separate and independent bugs (that affected different cases) have been
1306        fixed.
1307    
1308    39. Refactored the code to abolish the use of different opcodes for small
1309        capturing bracket numbers. This is a tidy that I avoided doing when I
1310        removed the limit on the number of capturing brackets for 3.5 back in 2001.
1311        The new approach is not only tidier, it makes it possible to reduce the
1312        memory needed to fix the previous bug (38).
1313    
1314    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1315        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1316        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1317        mode.
1318    
1319    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1320        report.
1321    
1322    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1323        copy construction and assignment in the C++ wrapper.
1324    
1325    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1326        couple of bugs that could have given wrong results in the "--newline=crlf"
1327        case.
1328    
1329    44. Added a number of casts and did some reorganization of signed/unsigned int
1330        variables following suggestions from Dair Grant. Also renamed the variable
1331        "this" as "item" because it is a C++ keyword.
1332    
1333    45. Arranged for dftables to add
1334    
1335          #include "pcre_internal.h"
1336    
1337        to pcre_chartables.c because without it, gcc 4.x may remove the array
1338        definition from the final binary if PCRE is built into a static library and
1339        dead code stripping is activated.
1340    
1341    46. For an unanchored pattern, if a match attempt fails at the start of a
1342        newline sequence, and the newline setting is CRLF or ANY, and the next two
1343        characters are CRLF, advance by two characters instead of one.
1344    
1345    
1346    Version 6.7 04-Jul-06
1347    ---------------------
1348    
1349     1. In order to handle tests when input lines are enormously long, pcretest has
1350        been re-factored so that it automatically extends its buffers when
1351        necessary. The code is crude, but this _is_ just a test program. The
1352        default size has been increased from 32K to 50K.
1353    
1354     2. The code in pcre_study() was using the value of the re argument before
1355        testing it for NULL. (Of course, in any sensible call of the function, it
1356        won't be NULL.)
1357    
1358     3. The memmove() emulation function in pcre_internal.h, which is used on
1359        systems that lack both memmove() and bcopy() - that is, hardly ever -
1360        was missing a "static" storage class specifier.
1361    
1362     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1363        containing an extended class (one that cannot be represented by a bitmap
1364        because it contains high-valued characters or Unicode property items, e.g.
1365        [\pZ]). Almost always one would set UTF-8 mode when processing such a
1366        pattern, but PCRE should not loop if you do not (it no longer does).
1367        [Detail: two cases were found: (a) a repeated subpattern containing an
1368        extended class; (b) a recursive reference to a subpattern that followed a
1369        previous extended class. It wasn't skipping over the extended class
1370        correctly when UTF-8 mode was not set.]
1371    
1372     5. A negated single-character class was not being recognized as fixed-length
1373        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1374        compile error "lookbehind assertion is not fixed length".
1375    
1376     6. The RunPerlTest auxiliary script was showing an unexpected difference
1377        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1378        write a Perl script that can interpret lines of an input file either as
1379        byte characters or as UTF-8, which is what "perltest" was being required to
1380        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1381        can't do is switch easily at run time between having the "use utf8;" pragma
1382        or not. In the end, I fudged it by using the RunPerlTest script to insert
1383        "use utf8;" explicitly for the UTF-8 tests.
1384    
1385     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1386        the end of the subject string, contrary to the documentation and to what
1387        Perl does. This was true of both matching functions. Now it matches only at
1388        the start of the subject and immediately after *internal* newlines.
1389    
1390     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1391        a pointer to an int instead of a pointer to an unsigned long int. This
1392        caused problems on 64-bit systems.
1393    
1394     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1395        instance of the 'standard' template library not being so standard".
1396    
1397    10. There was no check on the number of named subpatterns nor the maximum
1398        length of a subpattern name. The product of these values is used to compute
1399        the size of the memory block for a compiled pattern. By supplying a very
1400        long subpattern name and a large number of named subpatterns, the size
1401        computation could be caused to overflow. This is now prevented by limiting
1402        the length of names to 32 characters, and the number of named subpatterns
1403        to 10,000.
1404    
1405    11. Subpatterns that are repeated with specific counts have to be replicated in
1406        the compiled pattern. The size of memory for this was computed from the
1407        length of the subpattern and the repeat count. The latter is limited to
1408        65535, but there was no limit on the former, meaning that integer overflow
1409        could in principle occur. The compiled length of a repeated subpattern is
1410        now limited to 30,000 bytes in order to prevent this.
1411    
1412    12. Added the optional facility to have named substrings with the same name.
1413    
1414    13. Added the ability to use a named substring as a condition, using the
1415        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1416        are numbers (not recommended). Forward references are permitted.
1417    
1418    14. Added forward references in named backreferences (if you see what I mean).
1419    
1420    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1421        pattern could run off the end of the subject. For example, the pattern
1422        "(?s)(.{1,5})"8 did this with the subject "ab".
1423    
1424    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1425        PCRE_CASELESS was set when matching characters that were quantified with ?
1426        or *.
1427    
1428    17. A character class other than a single negated character that had a minimum
1429        but no maximum quantifier - for example [ab]{6,} - was not handled
1430        correctly by pce_dfa_exec(). It would match only one character.
1431    
1432    18. A valid (though odd) pattern that looked like a POSIX character
1433        class but used an invalid character after [ (for example [[,abc,]]) caused
1434        pcre_compile() to give the error "Failed: internal error: code overflow" or
1435        in some cases to crash with a glibc free() error. This could even happen if
1436        the pattern terminated after [[ but there just happened to be a sequence of
1437        letters, a binary zero, and a closing ] in the memory that followed.
1438    
1439    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1440        over the years. Originally (before any Unicode support), just the bottom 8
1441        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1442        output from "man perlunicode" includes this:
1443    
1444          The regular expression compiler produces polymorphic opcodes.  That
1445          is, the pattern adapts to the data and automatically switches to
1446          the Unicode character scheme when presented with Unicode data--or
1447          instead uses a traditional byte scheme when presented with byte
1448          data.
1449    
1450        Sadly, a wide octal escape does not cause a switch, and in a string with
1451        no other multibyte characters, these octal escapes are treated as before.
1452        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
1453        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1454        Unicode string.
1455    
1456        I have not perpetrated such confusion in PCRE. Up till now, it took just
1457        the bottom 8 bits, as in old Perl. I have now made octal escapes with
1458        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1459        translate to the appropriate multibyte character.
1460    
1461    29. Applied some refactoring to reduce the number of warnings from Microsoft
1462        and Borland compilers. This has included removing the fudge introduced
1463        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1464        a warning about an unused variable.
1465    
1466    21. PCRE has not included VT (character 0x0b) in the set of whitespace
1467        characters since release 4.0, because Perl (from release 5.004) does not.
1468        [Or at least, is documented not to: some releases seem to be in conflict
1469        with the documentation.] However, when a pattern was studied with
1470        pcre_study() and all its branches started with \s, PCRE still included VT
1471        as a possible starting character. Of course, this did no harm; it just
1472        caused an unnecessary match attempt.
1473    
1474    22. Removed a now-redundant internal flag bit that recorded the fact that case
1475        dependency changed within the pattern. This was once needed for "required
1476        byte" processing, but is no longer used. This recovers a now-scarce options
1477        bit. Also moved the least significant internal flag bit to the most-
1478        significant bit of the word, which was not previously used (hangover from
1479        the days when it was an int rather than a uint) to free up another bit for
1480        the future.
1481    
1482    23. Added support for CRLF line endings as well as CR and LF. As well as the
1483        default being selectable at build time, it can now be changed at runtime
1484        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1485        specify that it is scanning data with non-default line endings.
1486    
1487    24. Changed the definition of CXXLINK to make it agree with the definition of
1488        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1489    
1490    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1491        recursions. This makes a big different to stack usage for some patterns.
1492    
1493    26. If a subpattern containing a named recursion or subroutine reference such
1494        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1495        the space required for the compiled pattern went wrong and gave too small a
1496        value. Depending on the environment, this could lead to "Failed: internal
1497        error: code overflow at offset 49" or "glibc detected double free or
1498        corruption" errors.
1499    
1500    27. Applied patches from Google (a) to support the new newline modes and (b) to
1501        advance over multibyte UTF-8 characters in GlobalReplace.
1502    
1503    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1504        difference for some implementation of PCRE in some Windows version.
1505    
1506    29. Added some extra testing facilities to pcretest:
1507    
1508        \q<number>   in a data line sets the "match limit" value
1509        \Q<number>   in a data line sets the "match recursion limt" value
1510        -S <number>  sets the stack size, where <number> is in megabytes
1511    
1512        The -S option isn't available for Windows.
1513    
1514    
1515    Version 6.6 06-Feb-06
1516    ---------------------
1517    
1518     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1519        in pcreposix.h. I have copied the definition from pcre.h.
1520    
1521     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1522        because pcre.h is no longer a built file.
1523    
1524     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1525        not normally included in the compiled code.
1526    
1527    
1528    Version 6.5 01-Feb-06
1529    ---------------------
1530    
1531     1. When using the partial match feature with pcre_dfa_exec(), it was not
1532        anchoring the second and subsequent partial matches at the new starting
1533        point. This could lead to incorrect results. For example, with the pattern
1534        /1234/, partially matching against "123" and then "a4" gave a match.
