/[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 478 by ph10, Sun Jan 3 16:05:13 2010 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    Version 8.01 11-Dec-09
5    ----------------------
6    
7    1.  If a pattern contained a conditional subpattern with only one branch (in
8        particular, this includes all (DEFINE) patterns), a call to pcre_study()
9        computed the wrong minimum data length (which is of course zero for such
10        subpatterns).
11    
12    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
13        the pattern is reset in the first branch, pcre_compile() failed with
14        "internal error: code overflow at offset...". This happened only when
15        the reset was to the original external option setting. (An optimization
16        abstracts leading options settings into an external setting, which was the
17        cause of this.)
18    
19    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
20        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
21        assertion pattern did not match (meaning that the assertion was true), it
22        was incorrectly treated as false if the SKIP had been reached during the
23        matching. This also applied to assertions used as conditions.
24    
25    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
26        assertion subpattern, including such a pattern used as a condition,
27        unpredictable results occurred, instead of the error return
28        PCRE_ERROR_DFA_UITEM.
29    
30    5.  The C++ GlobalReplace function was not working like Perl for the special
31        situation when an empty string is matched. It now does the fancy magic
32        stuff that is necessary.
33    
34    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
35        removed. (These were left over from very, very early versions of PCRE.)
36    
37    7.  Some cosmetic changes to the code to make life easier when compiling it
38        as part of something else:
39    
40        (a) Change DEBUG to PCRE_DEBUG.
41    
42        (b) In pcre_compile(), rename the member of the "branch_chain" structure
43            called "current" as "current_branch", to prevent a collision with the
44            Linux macro when compiled as a kernel module.
45    
46        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
47            prevent a collision with the Linux macro when compiled as a kernel
48            module.
49    
50    8.  In pcre_compile() there are some checks for integer overflows that used to
51        cast potentially large values to (double). This has been changed to that
52        when building, a check for int64_t is made, and if it is found, it is used
53        instead, thus avoiding the use of floating point arithmetic. (There is no
54        other use of FP in PCRE.) If int64_t is not found, the fallback is to
55        double.
56    
57    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
58        2005 (difference between two addresses compared to an unsigned value).
59    
60    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
61        custom one, because of the following reported problem in Windows:
62    
63          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
64            under Win32.
65          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
66            therefore missing the function definition.
67          - The compiler thus generates a "C" signature for the test function.
68          - The linker fails to find the "C" function.
69          - PCRE fails to configure if asked to do so against libbz2.
70    
71    
72    Version 8.00 19-Oct-09
73    ----------------------
74    
75    1.  The table for translating pcre_compile() error codes into POSIX error codes
76        was out-of-date, and there was no check on the pcre_compile() error code
77        being within the table. This could lead to an OK return being given in
78        error.
79    
80    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
81        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
82        in a Windows environment.
83    
84    3.  The pcregrep --count option prints the count for each file even when it is
85        zero, as does GNU grep. However, pcregrep was also printing all files when
86        --files-with-matches was added. Now, when both options are given, it prints
87        counts only for those files that have at least one match. (GNU grep just
88        prints the file name in this circumstance, but including the count seems
89        more useful - otherwise, why use --count?) Also ensured that the
90        combination -clh just lists non-zero counts, with no names.
91    
92    4.  The long form of the pcregrep -F option was incorrectly implemented as
93        --fixed_strings instead of --fixed-strings. This is an incompatible change,
94        but it seems right to fix it, and I didn't think it was worth preserving
95        the old behaviour.
96    
97    5.  The command line items --regex=pattern and --regexp=pattern were not
98        recognized by pcregrep, which required --regex pattern or --regexp pattern
99        (with a space rather than an '='). The man page documented the '=' forms,
100        which are compatible with GNU grep; these now work.
101    
102    6.  No libpcreposix.pc file was created for pkg-config; there was just
103        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
104    
105    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
106        when UCP support is not needed, by modifying the Python script that
107        generates it from Unicode data files. This should not matter if the module
108        is correctly used as a library, but I received one complaint about 50K of
109        unwanted data. My guess is that the person linked everything into his
110        program rather than using a library. Anyway, it does no harm.
111    
112    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
113        was a minimum greater than 1 for a wide character in a possessive
114        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
115        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
116        character. Chaos in the form of incorrect output or a compiling loop could
117        result.
118    
119    9.  The restrictions on what a pattern can contain when partial matching is
120        requested for pcre_exec() have been removed. All patterns can now be
121        partially matched by this function. In addition, if there are at least two
122        slots in the offset vector, the offset of the earliest inspected character
123        for the match and the offset of the end of the subject are set in them when
124        PCRE_ERROR_PARTIAL is returned.
125    
126    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
127        synonymous with PCRE_PARTIAL, for backwards compatibility, and
128        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
129        and may be more useful for multi-segment matching.
130    
131    11. Partial matching with pcre_exec() is now more intuitive. A partial match
132        used to be given if ever the end of the subject was reached; now it is
133        given only if matching could not proceed because another character was
134        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
135        string "Z", which now yields "no match" instead of "partial match". In the
136        case of pcre_dfa_exec(), "no match" is given if every matching path for the
137        final character ended with (*FAIL).
138    
139    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
140        if the pattern had a "must contain" character that was already found in the
141        earlier partial match, unless partial matching was again requested. For
142        example, with the pattern /dog.(body)?/, the "must contain" character is
143        "g". If the first part-match was for the string "dog", restarting with
144        "sbody" failed. This bug has been fixed.
145    
146    13. The string returned by pcre_dfa_exec() after a partial match has been
147        changed so that it starts at the first inspected character rather than the
148        first character of the match. This makes a difference only if the pattern
149        starts with a lookbehind assertion or \b or \B (\K is not supported by
150        pcre_dfa_exec()). It's an incompatible change, but it makes the two
151        matching functions compatible, and I think it's the right thing to do.
152    
153    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
154        so that the demonstration program is easily available in environments where
155        PCRE has not been installed from source.
156    
157    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
158        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
159        library.
160    
161    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
162        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
163        is not the first non-POSIX option to be added. Clearly some people find
164        these options useful.
165    
166    17. If a caller to the POSIX matching function regexec() passes a non-zero
167        value for nmatch with a NULL value for pmatch, the value of
168        nmatch is forced to zero.
169    
170    18. RunGrepTest did not have a test for the availability of the -u option of
171        the diff command, as RunTest does. It now checks in the same way as
172        RunTest, and also checks for the -b option.
173    
174    19. If an odd number of negated classes containing just a single character
175        interposed, within parentheses, between a forward reference to a named
176        subpattern and the definition of the subpattern, compilation crashed with
177        an internal error, complaining that it could not find the referenced
178        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
179        [The bug was that it was starting one character too far in when skipping
180        over the character class, thus treating the ] as data rather than
181        terminating the class. This meant it could skip too much.]
182    
183    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
184        /g option in pcretest when the pattern contains \K, which makes it possible
185        to have an empty string match not at the start, even when the pattern is
186        anchored. Updated pcretest and pcredemo to use this option.
187    
188    21. If the maximum number of capturing subpatterns in a recursion was greater
189        than the maximum at the outer level, the higher number was returned, but
190        with unset values at the outer level. The correct (outer level) value is
191        now given.
192    
193    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
194        PCRE did not set those parentheses (unlike Perl). I have now found a way to
195        make it do so. The string so far is captured, making this feature
196        compatible with Perl.
197    
198    23. The tests have been re-organized, adding tests 11 and 12, to make it
199        possible to check the Perl 5.10 features against Perl 5.10.
200    
201    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
202        pattern matches a fixed length string. PCRE did not allow this; now it
203        does. Neither allows recursion.
204    
205    25. I finally figured out how to implement a request to provide the minimum
206        length of subject string that was needed in order to match a given pattern.
207        (It was back references and recursion that I had previously got hung up
208        on.) This code has now been added to pcre_study(); it finds a lower bound
209        to the length of subject needed. It is not necessarily the greatest lower
210        bound, but using it to avoid searching strings that are too short does give
211        some useful speed-ups. The value is available to calling programs via
212        pcre_fullinfo().
213    
214    26. While implementing 25, I discovered to my embarrassment that pcretest had
215        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
216        study optimizations had never been tested with that matching function.
217        Oops. What is worse, even when it was passed study data, there was a bug in
218        pcre_dfa_exec() that meant it never actually used it. Double oops. There
219        were also very few tests of studied patterns with pcre_dfa_exec().
220    
221    27. If (?| is used to create subpatterns with duplicate numbers, they are now
222        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
223        on the other side of the coin, they are no longer allowed to have different
224        names, because these cannot be distinguished in PCRE, and this has caused
225        confusion. (This is a difference from Perl.)
226    
227    28. When duplicate subpattern names are present (necessarily with different
228        numbers, as required by 27 above), and a test is made by name in a
229        conditional pattern, either for a subpattern having been matched, or for
230        recursion in such a pattern, all the associated numbered subpatterns are
231        tested, and the overall condition is true if the condition is true for any
232        one of them. This is the way Perl works, and is also more like the way
233        testing by number works.
234    
235    
236    Version 7.9 11-Apr-09
237    ---------------------
238    
239    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
240        (pcretest), all targets were linked against these libraries. This included
241        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
242        libraries. This caused unwanted dependencies to be created. This problem
243        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
244        pcretest is linked with readline.
245    
246    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
247        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
248        moved outside it again, because FALSE and TRUE are already defined in AIX,
249        but BOOL is not.
