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

Diff of /code/trunk/ChangeLog

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

revision 19 by nigel, Sat Feb 24 21:38:33 2007 UTC revision 437 by ph10, Sun Sep 6 16:34:42 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8        was out-of-date, and there was no check on the pcre_compile() error code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 \fInmatch\fP with a NULL value for \fIpmatch\fP, the value of
101        \fInmatch\fP 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    
108    Version 7.9 11-Apr-09
109    ---------------------
110    
111    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
112        (pcretest), all targets were linked against these libraries. This included
113        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
114        libraries. This caused unwanted dependencies to be created. This problem
115        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
116        pcretest is linked with readline.
117    
118    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
119        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
120        moved outside it again, because FALSE and TRUE are already defined in AIX,
121        but BOOL is not.
122    
123    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
124        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
125    
126    4.  The pcregrep documentation said spaces were inserted as well as colons (or
127        hyphens) following file names and line numbers when outputting matching
128        lines. This is not true; no spaces are inserted. I have also clarified the
129        wording for the --colour (or --color) option.
130    
131    5.  In pcregrep, when --colour was used with -o, the list of matching strings
132        was not coloured; this is different to GNU grep, so I have changed it to be
133        the same.
134    
135    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
136        each matching line was coloured. Now it goes on to look for further matches
137        of any of the test patterns, which is the same behaviour as GNU grep.
138    
139    7.  A pattern that could match an empty string could cause pcregrep to loop; it
140        doesn't make sense to accept an empty string match in pcregrep, so I have
141        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
142        seems to be how GNU grep behaves.
143    
144    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
145        start or after a newline", because the conditional assertion was not being
146        correctly handled. The rule now is that both the assertion and what follows
147        in the first alternative must satisfy the test.
148    
149    9.  If auto-callout was enabled in a pattern with a conditional group whose
150        condition was an assertion, PCRE could crash during matching, both with
151        pcre_exec() and pcre_dfa_exec().
152    
153    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
154        used for matching.
155    
156    11. Unicode property support in character classes was not working for
157        characters (bytes) greater than 127 when not in UTF-8 mode.
158    
159    12. Added the -M command line option to pcretest.
160    
161    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
162    
163    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
164    
165    16. Added comments and documentation about mis-use of no_arg in the C++
166        wrapper.
167    
168    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
169        from Martin Jerabek that uses macro names for all relevant character and
170        string constants.
171    
172    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
173        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
174        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
175        these, but not everybody uses configure.
176    
177    19. A conditional group that had only one branch was not being correctly
178        recognized as an item that could match an empty string. This meant that an
179        enclosing group might also not be so recognized, causing infinite looping
180        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
181        with the subject "ab", where knowledge that the repeated group can match
182        nothing is needed in order to break the loop.
183    
184    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
185        exec(), but without supplying a callout function, matching went wrong.
186    
187    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
188        leak if the size of the offset vector was greater than 30. When the vector
189        is smaller, the saved offsets during recursion go onto a local stack
190        vector, but for larger vectors malloc() is used. It was failing to free
191        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
192        error, in fact).
193    
194    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
195        heapframe that is used only when UTF-8 support is enabled. This caused no
196        problem, but was untidy.
197    
198    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
199        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
200        included within another project.
201    
202    24. Steven Van Ingelgem's patches to add more options to the CMake support,
203        slightly modified by me:
204    
205          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
206              not building pcregrep.
207    
208          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
209              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
210    
211    25. Forward references, both numeric and by name, in patterns that made use of
212        duplicate group numbers, could behave incorrectly or give incorrect errors,
213        because when scanning forward to find the reference group, PCRE was not
214        taking into account the duplicate group numbers. A pattern such as
215        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
216    
217    26. Changed a few more instances of "const unsigned char *" to USPTR, making
218        the feature of a custom pointer more persuasive (as requested by a user).
219    
220    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
221        pcretest.c, inside #ifndefs, because it seems they are sometimes already
222        pre-defined.
223    
224    28. Added support for (*UTF8) at the start of a pattern.
225    
226    29. Arrange for flags added by the "release type" setting in CMake to be shown
227        in the configuration summary.
228    
229    
230    Version 7.8 05-Sep-08
231    ---------------------
232    
233    1.  Replaced UCP searching code with optimized version as implemented for Ad
234        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
235        stage table and inline lookup instead of a function, giving speed ups of 2
236        to 5 times on some simple patterns that I tested. Permission was given to
237        distribute the MultiStage2.py script that generates the tables (it's not in
238        the tarball, but is in the Subversion repository).
239    
240    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
241        scripts.
242    
243    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
244        a group with a zero qualifier. The result of the study could be incorrect,
245        or the function might crash, depending on the pattern.
246    
247    4.  Caseless matching was not working for non-ASCII characters in back
248        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
249        It now works when Unicode Property Support is available.
250    
251    5.  In pcretest, an escape such as \x{de} in the data was always generating
252        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
253        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
254        truncation.
255    
256    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
257    
258    7.  Added two (int) casts to pcregrep when printing the difference of two
259        pointers, in case they are 64-bit values.
260    
261    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
262        test 2 if it fails.
263    
264    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
265        and a #define of that name to empty if it is not externally set. This is to
266        allow users of MSVC to set it if necessary.
267    
268    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
269        the convenience functions in the pcre_get.c source file.
270    
271    11. An option change at the start of a pattern that had top-level alternatives
272        could cause overwriting and/or a crash. This command provoked a crash in
273        some environments:
274    
275          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
276    
277        This potential security problem was recorded as CVE-2008-2371.
278    
279    12. For a pattern where the match had to start at the beginning or immediately
280        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
281        pcre_dfa_exec() could read past the end of the passed subject if there was
282        no match. To help with detecting such bugs (e.g. with valgrind), I modified
283        pcretest so that it places the subject at the end of its malloc-ed buffer.
284    
285    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
286        exec() might read past the end of the data buffer in UTF-8 mode.
287    
288    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
289        the data contained the byte 0x85 as part of a UTF-8 character within its
290        first line. This applied both to normal and DFA matching.
291    
292    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
293        /^[^d]*?$/8 failed to match "abc".
294    
295    16. Added a missing copyright notice to pcrecpp_internal.h.
296    
297    17. Make it more clear in the documentation that values returned from
298        pcre_exec() in ovector are byte offsets, not character counts.
299    
300    18. Tidied a few places to stop certain compilers from issuing warnings.
301    
302    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
303        supplied by Stefan Weber. I made a further small update for 7.8 because
304        there is a change of source arrangements: the pcre_searchfuncs.c module is
305        replaced by pcre_ucd.c.
306    
307    
308    Version 7.7 07-May-08
309    ---------------------
310    
311    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
312        a string to a long long, pretend we don't even have a long long." This is
313        done by checking for the strtoq, strtoll, and _strtoi64 functions.
314    
315    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
316        pre-7.6 versions, which defined a global no_arg variable instead of putting
317        it in the RE class. (See also #8 below.)
318    
319    3.  Remove a line of dead code, identified by coverity and reported by Nuno
320        Lopes.
321    
322    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
323    
324        (1) The include/exclude patterns were being applied to the whole pathnames
325            of files, instead of just to the final components.
326    
327        (2) If there was more than one level of directory, the subdirectories were
328            skipped unless they satisfied the include/exclude conditions. This is
329            inconsistent with GNU grep (and could even be seen as contrary to the
330            pcregrep specification - which I improved to make it absolutely clear).
331            The action now is always to scan all levels of directory, and just
332            apply the include/exclude patterns to regular files.
333    
334    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
335        --exclude_dir in the tests to avoid scanning .svn directories.
336    
337    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
338        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
339        doesn't support NULs in patterns.
340    
341    7.  Added some missing "const"s to declarations of static tables in
342        pcre_compile.c and pcre_dfa_exec.c.
343    
344    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
345        caused by fix #2  above. (Subsequently also a second patch to fix the
346        first patch. And a third patch - this was a messy problem.)
347    
348    9.  Applied Craig's patch to remove the use of push_back().
349    
350    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
351        matching function regexec().
352    
353    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
354        which, however, unlike Perl's \g{...}, are subroutine calls, not back
355        references. PCRE supports relative numbers with this syntax (I don't think
356        Oniguruma does).
357    
358    12. Previously, a group with a zero repeat such as (...){0} was completely
359        omitted from the compiled regex. However, this means that if the group
360        was called as a subroutine from elsewhere in the pattern, things went wrong
361        (an internal error was given). Such groups are now left in the compiled
362        pattern, with a new opcode that causes them to be skipped at execution
363        time.
364    
365    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
366        to the way PCRE behaves:
367    
368        (a) A lone ] character is dis-allowed (Perl treats it as data).
369    
370        (b) A back reference to an unmatched subpattern matches an empty string
371            (Perl fails the current match path).
372    
373        (c) A data ] in a character class must be notated as \] because if the
374            first data character in a class is ], it defines an empty class. (In
375            Perl it is not possible to have an empty class.) The empty class []
376            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
377            The negative empty class [^] matches any one character, independently
378            of the DOTALL setting.
379    
380    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
381        non-existent subpattern following a character class starting with ']' and
382        containing () gave an internal compiling error instead of "reference to
383        non-existent subpattern". Fortunately, when the pattern did exist, the
384        compiled code was correct. (When scanning forwards to check for the
385        existencd of the subpattern, it was treating the data ']' as terminating
386        the class, so got the count wrong. When actually compiling, the reference
387        was subsequently set up correctly.)
388    
389    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
390        it was being rejected as not supported by pcre_dfa_exec(), even though
391        other assertions are supported. I have made pcre_dfa_exec() support
392        (*FAIL).
393    
394    16. The implementation of 13c above involved the invention of a new opcode,
395        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
396        cannot be changed at match time, I realized I could make a small
397        improvement to matching performance by compiling OP_ALLANY instead of
398        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
399        on the OP_ANY path.
400    
401    17. Compiling pcretest on Windows with readline support failed without the
402        following two fixes: (1) Make the unistd.h include conditional on
403        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
404    
405    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
406        ncurses library to be included for pcretest when ReadLine support is
407        requested, but also to allow for it to be overridden. This patch came from
408        Daniel Bergström.
409    
410    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
411        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
412        any errors with the current Unicode tables. Thanks to Peter Kankowski for
413        spotting this.
414    
415    
416    Version 7.6 28-Jan-08
417    ---------------------
418    
419    1.  A character class containing a very large number of characters with
420        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
421        overflow.
422    
423    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
424        HAVE_LONG_LONG is not defined.
425    
426    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
427        bring them up to date and include new features. This patch includes:
428    
429        - Fixed PH's badly added libz and libbz2 support.
430        - Fixed a problem with static linking.
431        - Added pcredemo. [But later removed - see 7 below.]
432        - Fixed dftables problem and added an option.
433        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
434            HAVE_LONG_LONG.
435        - Added readline support for pcretest.
436        - Added an listing of the option settings after cmake has run.
437    
438    4.  A user submitted a patch to Makefile that makes it easy to create
439        "pcre.dll" under mingw when using Configure/Make. I added stuff to
440        Makefile.am that cause it to include this special target, without
441        affecting anything else. Note that the same mingw target plus all
442        the other distribution libraries and programs are now supported
443        when configuring with CMake (see 6 below) instead of with
444        Configure/Make.
445    
446    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
447        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
448        exported in the Windows port". It has not yet been confirmed that the patch
449        solves the problem, but it does no harm.
450    
451    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
452        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
453        with CMake, and also correct the comment about stack recursion.
454    
455    7.  Remove the automatic building of pcredemo from the ./configure system and
456        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
457        of a program that users should build themselves after PCRE is installed, so
458        building it automatically is not really right. What is more, it gave
459        trouble in some build environments.
460    
461    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
462    
463    
464    Version 7.5 10-Jan-08
465    ---------------------
466    
467    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
468        values in parens when parsing an RE using the C++ wrapper."
469    
470    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
471        Characters greater than 255 were excluded from the class instead of being
472        included.
473    
474    3.  The same bug as (2) above applied to negated POSIX classes such as
475        [:^space:].
476    
477    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
478        defined or documented. It seems to have been a typo for PCRE_STATIC, so
479        I have changed it.
480    
481    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
482        first named subpattern) and a construct such as (?&a) would reference the
483        first named subpattern whose name started with "a" (in other words, the
484        length check was missing). Both these problems are fixed. "Subpattern name
485        expected" is now given for (?&) (a zero-length name), and this patch also
486        makes it give the same error for \k'' (previously it complained that that
487        was a reference to a non-existent subpattern).
488    
489    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
490        this is right because (?- can be followed by option settings as well as by
491        digits. I have, however, made the messages clearer.
492    
493    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
494        than the number used in the conditional) now cause a compile-time error.
495        This is actually not compatible with Perl, which accepts such patterns, but
496        treats the conditional as always being FALSE (as PCRE used to), but it
497        seems to me that giving a diagnostic is better.
498    
499    8.  Change "alphameric" to the more common word "alphanumeric" in comments
500        and messages.
501    
502    9.  Fix two occurrences of "backslash" in comments that should have been
503        "backspace".
504    
505    10. Remove two redundant lines of code that can never be obeyed (their function
506        was moved elsewhere).
507    
508    11. The program that makes PCRE's Unicode character property table had a bug
509        which caused it to generate incorrect table entries for sequences of
510        characters that have the same character type, but are in different scripts.
511        It amalgamated them into a single range, with the script of the first of
512        them. In other words, some characters were in the wrong script. There were
513        thirteen such cases, affecting characters in the following ranges:
514    
515          U+002b0 - U+002c1
516          U+0060c - U+0060d
517          U+0061e - U+00612
518          U+0064b - U+0065e
519          U+0074d - U+0076d
520          U+01800 - U+01805
521          U+01d00 - U+01d77
522          U+01d9b - U+01dbf
523          U+0200b - U+0200f
524          U+030fc - U+030fe
525          U+03260 - U+0327f
526          U+0fb46 - U+0fbb1
527          U+10450 - U+1049d
528    
529    12. The -o option (show only the matching part of a line) for pcregrep was not
530        compatible with GNU grep in that, if there was more than one match in a
531        line, it showed only the first of them. It now behaves in the same way as
532        GNU grep.
533    
534    13. If the -o and -v options were combined for pcregrep, it printed a blank
535        line for every non-matching line. GNU grep prints nothing, and pcregrep now
536        does the same. The return code can be used to tell if there were any
537        non-matching lines.
538    
539    14. Added --file-offsets and --line-offsets to pcregrep.
540    
541    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
542        infinitely looping recursion. The bug was that positive lookaheads were not
543        being skipped when checking for a possible empty match (negative lookaheads
544        and both kinds of lookbehind were skipped).