1535    
1536     2. Changes to pcregrep:
1537    
1538        (a) All non-match returns from pcre_exec() were being treated as failures
1539            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1540            error message is output. Some extra information is given for the
1541            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
1542            probably the only errors that are likely to be caused by users (by
1543            specifying a regex that has nested indefinite repeats, for instance).
1544            If there are more than 20 of these errors, pcregrep is abandoned.
1545    
1546        (b) A binary zero was treated as data while matching, but terminated the
1547            output line if it was written out. This has been fixed: binary zeroes
1548            are now no different to any other data bytes.
1549    
1550        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1551            used to set a locale for matching. The --locale=xxxx long option has
1552            been added (no short equivalent) to specify a locale explicitly on the
1553            pcregrep command, overriding the environment variables.
1554    
1555        (d) When -B was used with -n, some line numbers in the output were one less
1556            than they should have been.
1557    
1558        (e) Added the -o (--only-matching) option.
1559    
1560        (f) If -A or -C was used with -c (count only), some lines of context were
1561            accidentally printed for the final match.
1562    
1563        (g) Added the -H (--with-filename) option.
1564    
1565        (h) The combination of options -rh failed to suppress file names for files
1566            that were found from directory arguments.
1567    
1568        (i) Added the -D (--devices) and -d (--directories) options.
1569    
1570        (j) Added the -F (--fixed-strings) option.
1571    
1572        (k) Allow "-" to be used as a file name for -f as well as for a data file.
1573    
1574        (l) Added the --colo(u)r option.
1575    
1576        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1577            is not present by default.
1578    
1579     3. A nasty bug was discovered in the handling of recursive patterns, that is,
1580        items such as (?R) or (?1), when the recursion could match a number of
1581        alternatives. If it matched one of the alternatives, but subsequently,
1582        outside the recursion, there was a failure, the code tried to back up into
1583        the recursion. However, because of the way PCRE is implemented, this is not
1584        possible, and the result was an incorrect result from the match.
1585    
1586        In order to prevent this happening, the specification of recursion has
1587        been changed so that all such subpatterns are automatically treated as
1588        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1589    
1590     4. I had overlooked the fact that, in some locales, there are characters for
1591        which isalpha() is true but neither isupper() nor islower() are true. In
1592        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1593        and ordfeminine) are like this. This affected the treatment of \w and \W
1594        when they appeared in character classes, but not when they appeared outside
1595        a character class. The bit map for "word" characters is now created
1596        separately from the results of isalnum() instead of just taking it from the
1597        upper, lower, and digit maps. (Plus the underscore character, of course.)
1598    
1599     5. The above bug also affected the handling of POSIX character classes such as
1600        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1601        permanent tables. Instead, the bit maps for such a class were previously
1602        created as the appropriate unions of the upper, lower, and digit bitmaps.
1603        Now they are created by subtraction from the [[:word:]] class, which has
1604        its own bitmap.
1605    
1606     6. The [[:blank:]] character class matches horizontal, but not vertical space.
1607        It is created by subtracting the vertical space characters (\x09, \x0a,
1608        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1609        subtraction was done in the overall bitmap for a character class, meaning
1610        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1611        be recognized. This bug has been fixed.
1612    
1613     7. Patches from the folks at Google:
1614    
1615          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1616          real life, but is still worth protecting against".
1617    
1618          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1619          regular expressions".
1620    
1621          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1622          have it.
1623    
1624          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1625          "configure" and the latter not, in order to fix a problem somebody had
1626          with compiling the Arg class on HP-UX.
1627    
1628          (e) Improve the error-handling of the C++ wrapper a little bit.
1629    
1630          (f) New tests for checking recursion limiting.
1631    
1632     8. The pcre_memmove() function, which is used only if the environment does not
1633        have a standard memmove() function (and is therefore rarely compiled),
1634        contained two bugs: (a) use of int instead of size_t, and (b) it was not
1635        returning a result (though PCRE never actually uses the result).
1636    
1637     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1638        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1639        returned instead of calling malloc() with an overflowing number that would
1640        most likely cause subsequent chaos.
1641    
1642    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1643    
1644    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1645        with this option is matched, the nmatch and pmatch options of regexec() are
1646        ignored.
1647    
1648    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1649        provided in case anyone wants to the the POSIX interface with UTF-8
1650        strings.
1651    
1652    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1653        C++ linking (needed for some HP-UX environments).
1654    
1655    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1656        (unused parameter) and in the pcre_printint() function (omitted "default"
1657        switch label when the default is to do nothing).
1658    
1659    15. Added some code to make it possible, when PCRE is compiled as a C++
1660        library, to replace subject pointers for pcre_exec() with a smart pointer
1661        class, thus making it possible to process discontinuous strings.
1662    
1663    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1664        much the same function. They were added by different people who were trying
1665        to make PCRE easy to compile on non-Unix systems. It has been suggested
1666        that PCRE_EXPORT be abolished now that there is more automatic apparatus
1667        for compiling on Windows systems. I have therefore replaced it with
1668        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1669        defaults to "extern" for C or "extern C" for C++, which works fine on
1670        Unix-like systems. It is now possible to override the value of PCRE_DATA_
1671        SCOPE with something explicit in config.h. In addition:
1672    
1673        (a) pcreposix.h still had just "extern" instead of either of these macros;
1674            I have replaced it with PCRE_DATA_SCOPE.
1675    
1676        (b) Functions such as _pcre_xclass(), which are internal to the library,
1677            but external in the C sense, all had PCRE_EXPORT in their definitions.
1678            This is apparently wrong for the Windows case, so I have removed it.
1679            (It makes no difference on Unix-like systems.)
1680    
1681    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1682        of recursive calls to match(). This is different to MATCH_LIMIT because
1683        that limits the total number of calls to match(), not all of which increase
1684        the depth of recursion. Limiting the recursion depth limits the amount of
1685        stack (or heap if NO_RECURSE is set) that is used. The default can be set
1686        when PCRE is compiled, and changed at run time. A patch from Google adds
1687        this functionality to the C++ interface.
1688    
1689    18. Changes to the handling of Unicode character properties:
1690    
1691        (a) Updated the table to Unicode 4.1.0.
1692    
1693        (b) Recognize characters that are not in the table as "Cn" (undefined).
1694    
1695        (c) I revised the way the table is implemented to a much improved format
1696            which includes recognition of ranges. It now supports the ranges that
1697            are defined in UnicodeData.txt, and it also amalgamates other
1698            characters into ranges. This has reduced the number of entries in the
1699            table from around 16,000 to around 3,000, thus reducing its size
1700            considerably. I realized I did not need to use a tree structure after
1701            all - a binary chop search is just as efficient. Having reduced the
1702            number of entries, I extended their size from 6 bytes to 8 bytes to
1703            allow for more data.
1704    
1705        (d) Added support for Unicode script names via properties such as \p{Han}.
1706    
1707    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1708        matching that character.
1709    
1710    20. When matching a repeated Unicode property with a minimum greater than zero,
1711        (for example \pL{2,}), PCRE could look past the end of the subject if it
1712        reached it while seeking the minimum number of characters. This could
1713        happen only if some of the characters were more than one byte long, because
1714        there is a check for at least the minimum number of bytes.
1715    
1716    21. Refactored the implementation of \p and \P so as to be more general, to
1717        allow for more different types of property in future. This has changed the
1718        compiled form incompatibly. Anybody with saved compiled patterns that use
1719        \p or \P will have to recompile them.
1720    
1721    22. Added "Any" and "L&" to the supported property types.
1722    
1723    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1724        but give a compile time error if the value is greater than 0xff.
1725    
1726    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1727        accidentally not being installed or uninstalled.
1728    
1729    25. The pcre.h file was built from pcre.h.in, but the only changes that were
1730        made were to insert the current release number. This seemed silly, because
1731        it made things harder for people building PCRE on systems that don't run
1732        "configure". I have turned pcre.h into a distributed file, no longer built
1733        by "configure", with the version identification directly included. There is
1734        no longer a pcre.h.in file.
1735    
1736        However, this change necessitated a change to the pcre-config script as
1737        well. It is built from pcre-config.in, and one of the substitutions was the
1738        release number. I have updated configure.ac so that ./configure now finds
1739        the release number by grepping pcre.h.
1740    
1741    26. Added the ability to run the tests under valgrind.
1742    
1743    
1744    Version 6.4 05-Sep-05
1745    ---------------------
1746    
1747     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1748        "--" to be printed when multiple files were scanned, even when none of the
1749        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1750        consider it to be a bug, and have restored the previous behaviour.
1751    
1752     2. A couple of code tidies to get rid of compiler warnings.
1753    
1754     3. The pcretest program used to cheat by referring to symbols in the library
1755        whose names begin with _pcre_. These are internal symbols that are not
1756        really supposed to be visible externally, and in some environments it is
1757        possible to suppress them. The cheating is now confined to including
1758        certain files from the library's source, which is a bit cleaner.
1759    
1760     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1761        file's purpose clearer.
1762    
1763     5. Reorganized pcre_ucp_findchar().
1764    
1765    
1766    Version 6.3 15-Aug-05
1767    ---------------------
1768    
1769     1. The file libpcre.pc.in did not have general read permission in the tarball.
1770    
1771     2. There were some problems when building without C++ support:
1772    
1773        (a) If C++ support was not built, "make install" and "make test" still
1774            tried to test it.