250    
251    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
252        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
253    
254    4.  The pcregrep documentation said spaces were inserted as well as colons (or
255        hyphens) following file names and line numbers when outputting matching
256        lines. This is not true; no spaces are inserted. I have also clarified the
257        wording for the --colour (or --color) option.
258    
259    5.  In pcregrep, when --colour was used with -o, the list of matching strings
260        was not coloured; this is different to GNU grep, so I have changed it to be
261        the same.
262    
263    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
264        each matching line was coloured. Now it goes on to look for further matches
265        of any of the test patterns, which is the same behaviour as GNU grep.
266    
267    7.  A pattern that could match an empty string could cause pcregrep to loop; it
268        doesn't make sense to accept an empty string match in pcregrep, so I have
269        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
270        seems to be how GNU grep behaves.
271    
272    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
273        start or after a newline", because the conditional assertion was not being
274        correctly handled. The rule now is that both the assertion and what follows
275        in the first alternative must satisfy the test.
276    
277    9.  If auto-callout was enabled in a pattern with a conditional group whose
278        condition was an assertion, PCRE could crash during matching, both with
279        pcre_exec() and pcre_dfa_exec().
280    
281    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
282        used for matching.
283    
284    11. Unicode property support in character classes was not working for
285        characters (bytes) greater than 127 when not in UTF-8 mode.
286    
287    12. Added the -M command line option to pcretest.
288    
289    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
290    
291    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
292    
293    16. Added comments and documentation about mis-use of no_arg in the C++
294        wrapper.
295    
296    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
297        from Martin Jerabek that uses macro names for all relevant character and
298        string constants.
299    
300    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
301        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
302        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
303        these, but not everybody uses configure.
304    
305    19. A conditional group that had only one branch was not being correctly
306        recognized as an item that could match an empty string. This meant that an
307        enclosing group might also not be so recognized, causing infinite looping
308        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
309        with the subject "ab", where knowledge that the repeated group can match
310        nothing is needed in order to break the loop.
311    
312    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
313        exec(), but without supplying a callout function, matching went wrong.
314    
315    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
316        leak if the size of the offset vector was greater than 30. When the vector
317        is smaller, the saved offsets during recursion go onto a local stack
318        vector, but for larger vectors malloc() is used. It was failing to free
319        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
320        error, in fact).
321    
322    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
323        heapframe that is used only when UTF-8 support is enabled. This caused no
324        problem, but was untidy.
325    
326    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
327        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
328        included within another project.
329    
330    24. Steven Van Ingelgem's patches to add more options to the CMake support,
331        slightly modified by me:
332    
333          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
334              not building pcregrep.
335    
336          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
337              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
338    
339    25. Forward references, both numeric and by name, in patterns that made use of
340        duplicate group numbers, could behave incorrectly or give incorrect errors,
341        because when scanning forward to find the reference group, PCRE was not
342        taking into account the duplicate group numbers. A pattern such as
343        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
344    
345    26. Changed a few more instances of "const unsigned char *" to USPTR, making
346        the feature of a custom pointer more persuasive (as requested by a user).
347    
348    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
349        pcretest.c, inside #ifndefs, because it seems they are sometimes already
350        pre-defined.
351    
352    28. Added support for (*UTF8) at the start of a pattern.
353    
354    29. Arrange for flags added by the "release type" setting in CMake to be shown
355        in the configuration summary.
356    
357    
358    Version 7.8 05-Sep-08
359    ---------------------
360    
361    1.  Replaced UCP searching code with optimized version as implemented for Ad
362        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
363        stage table and inline lookup instead of a function, giving speed ups of 2
364        to 5 times on some simple patterns that I tested. Permission was given to
365        distribute the MultiStage2.py script that generates the tables (it's not in
366        the tarball, but is in the Subversion repository).
367    
368    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
369        scripts.
370    
371    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
372        a group with a zero qualifier. The result of the study could be incorrect,
373        or the function might crash, depending on the pattern.
374    
375    4.  Caseless matching was not working for non-ASCII characters in back
376        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
377        It now works when Unicode Property Support is available.
378    
379    5.  In pcretest, an escape such as \x{de} in the data was always generating
380        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
381        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
382        truncation.
383    
384    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
385    
386    7.  Added two (int) casts to pcregrep when printing the difference of two
387        pointers, in case they are 64-bit values.
388    
389    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
390        test 2 if it fails.
391    
392    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
393        and a #define of that name to empty if it is not externally set. This is to
394        allow users of MSVC to set it if necessary.
395    
396    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
397        the convenience functions in the pcre_get.c source file.
398    
399    11. An option change at the start of a pattern that had top-level alternatives
400        could cause overwriting and/or a crash. This command provoked a crash in
401        some environments:
402    
403          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
404    
405        This potential security problem was recorded as CVE-2008-2371.
406    
407    12. For a pattern where the match had to start at the beginning or immediately
408        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
409        pcre_dfa_exec() could read past the end of the passed subject if there was
410        no match. To help with detecting such bugs (e.g. with valgrind), I modified
411        pcretest so that it places the subject at the end of its malloc-ed buffer.
412    
413    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
414        exec() might read past the end of the data buffer in UTF-8 mode.
415    
416    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
417        the data contained the byte 0x85 as part of a UTF-8 character within its
418        first line. This applied both to normal and DFA matching.
419    
420    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
421        /^[^d]*?$/8 failed to match "abc".
422    
423    16. Added a missing copyright notice to pcrecpp_internal.h.
424    
425    17. Make it more clear in the documentation that values returned from
426        pcre_exec() in ovector are byte offsets, not character counts.
427    
428    18. Tidied a few places to stop certain compilers from issuing warnings.
429    
430    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
431        supplied by Stefan Weber. I made a further small update for 7.8 because
432        there is a change of source arrangements: the pcre_searchfuncs.c module is
433        replaced by pcre_ucd.c.
434    
435    
436    Version 7.7 07-May-08
437    ---------------------
438    
439    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
440        a string to a long long, pretend we don't even have a long long." This is
441        done by checking for the strtoq, strtoll, and _strtoi64 functions.
442    
443    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
444        pre-7.6 versions, which defined a global no_arg variable instead of putting
445        it in the RE class. (See also #8 below.)
446    
447    3.  Remove a line of dead code, identified by coverity and reported by Nuno
448        Lopes.
449    
450    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
451    
452        (1) The include/exclude patterns were being applied to the whole pathnames
453            of files, instead of just to the final components.
454    
455        (2) If there was more than one level of directory, the subdirectories were
456            skipped unless they satisfied the include/exclude conditions. This is
457            inconsistent with GNU grep (and could even be seen as contrary to the
458            pcregrep specification - which I improved to make it absolutely clear).
459            The action now is always to scan all levels of directory, and just
460            apply the include/exclude patterns to regular files.
461    
462    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
463        --exclude_dir in the tests to avoid scanning .svn directories.
464    
465    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
466        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
467        doesn't support NULs in patterns.
468    
469    7.  Added some missing "const"s to declarations of static tables in
470        pcre_compile.c and pcre_dfa_exec.c.
471    
472    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
473        caused by fix #2  above. (Subsequently also a second patch to fix the
474        first patch. And a third patch - this was a messy problem.)
475    
476    9.  Applied Craig's patch to remove the use of push_back().
477    
478    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
479        matching function regexec().
480    
481    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
482        which, however, unlike Perl's \g{...}, are subroutine calls, not back
483        references. PCRE supports relative numbers with this syntax (I don't think
484        Oniguruma does).
485    
486    12. Previously, a group with a zero repeat such as (...){0} was completely
487        omitted from the compiled regex. However, this means that if the group
488        was called as a subroutine from elsewhere in the pattern, things went wrong
489        (an internal error was given). Such groups are now left in the compiled
490        pattern, with a new opcode that causes them to be skipped at execution
491        time.
492    
493    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
494        to the way PCRE behaves:
495    
496        (a) A lone ] character is dis-allowed (Perl treats it as data).
497    
498        (b) A back reference to an unmatched subpattern matches an empty string
499            (Perl fails the current match path).
500    
501        (c) A data ] in a character class must be notated as \] because if the
502            first data character in a class is ], it defines an empty class. (In
503            Perl it is not possible to have an empty class.) The empty class []
504            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
505            The negative empty class [^] matches any one character, independently
506            of the DOTALL setting.
507    
508    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
509        non-existent subpattern following a character class starting with ']' and
510        containing () gave an internal compiling error instead of "reference to
511        non-existent subpattern". Fortunately, when the pattern did exist, the
512        compiled code was correct. (When scanning forwards to check for the
513        existencd of the subpattern, it was treating the data ']' as terminating
514        the class, so got the count wrong. When actually compiling, the reference
515        was subsequently set up correctly.)
516    
517    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
518        it was being rejected as not supported by pcre_dfa_exec(), even though
519        other assertions are supported. I have made pcre_dfa_exec() support
520        (*FAIL).
521    
522    16. The implementation of 13c above involved the invention of a new opcode,
523        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
524        cannot be changed at match time, I realized I could make a small
525        improvement to matching performance by compiling OP_ALLANY instead of
526        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
527        on the OP_ANY path.
528    
529    17. Compiling pcretest on Windows with readline support failed without the
530        following two fixes: (1) Make the unistd.h include conditional on
531        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
532    
533    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
534        ncurses library to be included for pcretest when ReadLine support is
535        requested, but also to allow for it to be overridden. This patch came from
536        Daniel Bergström.
537    
538    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
539        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
540        any errors with the current Unicode tables. Thanks to Peter Kankowski for
541        spotting this.
542    
543    
544    Version 7.6 28-Jan-08
545    ---------------------
546    
547    1.  A character class containing a very large number of characters with
548        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
549        overflow.