545    
546    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
547        inclusion of <windows.h> to before rather than after the definition of
548        INVALID_FILE_ATTRIBUTES (patch from David Byron).
549    
550    17. Specifying a possessive quantifier with a specific limit for a Unicode
551        character property caused pcre_compile() to compile bad code, which led at
552        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
553        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
554        caused the error; without that there was no problem.
555    
556    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
557    
558    19. Added --enable-pcretest-libreadline.
559    
560    20. In pcrecpp.cc, the variable 'count' was incremented twice in
561        RE::GlobalReplace(). As a result, the number of replacements returned was
562        double what it should be. I removed one of the increments, but Craig sent a
563        later patch that removed the other one (the right fix) and added unit tests
564        that check the return values (which was not done before).
565    
566    21. Several CMake things:
567    
568        (1) Arranged that, when cmake is used on Unix, the libraries end up with
569            the names libpcre and libpcreposix, not just pcre and pcreposix.
570    
571        (2) The above change means that pcretest and pcregrep are now correctly
572            linked with the newly-built libraries, not previously installed ones.
573    
574        (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
575    
576    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
577        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
578        UTF-8 newline character). The key issue is that the pattern starts .*;
579        this means that the match must be either at the beginning, or after a
580        newline. The bug was in the code for advancing after a failed match and
581        checking that the new position followed a newline. It was not taking
582        account of UTF-8 characters correctly.
583    
584    23. PCRE was behaving differently from Perl in the way it recognized POSIX
585        character classes. PCRE was not treating the sequence [:...:] as a
586        character class unless the ... were all letters. Perl, however, seems to
587        allow any characters between [: and :], though of course it rejects as
588        unknown any "names" that contain non-letters, because all the known class
589        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
590        for example, whereas PCRE did not - it did not recognize a POSIX character
591        class. This seemed a bit dangerous, so the code has been changed to be
592        closer to Perl. The behaviour is not identical to Perl, because PCRE will
593        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
594        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
595        Perl does, and where it didn't before.
596    
597    24. Rewrite so as to remove the single use of %n from pcregrep because in some
598        Windows environments %n is disabled by default.
599    
600    
601    Version 7.4 21-Sep-07
602    ---------------------
603    
604    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
605        means that a class such as [\s] counted as "explicit reference to CR or
606        LF". That isn't really right - the whole point of the change was to try to
607        help when there was an actual mention of one of the two characters. So now
608        the change happens only if \r or \n (or a literal CR or LF) character is
609        encountered.
610    
611    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
612        of both had grown to the point where there were only 3 bits left.
613        Fortunately, there was spare space in the data structure, and so I have
614        moved the internal flags into a new 16-bit field to free up more option
615        bits.
616    
617    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
618        but did not set the internal JCHANGED flag - either of these is enough to
619        control the way the "get" function works - but the PCRE_INFO_JCHANGED
620        facility is supposed to tell if (?J) was ever used, so now (?J) at the
621        start sets both bits.
622    
623    4.  Added options (at build time, compile time, exec time) to change \R from
624        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
625    
626    5.  doc/pcresyntax.html was missing from the distribution.
627    
628    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
629        compatibility, even though it is no longer used.
630    
631    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
632        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
633        windows.h file is present (where different names are used). [This was
634        reversed later after testing - see 16 below.]
635    
636    8.  Changed all #include <config.h> to #include "config.h". There were also
637        some further <pcre.h> cases that I changed to "pcre.h".
638    
639    9.  When pcregrep was used with the --colour option, it missed the line ending
640        sequence off the lines that it output.
641    
642    10. It was pointed out to me that arrays of string pointers cause lots of
643        relocations when a shared library is dynamically loaded. A technique of
644        using a single long string with a table of offsets can drastically reduce
645        these. I have refactored PCRE in four places to do this. The result is
646        dramatic:
647    
648          Originally:                          290
649          After changing UCP table:            187
650          After changing error message table:   43
651          After changing table of "verbs"       36
652          After changing table of Posix names   22
653    
654        Thanks to the folks working on Gregex for glib for this insight.
655    
656    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
657        unicode-properties was also set.
658    
659    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
660    
661    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
662        checked only for CRLF.
663    
664    14. Added casts to pcretest.c to avoid compiler warnings.
665    
666    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
667    
668    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
669        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
670        entirely. This removes changes made in 7 above.
671    
672    17. The CMake files have been updated, and there is now more information about
673        building with CMake in the NON-UNIX-USE document.
674    
675    
676    Version 7.3 28-Aug-07
677    ---------------------
678    
679     1. In the rejigging of the build system that eventually resulted in 7.1, the
680        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
681        brackets there is not right, since it causes compilers to look for an
682        installed pcre.h, not the version that is in the source that is being
683        compiled (which of course may be different). I have changed it back to:
684    
685          #include "pcre.h"
686    
687        I have a vague recollection that the change was concerned with compiling in
688        different directories, but in the new build system, that is taken care of
689        by the VPATH setting the Makefile.
690    
691     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
692        when the subject happened to end in the byte 0x85 (e.g. if the last
693        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
694        characters but of course it shouldn't be taken as a newline when it is part
695        of another character. The bug was that, for an unlimited repeat of . in
696        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
697        characters when looking for a newline.
698    
699     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
700    
701     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
702        in debug output.
703    
704     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
705        long printing in the pcrecpp unittest when running under MinGW.
706    
707     6. ESC_K was left out of the EBCDIC table.
708    
709     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
710        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
711        limit also applies to "virtual nesting" when a pattern is recursive, and in
712        this case 1000 isn't so big. I have been able to remove this limit at the
713        expense of backing off one optimization in certain circumstances. Normally,
714        when pcre_exec() would call its internal match() function recursively and
715        immediately return the result unconditionally, it uses a "tail recursion"
716        feature to save stack. However, when a subpattern that can match an empty
717        string has an unlimited repetition quantifier, it no longer makes this
718        optimization. That gives it a stack frame in which to save the data for
719        checking that an empty string has been matched. Previously this was taken
720        from the 1000-entry workspace that had been reserved. So now there is no
721        explicit limit, but more stack is used.
722    
723     8. Applied Daniel's patches to solve problems with the import/export magic
724        syntax that is required for Windows, and which was going wrong for the
725        pcreposix and pcrecpp parts of the library. These were overlooked when this
726        problem was solved for the main library.
727    
728     9. There were some crude static tests to avoid integer overflow when computing
729        the size of patterns that contain repeated groups with explicit upper
730        limits. As the maximum quantifier is 65535, the maximum group length was
731        set at 30,000 so that the product of these two numbers did not overflow a
732        32-bit integer. However, it turns out that people want to use groups that
733        are longer than 30,000 bytes (though not repeat them that many times).
734        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
735        made it possible to implement the integer overflow checks in a much more
736        dynamic way, which I have now done. The artificial limitation on group
737        length has been removed - we now have only the limit on the total length of
738        the compiled pattern, which depends on the LINK_SIZE setting.
739    
740    10. Fixed a bug in the documentation for get/copy named substring when
741        duplicate names are permitted. If none of the named substrings are set, the
742        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
743        empty string.
744    
745    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
746        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
747        because the ] is interpreted as the first data character and the
748        terminating ] is not found. PCRE has been made compatible with Perl in this
749        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
750        cause memory overwriting.
751    
752    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
753        string has been matched (to stop an infinite loop). It was not recognizing
754        a conditional subpattern that could match an empty string if that
755        subpattern was within another subpattern. For example, it looped when
756        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
757        condition was not nested. This bug has been fixed.
758    
759    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
760        past the start of the subject in the presence of bytes with the top bit
761        set, for example "\x8aBCD".
762    
763    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
764        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
765    
766    14. Optimized (?!) to (*FAIL).
767    
768    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
769        This restricts code points to be within the range 0 to 0x10FFFF, excluding
770        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
771        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
772        does: it's just the validity check that is more restrictive.
773    
774    16. Inserted checks for integer overflows during escape sequence (backslash)
775        processing, and also fixed erroneous offset values for syntax errors during
776        backslash processing.
777    
778    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
779        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
780    
781    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
782        caused an overrun.
783    
784    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
785        something other than just ASCII characters) inside a group that had an
786        unlimited repeat caused a loop at compile time (while checking to see
787        whether the group could match an empty string).
788    
789    20. Debugging a pattern containing \p or \P could cause a crash. For example,
790        [\P{Any}] did so. (Error in the code for printing property names.)
791    
792    21. An orphan \E inside a character class could cause a crash.
793    
794    22. A repeated capturing bracket such as (A)? could cause a wild memory
795        reference during compilation.
796    
797    23. There are several functions in pcre_compile() that scan along a compiled
798        expression for various reasons (e.g. to see if it's fixed length for look
799        behind). There were bugs in these functions when a repeated \p or \P was
800        present in the pattern. These operators have additional parameters compared
801        with \d, etc, and these were not being taken into account when moving along
802        the compiled data. Specifically:
803    
804        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
805            length.
806    
807        (b) An item such as \pL+ within a repeated group could cause crashes or
808            loops.
809    
810        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
811            "reference to non-existent subpattern" error.
812    
813        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
814    
815    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
816        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
817    
818    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
819    
820    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
821        character were causing crashes (broken optimization).
822    
823    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
824        \p or \P) caused a compile-time loop.
825    
826    28. More problems have arisen in unanchored patterns when CRLF is a valid line
827        break. For example, the unstudied pattern [\r\n]A does not match the string
828        "\r\nA" because change 7.0/46 below moves the current point on by two
829        characters after failing to match at the start. However, the pattern \nA
830        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
831        the same is true. There doesn't seem any very clean way out of this, but
832        what I have chosen to do makes the common cases work: PCRE now takes note
833        of whether there can be an explicit match for \r or \n anywhere in the
834        pattern, and if so, 7.0/46 no longer applies. As part of this change,
835        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
836        pattern has explicit CR or LF references.
837    
838    29. Added (*CR) etc for changing newline setting at start of pattern.
839    
840    
841    Version 7.2 19-Jun-07
842    ---------------------
843    
844     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
845        which is apparently normally available under Windows.
846    
847     2. Re-jig the pcregrep tests with different newline settings in an attempt
848        to make them independent of the local environment's newline setting.
849    
850     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
851    
852     4. Some of the "internals" tests were previously cut out when the link size
853        was not 2, because the output contained actual offsets. The recent new
854        "Z" feature of pcretest means that these can be cut out, making the tests
855        usable with all link sizes.
856    
857     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
858        stack recursion. This gives a massive performance boost under BSD, but just
859        a small improvement under Linux. However, it saves one field in the frame
860        in all cases.
861    
862     6. Added more features from the forthcoming Perl 5.10:
863    
864        (a) (?-n) (where n is a string of digits) is a relative subroutine or
865            recursion call. It refers to the nth most recently opened parentheses.
866    
867        (b) (?+n) is also a relative subroutine call; it refers to the nth next
868            to be opened parentheses.
869    
870        (c) Conditions that refer to capturing parentheses can be specified
871            relatively, for example, (?(-2)... or (?(+3)...
872    
873        (d) \K resets the start of the current match so that everything before
874            is not part of it.
875    
876        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
877    
878        (f) \g{name} is another synonym - part of Perl 5.10's unification of
879            reference syntax.
880    
881        (g) (?| introduces a group in which the numbering of parentheses in each
882            alternative starts with the same number.
883    
884        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
885    
886     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
887        PCRE_INFO_JCHANGED.
888    
889     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
890        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
891        for detecting groups that can match an empty string.
892    
893     9. A pattern with a very large number of alternatives (more than several
894        hundred) was running out of internal workspace during the pre-compile
895        phase, where pcre_compile() figures out how much memory will be needed. A
896        bit of new cunning has reduced the workspace needed for groups with
897        alternatives. The 1000-alternative test pattern now uses 12 bytes of
898        workspace instead of running out of the 4096 that are available.
899    
900    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
901    
902    11. Applied patch from Google to remove an optimization that didn't quite work.
903        The report of the bug said:
904    
905          pcrecpp::RE("a*").FullMatch("aaa") matches, while
906          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
907          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
908    
909    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
910        it matched the wrong number of bytes.
911    
912    
913    Version 7.1 24-Apr-07
914    ---------------------
915    
916     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
917        that is more "standard", making use of automake and other Autotools. There
918        is some re-arrangement of the files and adjustment of comments consequent
919        on this.
920    
921     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
922        for recursive directory scanning broke on some systems because the files
923        are not scanned in any specific order and on different systems the order
924        was different. A call to "sort" has been inserted into RunGrepTest for the
925        approprate test as a short-term fix. In the longer term there may be an
926        alternative.
927    
928     3. I had an email from Eric Raymond about problems translating some of PCRE's
929        man pages to HTML (despite the fact that I distribute HTML pages, some
930        people do their own conversions for various reasons). The problems
931        concerned the use of low-level troff macros .br and .in. I have therefore
932        removed all such uses from the man pages (some were redundant, some could
933        be replaced by .nf/.fi pairs). The 132html script that I use to generate
934        HTML has been updated to handle .nf/.fi and to complain if it encounters
935        .br or .in.
936    
937     4. Updated comments in configure.ac that get placed in config.h.in and also
938        arranged for config.h to be included in the distribution, with the name
939        config.h.generic, for the benefit of those who have to compile without
940        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
941    
942     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
943        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
944        makevp.bat for the current PCRE, using the additional files
945        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
946    
947     6. A Windows user reported a minor discrepancy with test 2, which turned out
948        to be caused by a trailing space on an input line that had got lost in his
949        copy. The trailing space was an accident, so I've just removed it.
950    
951     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
952        that is needed.
953    
954     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
955        as "const" (a) because they are and (b) because it helps the PHP
956        maintainers who have recently made a script to detect big data structures
957        in the php code that should be moved to the .rodata section. I remembered
958        to update Builducptable as well, so it won't revert if ucptable.h is ever
959        re-created.
960    
961     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
962        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
963        order to be able to cut out the UTF-8 tables in the latter when UTF-8
964        support is not required. This saves 1.5-2K of code, which is important in
965        some applications.
966    
967        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
968        so as not to refer to the tables, even though these functions will never be
969        called when UTF-8 support is disabled. Otherwise there are problems with a
970        shared library.
971    
972    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
973    
974        (a) It was defining its arguments as char * instead of void *.
975    
976        (b) It was assuming that all moves were upwards in memory; this was true
977            a long time ago when I wrote it, but is no longer the case.
978    
979        The emulated memove() is provided for those environments that have neither
980        memmove() nor bcopy(). I didn't think anyone used it these days, but that
981        is clearly not the case, as these two bugs were recently reported.
982    
983    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
984        and Detrail to create the HTML documentation, the .txt form of the man
985        pages, and it removes trailing spaces from listed files. It also creates
986        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
987        case, it wraps all the #defines with #ifndefs. This script should be run
988        before "make dist".
989    
990    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
991        with Unicode property support.