1775    
1776        (b) There were problems when the value of CXX was explicitly set. Some
1777            changes have been made to try to fix these, and ...
1778    
1779        (c) --disable-cpp can now be used to explicitly disable C++ support.
1780    
1781        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1782            backslash in a target when C++ was disabled. This confuses some
1783            versions of "make", apparently. Using an intermediate variable solves
1784            this. (Same for CPP_LOBJ.)
1785    
1786     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1787        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1788        necessary on certain architectures.
1789    
1790     4. Added a setting of -export-symbols-regex to the link command to remove
1791        those symbols that are exported in the C sense, but actually are local
1792        within the library, and not documented. Their names all begin with
1793        "_pcre_". This is not a perfect job, because (a) we have to except some
1794        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1795        available (and never for static libraries). I have made a note to try to
1796        find a way round (a) in the future.
1797    
1798    
1799    Version 6.2 01-Aug-05
1800    ---------------------
1801    
1802     1. There was no test for integer overflow of quantifier values. A construction
1803        such as {1111111111111111} would give undefined results. What is worse, if
1804        a minimum quantifier for a parenthesized subpattern overflowed and became
1805        negative, the calculation of the memory size went wrong. This could have
1806        led to memory overwriting.
1807    
1808     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1809    
1810     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1811        operating environments where this matters.
1812    
1813     4. Applied Giuseppe Maxia's patch to add additional features for controlling
1814        PCRE options from within the C++ wrapper.
1815    
1816     5. Named capturing subpatterns were not being correctly counted when a pattern
1817        was compiled. This caused two problems: (a) If there were more than 100
1818        such subpatterns, the calculation of the memory needed for the whole
1819        compiled pattern went wrong, leading to an overflow error. (b) Numerical
1820        back references of the form \12, where the number was greater than 9, were
1821        not recognized as back references, even though there were sufficient
1822        previous subpatterns.
1823    
1824     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1825        versions of gcc, e.g. 2.95.4.
1826    
1827    
1828    Version 6.1 21-Jun-05
1829    ---------------------
1830    
1831     1. There was one reference to the variable "posix" in pcretest.c that was not
1832        surrounded by "#if !defined NOPOSIX".
1833    
1834     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1835        the cross-check on the old pcre_info() function, for the benefit of the
1836        cut-down version of PCRE that is currently imported into Exim.
1837    
1838     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1839        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1840        patterns that start (?i) but I don't think that matters. The use of (?i) is
1841        just an example; this all applies to the other options as well.
1842    
1843     4. Since libtool seems to echo the compile commands it is issuing, the output
1844        from "make" can be reduced a bit by putting "@" in front of each libtool
1845        compile command.
1846    
1847     5. Patch from the folks at Google for configure.in to be a bit more thorough
1848        in checking for a suitable C++ installation before trying to compile the
1849        C++ stuff. This should fix a reported problem when a compiler was present,
1850        but no suitable headers.
1851    
1852     6. The man pages all had just "PCRE" as their title. I have changed them to
1853        be the relevant file name. I have also arranged that these names are
1854        retained in the file doc/pcre.txt, which is a concatenation in text format
1855        of all the man pages except the little individual ones for each function.
1856    
1857     7. The NON-UNIX-USE file had not been updated for the different set of source
1858        files that come with release 6. I also added a few comments about the C++
1859        wrapper.
1860    
1861    
1862    Version 6.0 07-Jun-05
1863    ---------------------
1864    
1865     1. Some minor internal re-organization to help with my DFA experiments.
1866    
1867     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1868        didn't matter for the library itself when fully configured, but did matter
1869        when compiling without UCP support, or within Exim, where the ucp files are
1870        not imported.
1871    
1872     3. Refactoring of the library code to split up the various functions into
1873        different source modules. The addition of the new DFA matching code (see
1874        below) to a single monolithic source would have made it really too
1875        unwieldy, quite apart from causing all the code to be include in a
1876        statically linked application, when only some functions are used. This is
1877        relevant even without the DFA addition now that patterns can be compiled in
1878        one application and matched in another.
1879    
1880        The downside of splitting up is that there have to be some external
1881        functions and data tables that are used internally in different modules of
1882        the library but which are not part of the API. These have all had their
1883        names changed to start with "_pcre_" so that they are unlikely to clash
1884        with other external names.
1885    
1886     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1887        a different (DFA) algorithm. Although it is slower than the original
1888        function, it does have some advantages for certain types of matching
1889        problem.
1890    
1891     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1892        including restarting after a partial match.
1893    
1894     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1895        defined when compiling for Windows was sent to me. I have put it into the
1896        code, though I have no means of testing or verifying it.
1897    
1898     7. Added the pcre_refcount() auxiliary function.
1899    
1900     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1901        match before or at the first newline in the subject string. In pcretest,
1902        the /f option on a pattern can be used to set this.
1903    
1904     9. A repeated \w when used in UTF-8 mode with characters greater than 256
1905        would behave wrongly. This has been present in PCRE since release 4.0.
1906    
1907    10. A number of changes to the pcregrep command:
1908    
1909        (a) Refactored how -x works; insert ^(...)$ instead of setting
1910            PCRE_ANCHORED and checking the length, in preparation for adding
1911            something similar for -w.
1912    
1913        (b) Added the -w (match as a word) option.
1914    
1915        (c) Refactored the way lines are read and buffered so as to have more
1916            than one at a time available.
1917    
1918        (d) Implemented a pcregrep test script.
1919    
1920        (e) Added the -M (multiline match) option. This allows patterns to match
1921            over several lines of the subject. The buffering ensures that at least
1922            8K, or the rest of the document (whichever is the shorter) is available
1923            for matching (and similarly the previous 8K for lookbehind assertions).
1924    
1925        (f) Changed the --help output so that it now says
1926    
1927              -w, --word-regex(p)
1928    
1929            instead of two lines, one with "regex" and the other with "regexp"
1930            because that confused at least one person since the short forms are the
1931            same. (This required a bit of code, as the output is generated
1932            automatically from a table. It wasn't just a text change.)
1933    
1934        (g) -- can be used to terminate pcregrep options if the next thing isn't an
1935            option but starts with a hyphen. Could be a pattern or a path name
1936            starting with a hyphen, for instance.
1937    
1938        (h) "-" can be given as a file name to represent stdin.
1939    
1940        (i) When file names are being printed, "(standard input)" is used for
1941            the standard input, for compatibility with GNU grep. Previously
1942            "<stdin>" was used.
1943    
1944        (j) The option --label=xxx can be used to supply a name to be used for
1945            stdin when file names are being printed. There is no short form.
1946    
1947        (k) Re-factored the options decoding logic because we are going to add
1948            two more options that take data. Such options can now be given in four
1949            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1950    
1951        (l) Added the -A, -B, and -C options for requesting that lines of context
1952            around matches be printed.
1953    
1954        (m) Added the -L option to print the names of files that do not contain
1955            any matching lines, that is, the complement of -l.
1956    
1957        (n) The return code is 2 if any file cannot be opened, but pcregrep does
1958            continue to scan other files.
1959    
1960        (o) The -s option was incorrectly implemented. For compatibility with other
1961            greps, it now suppresses the error message for a non-existent or non-
1962            accessible file (but not the return code). There is a new option called
1963            -q that suppresses the output of matching lines, which was what -s was
1964            previously doing.
1965    
1966        (p) Added --include and --exclude options to specify files for inclusion
1967            and exclusion when recursing.
1968    
1969    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1970        Hopefully, it now does.
1971    
1972    12. Missing cast in pcre_study().
1973    
1974    13. Added an "uninstall" target to the makefile.
1975    
1976    14. Replaced "extern" in the function prototypes in Makefile.in with
1977        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1978        world, but is set differently for Windows.
1979    
1980    15. Added a second compiling function called pcre_compile2(). The only
1981        difference is that it has an extra argument, which is a pointer to an
1982        integer error code. When there is a compile-time failure, this is set
1983        non-zero, in addition to the error test pointer being set to point to an
1984        error message. The new argument may be NULL if no error number is required
1985        (but then you may as well call pcre_compile(), which is now just a
1986        wrapper). This facility is provided because some applications need a
1987        numeric error indication, but it has also enabled me to tidy up the way
1988        compile-time errors are handled in the POSIX wrapper.
1989    
1990    16. Added VPATH=.libs to the makefile; this should help when building with one
1991        prefix path and installing with another. (Or so I'm told by someone who
1992        knows more about this stuff than I do.)
1993    
1994    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1995        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1996        match everything, including newlines. This is not POSIX-compatible, but
1997        somebody wanted the feature. From pcretest it can be activated by using
1998        both the P and the s flags.
1999    
2000    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
2001    
2002    19. libpcre.pc was being incorrectly installed as executable.
2003    
2004    20. A couple of places in pcretest check for end-of-line by looking for '\n';
2005        it now also looks for '\r' so that it will work unmodified on Windows.
2006    
2007    21. Added Google's contributed C++ wrapper to the distribution.
2008    
2009    22. Added some untidy missing memory free() calls in pcretest, to keep
2010        Electric Fence happy when testing.
2011    
2012    
2013    
2014    Version 5.0 13-Sep-04
2015    ---------------------
2016    
2017     1. Internal change: literal characters are no longer packed up into items
2018        containing multiple characters in a single byte-string. Each character
2019        is now matched using a separate opcode. However, there may be more than one
2020        byte in the character in UTF-8 mode.