550    
551    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
552        HAVE_LONG_LONG is not defined.
553    
554    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
555        bring them up to date and include new features. This patch includes:
556    
557        - Fixed PH's badly added libz and libbz2 support.
558        - Fixed a problem with static linking.
559        - Added pcredemo. [But later removed - see 7 below.]
560        - Fixed dftables problem and added an option.
561        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
562            HAVE_LONG_LONG.
563        - Added readline support for pcretest.
564        - Added an listing of the option settings after cmake has run.
565    
566    4.  A user submitted a patch to Makefile that makes it easy to create
567        "pcre.dll" under mingw when using Configure/Make. I added stuff to
568        Makefile.am that cause it to include this special target, without
569        affecting anything else. Note that the same mingw target plus all
570        the other distribution libraries and programs are now supported
571        when configuring with CMake (see 6 below) instead of with
572        Configure/Make.
573    
574    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
575        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
576        exported in the Windows port". It has not yet been confirmed that the patch
577        solves the problem, but it does no harm.
578    
579    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
580        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
581        with CMake, and also correct the comment about stack recursion.
582    
583    7.  Remove the automatic building of pcredemo from the ./configure system and
584        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
585        of a program that users should build themselves after PCRE is installed, so
586        building it automatically is not really right. What is more, it gave
587        trouble in some build environments.
588    
589    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
590    
591    
592    Version 7.5 10-Jan-08
593    ---------------------
594    
595    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
596        values in parens when parsing an RE using the C++ wrapper."
597    
598    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
599        Characters greater than 255 were excluded from the class instead of being
600        included.
601    
602    3.  The same bug as (2) above applied to negated POSIX classes such as
603        [:^space:].
604    
605    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
606        defined or documented. It seems to have been a typo for PCRE_STATIC, so
607        I have changed it.
608    
609    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
610        first named subpattern) and a construct such as (?&a) would reference the
611        first named subpattern whose name started with "a" (in other words, the
612        length check was missing). Both these problems are fixed. "Subpattern name
613        expected" is now given for (?&) (a zero-length name), and this patch also
614        makes it give the same error for \k'' (previously it complained that that
615        was a reference to a non-existent subpattern).
616    
617    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
618        this is right because (?- can be followed by option settings as well as by
619        digits. I have, however, made the messages clearer.
620    
621    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
622        than the number used in the conditional) now cause a compile-time error.
623        This is actually not compatible with Perl, which accepts such patterns, but
624        treats the conditional as always being FALSE (as PCRE used to), but it
625        seems to me that giving a diagnostic is better.
626    
627    8.  Change "alphameric" to the more common word "alphanumeric" in comments
628        and messages.
629    
630    9.  Fix two occurrences of "backslash" in comments that should have been
631        "backspace".
632    
633    10. Remove two redundant lines of code that can never be obeyed (their function
634        was moved elsewhere).
635    
636    11. The program that makes PCRE's Unicode character property table had a bug
637        which caused it to generate incorrect table entries for sequences of
638        characters that have the same character type, but are in different scripts.
639        It amalgamated them into a single range, with the script of the first of
640        them. In other words, some characters were in the wrong script. There were
641        thirteen such cases, affecting characters in the following ranges:
642    
643          U+002b0 - U+002c1
644          U+0060c - U+0060d
645          U+0061e - U+00612
646          U+0064b - U+0065e
647          U+0074d - U+0076d
648          U+01800 - U+01805
649          U+01d00 - U+01d77
650          U+01d9b - U+01dbf
651          U+0200b - U+0200f
652          U+030fc - U+030fe
653          U+03260 - U+0327f
654          U+0fb46 - U+0fbb1
655          U+10450 - U+1049d
656    
657    12. The -o option (show only the matching part of a line) for pcregrep was not
658        compatible with GNU grep in that, if there was more than one match in a
659        line, it showed only the first of them. It now behaves in the same way as
660        GNU grep.
661    
662    13. If the -o and -v options were combined for pcregrep, it printed a blank
663        line for every non-matching line. GNU grep prints nothing, and pcregrep now
664        does the same. The return code can be used to tell if there were any
665        non-matching lines.
666    
667    14. Added --file-offsets and --line-offsets to pcregrep.
668    
669    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
670        infinitely looping recursion. The bug was that positive lookaheads were not
671        being skipped when checking for a possible empty match (negative lookaheads
672        and both kinds of lookbehind were skipped).
673    
674    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
675        inclusion of <windows.h> to before rather than after the definition of
676        INVALID_FILE_ATTRIBUTES (patch from David Byron).
677    
678    17. Specifying a possessive quantifier with a specific limit for a Unicode
679        character property caused pcre_compile() to compile bad code, which led at
680        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
681        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
682        caused the error; without that there was no problem.
683    
684    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
685    
686    19. Added --enable-pcretest-libreadline.
687    
688    20. In pcrecpp.cc, the variable 'count' was incremented twice in
689        RE::GlobalReplace(). As a result, the number of replacements returned was
690        double what it should be. I removed one of the increments, but Craig sent a
691        later patch that removed the other one (the right fix) and added unit tests
692        that check the return values (which was not done before).
693    
694    21. Several CMake things:
695    
696        (1) Arranged that, when cmake is used on Unix, the libraries end up with
697            the names libpcre and libpcreposix, not just pcre and pcreposix.
698    
699        (2) The above change means that pcretest and pcregrep are now correctly
700            linked with the newly-built libraries, not previously installed ones.
701    
702        (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
703    
704    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
705        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
706        UTF-8 newline character). The key issue is that the pattern starts .*;
707        this means that the match must be either at the beginning, or after a
708        newline. The bug was in the code for advancing after a failed match and
709        checking that the new position followed a newline. It was not taking
710        account of UTF-8 characters correctly.
711    
712    23. PCRE was behaving differently from Perl in the way it recognized POSIX
713        character classes. PCRE was not treating the sequence [:...:] as a
714        character class unless the ... were all letters. Perl, however, seems to
715        allow any characters between [: and :], though of course it rejects as
716        unknown any "names" that contain non-letters, because all the known class
717        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
718        for example, whereas PCRE did not - it did not recognize a POSIX character
719        class. This seemed a bit dangerous, so the code has been changed to be
720        closer to Perl. The behaviour is not identical to Perl, because PCRE will
721        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
722        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
723        Perl does, and where it didn't before.
724    
725    24. Rewrite so as to remove the single use of %n from pcregrep because in some
726        Windows environments %n is disabled by default.
727    
728    
729    Version 7.4 21-Sep-07
730    ---------------------
731    
732    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
733        means that a class such as [\s] counted as "explicit reference to CR or
734        LF". That isn't really right - the whole point of the change was to try to
735        help when there was an actual mention of one of the two characters. So now
736        the change happens only if \r or \n (or a literal CR or LF) character is
737        encountered.
738    
739    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
740        of both had grown to the point where there were only 3 bits left.
741        Fortunately, there was spare space in the data structure, and so I have
742        moved the internal flags into a new 16-bit field to free up more option
743        bits.
744    
745    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
746        but did not set the internal JCHANGED flag - either of these is enough to
747        control the way the "get" function works - but the PCRE_INFO_JCHANGED
748        facility is supposed to tell if (?J) was ever used, so now (?J) at the
749        start sets both bits.
750    
751    4.  Added options (at build time, compile time, exec time) to change \R from
752        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
753    
754    5.  doc/pcresyntax.html was missing from the distribution.
755    
756    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
757        compatibility, even though it is no longer used.
758    
759    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
760        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
761        windows.h file is present (where different names are used). [This was
762        reversed later after testing - see 16 below.]
763    
764    8.  Changed all #include <config.h> to #include "config.h". There were also
765        some further <pcre.h> cases that I changed to "pcre.h".
766    
767    9.  When pcregrep was used with the --colour option, it missed the line ending
768        sequence off the lines that it output.
769    
770    10. It was pointed out to me that arrays of string pointers cause lots of
771        relocations when a shared library is dynamically loaded. A technique of
772        using a single long string with a table of offsets can drastically reduce
773        these. I have refactored PCRE in four places to do this. The result is
774        dramatic:
775    
776          Originally:                          290
777          After changing UCP table:            187
778          After changing error message table:   43
779          After changing table of "verbs"       36
780          After changing table of Posix names   22
781    
782        Thanks to the folks working on Gregex for glib for this insight.
783    
784    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
785        unicode-properties was also set.
786    
787    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
788    
789    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
790        checked only for CRLF.
791    
792    14. Added casts to pcretest.c to avoid compiler warnings.
793    
794    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
795    
796    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
797        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
798        entirely. This removes changes made in 7 above.
799    
800    17. The CMake files have been updated, and there is now more information about
801        building with CMake in the NON-UNIX-USE document.
802    
803    
804    Version 7.3 28-Aug-07
805    ---------------------
806    
807     1. In the rejigging of the build system that eventually resulted in 7.1, the
808        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
809        brackets there is not right, since it causes compilers to look for an
810        installed pcre.h, not the version that is in the source that is being
811        compiled (which of course may be different). I have changed it back to:
812    
813          #include "pcre.h"
814    
815        I have a vague recollection that the change was concerned with compiling in
816        different directories, but in the new build system, that is taken care of
817        by the VPATH setting the Makefile.
818    
819     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
820        when the subject happened to end in the byte 0x85 (e.g. if the last
821        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
822        characters but of course it shouldn't be taken as a newline when it is part
823        of another character. The bug was that, for an unlimited repeat of . in
824        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
825        characters when looking for a newline.
826    
827     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
828    
829     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
830        in debug output.