992    
993        (a) For a maximizing quantifier, if the two different cases of the
994            character were of different lengths in their UTF-8 codings (there are
995            some cases like this - I found 11), and the matching function had to
996            back up over a mixture of the two cases, it incorrectly assumed they
997            were both the same length.
998    
999        (b) When PCRE was configured to use the heap rather than the stack for
1000            recursion during matching, it was not correctly preserving the data for
1001            the other case of a UTF-8 character when checking ahead for a match
1002            while processing a minimizing repeat. If the check also involved
1003            matching a wide character, but failed, corruption could cause an
1004            erroneous result when trying to check for a repeat of the original
1005            character.
1006    
1007    13. Some tidying changes to the testing mechanism:
1008    
1009        (a) The RunTest script now detects the internal link size and whether there
1010            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1011            values substituted by "configure". (The RunGrepTest script already did
1012            this for UTF-8.) The configure.ac script no longer substitutes the
1013            relevant variables.
1014    
1015        (b) The debugging options /B and /D in pcretest show the compiled bytecode
1016            with length and offset values. This means that the output is different
1017            for different internal link sizes. Test 2 is skipped for link sizes
1018            other than 2 because of this, bypassing the problem. Unfortunately,
1019            there was also a test in test 3 (the locale tests) that used /B and
1020            failed for link sizes other than 2. Rather than cut the whole test out,
1021            I have added a new /Z option to pcretest that replaces the length and
1022            offset values with spaces. This is now used to make test 3 independent
1023            of link size. (Test 2 will be tidied up later.)
1024    
1025    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1026        segmentation fault instead of returning the appropriate error message.
1027    
1028    15. In multiline mode when the newline sequence was set to "any", the pattern
1029        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1030        This doesn't seem right; it now treats the CRLF combination as the line
1031        ending, and so does not match in that case. It's only a pattern such as ^$
1032        that would hit this one: something like ^ABC$ would have failed after \r
1033        and then tried again after \r\n.
1034    
1035    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1036        in an attempt to make files that differ only in their line terminators
1037        compare equal. This works on Linux.
1038    
1039    17. Under certain error circumstances pcregrep might try to free random memory
1040        as it exited. This is now fixed, thanks to valgrind.
1041    
1042    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1043        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1044        was because its rules for how to advance for /g after matching an empty
1045        string at the end of a line did not allow for this case. They now check for
1046        it specially.
1047    
1048    20. pcretest is supposed to handle patterns and data of any length, by
1049        extending its buffers when necessary. It was getting this wrong when the
1050        buffer for a data line had to be extended.
1051    
1052    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1053        CRLF as a newline sequence.
1054    
1055    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1056        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1057        I have nevertheless tidied it up.
1058    
1059    23. Added some casts to kill warnings from HP-UX ia64 compiler.
1060    
1061    24. Added a man page for pcre-config.
1062    
1063    
1064    Version 7.0 19-Dec-06
1065    ---------------------
1066    
1067     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1068        moving to gcc 4.1.1.
1069    
1070     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1071        sys/time.h, which is documented as needed for this function. It doesn't
1072        seem to matter on Linux, but it showed up on some releases of OS X.
1073    
1074     3. It seems that there are systems where bytes whose values are greater than
1075        127 match isprint() in the "C" locale. The "C" locale should be the
1076        default when a C program starts up. In most systems, only ASCII printing
1077        characters match isprint(). This difference caused the output from pcretest
1078        to vary, making some of the tests fail. I have changed pcretest so that:
1079    
1080        (a) When it is outputting text in the compiled version of a pattern, bytes
1081            other than 32-126 are always shown as hex escapes.
1082    
1083        (b) When it is outputting text that is a matched part of a subject string,
1084            it does the same, unless a different locale has been set for the match
1085            (using the /L modifier). In this case, it uses isprint() to decide.
1086    
1087     4. Fixed a major bug that caused incorrect computation of the amount of memory
1088        required for a compiled pattern when options that changed within the
1089        pattern affected the logic of the preliminary scan that determines the
1090        length. The relevant options are -x, and -i in UTF-8 mode. The result was
1091        that the computed length was too small. The symptoms of this bug were
1092        either the PCRE error "internal error: code overflow" from pcre_compile(),
1093        or a glibc crash with a message such as "pcretest: free(): invalid next
1094        size (fast)". Examples of patterns that provoked this bug (shown in
1095        pcretest format) are:
1096    
1097          /(?-x: )/x
1098          /(?x)(?-x: \s*#\s*)/
1099          /((?i)[\x{c0}])/8
1100          /(?i:[\x{c0}])/8
1101    
1102        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1103        is now done differently.
1104    
1105     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1106        wrapper classes; (b) implement a new function in the C++ scanner that is
1107        more efficient than the old way of doing things because it avoids levels of
1108        recursion in the regex matching; (c) add a paragraph to the documentation
1109        for the FullMatch() function.
1110    
1111     6. The escape sequence \n was being treated as whatever was defined as
1112        "newline". Not only was this contrary to the documentation, which states
1113        that \n is character 10 (hex 0A), but it also went horribly wrong when
1114        "newline" was defined as CRLF. This has been fixed.
1115    
1116     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1117        was being set to -1 for the "end of line" case (supposedly a value that no
1118        character can have). Though this value is never used (the check for end of
1119        line is "zero bytes in current character"), it caused compiler complaints.
1120        I've changed it to 0xffffffff.
1121    
1122     8. In pcre_version.c, the version string was being built by a sequence of
1123        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1124        string (as it is for production releases) called a macro with an empty
1125        argument. The C standard says the result of this is undefined. The gcc
1126        compiler treats it as an empty string (which was what was wanted) but it is
1127        reported that Visual C gives an error. The source has been hacked around to
1128        avoid this problem.
1129    
1130     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1131        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1132        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1133        of them did).
1134    
1135    10. Originally, pcretest opened its input and output without "b"; then I was
1136        told that "b" was needed in some environments, so it was added for release
1137        5.0 to both the input and output. (It makes no difference on Unix-like
1138        systems.) Later I was told that it is wrong for the input on Windows. I've
1139        now abstracted the modes into two macros, to make it easier to fiddle with
1140        them, and removed "b" from the input mode under Windows.
1141    
1142    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1143    
1144    12. Added -help and --help to pcretest as an official way of being reminded
1145        of the options.
1146    
1147    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1148        and pcrecpp.cc because they annoy compilers at high warning levels.
1149    
1150    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1151    
1152    15. Fixed an occurrence of == in configure.ac that should have been = (shell
1153        scripts are not C programs :-) and which was not noticed because it works
1154        on Linux.
1155    
1156    16. pcretest is supposed to handle any length of pattern and data line (as one
1157        line or as a continued sequence of lines) by extending its input buffer if
1158        necessary. This feature was broken for very long pattern lines, leading to
1159        a string of junk being passed to pcre_compile() if the pattern was longer
1160        than about 50K.
1161    
1162    17. I have done a major re-factoring of the way pcre_compile() computes the
1163        amount of memory needed for a compiled pattern. Previously, there was code
1164        that made a preliminary scan of the pattern in order to do this. That was
1165        OK when PCRE was new, but as the facilities have expanded, it has become
1166        harder and harder to keep it in step with the real compile phase, and there
1167        have been a number of bugs (see for example, 4 above). I have now found a
1168        cunning way of running the real compile function in a "fake" mode that
1169        enables it to compute how much memory it would need, while actually only
1170        ever using a few hundred bytes of working memory and without too many
1171        tests of the mode. This should make future maintenance and development
1172        easier. A side effect of this work is that the limit of 200 on the nesting
1173        depth of parentheses has been removed (though this was never a serious
1174        limitation, I suspect). However, there is a downside: pcre_compile() now
1175        runs more slowly than before (30% or more, depending on the pattern). I
1176        hope this isn't a big issue. There is no effect on runtime performance.
1177    
1178    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1179        newline (only possible for the last line of a file) and it was a
1180        pattern that set a locale (followed by /Lsomething), pcretest crashed.
1181    
1182    19. Added additional timing features to pcretest. (1) The -tm option now times
1183        matching only, not compiling. (2) Both -t and -tm can be followed, as a
1184        separate command line item, by a number that specifies the number of
1185        repeats to use when timing. The default is 50000; this gives better
1186        precision, but takes uncomfortably long for very large patterns.
1187    
1188    20. Extended pcre_study() to be more clever in cases where a branch of a
1189        subpattern has no definite first character. For example, (a*|b*)[cd] would
1190        previously give no result from pcre_study(). Now it recognizes that the
1191        first character must be a, b, c, or d.
1192    
1193    21. There was an incorrect error "recursive call could loop indefinitely" if
1194        a subpattern (or the entire pattern) that was being tested for matching an
1195        empty string contained only one non-empty item after a nested subpattern.
1196        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1197        incorrectly, because the \d was being skipped in the check.
1198    
1199    22. The pcretest program now has a new pattern option /B and a command line
1200        option -b, which is equivalent to adding /B to every pattern. This causes
1201        it to show the compiled bytecode, without the additional information that
1202        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1203        is the same as /B/I).
1204    
1205    23. A new optimization is now able automatically to treat some sequences such
1206        as a*b as a*+b. More specifically, if something simple (such as a character
1207        or a simple class like \d) has an unlimited quantifier, and is followed by
1208        something that cannot possibly match the quantified thing, the quantifier
1209        is automatically "possessified".
1210    
1211    24. A recursive reference to a subpattern whose number was greater than 39
1212        went wrong under certain circumstances in UTF-8 mode. This bug could also
1213        have affected the operation of pcre_study().
1214    
1215    25. Realized that a little bit of performance could be had by replacing
1216        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1217    
1218    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1219    
1220    27. Possessive quantifiers such as a++ were previously implemented by turning
1221        them into atomic groups such as ($>a+). Now they have their own opcodes,
1222        which improves performance. This includes the automatically created ones
1223        from 23 above.
1224    
1225    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1226        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1227        the first matched character to be a colon. This applied both to named and
1228        numbered groups.
1229    
1230    29. The ucpinternal.h header file was missing its idempotency #ifdef.
1231    
1232    30. I was sent a "project" file called libpcre.a.dev which I understand makes
1233        building PCRE on Windows easier, so I have included it in the distribution.
1234    
1235    31. There is now a check in pcretest against a ridiculously large number being
1236        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1237        loop, the loop is abandoned.
1238    
1239    32. Forward references to subpatterns in conditions such as (?(2)...) where
1240        subpattern 2 is defined later cause pcre_compile() to search forwards in
1241        the pattern for the relevant set of parentheses. This search went wrong
1242        when there were unescaped parentheses in a character class, parentheses
1243        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1244    
1245    33. "Subroutine" calls and backreferences were previously restricted to
1246        referencing subpatterns earlier in the regex. This restriction has now
1247        been removed.
1248    
1249    34. Added a number of extra features that are going to be in Perl 5.10. On the
1250        whole, these are just syntactic alternatives for features that PCRE had
1251        previously implemented using the Python syntax or my own invention. The
1252        other formats are all retained for compatibility.
1253    
1254        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1255            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1256            also .NET compatible.
1257    
1258        (b) A recursion or subroutine call to a named group can now be defined as
1259            (?&name) as well as (?P>name).
1260    
1261        (c) A backreference to a named group can now be defined as \k<name> or
1262            \k'name' as well as (?P=name). The new forms, as well as being in Perl
1263            5.10, are also .NET compatible.
1264    
1265        (d) A conditional reference to a named group can now use the syntax
1266            (?(<name>) or (?('name') as well as (?(name).
1267    
1268        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1269            groups (named and numbered) that are never evaluated inline, but can be
1270            called as "subroutines" from elsewhere. In effect, the DEFINE condition
1271            is always false. There may be only one alternative in such a group.
1272    
1273        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1274            as the simple (?(R). The condition is true only if the most recent
1275            recursion is that of the given number or name. It does not search out
1276            through the entire recursion stack.
1277    
1278        (g) The escape \gN or \g{N} has been added, where N is a positive or
1279            negative number, specifying an absolute or relative reference.
1280    
1281    35. Tidied to get rid of some further signed/unsigned compiler warnings and
1282        some "unreachable code" warnings.
1283    
1284    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1285        things, this adds five new scripts.
1286    
1287    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1288        There were also incompatibilities regarding the handling of \Q..\E inside
1289        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1290        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1291    
1292    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1293        matches an empty string, and forcibly breaks the loop. There were bugs in
1294        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
1295        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
1296        separate and independent bugs (that affected different cases) have been
1297        fixed.
1298    
1299    39. Refactored the code to abolish the use of different opcodes for small
1300        capturing bracket numbers. This is a tidy that I avoided doing when I
1301        removed the limit on the number of capturing brackets for 3.5 back in 2001.
1302        The new approach is not only tidier, it makes it possible to reduce the
1303        memory needed to fix the previous bug (38).
1304    
1305    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1306        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1307        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1308        mode.
1309    
1310    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1311        report.
1312    
1313    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1314        copy construction and assignment in the C++ wrapper.
1315    
1316    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1317        couple of bugs that could have given wrong results in the "--newline=crlf"
1318        case.
1319    
1320    44. Added a number of casts and did some reorganization of signed/unsigned int
1321        variables following suggestions from Dair Grant. Also renamed the variable
1322        "this" as "item" because it is a C++ keyword.
1323    
1324    45. Arranged for dftables to add
1325    
1326          #include "pcre_internal.h"
1327    
1328        to pcre_chartables.c because without it, gcc 4.x may remove the array
1329        definition from the final binary if PCRE is built into a static library and
1330        dead code stripping is activated.
1331    
1332    46. For an unanchored pattern, if a match attempt fails at the start of a
1333        newline sequence, and the newline setting is CRLF or ANY, and the next two
1334        characters are CRLF, advance by two characters instead of one.
1335    
1336    
1337    Version 6.7 04-Jul-06
1338    ---------------------
1339    
1340     1. In order to handle tests when input lines are enormously long, pcretest has
1341        been re-factored so that it automatically extends its buffers when
1342        necessary. The code is crude, but this _is_ just a test program. The
1343        default size has been increased from 32K to 50K.
1344    
1345     2. The code in pcre_study() was using the value of the re argument before
1346        testing it for NULL. (Of course, in any sensible call of the function, it
1347        won't be NULL.)
1348    
1349     3. The memmove() emulation function in pcre_internal.h, which is used on
1350        systems that lack both memmove() and bcopy() - that is, hardly ever -
1351        was missing a "static" storage class specifier.
1352    
1353     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1354        containing an extended class (one that cannot be represented by a bitmap
1355        because it contains high-valued characters or Unicode property items, e.g.
1356        [\pZ]). Almost always one would set UTF-8 mode when processing such a
1357        pattern, but PCRE should not loop if you do not (it no longer does).