2021    
2022     2. The pcre_callout_block structure has two new fields: pattern_position and
2023        next_item_length. These contain the offset in the pattern to the next match
2024        item, and its length, respectively.
2025    
2026     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
2027        insertion of callouts before each pattern item. Added the /C option to
2028        pcretest to make use of this.
2029    
2030     4. On the advice of a Windows user, the lines
2031    
2032          #if defined(_WIN32) || defined(WIN32)
2033          _setmode( _fileno( stdout ), 0x8000 );
2034          #endif  /* defined(_WIN32) || defined(WIN32) */
2035    
2036        have been added to the source of pcretest. This apparently does useful
2037        magic in relation to line terminators.
2038    
2039     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
2040        for the benefit of those environments where the "b" makes a difference.
2041    
2042     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
2043        to know about it. I have put a hack into configure.in that adds in code
2044        to set GCC=yes if CC=icc. This seems to end up at a point in the
2045        generated configure script that is early enough to affect the setting of
2046        compiler options, which is what is needed, but I have no means of testing
2047        whether it really works. (The user who reported this had patched the
2048        generated configure script, which of course I cannot do.)
2049    
2050        LATER: After change 22 below (new libtool files), the configure script
2051        seems to know about icc (and also ecc). Therefore, I have commented out
2052        this hack in configure.in.
2053    
2054     7. Added support for pkg-config (2 patches were sent in).
2055    
2056     8. Negated POSIX character classes that used a combination of internal tables
2057        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2058        [[:^ascii]]. Typically, they would match almost any characters. The other
2059        POSIX classes were not broken in this way.
2060    
2061     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2062        to find the match, as PCRE was deluded into thinking that the match had to
2063        start at the start point or following a newline. The same bug applied to
2064        patterns with negative forward assertions or any backward assertions
2065        preceding ".*" at the start, unless the pattern required a fixed first
2066        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2067    
2068    10. In UTF-8 mode, when moving forwards in the subject after a failed match
2069        starting at the last subject character, bytes beyond the end of the subject
2070        string were read.
2071    
2072    11. Renamed the variable "class" as "classbits" to make life easier for C++
2073        users. (Previously there was a macro definition, but it apparently wasn't
2074        enough.)
2075    
2076    12. Added the new field "tables" to the extra data so that tables can be passed
2077        in at exec time, or the internal tables can be re-selected. This allows
2078        a compiled regex to be saved and re-used at a later time by a different
2079        program that might have everything at different addresses.
2080    
2081    13. Modified the pcre-config script so that, when run on Solaris, it shows a
2082        -R library as well as a -L library.
2083    
2084    14. The debugging options of pcretest (-d on the command line or D on a
2085        pattern) showed incorrect output for anything following an extended class
2086        that contained multibyte characters and which was followed by a quantifier.
2087    
2088    15. Added optional support for general category Unicode character properties
2089        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2090        support. It adds about 90K to the size of the library. The meanings of the
2091        inbuilt class escapes such as \d and \s have NOT been changed.
2092    
2093    16. Updated pcredemo.c to include calls to free() to release the memory for the
2094        compiled pattern.
2095    
2096    17. The generated file chartables.c was being created in the source directory
2097        instead of in the building directory. This caused the build to fail if the
2098        source directory was different from the building directory, and was
2099        read-only.
2100    
2101    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2102        file. No doubt somebody will tell me if they don't make sense... Also added
2103        Dan Mooney's comments about building on OpenVMS.
2104    
2105    19. Added support for partial matching via the PCRE_PARTIAL option for
2106        pcre_exec() and the \P data escape in pcretest.
2107    
2108    20. Extended pcretest with 3 new pattern features:
2109    
2110        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
2111              write the compiled pattern to the file whose name is "rest-of-line".
2112              This is a straight binary dump of the data, with the saved pointer to
2113              the character tables forced to be NULL. The study data, if any, is
2114              written too. After writing, pcretest reads a new pattern.
2115    
2116        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2117              compiled pattern from the given file. There must not be any
2118              occurrences of "<" in the file name (pretty unlikely); if there are,
2119              pcretest will instead treat the initial "<" as a pattern delimiter.
2120              After reading in the pattern, pcretest goes on to read data lines as
2121              usual.
2122    
2123        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2124              and 16-bit fields in a compiled pattern, to simulate a pattern that
2125              was compiled on a host of opposite endianness.
2126    
2127    21. The pcre-exec() function can now cope with patterns that were compiled on
2128        hosts of opposite endianness, with this restriction:
2129    
2130          As for any compiled expression that is saved and used later, the tables
2131          pointer field cannot be preserved; the extra_data field in the arguments
2132          to pcre_exec() should be used to pass in a tables address if a value
2133          other than the default internal tables were used at compile time.
2134    
2135    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2136        now diagnosed as an error. Previously, most of the time, a negative number
2137        would have been treated as zero, but if in addition "ovector" was passed as
2138        NULL, a crash could occur.
2139    
2140    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2141        new versions from the libtool 1.5 distribution (the last one is a copy of
2142        a file called libtool.m4). This seems to have fixed the need to patch
2143        "configure" to support Darwin 1.3 (which I used to do). However, I still
2144        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2145        workstation).
2146    
2147    24. Changed the PCRE licence to be the more standard "BSD" licence.
2148    
2149    
2150    Version 4.5 01-Dec-03
2151    ---------------------
2152    
2153     1. There has been some re-arrangement of the code for the match() function so
2154        that it can be compiled in a version that does not call itself recursively.
2155        Instead, it keeps those local variables that need separate instances for
2156        each "recursion" in a frame on the heap, and gets/frees frames whenever it
2157        needs to "recurse". Keeping track of where control must go is done by means
2158        of setjmp/longjmp. The whole thing is implemented by a set of macros that
2159        hide most of the details from the main code, and operates only if
2160        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2161        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2162        operating.
2163    
2164        To make it easier for callers to provide specially tailored get/free
2165        functions for this usage, two new functions, pcre_stack_malloc, and
2166        pcre_stack_free, are used. They are always called in strict stacking order,
2167        and the size of block requested is always the same.
2168    
2169        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2170        PCRE has been compiled to use the stack or the heap for recursion. The
2171        -C option of pcretest uses this to show which version is compiled.
2172    
2173        A new data escape \S, is added to pcretest; it causes the amounts of store
2174        obtained and freed by both kinds of malloc/free at match time to be added
2175        to the output.
2176    
2177     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2178        what's available on my current Linux desktop machine.
2179    
2180     3. When matching a UTF-8 string, the test for a valid string at the start has
2181        been extended. If start_offset is not zero, PCRE now checks that it points
2182        to a byte that is the start of a UTF-8 character. If not, it returns
2183        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2184        this is necessary because there may be backward assertions in the pattern.
2185        When matching the same subject several times, it may save resources to use
2186        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2187    
2188     4. The code for checking the validity of UTF-8 strings has been tightened so
2189        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2190        containing "overlong sequences".
2191    
2192     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2193        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2194        should have been "&", but it just so happened that all the cases this let
2195        through by mistake were picked up later in the function.
2196    
2197     6. I had used a variable called "isblank" - this is a C99 function, causing
2198        some compilers to warn. To avoid this, I renamed it (as "blankclass").
2199    
2200     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2201        prompting; (b) run "./pcretest /dev/null" at the start of the test script
2202        so the version is shown; (c) stop "make test" echoing "./RunTest".
2203    
2204     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2205    
2206     9. The prototype for memmove() for systems that don't have it was using
2207        size_t, but the inclusion of the header that defines size_t was later. I've
2208        moved the #includes for the C headers earlier to avoid this.
2209    
2210    10. Added some adjustments to the code to make it easier to compiler on certain
2211        special systems:
2212    
2213          (a) Some "const" qualifiers were missing.
2214          (b) Added the macro EXPORT before all exported functions; by default this
2215              is defined to be empty.
2216          (c) Changed the dftables auxiliary program (that builds chartables.c) so
2217              that it reads its output file name as an argument instead of writing
2218              to the standard output and assuming this can be redirected.
2219    
2220    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2221        class containing characters with values greater than 255, PCRE compilation
2222        went into a loop.
2223    
2224    12. A recursive reference to a subpattern that was within another subpattern
2225        that had a minimum quantifier of zero caused PCRE to crash. For example,
2226        (x(y(?2))z)? provoked this bug with a subject that got as far as the
2227        recursion. If the recursively-called subpattern itself had a zero repeat,
2228        that was OK.
2229    
2230    13. In pcretest, the buffer for reading a data line was set at 30K, but the
2231        buffer into which it was copied (for escape processing) was still set at
2232        1024, so long lines caused crashes.
2233    
2234    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2235        "internal error: code overflow...". This applied to any character class
2236        that was followed by a possessive quantifier.
2237    
2238    15. Modified the Makefile to add libpcre.la as a prerequisite for
2239        libpcreposix.la because I was told this is needed for a parallel build to
2240        work.
2241    
2242    16. If a pattern that contained .* following optional items at the start was
2243        studied, the wrong optimizing data was generated, leading to matching
2244        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2245        matching string must start with a or b or c. The correct conclusion for
2246        this pattern is that a match can start with any character.
2247    
2248    
2249    Version 4.4 13-Aug-03
2250    ---------------------
2251    
2252     1. In UTF-8 mode, a character class containing characters with values between
2253        127 and 255 was not handled correctly if the compiled pattern was studied.