831    
832     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
833        long printing in the pcrecpp unittest when running under MinGW.
834    
835     6. ESC_K was left out of the EBCDIC table.
836    
837     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
838        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
839        limit also applies to "virtual nesting" when a pattern is recursive, and in
840        this case 1000 isn't so big. I have been able to remove this limit at the
841        expense of backing off one optimization in certain circumstances. Normally,
842        when pcre_exec() would call its internal match() function recursively and
843        immediately return the result unconditionally, it uses a "tail recursion"
844        feature to save stack. However, when a subpattern that can match an empty
845        string has an unlimited repetition quantifier, it no longer makes this
846        optimization. That gives it a stack frame in which to save the data for
847        checking that an empty string has been matched. Previously this was taken
848        from the 1000-entry workspace that had been reserved. So now there is no
849        explicit limit, but more stack is used.
850    
851     8. Applied Daniel's patches to solve problems with the import/export magic
852        syntax that is required for Windows, and which was going wrong for the
853        pcreposix and pcrecpp parts of the library. These were overlooked when this
854        problem was solved for the main library.
855    
856     9. There were some crude static tests to avoid integer overflow when computing
857        the size of patterns that contain repeated groups with explicit upper
858        limits. As the maximum quantifier is 65535, the maximum group length was
859        set at 30,000 so that the product of these two numbers did not overflow a
860        32-bit integer. However, it turns out that people want to use groups that
861        are longer than 30,000 bytes (though not repeat them that many times).
862        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
863        made it possible to implement the integer overflow checks in a much more
864        dynamic way, which I have now done. The artificial limitation on group
865        length has been removed - we now have only the limit on the total length of
866        the compiled pattern, which depends on the LINK_SIZE setting.
867    
868    10. Fixed a bug in the documentation for get/copy named substring when
869        duplicate names are permitted. If none of the named substrings are set, the
870        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
871        empty string.
872    
873    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
874        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
875        because the ] is interpreted as the first data character and the
876        terminating ] is not found. PCRE has been made compatible with Perl in this
877        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
878        cause memory overwriting.
879    
880    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
881        string has been matched (to stop an infinite loop). It was not recognizing
882        a conditional subpattern that could match an empty string if that
883        subpattern was within another subpattern. For example, it looped when
884        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
885        condition was not nested. This bug has been fixed.
886    
887    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
888        past the start of the subject in the presence of bytes with the top bit
889        set, for example "\x8aBCD".
890    
891    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
892        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
893    
894    14. Optimized (?!) to (*FAIL).
895    
896    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
897        This restricts code points to be within the range 0 to 0x10FFFF, excluding
898        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
899        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
900        does: it's just the validity check that is more restrictive.
901    
902    16. Inserted checks for integer overflows during escape sequence (backslash)
903        processing, and also fixed erroneous offset values for syntax errors during
904        backslash processing.
905    
906    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
907        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
908    
909    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
910        caused an overrun.
911    
912    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
913        something other than just ASCII characters) inside a group that had an
914        unlimited repeat caused a loop at compile time (while checking to see
915        whether the group could match an empty string).
916    
917    20. Debugging a pattern containing \p or \P could cause a crash. For example,
918        [\P{Any}] did so. (Error in the code for printing property names.)
919    
920    21. An orphan \E inside a character class could cause a crash.
921    
922    22. A repeated capturing bracket such as (A)? could cause a wild memory
923        reference during compilation.
924    
925    23. There are several functions in pcre_compile() that scan along a compiled
926        expression for various reasons (e.g. to see if it's fixed length for look
927        behind). There were bugs in these functions when a repeated \p or \P was
928        present in the pattern. These operators have additional parameters compared
929        with \d, etc, and these were not being taken into account when moving along
930        the compiled data. Specifically:
931    
932        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
933            length.
934    
935        (b) An item such as \pL+ within a repeated group could cause crashes or
936            loops.
937    
938        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
939            "reference to non-existent subpattern" error.
940    
941        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
942    
943    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
944        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
945    
946    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
947    
948    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
949        character were causing crashes (broken optimization).
950    
951    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
952        \p or \P) caused a compile-time loop.
953    
954    28. More problems have arisen in unanchored patterns when CRLF is a valid line
955        break. For example, the unstudied pattern [\r\n]A does not match the string
956        "\r\nA" because change 7.0/46 below moves the current point on by two
957        characters after failing to match at the start. However, the pattern \nA
958        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
959        the same is true. There doesn't seem any very clean way out of this, but
960        what I have chosen to do makes the common cases work: PCRE now takes note
961        of whether there can be an explicit match for \r or \n anywhere in the
962        pattern, and if so, 7.0/46 no longer applies. As part of this change,
963        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
964        pattern has explicit CR or LF references.
965    
966    29. Added (*CR) etc for changing newline setting at start of pattern.
967    
968    
969    Version 7.2 19-Jun-07
970    ---------------------
971    
972     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
973        which is apparently normally available under Windows.
974    
975     2. Re-jig the pcregrep tests with different newline settings in an attempt
976        to make them independent of the local environment's newline setting.
977    
978     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
979    
980     4. Some of the "internals" tests were previously cut out when the link size
981        was not 2, because the output contained actual offsets. The recent new
982        "Z" feature of pcretest means that these can be cut out, making the tests
983        usable with all link sizes.
984    
985     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
986        stack recursion. This gives a massive performance boost under BSD, but just
987        a small improvement under Linux. However, it saves one field in the frame
988        in all cases.
989    
990     6. Added more features from the forthcoming Perl 5.10:
991    
992        (a) (?-n) (where n is a string of digits) is a relative subroutine or
993            recursion call. It refers to the nth most recently opened parentheses.
994    
995        (b) (?+n) is also a relative subroutine call; it refers to the nth next
996            to be opened parentheses.
997    
998        (c) Conditions that refer to capturing parentheses can be specified
999            relatively, for example, (?(-2)... or (?(+3)...
1000    
1001        (d) \K resets the start of the current match so that everything before
1002            is not part of it.
1003    
1004        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1005    
1006        (f) \g{name} is another synonym - part of Perl 5.10's unification of
1007            reference syntax.
1008    
1009        (g) (?| introduces a group in which the numbering of parentheses in each
1010            alternative starts with the same number.
1011    
1012        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
1013    
1014     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
1015        PCRE_INFO_JCHANGED.
1016    
1017     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
1018        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
1019        for detecting groups that can match an empty string.
1020    
1021     9. A pattern with a very large number of alternatives (more than several
1022        hundred) was running out of internal workspace during the pre-compile
1023        phase, where pcre_compile() figures out how much memory will be needed. A
1024        bit of new cunning has reduced the workspace needed for groups with
1025        alternatives. The 1000-alternative test pattern now uses 12 bytes of
1026        workspace instead of running out of the 4096 that are available.
1027    
1028    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
1029    
1030    11. Applied patch from Google to remove an optimization that didn't quite work.
1031        The report of the bug said:
1032    
1033          pcrecpp::RE("a*").FullMatch("aaa") matches, while
1034          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
1035          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
1036    
1037    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
1038        it matched the wrong number of bytes.
1039    
1040    
1041    Version 7.1 24-Apr-07
1042    ---------------------
1043    
1044     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
1045        that is more "standard", making use of automake and other Autotools. There
1046        is some re-arrangement of the files and adjustment of comments consequent
1047        on this.
1048    
1049     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
1050        for recursive directory scanning broke on some systems because the files
1051        are not scanned in any specific order and on different systems the order
1052        was different. A call to "sort" has been inserted into RunGrepTest for the
1053        approprate test as a short-term fix. In the longer term there may be an
1054        alternative.
1055    
1056     3. I had an email from Eric Raymond about problems translating some of PCRE's
1057        man pages to HTML (despite the fact that I distribute HTML pages, some
1058        people do their own conversions for various reasons). The problems
1059        concerned the use of low-level troff macros .br and .in. I have therefore
1060        removed all such uses from the man pages (some were redundant, some could
1061        be replaced by .nf/.fi pairs). The 132html script that I use to generate
1062        HTML has been updated to handle .nf/.fi and to complain if it encounters
1063        .br or .in.
1064    
1065     4. Updated comments in configure.ac that get placed in config.h.in and also
1066        arranged for config.h to be included in the distribution, with the name
1067        config.h.generic, for the benefit of those who have to compile without
1068        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
1069    
1070     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
1071        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
1072        makevp.bat for the current PCRE, using the additional files
1073        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
1074    
1075     6. A Windows user reported a minor discrepancy with test 2, which turned out
1076        to be caused by a trailing space on an input line that had got lost in his
1077        copy. The trailing space was an accident, so I've just removed it.
1078    
1079     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
1080        that is needed.
1081    
1082     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
1083        as "const" (a) because they are and (b) because it helps the PHP
1084        maintainers who have recently made a script to detect big data structures
1085        in the php code that should be moved to the .rodata section. I remembered
1086        to update Builducptable as well, so it won't revert if ucptable.h is ever
1087        re-created.
1088    
1089     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
1090        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
1091        order to be able to cut out the UTF-8 tables in the latter when UTF-8
1092        support is not required. This saves 1.5-2K of code, which is important in
1093        some applications.
1094    
1095        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
1096        so as not to refer to the tables, even though these functions will never be
1097        called when UTF-8 support is disabled. Otherwise there are problems with a
1098        shared library.
1099    
1100    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
1101    
1102        (a) It was defining its arguments as char * instead of void *.
1103    
1104        (b) It was assuming that all moves were upwards in memory; this was true
1105            a long time ago when I wrote it, but is no longer the case.