1358        [Detail: two cases were found: (a) a repeated subpattern containing an
1359        extended class; (b) a recursive reference to a subpattern that followed a
1360        previous extended class. It wasn't skipping over the extended class
1361        correctly when UTF-8 mode was not set.]
1362    
1363     5. A negated single-character class was not being recognized as fixed-length
1364        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1365        compile error "lookbehind assertion is not fixed length".
1366    
1367     6. The RunPerlTest auxiliary script was showing an unexpected difference
1368        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1369        write a Perl script that can interpret lines of an input file either as
1370        byte characters or as UTF-8, which is what "perltest" was being required to
1371        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1372        can't do is switch easily at run time between having the "use utf8;" pragma
1373        or not. In the end, I fudged it by using the RunPerlTest script to insert
1374        "use utf8;" explicitly for the UTF-8 tests.
1375    
1376     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1377        the end of the subject string, contrary to the documentation and to what
1378        Perl does. This was true of both matching functions. Now it matches only at
1379        the start of the subject and immediately after *internal* newlines.
1380    
1381     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1382        a pointer to an int instead of a pointer to an unsigned long int. This
1383        caused problems on 64-bit systems.
1384    
1385     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1386        instance of the 'standard' template library not being so standard".
1387    
1388    10. There was no check on the number of named subpatterns nor the maximum
1389        length of a subpattern name. The product of these values is used to compute
1390        the size of the memory block for a compiled pattern. By supplying a very
1391        long subpattern name and a large number of named subpatterns, the size
1392        computation could be caused to overflow. This is now prevented by limiting
1393        the length of names to 32 characters, and the number of named subpatterns
1394        to 10,000.
1395    
1396    11. Subpatterns that are repeated with specific counts have to be replicated in
1397        the compiled pattern. The size of memory for this was computed from the
1398        length of the subpattern and the repeat count. The latter is limited to
1399        65535, but there was no limit on the former, meaning that integer overflow
1400        could in principle occur. The compiled length of a repeated subpattern is
1401        now limited to 30,000 bytes in order to prevent this.
1402    
1403    12. Added the optional facility to have named substrings with the same name.
1404    
1405    13. Added the ability to use a named substring as a condition, using the
1406        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1407        are numbers (not recommended). Forward references are permitted.
1408    
1409    14. Added forward references in named backreferences (if you see what I mean).
1410    
1411    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1412        pattern could run off the end of the subject. For example, the pattern
1413        "(?s)(.{1,5})"8 did this with the subject "ab".
1414    
1415    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1416        PCRE_CASELESS was set when matching characters that were quantified with ?
1417        or *.
1418    
1419    17. A character class other than a single negated character that had a minimum
1420        but no maximum quantifier - for example [ab]{6,} - was not handled
1421        correctly by pce_dfa_exec(). It would match only one character.
1422    
1423    18. A valid (though odd) pattern that looked like a POSIX character
1424        class but used an invalid character after [ (for example [[,abc,]]) caused
1425        pcre_compile() to give the error "Failed: internal error: code overflow" or
1426        in some cases to crash with a glibc free() error. This could even happen if
1427        the pattern terminated after [[ but there just happened to be a sequence of
1428        letters, a binary zero, and a closing ] in the memory that followed.
1429    
1430    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1431        over the years. Originally (before any Unicode support), just the bottom 8
1432        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1433        output from "man perlunicode" includes this:
1434    
1435          The regular expression compiler produces polymorphic opcodes.  That
1436          is, the pattern adapts to the data and automatically switches to
1437          the Unicode character scheme when presented with Unicode data--or
1438          instead uses a traditional byte scheme when presented with byte
1439          data.
1440    
1441        Sadly, a wide octal escape does not cause a switch, and in a string with
1442        no other multibyte characters, these octal escapes are treated as before.
1443        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
1444        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1445        Unicode string.
1446    
1447        I have not perpetrated such confusion in PCRE. Up till now, it took just
1448        the bottom 8 bits, as in old Perl. I have now made octal escapes with
1449        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1450        translate to the appropriate multibyte character.
1451    
1452    29. Applied some refactoring to reduce the number of warnings from Microsoft
1453        and Borland compilers. This has included removing the fudge introduced
1454        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1455        a warning about an unused variable.
1456    
1457    21. PCRE has not included VT (character 0x0b) in the set of whitespace
1458        characters since release 4.0, because Perl (from release 5.004) does not.
1459        [Or at least, is documented not to: some releases seem to be in conflict
1460        with the documentation.] However, when a pattern was studied with
1461        pcre_study() and all its branches started with \s, PCRE still included VT
1462        as a possible starting character. Of course, this did no harm; it just
1463        caused an unnecessary match attempt.
1464    
1465    22. Removed a now-redundant internal flag bit that recorded the fact that case
1466        dependency changed within the pattern. This was once needed for "required
1467        byte" processing, but is no longer used. This recovers a now-scarce options
1468        bit. Also moved the least significant internal flag bit to the most-
1469        significant bit of the word, which was not previously used (hangover from
1470        the days when it was an int rather than a uint) to free up another bit for
1471        the future.
1472    
1473    23. Added support for CRLF line endings as well as CR and LF. As well as the
1474        default being selectable at build time, it can now be changed at runtime
1475        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1476        specify that it is scanning data with non-default line endings.
1477    
1478    24. Changed the definition of CXXLINK to make it agree with the definition of
1479        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1480    
1481    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1482        recursions. This makes a big different to stack usage for some patterns.
1483    
1484    26. If a subpattern containing a named recursion or subroutine reference such
1485        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1486        the space required for the compiled pattern went wrong and gave too small a
1487        value. Depending on the environment, this could lead to "Failed: internal
1488        error: code overflow at offset 49" or "glibc detected double free or
1489        corruption" errors.
1490    
1491    27. Applied patches from Google (a) to support the new newline modes and (b) to
1492        advance over multibyte UTF-8 characters in GlobalReplace.
1493    
1494    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1495        difference for some implementation of PCRE in some Windows version.
1496    
1497    29. Added some extra testing facilities to pcretest:
1498    
1499        \q<number>   in a data line sets the "match limit" value
1500        \Q<number>   in a data line sets the "match recursion limt" value
1501        -S <number>  sets the stack size, where <number> is in megabytes
1502    
1503        The -S option isn't available for Windows.
1504    
1505    
1506    Version 6.6 06-Feb-06
1507    ---------------------
1508    
1509     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1510        in pcreposix.h. I have copied the definition from pcre.h.
1511    
1512     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1513        because pcre.h is no longer a built file.
1514    
1515     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1516        not normally included in the compiled code.
1517    
1518    
1519    Version 6.5 01-Feb-06
1520    ---------------------
1521    
1522     1. When using the partial match feature with pcre_dfa_exec(), it was not
1523        anchoring the second and subsequent partial matches at the new starting
1524        point. This could lead to incorrect results. For example, with the pattern
1525        /1234/, partially matching against "123" and then "a4" gave a match.
1526    
1527     2. Changes to pcregrep:
1528    
1529        (a) All non-match returns from pcre_exec() were being treated as failures
1530            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1531            error message is output. Some extra information is given for the
1532            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
1533            probably the only errors that are likely to be caused by users (by
1534            specifying a regex that has nested indefinite repeats, for instance).
1535            If there are more than 20 of these errors, pcregrep is abandoned.
1536    
1537        (b) A binary zero was treated as data while matching, but terminated the
1538            output line if it was written out. This has been fixed: binary zeroes
1539            are now no different to any other data bytes.
1540    
1541        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1542            used to set a locale for matching. The --locale=xxxx long option has
1543            been added (no short equivalent) to specify a locale explicitly on the
1544            pcregrep command, overriding the environment variables.
1545    
1546        (d) When -B was used with -n, some line numbers in the output were one less
1547            than they should have been.
1548    
1549        (e) Added the -o (--only-matching) option.
1550    
1551        (f) If -A or -C was used with -c (count only), some lines of context were
1552            accidentally printed for the final match.
1553    
1554        (g) Added the -H (--with-filename) option.
1555    
1556        (h) The combination of options -rh failed to suppress file names for files
1557            that were found from directory arguments.
1558    
1559        (i) Added the -D (--devices) and -d (--directories) options.
1560    
1561        (j) Added the -F (--fixed-strings) option.
1562    
1563        (k) Allow "-" to be used as a file name for -f as well as for a data file.
1564    
1565        (l) Added the --colo(u)r option.
1566    
1567        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1568            is not present by default.
1569    
1570     3. A nasty bug was discovered in the handling of recursive patterns, that is,
1571        items such as (?R) or (?1), when the recursion could match a number of
1572        alternatives. If it matched one of the alternatives, but subsequently,
1573        outside the recursion, there was a failure, the code tried to back up into
1574        the recursion. However, because of the way PCRE is implemented, this is not
1575        possible, and the result was an incorrect result from the match.
1576    
1577        In order to prevent this happening, the specification of recursion has
1578        been changed so that all such subpatterns are automatically treated as
1579        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1580    
1581     4. I had overlooked the fact that, in some locales, there are characters for
1582        which isalpha() is true but neither isupper() nor islower() are true. In
1583        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1584        and ordfeminine) are like this. This affected the treatment of \w and \W
1585        when they appeared in character classes, but not when they appeared outside
1586        a character class. The bit map for "word" characters is now created
1587        separately from the results of isalnum() instead of just taking it from the
1588        upper, lower, and digit maps. (Plus the underscore character, of course.)
1589    
1590     5. The above bug also affected the handling of POSIX character classes such as
1591        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1592        permanent tables. Instead, the bit maps for such a class were previously
1593        created as the appropriate unions of the upper, lower, and digit bitmaps.
1594        Now they are created by subtraction from the [[:word:]] class, which has
1595        its own bitmap.
1596    
1597     6. The [[:blank:]] character class matches horizontal, but not vertical space.
1598        It is created by subtracting the vertical space characters (\x09, \x0a,
1599        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1600        subtraction was done in the overall bitmap for a character class, meaning
1601        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1602        be recognized. This bug has been fixed.
1603    
1604     7. Patches from the folks at Google:
1605    
1606          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1607          real life, but is still worth protecting against".
1608    
1609          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1610          regular expressions".
1611    
1612          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1613          have it.
1614    
1615          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1616          "configure" and the latter not, in order to fix a problem somebody had
1617          with compiling the Arg class on HP-UX.
1618    
1619          (e) Improve the error-handling of the C++ wrapper a little bit.
1620    
1621          (f) New tests for checking recursion limiting.
1622    
1623     8. The pcre_memmove() function, which is used only if the environment does not
1624        have a standard memmove() function (and is therefore rarely compiled),
1625        contained two bugs: (a) use of int instead of size_t, and (b) it was not
1626        returning a result (though PCRE never actually uses the result).
1627    
1628     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1629        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1630        returned instead of calling malloc() with an overflowing number that would
1631        most likely cause subsequent chaos.
1632    
1633    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1634    
1635    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1636        with this option is matched, the nmatch and pmatch options of regexec() are
1637        ignored.
1638    
1639    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1640        provided in case anyone wants to the the POSIX interface with UTF-8
1641        strings.
1642    
1643    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1644        C++ linking (needed for some HP-UX environments).
1645    
1646    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1647        (unused parameter) and in the pcre_printint() function (omitted "default"
1648        switch label when the default is to do nothing).
1649    
1650    15. Added some code to make it possible, when PCRE is compiled as a C++
1651        library, to replace subject pointers for pcre_exec() with a smart pointer
1652        class, thus making it possible to process discontinuous strings.
1653    
1654    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1655        much the same function. They were added by different people who were trying
1656        to make PCRE easy to compile on non-Unix systems. It has been suggested
1657        that PCRE_EXPORT be abolished now that there is more automatic apparatus
1658        for compiling on Windows systems. I have therefore replaced it with
1659        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1660        defaults to "extern" for C or "extern C" for C++, which works fine on
1661        Unix-like systems. It is now possible to override the value of PCRE_DATA_
1662        SCOPE with something explicit in config.h. In addition:
1663    
1664        (a) pcreposix.h still had just "extern" instead of either of these macros;
1665            I have replaced it with PCRE_DATA_SCOPE.
1666    
1667        (b) Functions such as _pcre_xclass(), which are internal to the library,
1668            but external in the C sense, all had PCRE_EXPORT in their definitions.
1669            This is apparently wrong for the Windows case, so I have removed it.
1670            (It makes no difference on Unix-like systems.)
1671    
1672    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1673        of recursive calls to match(). This is different to MATCH_LIMIT because
1674        that limits the total number of calls to match(), not all of which increase
1675        the depth of recursion. Limiting the recursion depth limits the amount of
1676        stack (or heap if NO_RECURSE is set) that is used. The default can be set
1677        when PCRE is compiled, and changed at run time. A patch from Google adds
1678        this functionality to the C++ interface.
1679    
1680    18. Changes to the handling of Unicode character properties:
1681    
1682        (a) Updated the table to Unicode 4.1.0.
1683    
1684        (b) Recognize characters that are not in the table as "Cn" (undefined).
1685    
1686        (c) I revised the way the table is implemented to a much improved format
1687            which includes recognition of ranges. It now supports the ranges that
1688            are defined in UnicodeData.txt, and it also amalgamates other
1689            characters into ranges. This has reduced the number of entries in the
1690            table from around 16,000 to around 3,000, thus reducing its size
1691            considerably. I realized I did not need to use a tree structure after
1692            all - a binary chop search is just as efficient. Having reduced the
1693            number of entries, I extended their size from 6 bytes to 8 bytes to
1694            allow for more data.
1695    
1696        (d) Added support for Unicode script names via properties such as \p{Han}.
1697    
1698    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1699        matching that character.
1700    
1701    20. When matching a repeated Unicode property with a minimum greater than zero,
1702        (for example \pL{2,}), PCRE could look past the end of the subject if it
1703        reached it while seeking the minimum number of characters. This could
1704        happen only if some of the characters were more than one byte long, because
1705        there is a check for at least the minimum number of bytes.
1706    
1707    21. Refactored the implementation of \p and \P so as to be more general, to
1708        allow for more different types of property in future. This has changed the
1709        compiled form incompatibly. Anybody with saved compiled patterns that use
1710        \p or \P will have to recompile them.
1711    
1712    22. Added "Any" and "L&" to the supported property types.
1713    
1714    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1715        but give a compile time error if the value is greater than 0xff.
1716    
1717    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1718        accidentally not being installed or uninstalled.
1719    
1720    25. The pcre.h file was built from pcre.h.in, but the only changes that were
1721        made were to insert the current release number. This seemed silly, because
1722        it made things harder for people building PCRE on systems that don't run
1723        "configure". I have turned pcre.h into a distributed file, no longer built
1724        by "configure", with the version identification directly included. There is
1725        no longer a pcre.h.in file.
