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

Diff of /code/trunk/ChangeLog

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

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

Legend:
Removed from v.51  
changed lines
  Added in v.411

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12