1106    
1107        The emulated memove() is provided for those environments that have neither
1108        memmove() nor bcopy(). I didn't think anyone used it these days, but that
1109        is clearly not the case, as these two bugs were recently reported.
1110    
1111    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
1112        and Detrail to create the HTML documentation, the .txt form of the man
1113        pages, and it removes trailing spaces from listed files. It also creates
1114        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
1115        case, it wraps all the #defines with #ifndefs. This script should be run
1116        before "make dist".
1117    
1118    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1119        with Unicode property support.
1120    
1121        (a) For a maximizing quantifier, if the two different cases of the
1122            character were of different lengths in their UTF-8 codings (there are
1123            some cases like this - I found 11), and the matching function had to
1124            back up over a mixture of the two cases, it incorrectly assumed they
1125            were both the same length.
1126    
1127        (b) When PCRE was configured to use the heap rather than the stack for
1128            recursion during matching, it was not correctly preserving the data for
1129            the other case of a UTF-8 character when checking ahead for a match
1130            while processing a minimizing repeat. If the check also involved
1131            matching a wide character, but failed, corruption could cause an
1132            erroneous result when trying to check for a repeat of the original
1133            character.
1134    
1135    13. Some tidying changes to the testing mechanism:
1136    
1137        (a) The RunTest script now detects the internal link size and whether there
1138            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1139            values substituted by "configure". (The RunGrepTest script already did
1140            this for UTF-8.) The configure.ac script no longer substitutes the
1141            relevant variables.
1142    
1143        (b) The debugging options /B and /D in pcretest show the compiled bytecode
1144            with length and offset values. This means that the output is different
1145            for different internal link sizes. Test 2 is skipped for link sizes
1146            other than 2 because of this, bypassing the problem. Unfortunately,
1147            there was also a test in test 3 (the locale tests) that used /B and
1148            failed for link sizes other than 2. Rather than cut the whole test out,
1149            I have added a new /Z option to pcretest that replaces the length and
1150            offset values with spaces. This is now used to make test 3 independent
1151            of link size. (Test 2 will be tidied up later.)
1152    
1153    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1154        segmentation fault instead of returning the appropriate error message.
1155    
1156    15. In multiline mode when the newline sequence was set to "any", the pattern
1157        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1158        This doesn't seem right; it now treats the CRLF combination as the line
1159        ending, and so does not match in that case. It's only a pattern such as ^$
1160        that would hit this one: something like ^ABC$ would have failed after \r
1161        and then tried again after \r\n.
1162    
1163    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1164        in an attempt to make files that differ only in their line terminators
1165        compare equal. This works on Linux.
1166    
1167    17. Under certain error circumstances pcregrep might try to free random memory
1168        as it exited. This is now fixed, thanks to valgrind.
1169    
1170    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1171        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1172        was because its rules for how to advance for /g after matching an empty
1173        string at the end of a line did not allow for this case. They now check for
1174        it specially.
1175    
1176    20. pcretest is supposed to handle patterns and data of any length, by
1177        extending its buffers when necessary. It was getting this wrong when the
1178        buffer for a data line had to be extended.
1179    
1180    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1181        CRLF as a newline sequence.
1182    
1183    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1184        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1185        I have nevertheless tidied it up.
1186    
1187    23. Added some casts to kill warnings from HP-UX ia64 compiler.
1188    
1189    24. Added a man page for pcre-config.
1190    
1191    
1192    Version 7.0 19-Dec-06
1193    ---------------------
1194    
1195     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1196        moving to gcc 4.1.1.
1197    
1198     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1199        sys/time.h, which is documented as needed for this function. It doesn't
1200        seem to matter on Linux, but it showed up on some releases of OS X.
1201    
1202     3. It seems that there are systems where bytes whose values are greater than
1203        127 match isprint() in the "C" locale. The "C" locale should be the
1204        default when a C program starts up. In most systems, only ASCII printing
1205        characters match isprint(). This difference caused the output from pcretest
1206        to vary, making some of the tests fail. I have changed pcretest so that:
1207    
1208        (a) When it is outputting text in the compiled version of a pattern, bytes
1209            other than 32-126 are always shown as hex escapes.
1210    
1211        (b) When it is outputting text that is a matched part of a subject string,
1212            it does the same, unless a different locale has been set for the match
1213            (using the /L modifier). In this case, it uses isprint() to decide.
1214    
1215     4. Fixed a major bug that caused incorrect computation of the amount of memory
1216        required for a compiled pattern when options that changed within the
1217        pattern affected the logic of the preliminary scan that determines the
1218        length. The relevant options are -x, and -i in UTF-8 mode. The result was
1219        that the computed length was too small. The symptoms of this bug were
1220        either the PCRE error "internal error: code overflow" from pcre_compile(),
1221        or a glibc crash with a message such as "pcretest: free(): invalid next
1222        size (fast)". Examples of patterns that provoked this bug (shown in
1223        pcretest format) are:
1224    
1225          /(?-x: )/x
1226          /(?x)(?-x: \s*#\s*)/
1227          /((?i)[\x{c0}])/8
1228          /(?i:[\x{c0}])/8
1229    
1230        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1231        is now done differently.
1232    
1233     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1234        wrapper classes; (b) implement a new function in the C++ scanner that is
1235        more efficient than the old way of doing things because it avoids levels of
1236        recursion in the regex matching; (c) add a paragraph to the documentation
1237        for the FullMatch() function.
1238    
1239     6. The escape sequence \n was being treated as whatever was defined as
1240        "newline". Not only was this contrary to the documentation, which states
1241        that \n is character 10 (hex 0A), but it also went horribly wrong when
1242        "newline" was defined as CRLF. This has been fixed.
1243    
1244     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1245        was being set to -1 for the "end of line" case (supposedly a value that no
1246        character can have). Though this value is never used (the check for end of
1247        line is "zero bytes in current character"), it caused compiler complaints.
1248        I've changed it to 0xffffffff.
1249    
1250     8. In pcre_version.c, the version string was being built by a sequence of
1251        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1252        string (as it is for production releases) called a macro with an empty
1253        argument. The C standard says the result of this is undefined. The gcc
1254        compiler treats it as an empty string (which was what was wanted) but it is
1255        reported that Visual C gives an error. The source has been hacked around to
1256        avoid this problem.
1257    
1258     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1259        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1260        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1261        of them did).
1262    
1263    10. Originally, pcretest opened its input and output without "b"; then I was
1264        told that "b" was needed in some environments, so it was added for release
1265        5.0 to both the input and output. (It makes no difference on Unix-like
1266        systems.) Later I was told that it is wrong for the input on Windows. I've
1267        now abstracted the modes into two macros, to make it easier to fiddle with
1268        them, and removed "b" from the input mode under Windows.
1269    
1270    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1271    
1272    12. Added -help and --help to pcretest as an official way of being reminded
1273        of the options.
1274    
1275    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1276        and pcrecpp.cc because they annoy compilers at high warning levels.
1277    
1278    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1279    
1280    15. Fixed an occurrence of == in configure.ac that should have been = (shell
1281        scripts are not C programs :-) and which was not noticed because it works
1282        on Linux.
1283    
1284    16. pcretest is supposed to handle any length of pattern and data line (as one
1285        line or as a continued sequence of lines) by extending its input buffer if
1286        necessary. This feature was broken for very long pattern lines, leading to
1287        a string of junk being passed to pcre_compile() if the pattern was longer
1288        than about 50K.
1289    
1290    17. I have done a major re-factoring of the way pcre_compile() computes the
1291        amount of memory needed for a compiled pattern. Previously, there was code
1292        that made a preliminary scan of the pattern in order to do this. That was
1293        OK when PCRE was new, but as the facilities have expanded, it has become
1294        harder and harder to keep it in step with the real compile phase, and there
1295        have been a number of bugs (see for example, 4 above). I have now found a
1296        cunning way of running the real compile function in a "fake" mode that
1297        enables it to compute how much memory it would need, while actually only
1298        ever using a few hundred bytes of working memory and without too many
1299        tests of the mode. This should make future maintenance and development
1300        easier. A side effect of this work is that the limit of 200 on the nesting
1301        depth of parentheses has been removed (though this was never a serious
1302        limitation, I suspect). However, there is a downside: pcre_compile() now
1303        runs more slowly than before (30% or more, depending on the pattern). I
1304        hope this isn't a big issue. There is no effect on runtime performance.
1305    
1306    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1307        newline (only possible for the last line of a file) and it was a
1308        pattern that set a locale (followed by /Lsomething), pcretest crashed.
1309    
1310    19. Added additional timing features to pcretest. (1) The -tm option now times
1311        matching only, not compiling. (2) Both -t and -tm can be followed, as a
1312        separate command line item, by a number that specifies the number of
1313        repeats to use when timing. The default is 50000; this gives better
1314        precision, but takes uncomfortably long for very large patterns.
1315    
1316    20. Extended pcre_study() to be more clever in cases where a branch of a
1317        subpattern has no definite first character. For example, (a*|b*)[cd] would
1318        previously give no result from pcre_study(). Now it recognizes that the
1319        first character must be a, b, c, or d.
1320    
1321    21. There was an incorrect error "recursive call could loop indefinitely" if
1322        a subpattern (or the entire pattern) that was being tested for matching an
1323        empty string contained only one non-empty item after a nested subpattern.
1324        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1325        incorrectly, because the \d was being skipped in the check.
1326    
1327    22. The pcretest program now has a new pattern option /B and a command line
1328        option -b, which is equivalent to adding /B to every pattern. This causes
1329        it to show the compiled bytecode, without the additional information that
1330        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1331        is the same as /B/I).