1726    
1727        However, this change necessitated a change to the pcre-config script as
1728        well. It is built from pcre-config.in, and one of the substitutions was the
1729        release number. I have updated configure.ac so that ./configure now finds
1730        the release number by grepping pcre.h.
1731    
1732    26. Added the ability to run the tests under valgrind.
1733    
1734    
1735    Version 6.4 05-Sep-05
1736    ---------------------
1737    
1738     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1739        "--" to be printed when multiple files were scanned, even when none of the
1740        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1741        consider it to be a bug, and have restored the previous behaviour.
1742    
1743     2. A couple of code tidies to get rid of compiler warnings.
1744    
1745     3. The pcretest program used to cheat by referring to symbols in the library
1746        whose names begin with _pcre_. These are internal symbols that are not
1747        really supposed to be visible externally, and in some environments it is
1748        possible to suppress them. The cheating is now confined to including
1749        certain files from the library's source, which is a bit cleaner.
1750    
1751     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1752        file's purpose clearer.
1753    
1754     5. Reorganized pcre_ucp_findchar().
1755    
1756    
1757    Version 6.3 15-Aug-05
1758    ---------------------
1759    
1760     1. The file libpcre.pc.in did not have general read permission in the tarball.
1761    
1762     2. There were some problems when building without C++ support:
1763    
1764        (a) If C++ support was not built, "make install" and "make test" still
1765            tried to test it.
1766    
1767        (b) There were problems when the value of CXX was explicitly set. Some
1768            changes have been made to try to fix these, and ...
1769    
1770        (c) --disable-cpp can now be used to explicitly disable C++ support.
1771    
1772        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1773            backslash in a target when C++ was disabled. This confuses some
1774            versions of "make", apparently. Using an intermediate variable solves
1775            this. (Same for CPP_LOBJ.)
1776    
1777     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1778        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1779        necessary on certain architectures.
1780    
1781     4. Added a setting of -export-symbols-regex to the link command to remove
1782        those symbols that are exported in the C sense, but actually are local
1783        within the library, and not documented. Their names all begin with
1784        "_pcre_". This is not a perfect job, because (a) we have to except some
1785        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1786        available (and never for static libraries). I have made a note to try to
1787        find a way round (a) in the future.
1788    
1789    
1790    Version 6.2 01-Aug-05
1791    ---------------------
1792    
1793     1. There was no test for integer overflow of quantifier values. A construction
1794        such as {1111111111111111} would give undefined results. What is worse, if
1795        a minimum quantifier for a parenthesized subpattern overflowed and became
1796        negative, the calculation of the memory size went wrong. This could have
1797        led to memory overwriting.
1798    
1799     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1800    
1801     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1802        operating environments where this matters.
1803    
1804     4. Applied Giuseppe Maxia's patch to add additional features for controlling
1805        PCRE options from within the C++ wrapper.
1806    
1807     5. Named capturing subpatterns were not being correctly counted when a pattern
1808        was compiled. This caused two problems: (a) If there were more than 100
1809        such subpatterns, the calculation of the memory needed for the whole
1810        compiled pattern went wrong, leading to an overflow error. (b) Numerical
1811        back references of the form \12, where the number was greater than 9, were
1812        not recognized as back references, even though there were sufficient
1813        previous subpatterns.
1814    
1815     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1816        versions of gcc, e.g. 2.95.4.
1817    
1818    
1819    Version 6.1 21-Jun-05
1820    ---------------------
1821    
1822     1. There was one reference to the variable "posix" in pcretest.c that was not
1823        surrounded by "#if !defined NOPOSIX".
1824    
1825     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1826        the cross-check on the old pcre_info() function, for the benefit of the
1827        cut-down version of PCRE that is currently imported into Exim.
1828    
1829     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1830        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1831        patterns that start (?i) but I don't think that matters. The use of (?i) is
1832        just an example; this all applies to the other options as well.
1833    
1834     4. Since libtool seems to echo the compile commands it is issuing, the output
1835        from "make" can be reduced a bit by putting "@" in front of each libtool
1836        compile command.
1837    
1838     5. Patch from the folks at Google for configure.in to be a bit more thorough
1839        in checking for a suitable C++ installation before trying to compile the
1840        C++ stuff. This should fix a reported problem when a compiler was present,
1841        but no suitable headers.
1842    
1843     6. The man pages all had just "PCRE" as their title. I have changed them to
1844        be the relevant file name. I have also arranged that these names are
1845        retained in the file doc/pcre.txt, which is a concatenation in text format
1846        of all the man pages except the little individual ones for each function.
1847    
1848     7. The NON-UNIX-USE file had not been updated for the different set of source
1849        files that come with release 6. I also added a few comments about the C++
1850        wrapper.
1851    
1852    
1853    Version 6.0 07-Jun-05
1854    ---------------------
1855    
1856     1. Some minor internal re-organization to help with my DFA experiments.
1857    
1858     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1859        didn't matter for the library itself when fully configured, but did matter
1860        when compiling without UCP support, or within Exim, where the ucp files are
1861        not imported.
1862    
1863     3. Refactoring of the library code to split up the various functions into
1864        different source modules. The addition of the new DFA matching code (see
1865        below) to a single monolithic source would have made it really too
1866        unwieldy, quite apart from causing all the code to be include in a
1867        statically linked application, when only some functions are used. This is
1868        relevant even without the DFA addition now that patterns can be compiled in
1869        one application and matched in another.
1870    
1871        The downside of splitting up is that there have to be some external
1872        functions and data tables that are used internally in different modules of
1873        the library but which are not part of the API. These have all had their
1874        names changed to start with "_pcre_" so that they are unlikely to clash
1875        with other external names.
1876    
1877     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1878        a different (DFA) algorithm. Although it is slower than the original
1879        function, it does have some advantages for certain types of matching
1880        problem.
1881    
1882     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1883        including restarting after a partial match.
1884    
1885     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1886        defined when compiling for Windows was sent to me. I have put it into the
1887        code, though I have no means of testing or verifying it.
1888    
1889     7. Added the pcre_refcount() auxiliary function.
1890    
1891     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1892        match before or at the first newline in the subject string. In pcretest,
1893        the /f option on a pattern can be used to set this.
1894    
1895     9. A repeated \w when used in UTF-8 mode with characters greater than 256
1896        would behave wrongly. This has been present in PCRE since release 4.0.
1897    
1898    10. A number of changes to the pcregrep command:
1899    
1900        (a) Refactored how -x works; insert ^(...)$ instead of setting
1901            PCRE_ANCHORED and checking the length, in preparation for adding
1902            something similar for -w.
1903    
1904        (b) Added the -w (match as a word) option.
1905    
1906        (c) Refactored the way lines are read and buffered so as to have more
1907            than one at a time available.
1908    
1909        (d) Implemented a pcregrep test script.
1910    
1911        (e) Added the -M (multiline match) option. This allows patterns to match
1912            over several lines of the subject. The buffering ensures that at least
1913            8K, or the rest of the document (whichever is the shorter) is available
1914            for matching (and similarly the previous 8K for lookbehind assertions).
1915    
1916        (f) Changed the --help output so that it now says
1917    
1918              -w, --word-regex(p)
1919    
1920            instead of two lines, one with "regex" and the other with "regexp"
1921            because that confused at least one person since the short forms are the
1922            same. (This required a bit of code, as the output is generated
1923            automatically from a table. It wasn't just a text change.)
1924    
1925        (g) -- can be used to terminate pcregrep options if the next thing isn't an
1926            option but starts with a hyphen. Could be a pattern or a path name
1927            starting with a hyphen, for instance.
1928    
1929        (h) "-" can be given as a file name to represent stdin.
1930    
1931        (i) When file names are being printed, "(standard input)" is used for
1932            the standard input, for compatibility with GNU grep. Previously
1933            "<stdin>" was used.
1934    
1935        (j) The option --label=xxx can be used to supply a name to be used for
1936            stdin when file names are being printed. There is no short form.
1937    
1938        (k) Re-factored the options decoding logic because we are going to add
1939            two more options that take data. Such options can now be given in four
1940            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1941    
1942        (l) Added the -A, -B, and -C options for requesting that lines of context
1943            around matches be printed.
1944    
1945        (m) Added the -L option to print the names of files that do not contain
1946            any matching lines, that is, the complement of -l.
1947    
1948        (n) The return code is 2 if any file cannot be opened, but pcregrep does
1949            continue to scan other files.
1950    
1951        (o) The -s option was incorrectly implemented. For compatibility with other
1952            greps, it now suppresses the error message for a non-existent or non-
1953            accessible file (but not the return code). There is a new option called
1954            -q that suppresses the output of matching lines, which was what -s was
1955            previously doing.
1956    
1957        (p) Added --include and --exclude options to specify files for inclusion
1958            and exclusion when recursing.
1959    
1960    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1961        Hopefully, it now does.
1962    
1963    12. Missing cast in pcre_study().
1964    
1965    13. Added an "uninstall" target to the makefile.
1966    
1967    14. Replaced "extern" in the function prototypes in Makefile.in with
1968        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1969        world, but is set differently for Windows.
1970    
1971    15. Added a second compiling function called pcre_compile2(). The only
1972        difference is that it has an extra argument, which is a pointer to an
1973        integer error code. When there is a compile-time failure, this is set
1974        non-zero, in addition to the error test pointer being set to point to an
1975        error message. The new argument may be NULL if no error number is required
1976        (but then you may as well call pcre_compile(), which is now just a
1977        wrapper). This facility is provided because some applications need a
1978        numeric error indication, but it has also enabled me to tidy up the way
1979        compile-time errors are handled in the POSIX wrapper.
1980    
1981    16. Added VPATH=.libs to the makefile; this should help when building with one
1982        prefix path and installing with another. (Or so I'm told by someone who
1983        knows more about this stuff than I do.)
1984    
1985    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1986        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1987        match everything, including newlines. This is not POSIX-compatible, but
1988        somebody wanted the feature. From pcretest it can be activated by using
1989        both the P and the s flags.
1990    
1991    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1992    
1993    19. libpcre.pc was being incorrectly installed as executable.
1994    
1995    20. A couple of places in pcretest check for end-of-line by looking for '\n';
1996        it now also looks for '\r' so that it will work unmodified on Windows.
1997    
1998    21. Added Google's contributed C++ wrapper to the distribution.
1999    
2000    22. Added some untidy missing memory free() calls in pcretest, to keep
2001        Electric Fence happy when testing.
2002    
2003    
2004    
2005    Version 5.0 13-Sep-04
2006    ---------------------
2007    
2008     1. Internal change: literal characters are no longer packed up into items
2009        containing multiple characters in a single byte-string. Each character
2010        is now matched using a separate opcode. However, there may be more than one
2011        byte in the character in UTF-8 mode.
2012    
2013     2. The pcre_callout_block structure has two new fields: pattern_position and
2014        next_item_length. These contain the offset in the pattern to the next match
2015        item, and its length, respectively.
2016    
2017     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
2018        insertion of callouts before each pattern item. Added the /C option to
2019        pcretest to make use of this.
2020    
2021     4. On the advice of a Windows user, the lines
2022    
2023          #if defined(_WIN32) || defined(WIN32)
2024          _setmode( _fileno( stdout ), 0x8000 );
2025          #endif  /* defined(_WIN32) || defined(WIN32) */
2026    
2027        have been added to the source of pcretest. This apparently does useful
2028        magic in relation to line terminators.
2029    
2030     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
2031        for the benefit of those environments where the "b" makes a difference.
2032    
2033     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
2034        to know about it. I have put a hack into configure.in that adds in code
2035        to set GCC=yes if CC=icc. This seems to end up at a point in the
2036        generated configure script that is early enough to affect the setting of
2037        compiler options, which is what is needed, but I have no means of testing
2038        whether it really works. (The user who reported this had patched the
2039        generated configure script, which of course I cannot do.)
2040    
2041        LATER: After change 22 below (new libtool files), the configure script
2042        seems to know about icc (and also ecc). Therefore, I have commented out
2043        this hack in configure.in.
2044    
2045     7. Added support for pkg-config (2 patches were sent in).
2046    
2047     8. Negated POSIX character classes that used a combination of internal tables
2048        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2049        [[:^ascii]]. Typically, they would match almost any characters. The other
2050        POSIX classes were not broken in this way.
2051    
2052     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2053        to find the match, as PCRE was deluded into thinking that the match had to
2054        start at the start point or following a newline. The same bug applied to
2055        patterns with negative forward assertions or any backward assertions
2056        preceding ".*" at the start, unless the pattern required a fixed first
2057        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2058    
2059    10. In UTF-8 mode, when moving forwards in the subject after a failed match
2060        starting at the last subject character, bytes beyond the end of the subject
2061        string were read.
2062    
2063    11. Renamed the variable "class" as "classbits" to make life easier for C++
2064        users. (Previously there was a macro definition, but it apparently wasn't
2065        enough.)
2066    
2067    12. Added the new field "tables" to the extra data so that tables can be passed
2068        in at exec time, or the internal tables can be re-selected. This allows
2069        a compiled regex to be saved and re-used at a later time by a different
2070        program that might have everything at different addresses.
2071    
2072    13. Modified the pcre-config script so that, when run on Solaris, it shows a
2073        -R library as well as a -L library.
2074    
2075    14. The debugging options of pcretest (-d on the command line or D on a
2076        pattern) showed incorrect output for anything following an extended class
2077        that contained multibyte characters and which was followed by a quantifier.
2078    
2079    15. Added optional support for general category Unicode character properties
2080        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2081        support. It adds about 90K to the size of the library. The meanings of the
2082        inbuilt class escapes such as \d and \s have NOT been changed.
2083    
2084    16. Updated pcredemo.c to include calls to free() to release the memory for the
2085        compiled pattern.
2086    
2087    17. The generated file chartables.c was being created in the source directory
2088        instead of in the building directory. This caused the build to fail if the
2089        source directory was different from the building directory, and was
2090        read-only.
2091    
2092    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2093        file. No doubt somebody will tell me if they don't make sense... Also added
2094        Dan Mooney's comments about building on OpenVMS.
2095    
2096    19. Added support for partial matching via the PCRE_PARTIAL option for
2097        pcre_exec() and the \P data escape in pcretest.
2098    
2099    20. Extended pcretest with 3 new pattern features:
2100    
2101        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
2102              write the compiled pattern to the file whose name is "rest-of-line".
2103              This is a straight binary dump of the data, with the saved pointer to
2104              the character tables forced to be NULL. The study data, if any, is
2105              written too. After writing, pcretest reads a new pattern.
2106    
2107        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2108              compiled pattern from the given file. There must not be any
2109              occurrences of "<" in the file name (pretty unlikely); if there are,
2110              pcretest will instead treat the initial "<" as a pattern delimiter.
2111              After reading in the pattern, pcretest goes on to read data lines as
2112              usual.
