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

Legend:
Removed from v.47  
changed lines
  Added in v.408

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12