1332    
1333    23. A new optimization is now able automatically to treat some sequences such
1334        as a*b as a*+b. More specifically, if something simple (such as a character
1335        or a simple class like \d) has an unlimited quantifier, and is followed by
1336        something that cannot possibly match the quantified thing, the quantifier
1337        is automatically "possessified".
1338    
1339    24. A recursive reference to a subpattern whose number was greater than 39
1340        went wrong under certain circumstances in UTF-8 mode. This bug could also
1341        have affected the operation of pcre_study().
1342    
1343    25. Realized that a little bit of performance could be had by replacing
1344        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1345    
1346    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1347    
1348    27. Possessive quantifiers such as a++ were previously implemented by turning
1349        them into atomic groups such as ($>a+). Now they have their own opcodes,
1350        which improves performance. This includes the automatically created ones
1351        from 23 above.
1352    
1353    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1354        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1355        the first matched character to be a colon. This applied both to named and
1356        numbered groups.
1357    
1358    29. The ucpinternal.h header file was missing its idempotency #ifdef.
1359    
1360    30. I was sent a "project" file called libpcre.a.dev which I understand makes
1361        building PCRE on Windows easier, so I have included it in the distribution.
1362    
1363    31. There is now a check in pcretest against a ridiculously large number being
1364        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1365        loop, the loop is abandoned.
1366    
1367    32. Forward references to subpatterns in conditions such as (?(2)...) where
1368        subpattern 2 is defined later cause pcre_compile() to search forwards in
1369        the pattern for the relevant set of parentheses. This search went wrong
1370        when there were unescaped parentheses in a character class, parentheses
1371        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1372    
1373    33. "Subroutine" calls and backreferences were previously restricted to
1374        referencing subpatterns earlier in the regex. This restriction has now
1375        been removed.
1376    
1377    34. Added a number of extra features that are going to be in Perl 5.10. On the
1378        whole, these are just syntactic alternatives for features that PCRE had
1379        previously implemented using the Python syntax or my own invention. The
1380        other formats are all retained for compatibility.
1381    
1382        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1383            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1384            also .NET compatible.
1385    
1386        (b) A recursion or subroutine call to a named group can now be defined as
1387            (?&name) as well as (?P>name).
1388    
1389        (c) A backreference to a named group can now be defined as \k<name> or
1390            \k'name' as well as (?P=name). The new forms, as well as being in Perl
1391            5.10, are also .NET compatible.
1392    
1393        (d) A conditional reference to a named group can now use the syntax
1394            (?(<name>) or (?('name') as well as (?(name).
1395    
1396        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1397            groups (named and numbered) that are never evaluated inline, but can be
1398            called as "subroutines" from elsewhere. In effect, the DEFINE condition
1399            is always false. There may be only one alternative in such a group.
1400    
1401        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1402            as the simple (?(R). The condition is true only if the most recent
1403            recursion is that of the given number or name. It does not search out
1404            through the entire recursion stack.
1405    
1406        (g) The escape \gN or \g{N} has been added, where N is a positive or
1407            negative number, specifying an absolute or relative reference.
1408    
1409    35. Tidied to get rid of some further signed/unsigned compiler warnings and
1410        some "unreachable code" warnings.
1411    
1412    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1413        things, this adds five new scripts.
1414    
1415    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1416        There were also incompatibilities regarding the handling of \Q..\E inside
1417        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1418        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1419    
1420    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1421        matches an empty string, and forcibly breaks the loop. There were bugs in
1422        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
1423        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
1424        separate and independent bugs (that affected different cases) have been
1425        fixed.
1426    
1427    39. Refactored the code to abolish the use of different opcodes for small
1428        capturing bracket numbers. This is a tidy that I avoided doing when I
1429        removed the limit on the number of capturing brackets for 3.5 back in 2001.
1430        The new approach is not only tidier, it makes it possible to reduce the
1431        memory needed to fix the previous bug (38).
1432    
1433    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1434        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1435        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1436        mode.
1437    
1438    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1439        report.
1440    
1441    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1442        copy construction and assignment in the C++ wrapper.
1443    
1444    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1445        couple of bugs that could have given wrong results in the "--newline=crlf"
1446        case.
1447    
1448    44. Added a number of casts and did some reorganization of signed/unsigned int
1449        variables following suggestions from Dair Grant. Also renamed the variable
1450        "this" as "item" because it is a C++ keyword.
1451    
1452    45. Arranged for dftables to add
1453    
1454          #include "pcre_internal.h"
1455    
1456        to pcre_chartables.c because without it, gcc 4.x may remove the array
1457        definition from the final binary if PCRE is built into a static library and
1458        dead code stripping is activated.
1459    
1460    46. For an unanchored pattern, if a match attempt fails at the start of a
1461        newline sequence, and the newline setting is CRLF or ANY, and the next two
1462        characters are CRLF, advance by two characters instead of one.
1463    
1464    
1465    Version 6.7 04-Jul-06
1466    ---------------------
1467    
1468     1. In order to handle tests when input lines are enormously long, pcretest has
1469        been re-factored so that it automatically extends its buffers when
1470        necessary. The code is crude, but this _is_ just a test program. The
1471        default size has been increased from 32K to 50K.
1472    
1473     2. The code in pcre_study() was using the value of the re argument before
1474        testing it for NULL. (Of course, in any sensible call of the function, it
1475        won't be NULL.)
1476    
1477     3. The memmove() emulation function in pcre_internal.h, which is used on
1478        systems that lack both memmove() and bcopy() - that is, hardly ever -
1479        was missing a "static" storage class specifier.
1480    
1481     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1482        containing an extended class (one that cannot be represented by a bitmap
1483        because it contains high-valued characters or Unicode property items, e.g.
1484        [\pZ]). Almost always one would set UTF-8 mode when processing such a
1485        pattern, but PCRE should not loop if you do not (it no longer does).
1486        [Detail: two cases were found: (a) a repeated subpattern containing an
1487        extended class; (b) a recursive reference to a subpattern that followed a
1488        previous extended class. It wasn't skipping over the extended class
1489        correctly when UTF-8 mode was not set.]
1490    
1491     5. A negated single-character class was not being recognized as fixed-length
1492        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1493        compile error "lookbehind assertion is not fixed length".
1494    
1495     6. The RunPerlTest auxiliary script was showing an unexpected difference
1496        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1497        write a Perl script that can interpret lines of an input file either as
1498        byte characters or as UTF-8, which is what "perltest" was being required to
1499        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1500        can't do is switch easily at run time between having the "use utf8;" pragma
1501        or not. In the end, I fudged it by using the RunPerlTest script to insert
1502        "use utf8;" explicitly for the UTF-8 tests.
1503    
1504     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1505        the end of the subject string, contrary to the documentation and to what
1506        Perl does. This was true of both matching functions. Now it matches only at
1507        the start of the subject and immediately after *internal* newlines.
1508    
1509     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1510        a pointer to an int instead of a pointer to an unsigned long int. This
1511        caused problems on 64-bit systems.
1512    
1513     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1514        instance of the 'standard' template library not being so standard".
1515    
1516    10. There was no check on the number of named subpatterns nor the maximum
1517        length of a subpattern name. The product of these values is used to compute
1518        the size of the memory block for a compiled pattern. By supplying a very
1519        long subpattern name and a large number of named subpatterns, the size
1520        computation could be caused to overflow. This is now prevented by limiting
1521        the length of names to 32 characters, and the number of named subpatterns
1522        to 10,000.
1523    
1524    11. Subpatterns that are repeated with specific counts have to be replicated in
1525        the compiled pattern. The size of memory for this was computed from the
1526        length of the subpattern and the repeat count. The latter is limited to
1527        65535, but there was no limit on the former, meaning that integer overflow
1528        could in principle occur. The compiled length of a repeated subpattern is
1529        now limited to 30,000 bytes in order to prevent this.
1530    
1531    12. Added the optional facility to have named substrings with the same name.
1532    
1533    13. Added the ability to use a named substring as a condition, using the
1534        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1535        are numbers (not recommended). Forward references are permitted.
1536    
1537    14. Added forward references in named backreferences (if you see what I mean).
1538    
1539    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1540        pattern could run off the end of the subject. For example, the pattern
1541        "(?s)(.{1,5})"8 did this with the subject "ab".
1542    
1543    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1544        PCRE_CASELESS was set when matching characters that were quantified with ?
1545        or *.
1546    
1547    17. A character class other than a single negated character that had a minimum
1548        but no maximum quantifier - for example [ab]{6,} - was not handled
1549        correctly by pce_dfa_exec(). It would match only one character.
1550    
1551    18. A valid (though odd) pattern that looked like a POSIX character
1552        class but used an invalid character after [ (for example [[,abc,]]) caused
1553        pcre_compile() to give the error "Failed: internal error: code overflow" or
1554        in some cases to crash with a glibc free() error. This could even happen if
1555        the pattern terminated after [[ but there just happened to be a sequence of
1556        letters, a binary zero, and a closing ] in the memory that followed.
1557    
1558    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1559        over the years. Originally (before any Unicode support), just the bottom 8
1560        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1561        output from "man perlunicode" includes this:
1562    
1563          The regular expression compiler produces polymorphic opcodes.  That
1564          is, the pattern adapts to the data and automatically switches to
1565          the Unicode character scheme when presented with Unicode data--or
1566          instead uses a traditional byte scheme when presented with byte
1567          data.
1568    
1569        Sadly, a wide octal escape does not cause a switch, and in a string with
1570        no other multibyte characters, these octal escapes are treated as before.