2254        In fixing this, I have also improved the studying algorithm for such
2255        classes (slightly).
2256    
2257     2. Three internal functions had redundant arguments passed to them. Removal
2258        might give a very teeny performance improvement.
2259    
2260     3. Documentation bug: the value of the capture_top field in a callout is *one
2261        more than* the number of the hightest numbered captured substring.
2262    
2263     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2264        in incorrectly linking with a previously installed version. They now link
2265        explicitly with libpcre.la.
2266    
2267     5. configure.in no longer needs to recognize Cygwin specially.
2268    
2269     6. A problem in pcre.in for Windows platforms is fixed.
2270    
2271     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2272        pcretest, it used to include the size of the study block as part of its
2273        output. Unfortunately, the structure contains a field that has a different
2274        size on different hardware architectures. This meant that the tests that
2275        showed this size failed. As the block is currently always of a fixed size,
2276        this information isn't actually particularly useful in pcretest output, so
2277        I have just removed it.
2278    
2279     8. Three pre-processor statements accidentally did not start in column 1.
2280        Sadly, there are *still* compilers around that complain, even though
2281        standard C has not required this for well over a decade. Sigh.
2282    
2283     9. In pcretest, the code for checking callouts passed small integers in the
2284        callout_data field, which is a void * field. However, some picky compilers
2285        complained about the casts involved for this on 64-bit systems. Now
2286        pcretest passes the address of the small integer instead, which should get
2287        rid of the warnings.
2288    
2289    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2290        both compile and run time, and gives an error if an invalid UTF-8 sequence
2291        is found. There is a option for disabling this check in cases where the
2292        string is known to be correct and/or the maximum performance is wanted.
2293    
2294    11. In response to a bug report, I changed one line in Makefile.in from
2295    
2296            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2297        to
2298            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2299    
2300        to look similar to other lines, but I have no way of telling whether this
2301        is the right thing to do, as I do not use Windows. No doubt I'll get told
2302        if it's wrong...
2303    
2304    
2305    Version 4.3 21-May-03
2306    ---------------------
2307    
2308    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2309       Makefile.
2310    
2311    2. Some refactoring to improve the quality of the code:
2312    
2313       (i)   The utf8_table... variables are now declared "const".
2314    
2315       (ii)  The code for \cx, which used the "case flipping" table to upper case
2316             lower case letters, now just substracts 32. This is ASCII-specific,
2317             but the whole concept of \cx is ASCII-specific, so it seems
2318             reasonable.
2319    
2320       (iii) PCRE was using its character types table to recognize decimal and
2321             hexadecimal digits in the pattern. This is silly, because it handles
2322             only 0-9, a-f, and A-F, but the character types table is locale-
2323             specific, which means strange things might happen. A private
2324             table is now used for this - though it costs 256 bytes, a table is
2325             much faster than multiple explicit tests. Of course, the standard
2326             character types table is still used for matching digits in subject
2327             strings against \d.
2328    
2329       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2330             ending in _t are). So I've renamed it as ESC_tee.
2331    
2332    3. The first argument for regexec() in the POSIX wrapper should have been
2333       defined as "const".
2334    
2335    4. Changed pcretest to use malloc() for its buffers so that they can be
2336       Electric Fenced for debugging.
2337    
2338    5. There were several places in the code where, in UTF-8 mode, PCRE would try
2339       to read one or more bytes before the start of the subject string. Often this
2340       had no effect on PCRE's behaviour, but in some circumstances it could
2341       provoke a segmentation fault.
2342    
2343    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2344       to try to read one or more bytes before the start of the subject string.
2345    
2346    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
2347       UTF-8 support could misbehave in various ways if the subject string
2348       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
2349       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
2350       back over UTF-8 characters.)
2351    
2352    
2353    Version 4.2 14-Apr-03
2354    ---------------------
2355    
2356    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
2357    
2358    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
2359         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
2360         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
2361         [WIN_PREFIX]: new variable, "cyg" for Cygwin
2362         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
2363           and BUILD_EXEEXT
2364         Note: automatic setting of the BUILD variables is not yet working
2365         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
2366           compile-time but not at link-time
2367         [LINK]: use for linking executables only
2368         make different versions for Windows and non-Windows
2369         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
2370           libraries
2371         [LINK_FOR_BUILD]: new variable
2372         [OBJEXT]: use throughout
2373         [EXEEXT]: use throughout
2374         <winshared>: new target
2375         <wininstall>: new target
2376         <dftables.o>: use native compiler
2377         <dftables>: use native linker
2378         <install>: handle Windows platform correctly
2379         <clean>: ditto
2380         <check>: ditto
2381         copy DLL to top builddir before testing
2382    
2383       As part of these changes, -no-undefined was removed again. This was reported
2384       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
2385       in any case.
2386    
2387    3. Some tidies to get rid of compiler warnings:
2388    
2389       . In the match_data structure, match_limit was an unsigned long int, whereas
2390         match_call_count was an int. I've made them both unsigned long ints.
2391    
2392       . In pcretest the fact that a const uschar * doesn't automatically cast to
2393         a void * provoked a warning.
2394    
2395       . Turning on some more compiler warnings threw up some "shadow" variables
2396         and a few more missing casts.
2397    
2398    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
2399       option, a class that contained a single character with a value between 128
2400       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
2401    
2402    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
2403       option, a class that contained several characters, but with at least one
2404       whose value was between 128 and 255 caused PCRE to crash.
2405    
2406    
2407    Version 4.1 12-Mar-03
2408    ---------------------
2409    
2410    1. Compiling with gcc -pedantic found a couple of places where casts were
2411    needed, and a string in dftables.c that was longer than standard compilers are
2412    required to support.
2413    
2414    2. Compiling with Sun's compiler found a few more places where the code could
2415    be tidied up in order to avoid warnings.
2416    
2417    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
2418    first of these names is deprecated in the latest Autoconf in favour of the name
2419    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
2420    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2421    analogy I have changed it to CFLAGS_FOR_BUILD.
2422    
2423    4. Added -no-undefined to the linking command in the Makefile, because this is
2424    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2425    linking step for the pcreposix library.
2426    
2427    5. PCRE was failing to diagnose the case of two named groups with the same
2428    name.
2429    
2430    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2431    literal character that is needed in the subject for a match, and scans along to
2432    ensure that it is present before embarking on the full matching process. This
2433    saves time in cases of nested unlimited repeats that are never going to match.
2434    Problem: the scan can take a lot of time if the subject is very long (e.g.
2435    megabytes), thus penalizing straightforward matches. It is now done only if the
2436    amount of subject to be scanned is less than 1000 bytes.
2437    
2438    7. A lesser problem with the same optimization is that it was recording the
2439    first character of an anchored pattern as "needed", thus provoking a search
2440    right along the subject, even when the first match of the pattern was going to
2441    fail. The "needed" character is now not set for anchored patterns, unless it
2442    follows something in the pattern that is of non-fixed length. Thus, it still
2443    fulfils its original purpose of finding quick non-matches in cases of nested
2444    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
2445    
2446    
2447    Version 4.0 17-Feb-03
2448    ---------------------
2449    
2450    1. If a comment in an extended regex that started immediately after a meta-item
2451    extended to the end of string, PCRE compiled incorrect data. This could lead to
2452    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
2453    
2454    2. Moved to autoconf 2.53 and libtool 1.4.2.
2455    
2456    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
2457    the special perltest8 script is no longer needed - all the tests can be run
2458    from a single perltest script.
2459    
2460    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
2461    by \s. It has now been removed in PCRE. This means it isn't recognized as
2462    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
2463    class [:space:] *does* include VT, thereby creating a mess.
2464    
2465    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
2466    space and tab.
2467    
2468    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
2469    its new features into the main test script, reducing the number of scripts.
2470    
2471    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
2472    were backward compatible, and made the (?i) apply to the whole pattern, as if
2473    /i were given. Now it behaves more logically, and applies the option setting
2474    only to what follows. PCRE has been changed to follow suit. However, if it
2475    finds options settings right at the start of the pattern, it extracts them into
2476    the global options, as before. Thus, they show up in the info data.
2477    
2478    8. Added support for the \Q...\E escape sequence. Characters in between are
2479    treated as literals. This is slightly different from Perl in that $ and @ are
2480    also handled as literals inside the quotes. In Perl, they will cause variable
2481    interpolation. Note the following examples:
2482    
2483        Pattern            PCRE matches      Perl matches
2484    
2485        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
2486        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
2487        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
2488    
2489    For compatibility with Perl, \Q...\E sequences are recognized inside character
2490    classes as well as outside them.
2491    
2492    9. Re-organized 3 code statements in pcretest to avoid "overflow in
2493    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2494    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2495    signed/unsigned warnings.
2496    
2497    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2498    option for pcretest, so I've replaced it by a simple function that does just
2499    that job.
2500    
2501    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2502    "pcregrep -".
2503    
2504    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2505    Java package. This provides some syntactic sugar for simple cases of what my
2506    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2507    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2508    item, you can use this simplified notation. Note that only makes sense with
2509    greedy quantifiers. Consequently, the use of the possessive quantifier forces
2510    greediness, whatever the setting of the PCRE_UNGREEDY option.
2511    
2512    13. A change of greediness default within a pattern was not taking effect at
2513    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2514    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2515    was abstracted outside.
