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

Diff of /code/trunk/ChangeLog

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

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

Legend:
Removed from v.47  
changed lines
  Added in v.397

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12