1571        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
1572        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1573        Unicode string.
1574    
1575        I have not perpetrated such confusion in PCRE. Up till now, it took just
1576        the bottom 8 bits, as in old Perl. I have now made octal escapes with
1577        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1578        translate to the appropriate multibyte character.
1579    
1580    29. Applied some refactoring to reduce the number of warnings from Microsoft
1581        and Borland compilers. This has included removing the fudge introduced
1582        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1583        a warning about an unused variable.
1584    
1585    21. PCRE has not included VT (character 0x0b) in the set of whitespace
1586        characters since release 4.0, because Perl (from release 5.004) does not.
1587        [Or at least, is documented not to: some releases seem to be in conflict
1588        with the documentation.] However, when a pattern was studied with
1589        pcre_study() and all its branches started with \s, PCRE still included VT
1590        as a possible starting character. Of course, this did no harm; it just
1591        caused an unnecessary match attempt.
1592    
1593    22. Removed a now-redundant internal flag bit that recorded the fact that case
1594        dependency changed within the pattern. This was once needed for "required
1595        byte" processing, but is no longer used. This recovers a now-scarce options
1596        bit. Also moved the least significant internal flag bit to the most-
1597        significant bit of the word, which was not previously used (hangover from
1598        the days when it was an int rather than a uint) to free up another bit for
1599        the future.
1600    
1601    23. Added support for CRLF line endings as well as CR and LF. As well as the
1602        default being selectable at build time, it can now be changed at runtime
1603        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1604        specify that it is scanning data with non-default line endings.
1605    
1606    24. Changed the definition of CXXLINK to make it agree with the definition of
1607        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1608    
1609    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1610        recursions. This makes a big different to stack usage for some patterns.
1611    
1612    26. If a subpattern containing a named recursion or subroutine reference such
1613        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1614        the space required for the compiled pattern went wrong and gave too small a
1615        value. Depending on the environment, this could lead to "Failed: internal
1616        error: code overflow at offset 49" or "glibc detected double free or
1617        corruption" errors.
1618    
1619    27. Applied patches from Google (a) to support the new newline modes and (b) to
1620        advance over multibyte UTF-8 characters in GlobalReplace.
1621    
1622    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1623        difference for some implementation of PCRE in some Windows version.
1624    
1625    29. Added some extra testing facilities to pcretest:
1626    
1627        \q<number>   in a data line sets the "match limit" value
1628        \Q<number>   in a data line sets the "match recursion limt" value
1629        -S <number>  sets the stack size, where <number> is in megabytes
1630    
1631        The -S option isn't available for Windows.
1632    
1633    
1634    Version 6.6 06-Feb-06
1635    ---------------------
1636    
1637     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1638        in pcreposix.h. I have copied the definition from pcre.h.
1639    
1640     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1641        because pcre.h is no longer a built file.
1642    
1643     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1644        not normally included in the compiled code.
1645    
1646    
1647    Version 6.5 01-Feb-06
1648    ---------------------
1649    
1650     1. When using the partial match feature with pcre_dfa_exec(), it was not
1651        anchoring the second and subsequent partial matches at the new starting
1652        point. This could lead to incorrect results. For example, with the pattern
1653        /1234/, partially matching against "123" and then "a4" gave a match.
1654    
1655     2. Changes to pcregrep:
1656    
1657        (a) All non-match returns from pcre_exec() were being treated as failures
1658            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1659            error message is output. Some extra information is given for the
1660            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
1661            probably the only errors that are likely to be caused by users (by
1662            specifying a regex that has nested indefinite repeats, for instance).
1663            If there are more than 20 of these errors, pcregrep is abandoned.
1664    
1665        (b) A binary zero was treated as data while matching, but terminated the
1666            output line if it was written out. This has been fixed: binary zeroes
1667            are now no different to any other data bytes.
1668    
1669        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1670            used to set a locale for matching. The --locale=xxxx long option has
1671            been added (no short equivalent) to specify a locale explicitly on the
1672            pcregrep command, overriding the environment variables.
1673    
1674        (d) When -B was used with -n, some line numbers in the output were one less
1675            than they should have been.
1676    
1677        (e) Added the -o (--only-matching) option.
1678    
1679        (f) If -A or -C was used with -c (count only), some lines of context were
1680            accidentally printed for the final match.
1681    
1682        (g) Added the -H (--with-filename) option.
1683    
1684        (h) The combination of options -rh failed to suppress file names for files
1685            that were found from directory arguments.
1686    
1687        (i) Added the -D (--devices) and -d (--directories) options.
1688    
1689        (j) Added the -F (--fixed-strings) option.
1690    
1691        (k) Allow "-" to be used as a file name for -f as well as for a data file.
1692    
1693        (l) Added the --colo(u)r option.
1694    
1695        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1696            is not present by default.
1697    
1698     3. A nasty bug was discovered in the handling of recursive patterns, that is,
1699        items such as (?R) or (?1), when the recursion could match a number of
1700        alternatives. If it matched one of the alternatives, but subsequently,
1701        outside the recursion, there was a failure, the code tried to back up into
1702        the recursion. However, because of the way PCRE is implemented, this is not
1703        possible, and the result was an incorrect result from the match.
1704    
1705        In order to prevent this happening, the specification of recursion has
1706        been changed so that all such subpatterns are automatically treated as
1707        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1708    
1709     4. I had overlooked the fact that, in some locales, there are characters for
1710        which isalpha() is true but neither isupper() nor islower() are true. In
1711        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1712        and ordfeminine) are like this. This affected the treatment of \w and \W
1713        when they appeared in character classes, but not when they appeared outside
1714        a character class. The bit map for "word" characters is now created
1715        separately from the results of isalnum() instead of just taking it from the
1716        upper, lower, and digit maps. (Plus the underscore character, of course.)
1717    
1718     5. The above bug also affected the handling of POSIX character classes such as
1719        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1720        permanent tables. Instead, the bit maps for such a class were previously
1721        created as the appropriate unions of the upper, lower, and digit bitmaps.
1722        Now they are created by subtraction from the [[:word:]] class, which has
1723        its own bitmap.
1724    
1725     6. The [[:blank:]] character class matches horizontal, but not vertical space.
1726        It is created by subtracting the vertical space characters (\x09, \x0a,
1727        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1728        subtraction was done in the overall bitmap for a character class, meaning
1729        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1730        be recognized. This bug has been fixed.
1731    
1732     7. Patches from the folks at Google:
1733    
1734          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1735          real life, but is still worth protecting against".
1736    
1737          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1738          regular expressions".
1739    
1740          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1741          have it.
1742    
1743          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1744          "configure" and the latter not, in order to fix a problem somebody had
1745          with compiling the Arg class on HP-UX.
1746    
1747          (e) Improve the error-handling of the C++ wrapper a little bit.
1748    
1749          (f) New tests for checking recursion limiting.
1750    
1751     8. The pcre_memmove() function, which is used only if the environment does not
1752        have a standard memmove() function (and is therefore rarely compiled),
1753        contained two bugs: (a) use of int instead of size_t, and (b) it was not
1754        returning a result (though PCRE never actually uses the result).
1755    
1756     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1757        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1758        returned instead of calling malloc() with an overflowing number that would
1759        most likely cause subsequent chaos.
1760    
1761    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1762    
1763    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1764        with this option is matched, the nmatch and pmatch options of regexec() are
1765        ignored.
1766    
1767    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1768        provided in case anyone wants to the the POSIX interface with UTF-8
1769        strings.
1770    
1771    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1772        C++ linking (needed for some HP-UX environments).
1773    
1774    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1775        (unused parameter) and in the pcre_printint() function (omitted "default"
1776        switch label when the default is to do nothing).
1777    
1778    15. Added some code to make it possible, when PCRE is compiled as a C++
1779        library, to replace subject pointers for pcre_exec() with a smart pointer
1780        class, thus making it possible to process discontinuous strings.
1781    
1782    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1783        much the same function. They were added by different people who were trying
1784        to make PCRE easy to compile on non-Unix systems. It has been suggested
1785        that PCRE_EXPORT be abolished now that there is more automatic apparatus
1786        for compiling on Windows systems. I have therefore replaced it with
1787        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1788        defaults to "extern" for C or "extern C" for C++, which works fine on
1789        Unix-like systems. It is now possible to override the value of PCRE_DATA_
1790        SCOPE with something explicit in config.h. In addition:
1791    
1792        (a) pcreposix.h still had just "extern" instead of either of these macros;
1793            I have replaced it with PCRE_DATA_SCOPE.
1794    
1795        (b) Functions such as _pcre_xclass(), which are internal to the library,
1796            but external in the C sense, all had PCRE_EXPORT in their definitions.
1797            This is apparently wrong for the Windows case, so I have removed it.
1798            (It makes no difference on Unix-like systems.)
1799    
1800    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1801        of recursive calls to match(). This is different to MATCH_LIMIT because
1802        that limits the total number of calls to match(), not all of which increase
1803        the depth of recursion. Limiting the recursion depth limits the amount of
1804        stack (or heap if NO_RECURSE is set) that is used. The default can be set
1805        when PCRE is compiled, and changed at run time. A patch from Google adds
1806        this functionality to the C++ interface.
1807    
1808    18. Changes to the handling of Unicode character properties:
1809    
1810        (a) Updated the table to Unicode 4.1.0.
1811    
1812        (b) Recognize characters that are not in the table as "Cn" (undefined).