2516    
2517    14. PCRE now supports the \G assertion. It is true when the current matching
2518    position is at the start point of the match. This differs from \A when the
2519    starting offset is non-zero. Used with the /g option of pcretest (or similar
2520    code), it works in the same way as it does for Perl's /g option. If all
2521    alternatives of a regex begin with \G, the expression is anchored to the start
2522    match position, and the "anchored" flag is set in the compiled expression.
2523    
2524    15. Some bugs concerning the handling of certain option changes within patterns
2525    have been fixed. These applied to options other than (?ims). For example,
2526    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2527    been the other way round. Some of this was related to change 7 above.
2528    
2529    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2530    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2531    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2532    POSIX classes only within a class (e.g. /[[:alpha:]]/).
2533    
2534    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2535    mode. Unlike ".", it always matches newline, whatever the setting of
2536    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2537    assertions. Perl allows it, but it doesn't (in general) work because it can't
2538    calculate the length of the lookbehind. At least, that's the case for Perl
2539    5.8.0 - I've been told they are going to document that it doesn't work in
2540    future.
2541    
2542    18. Added an error diagnosis for escapes that PCRE does not support: these are
2543    \L, \l, \N, \P, \p, \U, \u, and \X.
2544    
2545    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2546    reading past the end of the pattern in cases such as /[abcd/.
2547    
2548    20. PCRE was getting more memory than necessary for patterns with classes that
2549    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2550    
2551    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2552    compiling PCRE for use with Virtual Pascal.
2553    
2554    22. Small fix to the Makefile to make it work properly if the build is done
2555    outside the source tree.
2556    
2557    23. Added a new extension: a condition to go with recursion. If a conditional
2558    subpattern starts with (?(R) the "true" branch is used if recursion has
2559    happened, whereas the "false" branch is used only at the top level.
2560    
2561    24. When there was a very long string of literal characters (over 255 bytes
2562    without UTF support, over 250 bytes with UTF support), the computation of how
2563    much memory was required could be incorrect, leading to segfaults or other
2564    strange effects.
2565    
2566    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2567    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2568    there was a subsequent back reference to those brackets. This meant that, for
2569    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2570    possible to check for precisely this case. All we can do is abandon the
2571    optimization if .* occurs inside capturing brackets when there are any back
2572    references whatsoever. (See below for a better fix that came later.)
2573    
2574    26. The handling of the optimization for finding the first character of a
2575    non-anchored pattern, and for finding a character that is required later in the
2576    match were failing in some cases. This didn't break the matching; it just
2577    failed to optimize when it could. The way this is done has been re-implemented.
2578    
2579    27. Fixed typo in error message for invalid (?R item (it said "(?p").
2580    
2581    28. Added a new feature that provides some of the functionality that Perl
2582    provides with (?{...}). The facility is termed a "callout". The way it is done
2583    in PCRE is for the caller to provide an optional function, by setting
2584    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2585    global variable. By default it is unset, which disables all calling out. To get
2586    the function called, the regex must include (?C) at appropriate points. This
2587    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2588    This provides a means of identifying different callout points. When PCRE
2589    reaches such a point in the regex, if pcre_callout has been set, the external
2590    function is called. It is provided with data in a structure called
2591    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2592    matching continues; if it returns a non-zero value, the match at the current
2593    point fails. However, backtracking will occur if possible. [This was changed
2594    later and other features added - see item 49 below.]
2595    
2596    29. pcretest is upgraded to test the callout functionality. It provides a
2597    callout function that displays information. By default, it shows the start of
2598    the match and the current position in the text. There are some new data escapes
2599    to vary what happens:
2600    
2601        \C+         in addition, show current contents of captured substrings
2602        \C-         do not supply a callout function
2603        \C!n        return 1 when callout number n is reached
2604        \C!n!m      return 1 when callout number n is reached for the mth time
2605    
2606    30. If pcregrep was called with the -l option and just a single file name, it
2607    output "<stdin>" if a match was found, instead of the file name.
2608    
2609    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2610    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2611    pcre_exec(). This saves a malloc/free per call. The default value of
2612    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2613    when configuring.
2614    
2615    32. The default maximum size of a compiled pattern is 64K. There have been a
2616    few cases of people hitting this limit. The code now uses macros to handle the
2617    storing of links as offsets within the compiled pattern. It defaults to 2-byte
2618    links, but this can be changed to 3 or 4 bytes by --with-link-size when
2619    configuring. Tests 2 and 5 work only with 2-byte links because they output
2620    debugging information about compiled patterns.
2621    
2622    33. Internal code re-arrangements:
2623    
2624    (a) Moved the debugging function for printing out a compiled regex into
2625        its own source file (printint.c) and used #include to pull it into
2626        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2627        separate copies.
2628    
2629    (b) Defined the list of op-code names for debugging as a macro in
2630        internal.h so that it is next to the definition of the opcodes.
2631    
2632    (c) Defined a table of op-code lengths for simpler skipping along compiled
2633        code. This is again a macro in internal.h so that it is next to the
2634        definition of the opcodes.
2635    
2636    34. Added support for recursive calls to individual subpatterns, along the
2637    lines of Robin Houston's patch (but implemented somewhat differently).
2638    
2639    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2640    allow it to read and process whole directories in Win32. This code was
2641    contributed by Lionel Fourquaux; it has not been tested by me.
2642    
2643    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2644    used to name a group. Names consist of alphanumerics and underscores, and must
2645    be unique. Back references use the syntax (?P=name) and recursive calls use
2646    (?P>name) which is a PCRE extension to the Python extension. Groups still have
2647    numbers. The function pcre_fullinfo() can be used after compilation to extract
2648    a name/number map. There are three relevant calls:
2649    
2650      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
2651      PCRE_INFO_NAMECOUNT            yields the number of entries
2652      PCRE_INFO_NAMETABLE            yields a pointer to the map.
2653    
2654    The map is a vector of fixed-size entries. The size of each entry depends on
2655    the length of the longest name used. The first two bytes of each entry are the
2656    group number, most significant byte first. There follows the corresponding
2657    name, zero terminated. The names are in alphabetical order.
2658    
2659    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2660    case instead of 255. Making it the same both with and without UTF-8 support
2661    means that the same test output works with both.
2662    
2663    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2664    calling malloc() with a zero argument.
2665    
2666    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2667    optimization. I've improved things by keeping a bitmap of backreferences with
2668    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2669    fact referenced, the optimization can be applied. It is unlikely that a
2670    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2671    the match to follow \n) will appear inside brackets with a number greater than
2672    31, but if it does, any back reference > 31 suppresses the optimization.
2673    
2674    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2675    of disabling numbered capturing parentheses. Any opening parenthesis that is
2676    not followed by ? behaves as if it were followed by ?: but named parentheses
2677    can still be used for capturing (and they will acquire numbers in the usual
2678    way).
2679    
2680    41. Redesigned the return codes from the match() function into yes/no/error so
2681    that errors can be passed back from deep inside the nested calls. A malloc
2682    failure while inside a recursive subpattern call now causes the
2683    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2684    
2685    42. It is now possible to set a limit on the number of times the match()
2686    function is called in a call to pcre_exec(). This facility makes it possible to
2687    limit the amount of recursion and backtracking, though not in a directly
2688    obvious way, because the match() function is used in a number of different
2689    circumstances. The count starts from zero for each position in the subject
2690    string (for non-anchored patterns). The default limit is, for compatibility, a
2691    large number, namely 10 000 000. You can change this in two ways:
2692    
2693    (a) When configuring PCRE before making, you can use --with-match-limit=n
2694        to set a default value for the compiled library.
2695    
2696    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2697        a different value is set. See 45 below.
2698    
2699    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2700    
2701    43. Added a new function pcre_config(int, void *) to enable run-time extraction
2702    of things that can be changed at compile time. The first argument specifies
2703    what is wanted and the second points to where the information is to be placed.
2704    The current list of available information is:
2705    
2706      PCRE_CONFIG_UTF8
2707    
2708    The output is an integer that is set to one if UTF-8 support is available;
2709    otherwise it is set to zero.
2710    
2711      PCRE_CONFIG_NEWLINE
2712    
2713    The output is an integer that it set to the value of the code that is used for
2714    newline. It is either LF (10) or CR (13).
2715    
2716      PCRE_CONFIG_LINK_SIZE
2717    
2718    The output is an integer that contains the number of bytes used for internal
2719    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2720    
2721      PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
2722    
2723    The output is an integer that contains the threshold above which the POSIX
2724    interface uses malloc() for output vectors. See item 31 above.
2725    
2726      PCRE_CONFIG_MATCH_LIMIT
2727    
2728    The output is an unsigned integer that contains the default limit of the number
2729    of match() calls in a pcre_exec() execution. See 42 above.
2730    
2731    44. pcretest has been upgraded by the addition of the -C option. This causes it
2732    to extract all the available output from the new pcre_config() function, and to
2733    output it. The program then exits immediately.
2734    
2735    45. A need has arisen to pass over additional data with calls to pcre_exec() in
2736    order to support additional features. One way would have been to define
2737    pcre_exec2() (for example) with extra arguments, but this would not have been
2738    extensible, and would also have required all calls to the original function to
2739    be mapped to the new one. Instead, I have chosen to extend the mechanism that
2740    is used for passing in "extra" data from pcre_study().