2113    
2114        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2115              and 16-bit fields in a compiled pattern, to simulate a pattern that
2116              was compiled on a host of opposite endianness.
2117    
2118    21. The pcre-exec() function can now cope with patterns that were compiled on
2119        hosts of opposite endianness, with this restriction:
2120    
2121          As for any compiled expression that is saved and used later, the tables
2122          pointer field cannot be preserved; the extra_data field in the arguments
2123          to pcre_exec() should be used to pass in a tables address if a value
2124          other than the default internal tables were used at compile time.
2125    
2126    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2127        now diagnosed as an error. Previously, most of the time, a negative number
2128        would have been treated as zero, but if in addition "ovector" was passed as
2129        NULL, a crash could occur.
2130    
2131    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2132        new versions from the libtool 1.5 distribution (the last one is a copy of
2133        a file called libtool.m4). This seems to have fixed the need to patch
2134        "configure" to support Darwin 1.3 (which I used to do). However, I still
2135        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2136        workstation).
2137    
2138    24. Changed the PCRE licence to be the more standard "BSD" licence.
2139    
2140    
2141    Version 4.5 01-Dec-03
2142    ---------------------
2143    
2144     1. There has been some re-arrangement of the code for the match() function so
2145        that it can be compiled in a version that does not call itself recursively.
2146        Instead, it keeps those local variables that need separate instances for
2147        each "recursion" in a frame on the heap, and gets/frees frames whenever it
2148        needs to "recurse". Keeping track of where control must go is done by means
2149        of setjmp/longjmp. The whole thing is implemented by a set of macros that
2150        hide most of the details from the main code, and operates only if
2151        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2152        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2153        operating.
2154    
2155        To make it easier for callers to provide specially tailored get/free
2156        functions for this usage, two new functions, pcre_stack_malloc, and
2157        pcre_stack_free, are used. They are always called in strict stacking order,
2158        and the size of block requested is always the same.
2159    
2160        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2161        PCRE has been compiled to use the stack or the heap for recursion. The
2162        -C option of pcretest uses this to show which version is compiled.
2163    
2164        A new data escape \S, is added to pcretest; it causes the amounts of store
2165        obtained and freed by both kinds of malloc/free at match time to be added
2166        to the output.
2167    
2168     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2169        what's available on my current Linux desktop machine.
2170    
2171     3. When matching a UTF-8 string, the test for a valid string at the start has
2172        been extended. If start_offset is not zero, PCRE now checks that it points
2173        to a byte that is the start of a UTF-8 character. If not, it returns
2174        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2175        this is necessary because there may be backward assertions in the pattern.
2176        When matching the same subject several times, it may save resources to use
2177        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2178    
2179     4. The code for checking the validity of UTF-8 strings has been tightened so
2180        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2181        containing "overlong sequences".
2182    
2183     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2184        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2185        should have been "&", but it just so happened that all the cases this let
2186        through by mistake were picked up later in the function.
2187    
2188     6. I had used a variable called "isblank" - this is a C99 function, causing
2189        some compilers to warn. To avoid this, I renamed it (as "blankclass").
2190    
2191     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2192        prompting; (b) run "./pcretest /dev/null" at the start of the test script
2193        so the version is shown; (c) stop "make test" echoing "./RunTest".
2194    
2195     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2196    
2197     9. The prototype for memmove() for systems that don't have it was using
2198        size_t, but the inclusion of the header that defines size_t was later. I've
2199        moved the #includes for the C headers earlier to avoid this.
2200    
2201    10. Added some adjustments to the code to make it easier to compiler on certain
2202        special systems:
2203    
2204          (a) Some "const" qualifiers were missing.
2205          (b) Added the macro EXPORT before all exported functions; by default this
2206              is defined to be empty.
2207          (c) Changed the dftables auxiliary program (that builds chartables.c) so
2208              that it reads its output file name as an argument instead of writing
2209              to the standard output and assuming this can be redirected.
2210    
2211    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2212        class containing characters with values greater than 255, PCRE compilation
2213        went into a loop.
2214    
2215    12. A recursive reference to a subpattern that was within another subpattern
2216        that had a minimum quantifier of zero caused PCRE to crash. For example,
2217        (x(y(?2))z)? provoked this bug with a subject that got as far as the
2218        recursion. If the recursively-called subpattern itself had a zero repeat,
2219        that was OK.
2220    
2221    13. In pcretest, the buffer for reading a data line was set at 30K, but the
2222        buffer into which it was copied (for escape processing) was still set at
2223        1024, so long lines caused crashes.
2224    
2225    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2226        "internal error: code overflow...". This applied to any character class
2227        that was followed by a possessive quantifier.
2228    
2229    15. Modified the Makefile to add libpcre.la as a prerequisite for
2230        libpcreposix.la because I was told this is needed for a parallel build to
2231        work.
2232    
2233    16. If a pattern that contained .* following optional items at the start was
2234        studied, the wrong optimizing data was generated, leading to matching
2235        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2236        matching string must start with a or b or c. The correct conclusion for
2237        this pattern is that a match can start with any character.
2238    
2239    
2240    Version 4.4 13-Aug-03
2241    ---------------------
2242    
2243     1. In UTF-8 mode, a character class containing characters with values between
2244        127 and 255 was not handled correctly if the compiled pattern was studied.
2245        In fixing this, I have also improved the studying algorithm for such
2246        classes (slightly).
2247    
2248     2. Three internal functions had redundant arguments passed to them. Removal
2249        might give a very teeny performance improvement.
2250    
2251     3. Documentation bug: the value of the capture_top field in a callout is *one
2252        more than* the number of the hightest numbered captured substring.
2253    
2254     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2255        in incorrectly linking with a previously installed version. They now link
2256        explicitly with libpcre.la.
2257    
2258     5. configure.in no longer needs to recognize Cygwin specially.
2259    
2260     6. A problem in pcre.in for Windows platforms is fixed.
2261    
2262     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2263        pcretest, it used to include the size of the study block as part of its
2264        output. Unfortunately, the structure contains a field that has a different
2265        size on different hardware architectures. This meant that the tests that
2266        showed this size failed. As the block is currently always of a fixed size,
2267        this information isn't actually particularly useful in pcretest output, so
2268        I have just removed it.
2269    
2270     8. Three pre-processor statements accidentally did not start in column 1.
2271        Sadly, there are *still* compilers around that complain, even though
2272        standard C has not required this for well over a decade. Sigh.
2273    
2274     9. In pcretest, the code for checking callouts passed small integers in the
2275        callout_data field, which is a void * field. However, some picky compilers
2276        complained about the casts involved for this on 64-bit systems. Now
2277        pcretest passes the address of the small integer instead, which should get
2278        rid of the warnings.
2279    
2280    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2281        both compile and run time, and gives an error if an invalid UTF-8 sequence
2282        is found. There is a option for disabling this check in cases where the
2283        string is known to be correct and/or the maximum performance is wanted.
2284    
2285    11. In response to a bug report, I changed one line in Makefile.in from
2286    
2287            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2288        to
2289            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2290    
2291        to look similar to other lines, but I have no way of telling whether this
2292        is the right thing to do, as I do not use Windows. No doubt I'll get told
2293        if it's wrong...
2294    
2295    
2296    Version 4.3 21-May-03
2297    ---------------------
2298    
2299    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2300       Makefile.
2301    
2302    2. Some refactoring to improve the quality of the code:
2303    
2304       (i)   The utf8_table... variables are now declared "const".
2305    
2306       (ii)  The code for \cx, which used the "case flipping" table to upper case
2307             lower case letters, now just substracts 32. This is ASCII-specific,
2308             but the whole concept of \cx is ASCII-specific, so it seems
2309             reasonable.
2310    
2311       (iii) PCRE was using its character types table to recognize decimal and
2312             hexadecimal digits in the pattern. This is silly, because it handles
2313             only 0-9, a-f, and A-F, but the character types table is locale-
2314             specific, which means strange things might happen. A private
2315             table is now used for this - though it costs 256 bytes, a table is
2316             much faster than multiple explicit tests. Of course, the standard
2317             character types table is still used for matching digits in subject
2318             strings against \d.
2319    
2320       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2321             ending in _t are). So I've renamed it as ESC_tee.
2322    
2323    3. The first argument for regexec() in the POSIX wrapper should have been
2324       defined as "const".
2325    
2326    4. Changed pcretest to use malloc() for its buffers so that they can be
2327       Electric Fenced for debugging.
2328    
2329    5. There were several places in the code where, in UTF-8 mode, PCRE would try
2330       to read one or more bytes before the start of the subject string. Often this
2331       had no effect on PCRE's behaviour, but in some circumstances it could
2332       provoke a segmentation fault.
2333    
2334    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2335       to try to read one or more bytes before the start of the subject string.
2336    
2337    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
2338       UTF-8 support could misbehave in various ways if the subject string
2339       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
2340       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
2341       back over UTF-8 characters.)
2342    
2343    
2344    Version 4.2 14-Apr-03
2345    ---------------------
2346    
2347    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
2348    
2349    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
2350         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
2351         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
2352         [WIN_PREFIX]: new variable, "cyg" for Cygwin
2353         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
2354           and BUILD_EXEEXT
2355         Note: automatic setting of the BUILD variables is not yet working
2356         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
2357           compile-time but not at link-time
2358         [LINK]: use for linking executables only
2359         make different versions for Windows and non-Windows
2360         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
2361           libraries
2362         [LINK_FOR_BUILD]: new variable
2363         [OBJEXT]: use throughout
2364         [EXEEXT]: use throughout
2365         <winshared>: new target
2366         <wininstall>: new target
2367         <dftables.o>: use native compiler
2368         <dftables>: use native linker
2369         <install>: handle Windows platform correctly
2370         <clean>: ditto
2371         <check>: ditto
2372         copy DLL to top builddir before testing
2373    
2374       As part of these changes, -no-undefined was removed again. This was reported
2375       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
2376       in any case.
2377    
2378    3. Some tidies to get rid of compiler warnings:
2379    
2380       . In the match_data structure, match_limit was an unsigned long int, whereas
2381         match_call_count was an int. I've made them both unsigned long ints.
2382    
2383       . In pcretest the fact that a const uschar * doesn't automatically cast to
2384         a void * provoked a warning.
2385    
2386       . Turning on some more compiler warnings threw up some "shadow" variables
2387         and a few more missing casts.
2388    
2389    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
2390       option, a class that contained a single character with a value between 128
2391       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
2392    
2393    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
2394       option, a class that contained several characters, but with at least one
2395       whose value was between 128 and 255 caused PCRE to crash.
2396    
2397    
2398    Version 4.1 12-Mar-03
2399    ---------------------
2400    
2401    1. Compiling with gcc -pedantic found a couple of places where casts were
2402    needed, and a string in dftables.c that was longer than standard compilers are
2403    required to support.
2404    
2405    2. Compiling with Sun's compiler found a few more places where the code could
2406    be tidied up in order to avoid warnings.
2407    
2408    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
2409    first of these names is deprecated in the latest Autoconf in favour of the name
2410    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
2411    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2412    analogy I have changed it to CFLAGS_FOR_BUILD.
2413    
2414    4. Added -no-undefined to the linking command in the Makefile, because this is
2415    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2416    linking step for the pcreposix library.
2417    
2418    5. PCRE was failing to diagnose the case of two named groups with the same
2419    name.
2420    
2421    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2422    literal character that is needed in the subject for a match, and scans along to
2423    ensure that it is present before embarking on the full matching process. This
2424    saves time in cases of nested unlimited repeats that are never going to match.
2425    Problem: the scan can take a lot of time if the subject is very long (e.g.
2426    megabytes), thus penalizing straightforward matches. It is now done only if the
2427    amount of subject to be scanned is less than 1000 bytes.
2428    
2429    7. A lesser problem with the same optimization is that it was recording the
2430    first character of an anchored pattern as "needed", thus provoking a search
2431    right along the subject, even when the first match of the pattern was going to
2432    fail. The "needed" character is now not set for anchored patterns, unless it
2433    follows something in the pattern that is of non-fixed length. Thus, it still
2434    fulfils its original purpose of finding quick non-matches in cases of nested
2435    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
2436    
2437    
2438    Version 4.0 17-Feb-03
2439    ---------------------
2440    
2441    1. If a comment in an extended regex that started immediately after a meta-item
2442    extended to the end of string, PCRE compiled incorrect data. This could lead to
2443    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
2444    
2445    2. Moved to autoconf 2.53 and libtool 1.4.2.
2446    
2447    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
2448    the special perltest8 script is no longer needed - all the tests can be run
2449    from a single perltest script.
2450    
2451    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
2452    by \s. It has now been removed in PCRE. This means it isn't recognized as
2453    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
2454    class [:space:] *does* include VT, thereby creating a mess.
2455    
2456    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
2457    space and tab.
2458    
2459    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
2460    its new features into the main test script, reducing the number of scripts.
2461    
2462    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
2463    were backward compatible, and made the (?i) apply to the whole pattern, as if
2464    /i were given. Now it behaves more logically, and applies the option setting
2465    only to what follows. PCRE has been changed to follow suit. However, if it
2466    finds options settings right at the start of the pattern, it extracts them into
2467    the global options, as before. Thus, they show up in the info data.
2468    
2469    8. Added support for the \Q...\E escape sequence. Characters in between are
2470    treated as literals. This is slightly different from Perl in that $ and @ are
2471    also handled as literals inside the quotes. In Perl, they will cause variable
2472    interpolation. Note the following examples:
2473    
2474        Pattern            PCRE matches      Perl matches
2475    
2476        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
2477        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
2478        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
2479    
2480    For compatibility with Perl, \Q...\E sequences are recognized inside character
2481    classes as well as outside them.
2482    
2483    9. Re-organized 3 code statements in pcretest to avoid "overflow in
2484    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2485    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2486    signed/unsigned warnings.
2487    
2488    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2489    option for pcretest, so I've replaced it by a simple function that does just
2490    that job.
2491    
2492    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2493    "pcregrep -".
2494    
2495    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2496    Java package. This provides some syntactic sugar for simple cases of what my
2497    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2498    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2499    item, you can use this simplified notation. Note that only makes sense with
2500    greedy quantifiers. Consequently, the use of the possessive quantifier forces
2501    greediness, whatever the setting of the PCRE_UNGREEDY option.
2502    
2503    13. A change of greediness default within a pattern was not taking effect at
2504    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2505    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2506    was abstracted outside.
2507    
2508    14. PCRE now supports the \G assertion. It is true when the current matching
2509    position is at the start point of the match. This differs from \A when the
2510    starting offset is non-zero. Used with the /g option of pcretest (or similar
2511    code), it works in the same way as it does for Perl's /g option. If all
2512    alternatives of a regex begin with \G, the expression is anchored to the start
2513    match position, and the "anchored" flag is set in the compiled expression.