2741    
2742    The pcre_extra structure is now exposed and defined in pcre.h. It currently
2743    contains the following fields:
2744    
2745      flags         a bitmap indicating which of the following fields are set
2746      study_data    opaque data from pcre_study()
2747      match_limit   a way of specifying a limit on match() calls for a specific
2748                      call to pcre_exec()
2749      callout_data  data for callouts (see 49 below)
2750    
2751    The flag bits are also defined in pcre.h, and are
2752    
2753      PCRE_EXTRA_STUDY_DATA
2754      PCRE_EXTRA_MATCH_LIMIT
2755      PCRE_EXTRA_CALLOUT_DATA
2756    
2757    The pcre_study() function now returns one of these new pcre_extra blocks, with
2758    the actual study data pointed to by the study_data field, and the
2759    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2760    before. That is, this change is entirely upwards-compatible and requires no
2761    change to existing code.
2762    
2763    If you want to pass in additional data to pcre_exec(), you can either place it
2764    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2765    block.
2766    
2767    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2768    data string contains the escape sequence \M, pcretest calls pcre_exec() several
2769    times with different match limits, until it finds the minimum value needed for
2770    pcre_exec() to complete. The value is then output. This can be instructive; for
2771    most simple matches the number is quite small, but for pathological cases it
2772    gets very large very quickly.
2773    
2774    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2775    returns the size of the data block pointed to by the study_data field in a
2776    pcre_extra block, that is, the value that was passed as the argument to
2777    pcre_malloc() when PCRE was getting memory in which to place the information
2778    created by pcre_study(). The fourth argument should point to a size_t variable.
2779    pcretest has been extended so that this information is shown after a successful
2780    pcre_study() call when information about the compiled regex is being displayed.
2781    
2782    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2783    because what follows is always an absolute path. (Later: it turns out that this
2784    is more than cosmetic for MinGW, because it doesn't like empty path
2785    components.)
2786    
2787    49. Some changes have been made to the callout feature (see 28 above):
2788    
2789    (i)  A callout function now has three choices for what it returns:
2790    
2791           0  =>  success, carry on matching
2792         > 0  =>  failure at this point, but backtrack if possible
2793         < 0  =>  serious error, return this value from pcre_exec()
2794    
2795         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2796         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2797         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2798         use by callout functions. It will never be used by PCRE itself.
2799    
2800    (ii) The pcre_extra structure (see 45 above) has a void * field called
2801         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2802         pcre_callout_block structure has a field of the same name. The contents of
2803         the field passed in the pcre_extra structure are passed to the callout
2804         function in the corresponding field in the callout block. This makes it
2805         easier to use the same callout-containing regex from multiple threads. For
2806         testing, the pcretest program has a new data escape
2807    
2808           \C*n        pass the number n (may be negative) as callout_data
2809    
2810         If the callout function in pcretest receives a non-zero value as
2811         callout_data, it returns that value.
2812    
2813    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2814    there were some redundant $(CFLAGS) in commands that are now specified as
2815    $(LINK), which already includes $(CFLAGS).
2816    
2817    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2818    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2819    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2820    one-byte characters throughout. Note that case-insensitive matching applies
2821    only to characters whose values are less than 256. PCRE doesn't support the
2822    notion of cases for higher-valued characters.
2823    
2824    (i)   A character class whose characters are all within 0-255 is handled as
2825          a bit map, and the map is inverted for negative classes. Previously, a
2826          character > 255 always failed to match such a class; however it should
2827          match if the class was a negative one (e.g. [^ab]). This has been fixed.
2828    
2829    (ii)  A negated character class with a single character < 255 is coded as
2830          "not this character" (OP_NOT). This wasn't working properly when the test
2831          character was multibyte, either singly or repeated.
2832    
2833    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2834          mode, for example: \x{100}{2,3}.
2835    
2836    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2837          singly or repeated) now correctly test multibyte characters. However,
2838          PCRE doesn't recognize any characters with values greater than 255 as
2839          digits, spaces, or word characters. Such characters always match \D, \S,
2840          and \W, and never match \d, \s, or \w.
2841    
2842    (v)   Classes may now contain characters and character ranges with values
2843          greater than 255. For example: [ab\x{100}-\x{400}].
2844    
2845    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
2846          PCRE in UTF-8 mode.
2847    
2848    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2849    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2850    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2851    value.)
2852    
2853    53. The single man page has become too large. I have therefore split it up into
2854    a number of separate man pages. These also give rise to individual HTML pages;
2855    these are now put in a separate directory, and there is an index.html page that
2856    lists them all. Some hyperlinking between the pages has been installed.
2857    
2858    54. Added convenience functions for handling named capturing parentheses.
2859    
2860    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2861    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2862    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2863    are faulted.
2864    
2865    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2866    calling configure. These values are used when compiling the dftables.c program
2867    which is run to generate the source of the default character tables. They
2868    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2869    you will need to set these values.
2870    
2871    57. Updated the building process for Windows DLL, as provided by Fred Cox.
2872    
2873    
2874    Version 3.9 02-Jan-02
2875    ---------------------
2876    
2877    1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2878    
2879    2. If --disable-static was given, the building process failed when trying to
2880    build pcretest and pcregrep. (For some reason it was using libtool to compile
2881    them, which is not right, as they aren't part of the library.)
2882    
2883    
2884    Version 3.8 18-Dec-01
2885    ---------------------
2886    
2887    1. The experimental UTF-8 code was completely screwed up. It was packing the
2888    bytes in the wrong order. How dumb can you get?
2889    
2890    
2891    Version 3.7 29-Oct-01
2892    ---------------------
2893    
2894    1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2895    This caused pcretest, when used on the test data, to segfault. Unfortunately,
2896    this didn't happen under Solaris 8, where I normally test things.
2897    
2898    2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2899    doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2900    isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2901    this fix an hour or so after the initial 3.7 release.)
2902    
2903    
2904    Version 3.6 23-Oct-01
2905    ---------------------
2906    
2907    1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2908    offsets passed as NULL with zero offset count.
2909    
2910    2. The config.guess and config.sub files had not been updated when I moved to
2911    the latest autoconf.
2912    
2913    
2914    Version 3.5 15-Aug-01
2915    ---------------------
2916    
2917    1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2918    had been forgotten.
2919    
2920    2. By using declared but undefined structures, we can avoid using "void"
2921    definitions in pcre.h while keeping the internal definitions of the structures
2922    private.
2923    
2924    3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2925    user point of view, this means that both static and shared libraries are built
2926    by default, but this can be individually controlled. More of the work of
2927    handling this static/shared cases is now inside libtool instead of PCRE's make
2928    file.
2929    
2930    4. The pcretest utility is now installed along with pcregrep because it is
2931    useful for users (to test regexs) and by doing this, it automatically gets
2932    relinked by libtool. The documentation has been turned into a man page, so
2933    there are now .1, .txt, and .html versions in /doc.
2934    
2935    5. Upgrades to pcregrep:
2936       (i)   Added long-form option names like gnu grep.
2937       (ii)  Added --help to list all options with an explanatory phrase.
2938       (iii) Added -r, --recursive to recurse into sub-directories.
2939       (iv)  Added -f, --file to read patterns from a file.
2940    
2941    6. pcre_exec() was referring to its "code" argument before testing that
2942    argument for NULL (and giving an error if it was NULL).
2943    
2944    7. Upgraded Makefile.in to allow for compiling in a different directory from
2945    the source directory.
2946    
2947    8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2948    options bits, the pointer it was passed was to an int instead of to an unsigned
2949    long int. This mattered only on 64-bit systems.
2950    
2951    9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2952    generated) instead of pcre.in, which it its source. Also made the same change
2953    in several of the .c files.
2954    
2955    10. A new release of gcc defines printf() as a macro, which broke pcretest
2956    because it had an ifdef in the middle of a string argument for printf(). Fixed
2957    by using separate calls to printf().
2958    
2959    11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2960    script, to force use of CR or LF instead of \n in the source. On non-Unix
2961    systems, the value can be set in config.h.
2962    
2963    12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2964    absolute limit. Changed the text of the error message to make this clear, and
2965    likewise updated the man page.
2966    
2967    13. The limit of 99 on the number of capturing subpatterns has been removed.
2968    The new limit is 65535, which I hope will not be a "real" limit.
2969    
2970    
2971    Version 3.4 22-Aug-00
2972    ---------------------
2973    
2974    1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2975    
2976    2. Diagnose condition (?(0) as an error instead of crashing on matching.
2977    
2978    
2979    Version 3.3 01-Aug-00
2980    ---------------------
2981    
2982    1. If an octal character was given, but the value was greater than \377, it
2983    was not getting masked to the least significant bits, as documented. This could
2984    lead to crashes in some systems.
2985    
2986    2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2987    the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2988    
2989    3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2990    These just pass their arguments on to (pcre_free)(), but they are provided
2991    because some uses of PCRE bind it to non-C systems that can call its functions,
2992    but cannot call free() or pcre_free() directly.
2993    
2994    4. Add "make test" as a synonym for "make check". Corrected some comments in
2995    the Makefile.
2996    
2997    5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2998    Makefile.
2999    
3000    6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
3001    command called pgrep for grepping around the active processes.
3002    
3003    7. Added the beginnings of support for UTF-8 character strings.
3004    
3005    8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
3006    RANLIB to ./ltconfig so that they are used by libtool. I think these are all
3007    the relevant ones. (AR is not passed because ./ltconfig does its own figuring
3008    out for the ar command.)