2514    
2515    15. Some bugs concerning the handling of certain option changes within patterns
2516    have been fixed. These applied to options other than (?ims). For example,
2517    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2518    been the other way round. Some of this was related to change 7 above.
2519    
2520    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2521    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2522    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2523    POSIX classes only within a class (e.g. /[[:alpha:]]/).
2524    
2525    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2526    mode. Unlike ".", it always matches newline, whatever the setting of
2527    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2528    assertions. Perl allows it, but it doesn't (in general) work because it can't
2529    calculate the length of the lookbehind. At least, that's the case for Perl
2530    5.8.0 - I've been told they are going to document that it doesn't work in
2531    future.
2532    
2533    18. Added an error diagnosis for escapes that PCRE does not support: these are
2534    \L, \l, \N, \P, \p, \U, \u, and \X.
2535    
2536    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2537    reading past the end of the pattern in cases such as /[abcd/.
2538    
2539    20. PCRE was getting more memory than necessary for patterns with classes that
2540    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2541    
2542    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2543    compiling PCRE for use with Virtual Pascal.
2544    
2545    22. Small fix to the Makefile to make it work properly if the build is done
2546    outside the source tree.
2547    
2548    23. Added a new extension: a condition to go with recursion. If a conditional
2549    subpattern starts with (?(R) the "true" branch is used if recursion has
2550    happened, whereas the "false" branch is used only at the top level.
2551    
2552    24. When there was a very long string of literal characters (over 255 bytes
2553    without UTF support, over 250 bytes with UTF support), the computation of how
2554    much memory was required could be incorrect, leading to segfaults or other
2555    strange effects.
2556    
2557    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2558    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2559    there was a subsequent back reference to those brackets. This meant that, for
2560    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2561    possible to check for precisely this case. All we can do is abandon the
2562    optimization if .* occurs inside capturing brackets when there are any back
2563    references whatsoever. (See below for a better fix that came later.)
2564    
2565    26. The handling of the optimization for finding the first character of a
2566    non-anchored pattern, and for finding a character that is required later in the
2567    match were failing in some cases. This didn't break the matching; it just
2568    failed to optimize when it could. The way this is done has been re-implemented.
2569    
2570    27. Fixed typo in error message for invalid (?R item (it said "(?p").
2571    
2572    28. Added a new feature that provides some of the functionality that Perl
2573    provides with (?{...}). The facility is termed a "callout". The way it is done
2574    in PCRE is for the caller to provide an optional function, by setting
2575    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2576    global variable. By default it is unset, which disables all calling out. To get
2577    the function called, the regex must include (?C) at appropriate points. This
2578    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2579    This provides a means of identifying different callout points. When PCRE
2580    reaches such a point in the regex, if pcre_callout has been set, the external
2581    function is called. It is provided with data in a structure called
2582    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2583    matching continues; if it returns a non-zero value, the match at the current
2584    point fails. However, backtracking will occur if possible. [This was changed
2585    later and other features added - see item 49 below.]
2586    
2587    29. pcretest is upgraded to test the callout functionality. It provides a
2588    callout function that displays information. By default, it shows the start of
2589    the match and the current position in the text. There are some new data escapes
2590    to vary what happens:
2591    
2592        \C+         in addition, show current contents of captured substrings
2593        \C-         do not supply a callout function
2594        \C!n        return 1 when callout number n is reached
2595        \C!n!m      return 1 when callout number n is reached for the mth time
2596    
2597    30. If pcregrep was called with the -l option and just a single file name, it
2598    output "<stdin>" if a match was found, instead of the file name.
2599    
2600    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2601    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2602    pcre_exec(). This saves a malloc/free per call. The default value of
2603    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2604    when configuring.
2605    
2606    32. The default maximum size of a compiled pattern is 64K. There have been a
2607    few cases of people hitting this limit. The code now uses macros to handle the
2608    storing of links as offsets within the compiled pattern. It defaults to 2-byte
2609    links, but this can be changed to 3 or 4 bytes by --with-link-size when
2610    configuring. Tests 2 and 5 work only with 2-byte links because they output
2611    debugging information about compiled patterns.
2612    
2613    33. Internal code re-arrangements:
2614    
2615    (a) Moved the debugging function for printing out a compiled regex into
2616        its own source file (printint.c) and used #include to pull it into
2617        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2618        separate copies.
2619    
2620    (b) Defined the list of op-code names for debugging as a macro in
2621        internal.h so that it is next to the definition of the opcodes.
2622    
2623    (c) Defined a table of op-code lengths for simpler skipping along compiled
2624        code. This is again a macro in internal.h so that it is next to the
2625        definition of the opcodes.
2626    
2627    34. Added support for recursive calls to individual subpatterns, along the
2628    lines of Robin Houston's patch (but implemented somewhat differently).
2629    
2630    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2631    allow it to read and process whole directories in Win32. This code was
2632    contributed by Lionel Fourquaux; it has not been tested by me.
2633    
2634    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2635    used to name a group. Names consist of alphanumerics and underscores, and must
2636    be unique. Back references use the syntax (?P=name) and recursive calls use
2637    (?P>name) which is a PCRE extension to the Python extension. Groups still have
2638    numbers. The function pcre_fullinfo() can be used after compilation to extract
2639    a name/number map. There are three relevant calls:
2640    
2641      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
2642      PCRE_INFO_NAMECOUNT            yields the number of entries
2643      PCRE_INFO_NAMETABLE            yields a pointer to the map.
2644    
2645    The map is a vector of fixed-size entries. The size of each entry depends on
2646    the length of the longest name used. The first two bytes of each entry are the
2647    group number, most significant byte first. There follows the corresponding
2648    name, zero terminated. The names are in alphabetical order.
2649    
2650    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2651    case instead of 255. Making it the same both with and without UTF-8 support
2652    means that the same test output works with both.
2653    
2654    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2655    calling malloc() with a zero argument.
2656    
2657    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2658    optimization. I've improved things by keeping a bitmap of backreferences with
2659    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2660    fact referenced, the optimization can be applied. It is unlikely that a
2661    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2662    the match to follow \n) will appear inside brackets with a number greater than
2663    31, but if it does, any back reference > 31 suppresses the optimization.
2664    
2665    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2666    of disabling numbered capturing parentheses. Any opening parenthesis that is
2667    not followed by ? behaves as if it were followed by ?: but named parentheses
2668    can still be used for capturing (and they will acquire numbers in the usual
2669    way).
2670    
2671    41. Redesigned the return codes from the match() function into yes/no/error so
2672    that errors can be passed back from deep inside the nested calls. A malloc
2673    failure while inside a recursive subpattern call now causes the
2674    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2675    
2676    42. It is now possible to set a limit on the number of times the match()
2677    function is called in a call to pcre_exec(). This facility makes it possible to
2678    limit the amount of recursion and backtracking, though not in a directly
2679    obvious way, because the match() function is used in a number of different
2680    circumstances. The count starts from zero for each position in the subject
2681    string (for non-anchored patterns). The default limit is, for compatibility, a
2682    large number, namely 10 000 000. You can change this in two ways:
2683    
2684    (a) When configuring PCRE before making, you can use --with-match-limit=n
2685        to set a default value for the compiled library.
2686    
2687    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2688        a different value is set. See 45 below.
2689    
2690    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2691    
2692    43. Added a new function pcre_config(int, void *) to enable run-time extraction
2693    of things that can be changed at compile time. The first argument specifies
2694    what is wanted and the second points to where the information is to be placed.
2695    The current list of available information is:
2696    
2697      PCRE_CONFIG_UTF8
2698    
2699    The output is an integer that is set to one if UTF-8 support is available;
2700    otherwise it is set to zero.
2701    
2702      PCRE_CONFIG_NEWLINE
2703    
2704    The output is an integer that it set to the value of the code that is used for
2705    newline. It is either LF (10) or CR (13).
2706    
2707      PCRE_CONFIG_LINK_SIZE
2708    
2709    The output is an integer that contains the number of bytes used for internal
2710    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2711    
2712      PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
2713    
2714    The output is an integer that contains the threshold above which the POSIX
2715    interface uses malloc() for output vectors. See item 31 above.
2716    
2717      PCRE_CONFIG_MATCH_LIMIT
2718    
2719    The output is an unsigned integer that contains the default limit of the number
2720    of match() calls in a pcre_exec() execution. See 42 above.
2721    
2722    44. pcretest has been upgraded by the addition of the -C option. This causes it
2723    to extract all the available output from the new pcre_config() function, and to
2724    output it. The program then exits immediately.
2725    
2726    45. A need has arisen to pass over additional data with calls to pcre_exec() in
2727    order to support additional features. One way would have been to define
2728    pcre_exec2() (for example) with extra arguments, but this would not have been
2729    extensible, and would also have required all calls to the original function to
2730    be mapped to the new one. Instead, I have chosen to extend the mechanism that
2731    is used for passing in "extra" data from pcre_study().
2732    
2733    The pcre_extra structure is now exposed and defined in pcre.h. It currently
2734    contains the following fields:
2735    
2736      flags         a bitmap indicating which of the following fields are set
2737      study_data    opaque data from pcre_study()
2738      match_limit   a way of specifying a limit on match() calls for a specific
2739                      call to pcre_exec()
2740      callout_data  data for callouts (see 49 below)
2741    
2742    The flag bits are also defined in pcre.h, and are
2743    
2744      PCRE_EXTRA_STUDY_DATA
2745      PCRE_EXTRA_MATCH_LIMIT
2746      PCRE_EXTRA_CALLOUT_DATA
2747    
2748    The pcre_study() function now returns one of these new pcre_extra blocks, with
2749    the actual study data pointed to by the study_data field, and the
2750    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2751    before. That is, this change is entirely upwards-compatible and requires no
2752    change to existing code.
2753    
2754    If you want to pass in additional data to pcre_exec(), you can either place it
2755    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2756    block.
2757    
2758    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2759    data string contains the escape sequence \M, pcretest calls pcre_exec() several
2760    times with different match limits, until it finds the minimum value needed for
2761    pcre_exec() to complete. The value is then output. This can be instructive; for
2762    most simple matches the number is quite small, but for pathological cases it
2763    gets very large very quickly.
2764    
2765    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2766    returns the size of the data block pointed to by the study_data field in a
2767    pcre_extra block, that is, the value that was passed as the argument to
2768    pcre_malloc() when PCRE was getting memory in which to place the information
2769    created by pcre_study(). The fourth argument should point to a size_t variable.
2770    pcretest has been extended so that this information is shown after a successful
2771    pcre_study() call when information about the compiled regex is being displayed.
2772    
2773    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2774    because what follows is always an absolute path. (Later: it turns out that this
2775    is more than cosmetic for MinGW, because it doesn't like empty path
2776    components.)
2777    
2778    49. Some changes have been made to the callout feature (see 28 above):
2779    
2780    (i)  A callout function now has three choices for what it returns:
2781    
2782           0  =>  success, carry on matching
2783         > 0  =>  failure at this point, but backtrack if possible
2784         < 0  =>  serious error, return this value from pcre_exec()
2785    
2786         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2787         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2788         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2789         use by callout functions. It will never be used by PCRE itself.
2790    
2791    (ii) The pcre_extra structure (see 45 above) has a void * field called
2792         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2793         pcre_callout_block structure has a field of the same name. The contents of
2794         the field passed in the pcre_extra structure are passed to the callout
2795         function in the corresponding field in the callout block. This makes it
2796         easier to use the same callout-containing regex from multiple threads. For
2797         testing, the pcretest program has a new data escape
2798    
2799           \C*n        pass the number n (may be negative) as callout_data
2800    
2801         If the callout function in pcretest receives a non-zero value as
2802         callout_data, it returns that value.
2803    
2804    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2805    there were some redundant $(CFLAGS) in commands that are now specified as
2806    $(LINK), which already includes $(CFLAGS).
2807    
2808    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2809    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2810    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2811    one-byte characters throughout. Note that case-insensitive matching applies
2812    only to characters whose values are less than 256. PCRE doesn't support the
2813    notion of cases for higher-valued characters.
2814    
2815    (i)   A character class whose characters are all within 0-255 is handled as
2816          a bit map, and the map is inverted for negative classes. Previously, a
2817          character > 255 always failed to match such a class; however it should
2818          match if the class was a negative one (e.g. [^ab]). This has been fixed.
2819    
2820    (ii)  A negated character class with a single character < 255 is coded as
2821          "not this character" (OP_NOT). This wasn't working properly when the test
2822          character was multibyte, either singly or repeated.
2823    
2824    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2825          mode, for example: \x{100}{2,3}.
2826    
2827    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2828          singly or repeated) now correctly test multibyte characters. However,
2829          PCRE doesn't recognize any characters with values greater than 255 as
2830          digits, spaces, or word characters. Such characters always match \D, \S,
2831          and \W, and never match \d, \s, or \w.
2832    
2833    (v)   Classes may now contain characters and character ranges with values
2834          greater than 255. For example: [ab\x{100}-\x{400}].
2835    
2836    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
2837          PCRE in UTF-8 mode.
2838    
2839    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2840    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2841    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2842    value.)
2843    
2844    53. The single man page has become too large. I have therefore split it up into
2845    a number of separate man pages. These also give rise to individual HTML pages;
2846    these are now put in a separate directory, and there is an index.html page that
2847    lists them all. Some hyperlinking between the pages has been installed.
2848    
2849    54. Added convenience functions for handling named capturing parentheses.
2850    
2851    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2852    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2853    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2854    are faulted.
2855    
2856    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2857    calling configure. These values are used when compiling the dftables.c program
2858    which is run to generate the source of the default character tables. They
2859    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2860    you will need to set these values.
2861    
2862    57. Updated the building process for Windows DLL, as provided by Fred Cox.
2863    
2864    
2865    Version 3.9 02-Jan-02
2866    ---------------------
2867    
2868    1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2869    
2870    2. If --disable-static was given, the building process failed when trying to
2871    build pcretest and pcregrep. (For some reason it was using libtool to compile
2872    them, which is not right, as they aren't part of the library.)
2873    
2874    
2875    Version 3.8 18-Dec-01
2876    ---------------------
2877    
2878    1. The experimental UTF-8 code was completely screwed up. It was packing the
2879    bytes in the wrong order. How dumb can you get?
2880    
2881    
2882    Version 3.7 29-Oct-01
2883    ---------------------
2884    
2885    1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2886    This caused pcretest, when used on the test data, to segfault. Unfortunately,
2887    this didn't happen under Solaris 8, where I normally test things.
2888    
2889    2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2890    doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2891    isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2892    this fix an hour or so after the initial 3.7 release.)
2893    
2894    
2895    Version 3.6 23-Oct-01
2896    ---------------------
2897    
2898    1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2899    offsets passed as NULL with zero offset count.