3009    
3010    
3011    Version 3.2 12-May-00
3012    ---------------------
3013    
3014    This is purely a bug fixing release.
3015    
3016    1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
3017    of ZA. This was just one example of several cases that could provoke this bug,
3018    which was introduced by change 9 of version 2.00. The code for breaking
3019    infinite loops after an iteration that matches an empty string was't working
3020    correctly.
3021    
3022    2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
3023    when matched against abbab (for example). After matching an empty string, it
3024    wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
3025    caused it to match further down the string than it should.
3026    
3027    3. The code contained an inclusion of sys/types.h. It isn't clear why this
3028    was there because it doesn't seem to be needed, and it causes trouble on some
3029    systems, as it is not a Standard C header. It has been removed.
3030    
3031    4. Made 4 silly changes to the source to avoid stupid compiler warnings that
3032    were reported on the Macintosh. The changes were from
3033    
3034      while ((c = *(++ptr)) != 0 && c != '\n');
3035    to
3036      while ((c = *(++ptr)) != 0 && c != '\n') ;
3037    
3038    Totally extraordinary, but if that's what it takes...
3039    
3040    5. PCRE is being used in one environment where neither memmove() nor bcopy() is
3041    available. Added HAVE_BCOPY and an autoconf test for it; if neither
3042    HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
3043    assumes the way PCRE uses memmove() (always moving upwards).
3044    
3045    6. PCRE is being used in one environment where strchr() is not available. There
3046    was only one use in pcre.c, and writing it out to avoid strchr() probably gives
3047    faster code anyway.
3048    
3049    
3050    Version 3.1 09-Feb-00
3051    ---------------------
3052    
3053    The only change in this release is the fixing of some bugs in Makefile.in for
3054    the "install" target:
3055    
3056    (1) It was failing to install pcreposix.h.
3057    
3058    (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
3059    
3060    
3061    Version 3.0 01-Feb-00
3062    ---------------------
3063    
3064    1. Add support for the /+ modifier to perltest (to output $` like it does in
3065    pcretest).
3066    
3067    2. Add support for the /g modifier to perltest.
3068    
3069    3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
3070    matches null strings.
3071    
3072    4. Fix perltest so that it doesn't do unwanted things when fed an empty
3073    pattern. Perl treats empty patterns specially - it reuses the most recent
3074    pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
3075    effect.
3076    
3077    5. The POSIX interface was broken in that it was just handing over the POSIX
3078    captured string vector to pcre_exec(), but (since release 2.00) PCRE has
3079    required a bigger vector, with some working space on the end. This means that
3080    the POSIX wrapper now has to get and free some memory, and copy the results.
3081    
3082    6. Added some simple autoconf support, placing the test data and the
3083    documentation in separate directories, re-organizing some of the
3084    information files, and making it build pcre-config (a GNU standard). Also added
3085    libtool support for building PCRE as a shared library, which is now the
3086    default.
3087    
3088    7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
3089    09 are not valid octal constants. Single digits will be used for minor values
3090    less than 10.
3091    
3092    8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
3093    existing programs that set these in the POSIX interface can use PCRE without
3094    modification.
3095    
3096    9. Added a new function, pcre_fullinfo() with an extensible interface. It can
3097    return all that pcre_info() returns, plus additional data. The pcre_info()
3098    function is retained for compatibility, but is considered to be obsolete.
3099    
3100    10. Added experimental recursion feature (?R) to handle one common case that
3101    Perl 5.6 will be able to do with (?p{...}).
3102    
3103    11. Added support for POSIX character classes like [:alpha:], which Perl is
3104    adopting.
3105    
3106    
3107    Version 2.08 31-Aug-99
3108    ----------------------
3109    
3110    1. When startoffset was not zero and the pattern began with ".*", PCRE was not
3111    trying to match at the startoffset position, but instead was moving forward to
3112    the next newline as if a previous match had failed.
3113    
3114    2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
3115    and could get into a loop if a null string was matched other than at the start
3116    of the subject.
3117    
3118    3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
3119    be distinguished at compile time, and for completeness also added PCRE_DATE.
3120    
3121    5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
3122    in GnuWin32 environments.
3123    
3124    
3125    Version 2.07 29-Jul-99
3126    ----------------------
3127    
3128    1. The documentation is now supplied in plain text form and HTML as well as in
3129    the form of man page sources.
3130    
3131    2. C++ compilers don't like assigning (void *) values to other pointer types.
3132    In particular this affects malloc(). Although there is no problem in Standard
3133    C, I've put in casts to keep C++ compilers happy.
3134    
3135    3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
3136    should be (const char *).
3137    
3138    4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
3139    be useful for non-Unix systems who don't want to bother with the POSIX stuff.
3140    However, I haven't made this a standard facility. The documentation doesn't
3141    mention it, and the Makefile doesn't support it.
3142    
3143    5. The Makefile now contains an "install" target, with editable destinations at
3144    the top of the file. The pcretest program is not installed.
3145    
3146    6. pgrep -V now gives the PCRE version number and date.
3147    
3148    7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
3149    causing the entire string to be ignored, instead of just the last character.
3150    
3151    8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
3152    non-matching string, it can take a very, very long time, even for strings of
3153    quite modest length, because of the nested recursion. PCRE now does better in
3154    some of these cases. It does this by remembering the last required literal
3155    character in the pattern, and pre-searching the subject to ensure it is present
3156    before running the real match. In other words, it applies a heuristic to detect
3157    some types of certain failure quickly, and in the above example, if presented
3158    with a string that has no trailing " it gives "no match" very quickly.
3159    
3160    9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
3161    other alternatives are tried instead.
3162    
3163    
3164    Version 2.06 09-Jun-99
3165    ----------------------
3166    
3167    1. Change pcretest's output for amount of store used to show just the code
3168    space, because the remainder (the data block) varies in size between 32-bit and
3169    64-bit systems.
3170    
3171    2. Added an extra argument to pcre_exec() to supply an offset in the subject to
3172    start matching at. This allows lookbehinds to work when searching for multiple
3173    occurrences in a string.
3174    
3175    3. Added additional options to pcretest for testing multiple occurrences:
3176    
3177       /+   outputs the rest of the string that follows a match
3178       /g   loops for multiple occurrences, using the new startoffset argument
3179       /G   loops for multiple occurrences by passing an incremented pointer
3180    
3181    4. PCRE wasn't doing the "first character" optimization for patterns starting
3182    with \b or \B, though it was doing it for other lookbehind assertions. That is,
3183    it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
3184    the letter 'x'. On long subject strings, this gives a significant speed-up.
3185    
3186    
3187    Version 2.05 21-Apr-99
3188    ----------------------
3189    
3190    1. Changed the type of magic_number from int to long int so that it works
3191    properly on 16-bit systems.
3192    
3193    2. Fixed a bug which caused patterns starting with .* not to work correctly
3194    when the subject string contained newline characters. PCRE was assuming
3195    anchoring for such patterns in all cases, which is not correct because .* will
3196    not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
3197    DOTALL is set at top level; otherwise it knows that patterns starting with .*
3198    must be retried after every newline in the subject.
3199    
3200    
3201    Version 2.04 18-Feb-99
3202    ----------------------
3203    
3204    1. For parenthesized subpatterns with repeats whose minimum was zero, the
3205    computation of the store needed to hold the pattern was incorrect (too large).
3206    If such patterns were nested a few deep, this could multiply and become a real
3207    problem.
3208    
3209    2. Added /M option to pcretest to show the memory requirement of a specific
3210    pattern. Made -m a synonym of -s (which does this globally) for compatibility.
3211    
3212    3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
3213    compiled in such a way that the backtracking after subsequent failure was
3214    pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
3215    ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
3216    
3217    
3218    Version 2.03 02-Feb-99
3219    ----------------------
3220    
3221    1. Fixed typo and small mistake in man page.
3222    
3223    2. Added 4th condition (GPL supersedes if conflict) and created separate
3224    LICENCE file containing the conditions.
3225    
3226    3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
3227    Perl, that is the internal \ allows the delimiter to be included in the
3228    pattern. Locked out the use of \ as a delimiter. If \ immediately follows
3229    the final delimiter, add \ to the end of the pattern (to test the error).
3230    
3231    4. Added the convenience functions for extracting substrings after a successful
3232    match. Updated pcretest to make it able to test these functions.
3233    
3234    
3235    Version 2.02 14-Jan-99
3236    ----------------------
3237    
3238    1. Initialized the working variables associated with each extraction so that
3239    their saving and restoring doesn't refer to uninitialized store.
3240    
3241    2. Put dummy code into study.c in order to trick the optimizer of the IBM C
3242    compiler for OS/2 into generating correct code. Apparently IBM isn't going to
3243    fix the problem.
3244    
3245    3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
3246    calls, and wasn't printing the correct value for compiling calls. Increased the
3247    default value of LOOPREPEAT, and the number of significant figures in the
3248    times.
3249    
3250    4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
3251    
3252    5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
3253    a building problem on Windows NT with a FAT file system.
3254    
3255    
3256  Version 2.01 21-Oct-98  Version 2.01 21-Oct-98
3257  ----------------------  ----------------------

Legend:
Removed from v.25  
changed lines
  Added in v.438

  ViewVC Help
Powered by ViewVC 1.1.5