2900    
2901    2. The config.guess and config.sub files had not been updated when I moved to
2902    the latest autoconf.
2903    
2904    
2905    Version 3.5 15-Aug-01
2906    ---------------------
2907    
2908    1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2909    had been forgotten.
2910    
2911    2. By using declared but undefined structures, we can avoid using "void"
2912    definitions in pcre.h while keeping the internal definitions of the structures
2913    private.
2914    
2915    3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2916    user point of view, this means that both static and shared libraries are built
2917    by default, but this can be individually controlled. More of the work of
2918    handling this static/shared cases is now inside libtool instead of PCRE's make
2919    file.
2920    
2921    4. The pcretest utility is now installed along with pcregrep because it is
2922    useful for users (to test regexs) and by doing this, it automatically gets
2923    relinked by libtool. The documentation has been turned into a man page, so
2924    there are now .1, .txt, and .html versions in /doc.
2925    
2926    5. Upgrades to pcregrep:
2927       (i)   Added long-form option names like gnu grep.
2928       (ii)  Added --help to list all options with an explanatory phrase.
2929       (iii) Added -r, --recursive to recurse into sub-directories.
2930       (iv)  Added -f, --file to read patterns from a file.
2931    
2932    6. pcre_exec() was referring to its "code" argument before testing that
2933    argument for NULL (and giving an error if it was NULL).
2934    
2935    7. Upgraded Makefile.in to allow for compiling in a different directory from
2936    the source directory.
2937    
2938    8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2939    options bits, the pointer it was passed was to an int instead of to an unsigned
2940    long int. This mattered only on 64-bit systems.
2941    
2942    9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2943    generated) instead of pcre.in, which it its source. Also made the same change
2944    in several of the .c files.
2945    
2946    10. A new release of gcc defines printf() as a macro, which broke pcretest
2947    because it had an ifdef in the middle of a string argument for printf(). Fixed
2948    by using separate calls to printf().
2949    
2950    11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2951    script, to force use of CR or LF instead of \n in the source. On non-Unix
2952    systems, the value can be set in config.h.
2953    
2954    12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2955    absolute limit. Changed the text of the error message to make this clear, and
2956    likewise updated the man page.
2957    
2958    13. The limit of 99 on the number of capturing subpatterns has been removed.
2959    The new limit is 65535, which I hope will not be a "real" limit.
2960    
2961    
2962    Version 3.4 22-Aug-00
2963    ---------------------
2964    
2965    1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2966    
2967    2. Diagnose condition (?(0) as an error instead of crashing on matching.
2968    
2969    
2970    Version 3.3 01-Aug-00
2971    ---------------------
2972    
2973    1. If an octal character was given, but the value was greater than \377, it
2974    was not getting masked to the least significant bits, as documented. This could
2975    lead to crashes in some systems.
2976    
2977    2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2978    the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2979    
2980    3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2981    These just pass their arguments on to (pcre_free)(), but they are provided
2982    because some uses of PCRE bind it to non-C systems that can call its functions,
2983    but cannot call free() or pcre_free() directly.
2984    
2985    4. Add "make test" as a synonym for "make check". Corrected some comments in
2986    the Makefile.
2987    
2988    5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2989    Makefile.
2990    
2991    6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2992    command called pgrep for grepping around the active processes.
2993    
2994    7. Added the beginnings of support for UTF-8 character strings.
2995    
2996    8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2997    RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2998    the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2999    out for the ar command.)
3000    
3001    
3002    Version 3.2 12-May-00
3003    ---------------------
3004    
3005    This is purely a bug fixing release.
3006    
3007    1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
3008    of ZA. This was just one example of several cases that could provoke this bug,
3009    which was introduced by change 9 of version 2.00. The code for breaking
3010    infinite loops after an iteration that matches an empty string was't working
3011    correctly.
3012    
3013    2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
3014    when matched against abbab (for example). After matching an empty string, it
3015    wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
3016    caused it to match further down the string than it should.
3017    
3018    3. The code contained an inclusion of sys/types.h. It isn't clear why this
3019    was there because it doesn't seem to be needed, and it causes trouble on some
3020    systems, as it is not a Standard C header. It has been removed.
3021    
3022    4. Made 4 silly changes to the source to avoid stupid compiler warnings that
3023    were reported on the Macintosh. The changes were from
3024    
3025      while ((c = *(++ptr)) != 0 && c != '\n');
3026    to
3027      while ((c = *(++ptr)) != 0 && c != '\n') ;
3028    
3029    Totally extraordinary, but if that's what it takes...
3030    
3031    5. PCRE is being used in one environment where neither memmove() nor bcopy() is
3032    available. Added HAVE_BCOPY and an autoconf test for it; if neither
3033    HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
3034    assumes the way PCRE uses memmove() (always moving upwards).
3035    
3036    6. PCRE is being used in one environment where strchr() is not available. There
3037    was only one use in pcre.c, and writing it out to avoid strchr() probably gives
3038    faster code anyway.
3039    
3040    
3041    Version 3.1 09-Feb-00
3042    ---------------------
3043    
3044    The only change in this release is the fixing of some bugs in Makefile.in for
3045    the "install" target:
3046    
3047    (1) It was failing to install pcreposix.h.
3048    
3049    (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
3050    
3051    
3052    Version 3.0 01-Feb-00
3053    ---------------------
3054    
3055    1. Add support for the /+ modifier to perltest (to output $` like it does in
3056    pcretest).
3057    
3058    2. Add support for the /g modifier to perltest.
3059    
3060    3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
3061    matches null strings.
3062    
3063    4. Fix perltest so that it doesn't do unwanted things when fed an empty
3064    pattern. Perl treats empty patterns specially - it reuses the most recent
3065    pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
3066    effect.
3067    
3068    5. The POSIX interface was broken in that it was just handing over the POSIX
3069    captured string vector to pcre_exec(), but (since release 2.00) PCRE has
3070    required a bigger vector, with some working space on the end. This means that
3071    the POSIX wrapper now has to get and free some memory, and copy the results.
3072    
3073    6. Added some simple autoconf support, placing the test data and the
3074    documentation in separate directories, re-organizing some of the
3075    information files, and making it build pcre-config (a GNU standard). Also added
3076    libtool support for building PCRE as a shared library, which is now the
3077    default.
3078    
3079    7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
3080    09 are not valid octal constants. Single digits will be used for minor values
3081    less than 10.
3082    
3083    8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
3084    existing programs that set these in the POSIX interface can use PCRE without
3085    modification.
3086    
3087    9. Added a new function, pcre_fullinfo() with an extensible interface. It can
3088    return all that pcre_info() returns, plus additional data. The pcre_info()
3089    function is retained for compatibility, but is considered to be obsolete.
3090    
3091    10. Added experimental recursion feature (?R) to handle one common case that
3092    Perl 5.6 will be able to do with (?p{...}).
3093    
3094    11. Added support for POSIX character classes like [:alpha:], which Perl is
3095    adopting.
3096    
3097    
3098    Version 2.08 31-Aug-99
3099    ----------------------
3100    
3101    1. When startoffset was not zero and the pattern began with ".*", PCRE was not
3102    trying to match at the startoffset position, but instead was moving forward to
3103    the next newline as if a previous match had failed.
3104    
3105    2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
3106    and could get into a loop if a null string was matched other than at the start
3107    of the subject.
3108    
3109    3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
3110    be distinguished at compile time, and for completeness also added PCRE_DATE.
3111    
3112    5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
3113    in GnuWin32 environments.
3114    
3115    
3116    Version 2.07 29-Jul-99
3117    ----------------------
3118    
3119    1. The documentation is now supplied in plain text form and HTML as well as in
3120    the form of man page sources.
3121    
3122    2. C++ compilers don't like assigning (void *) values to other pointer types.
3123    In particular this affects malloc(). Although there is no problem in Standard
3124    C, I've put in casts to keep C++ compilers happy.
3125    
3126    3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
3127    should be (const char *).
3128    
3129    4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
3130    be useful for non-Unix systems who don't want to bother with the POSIX stuff.
3131    However, I haven't made this a standard facility. The documentation doesn't
3132    mention it, and the Makefile doesn't support it.
3133    
3134    5. The Makefile now contains an "install" target, with editable destinations at
3135    the top of the file. The pcretest program is not installed.
3136    
3137    6. pgrep -V now gives the PCRE version number and date.
3138    
3139    7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
3140    causing the entire string to be ignored, instead of just the last character.
3141    
3142    8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
3143    non-matching string, it can take a very, very long time, even for strings of
3144    quite modest length, because of the nested recursion. PCRE now does better in
3145    some of these cases. It does this by remembering the last required literal
3146    character in the pattern, and pre-searching the subject to ensure it is present
3147    before running the real match. In other words, it applies a heuristic to detect
3148    some types of certain failure quickly, and in the above example, if presented
3149    with a string that has no trailing " it gives "no match" very quickly.
3150    
3151    9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
3152    other alternatives are tried instead.
3153    
3154    
3155    Version 2.06 09-Jun-99
3156    ----------------------
3157    
3158    1. Change pcretest's output for amount of store used to show just the code
3159    space, because the remainder (the data block) varies in size between 32-bit and
3160    64-bit systems.
3161    
3162    2. Added an extra argument to pcre_exec() to supply an offset in the subject to
3163    start matching at. This allows lookbehinds to work when searching for multiple
3164    occurrences in a string.
3165    
3166    3. Added additional options to pcretest for testing multiple occurrences:
3167    
3168       /+   outputs the rest of the string that follows a match
3169       /g   loops for multiple occurrences, using the new startoffset argument
3170       /G   loops for multiple occurrences by passing an incremented pointer
3171    
3172    4. PCRE wasn't doing the "first character" optimization for patterns starting
3173    with \b or \B, though it was doing it for other lookbehind assertions. That is,
3174    it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
3175    the letter 'x'. On long subject strings, this gives a significant speed-up.
3176    
3177    
3178    Version 2.05 21-Apr-99
3179    ----------------------
3180    
3181    1. Changed the type of magic_number from int to long int so that it works
3182    properly on 16-bit systems.
3183    
3184    2. Fixed a bug which caused patterns starting with .* not to work correctly
3185    when the subject string contained newline characters. PCRE was assuming
3186    anchoring for such patterns in all cases, which is not correct because .* will
3187    not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
3188    DOTALL is set at top level; otherwise it knows that patterns starting with .*
3189    must be retried after every newline in the subject.
3190    
3191    
3192    Version 2.04 18-Feb-99
3193    ----------------------
3194    
3195    1. For parenthesized subpatterns with repeats whose minimum was zero, the
3196    computation of the store needed to hold the pattern was incorrect (too large).
3197    If such patterns were nested a few deep, this could multiply and become a real
3198    problem.
3199    
3200    2. Added /M option to pcretest to show the memory requirement of a specific
3201    pattern. Made -m a synonym of -s (which does this globally) for compatibility.
3202    
3203    3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
3204    compiled in such a way that the backtracking after subsequent failure was
3205    pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
3206    ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
3207    
3208    
3209    Version 2.03 02-Feb-99
3210    ----------------------
3211    
3212    1. Fixed typo and small mistake in man page.
3213    
3214    2. Added 4th condition (GPL supersedes if conflict) and created separate
3215    LICENCE file containing the conditions.
3216    
3217    3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
3218    Perl, that is the internal \ allows the delimiter to be included in the
3219    pattern. Locked out the use of \ as a delimiter. If \ immediately follows
3220    the final delimiter, add \ to the end of the pattern (to test the error).
3221    
3222    4. Added the convenience functions for extracting substrings after a successful
3223    match. Updated pcretest to make it able to test these functions.
3224    
3225    
3226    Version 2.02 14-Jan-99
3227    ----------------------
3228    
3229    1. Initialized the working variables associated with each extraction so that
3230    their saving and restoring doesn't refer to uninitialized store.
3231    
3232    2. Put dummy code into study.c in order to trick the optimizer of the IBM C
3233    compiler for OS/2 into generating correct code. Apparently IBM isn't going to
3234    fix the problem.
3235    
3236    3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
3237    calls, and wasn't printing the correct value for compiling calls. Increased the
3238    default value of LOOPREPEAT, and the number of significant figures in the
3239    times.
3240    
3241    4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
3242    
3243    5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
3244    a building problem on Windows NT with a FAT file system.
3245    
3246    
3247    Version 2.01 21-Oct-98
3248    ----------------------
3249    
3250    1. Changed the API for pcre_compile() to allow for the provision of a pointer
3251    to character tables built by pcre_maketables() in the current locale. If NULL
3252    is passed, the default tables are used.
3253    
3254    
3255    Version 2.00 24-Sep-98
3256    ----------------------
3257    
3258    1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
3259    it any more.
3260    
3261    2. Allow quantification of (?>) groups, and make it work correctly.
3262    
3263    3. The first character computation wasn't working for (?>) groups.
3264    
3265    4. Correct the implementation of \Z (it is permitted to match on the \n at the
3266    end of the subject) and add 5.005's \z, which really does match only at the
3267    very end of the subject.
3268    
3269    5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
3270    
3271    6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
3272    DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
3273    localized options. All options to pcre_study() were also removed.
3274    
3275    7. Add other new features from 5.005:
3276    
3277       $(?<=           positive lookbehind
3278       $(?<!           negative lookbehind
3279       (?imsx-imsx)    added the unsetting capability
3280                       such a setting is global if at outer level; local otherwise
3281       (?imsx-imsx:)   non-capturing groups with option setting
3282       (?(cond)re|re)  conditional pattern matching
3283    
3284       A backreference to itself in a repeated group matches the previous
3285       captured string.
3286    
3287    8. General tidying up of studying (both automatic and via "study")
3288    consequential on the addition of new assertions.
3289    
3290    9. As in 5.005, unlimited repeated groups that could match an empty substring
3291    are no longer faulted at compile time. Instead, the loop is forcibly broken at
3292    runtime if any iteration does actually match an empty substring.
3293    
3294    10. Include the RunTest script in the distribution.
3295    
3296    11. Added tests from the Perl 5.005_02 distribution. This showed up a few
3297    discrepancies, some of which were old and were also with respect to 5.004. They
3298    have now been fixed.
3299    
3300    
3301    Version 1.09 28-Apr-98
3302    ----------------------
3303    
3304    1. A negated single character class followed by a quantifier with a minimum
3305    value of one (e.g.  [^x]{1,6}  ) was not compiled correctly. This could lead to
3306    program crashes, or just wrong answers. This did not apply to negated classes
3307    containing more than one character, or to minima other than one.
3308    
3309    
3310  Version 1.08 27-Mar-98  Version 1.08 27-Mar-98
3311  ----------------------  ----------------------

Legend:
Removed from v.19  
changed lines
  Added in v.437

  ViewVC Help
Powered by ViewVC 1.1.5