1813    
1814        (c) I revised the way the table is implemented to a much improved format
1815            which includes recognition of ranges. It now supports the ranges that
1816            are defined in UnicodeData.txt, and it also amalgamates other
1817            characters into ranges. This has reduced the number of entries in the
1818            table from around 16,000 to around 3,000, thus reducing its size
1819            considerably. I realized I did not need to use a tree structure after
1820            all - a binary chop search is just as efficient. Having reduced the
1821            number of entries, I extended their size from 6 bytes to 8 bytes to
1822            allow for more data.
1823    
1824        (d) Added support for Unicode script names via properties such as \p{Han}.
1825    
1826    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1827        matching that character.
1828    
1829    20. When matching a repeated Unicode property with a minimum greater than zero,
1830        (for example \pL{2,}), PCRE could look past the end of the subject if it
1831        reached it while seeking the minimum number of characters. This could
1832        happen only if some of the characters were more than one byte long, because
1833        there is a check for at least the minimum number of bytes.
1834    
1835    21. Refactored the implementation of \p and \P so as to be more general, to
1836        allow for more different types of property in future. This has changed the
1837        compiled form incompatibly. Anybody with saved compiled patterns that use
1838        \p or \P will have to recompile them.
1839    
1840    22. Added "Any" and "L&" to the supported property types.
1841    
1842    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1843        but give a compile time error if the value is greater than 0xff.
1844    
1845    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1846        accidentally not being installed or uninstalled.
1847    
1848    25. The pcre.h file was built from pcre.h.in, but the only changes that were
1849        made were to insert the current release number. This seemed silly, because
1850        it made things harder for people building PCRE on systems that don't run
1851        "configure". I have turned pcre.h into a distributed file, no longer built
1852        by "configure", with the version identification directly included. There is
1853        no longer a pcre.h.in file.
1854    
1855        However, this change necessitated a change to the pcre-config script as
1856        well. It is built from pcre-config.in, and one of the substitutions was the
1857        release number. I have updated configure.ac so that ./configure now finds
1858        the release number by grepping pcre.h.
1859    
1860    26. Added the ability to run the tests under valgrind.
1861    
1862    
1863    Version 6.4 05-Sep-05
1864    ---------------------
1865    
1866     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1867        "--" to be printed when multiple files were scanned, even when none of the
1868        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1869        consider it to be a bug, and have restored the previous behaviour.
1870    
1871     2. A couple of code tidies to get rid of compiler warnings.
1872    
1873     3. The pcretest program used to cheat by referring to symbols in the library
1874        whose names begin with _pcre_. These are internal symbols that are not
1875        really supposed to be visible externally, and in some environments it is
1876        possible to suppress them. The cheating is now confined to including
1877        certain files from the library's source, which is a bit cleaner.
1878    
1879     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1880        file's purpose clearer.
1881    
1882     5. Reorganized pcre_ucp_findchar().
1883    
1884    
1885    Version 6.3 15-Aug-05
1886    ---------------------
1887    
1888     1. The file libpcre.pc.in did not have general read permission in the tarball.
1889    
1890     2. There were some problems when building without C++ support:
1891    
1892        (a) If C++ support was not built, "make install" and "make test" still
1893            tried to test it.
1894    
1895        (b) There were problems when the value of CXX was explicitly set. Some
1896            changes have been made to try to fix these, and ...
1897    
1898        (c) --disable-cpp can now be used to explicitly disable C++ support.
1899    
1900        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1901            backslash in a target when C++ was disabled. This confuses some
1902            versions of "make", apparently. Using an intermediate variable solves
1903            this. (Same for CPP_LOBJ.)
1904    
1905     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1906        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1907        necessary on certain architectures.
1908    
1909     4. Added a setting of -export-symbols-regex to the link command to remove
1910        those symbols that are exported in the C sense, but actually are local
1911        within the library, and not documented. Their names all begin with
1912        "_pcre_". This is not a perfect job, because (a) we have to except some
1913        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1914        available (and never for static libraries). I have made a note to try to
1915        find a way round (a) in the future.
1916    
1917    
1918    Version 6.2 01-Aug-05
1919    ---------------------
1920    
1921     1. There was no test for integer overflow of quantifier values. A construction
1922        such as {1111111111111111} would give undefined results. What is worse, if
1923        a minimum quantifier for a parenthesized subpattern overflowed and became
1924        negative, the calculation of the memory size went wrong. This could have
1925        led to memory overwriting.
1926    
1927     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1928    
1929     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1930        operating environments where this matters.
1931    
1932     4. Applied Giuseppe Maxia's patch to add additional features for controlling
1933        PCRE options from within the C++ wrapper.
1934    
1935     5. Named capturing subpatterns were not being correctly counted when a pattern
1936        was compiled. This caused two problems: (a) If there were more than 100
1937        such subpatterns, the calculation of the memory needed for the whole
1938        compiled pattern went wrong, leading to an overflow error. (b) Numerical
1939        back references of the form \12, where the number was greater than 9, were
1940        not recognized as back references, even though there were sufficient
1941        previous subpatterns.
1942    
1943     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1944        versions of gcc, e.g. 2.95.4.
1945    
1946    
1947    Version 6.1 21-Jun-05
1948    ---------------------
1949    
1950     1. There was one reference to the variable "posix" in pcretest.c that was not
1951        surrounded by "#if !defined NOPOSIX".
1952    
1953     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1954        the cross-check on the old pcre_info() function, for the benefit of the
1955        cut-down version of PCRE that is currently imported into Exim.
1956    
1957     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1958        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1959        patterns that start (?i) but I don't think that matters. The use of (?i) is
1960        just an example; this all applies to the other options as well.
1961    
1962     4. Since libtool seems to echo the compile commands it is issuing, the output
1963        from "make" can be reduced a bit by putting "@" in front of each libtool
1964        compile command.
1965    
1966     5. Patch from the folks at Google for configure.in to be a bit more thorough
1967        in checking for a suitable C++ installation before trying to compile the
1968        C++ stuff. This should fix a reported problem when a compiler was present,
1969        but no suitable headers.
1970    
1971     6. The man pages all had just "PCRE" as their title. I have changed them to
1972        be the relevant file name. I have also arranged that these names are
1973        retained in the file doc/pcre.txt, which is a concatenation in text format
1974        of all the man pages except the little individual ones for each function.
1975    
1976     7. The NON-UNIX-USE file had not been updated for the different set of source
1977        files that come with release 6. I also added a few comments about the C++
1978        wrapper.
1979    
1980    
1981    Version 6.0 07-Jun-05
1982    ---------------------
1983    
1984     1. Some minor internal re-organization to help with my DFA experiments.
1985    
1986     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1987        didn't matter for the library itself when fully configured, but did matter
1988        when compiling without UCP support, or within Exim, where the ucp files are
1989        not imported.
1990    
1991     3. Refactoring of the library code to split up the various functions into
1992        different source modules. The addition of the new DFA matching code (see
1993        below) to a single monolithic source would have made it really too
1994        unwieldy, quite apart from causing all the code to be include in a
1995        statically linked application, when only some functions are used. This is
1996        relevant even without the DFA addition now that patterns can be compiled in
1997        one application and matched in another.
1998    
1999        The downside of splitting up is that there have to be some external
2000        functions and data tables that are used internally in different modules of
2001        the library but which are not part of the API. These have all had their
2002        names changed to start with "_pcre_" so that they are unlikely to clash
2003        with other external names.
2004    
2005     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
2006        a different (DFA) algorithm. Although it is slower than the original
2007        function, it does have some advantages for certain types of matching
2008        problem.
2009    
2010     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
2011        including restarting after a partial match.
2012    
2013     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
2014        defined when compiling for Windows was sent to me. I have put it into the
2015        code, though I have no means of testing or verifying it.
2016    
2017     7. Added the pcre_refcount() auxiliary function.
2018    
2019     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
2020        match before or at the first newline in the subject string. In pcretest,
2021        the /f option on a pattern can be used to set this.
2022    
2023     9. A repeated \w when used in UTF-8 mode with characters greater than 256
2024        would behave wrongly. This has been present in PCRE since release 4.0.
2025    
2026    10. A number of changes to the pcregrep command:
2027    
2028        (a) Refactored how -x works; insert ^(...)$ instead of setting
2029            PCRE_ANCHORED and checking the length, in preparation for adding
2030            something similar for -w.
2031    
2032        (b) Added the -w (match as a word) option.
2033    
2034        (c) Refactored the way lines are read and buffered so as to have more
2035            than one at a time available.
2036    
2037        (d) Implemented a pcregrep test script.
2038    
2039        (e) Added the -M (multiline match) option. This allows patterns to match
2040            over several lines of the subject. The buffering ensures that at least
2041            8K, or the rest of the document (whichever is the shorter) is available
2042            for matching (and similarly the previous 8K for lookbehind assertions).
2043    
2044        (f) Changed the --help output so that it now says
2045    
2046              -w, --word-regex(p)
2047    
2048            instead of two lines, one with "regex" and the other with "regexp"
2049            because that confused at least one person since the short forms are the
2050            same. (This required a bit of code, as the output is generated
2051            automatically from a table. It wasn't just a text change.)
2052    
2053        (g) -- can be used to terminate pcregrep options if the next thing isn't an
2054            option but starts with a hyphen. Could be a pattern or a path name
2055            starting with a hyphen, for instance.
2056    
2057        (h) "-" can be given as a file name to represent stdin.
2058    
2059        (i) When file names are being printed, "(standard input)" is used for
2060            the standard input, for compatibility with GNU grep. Previously
2061            "<stdin>" was used.
2062    
2063        (j) The option --label=xxx can be used to supply a name to be used for
2064            stdin when file names are being printed. There is no short form.
2065    
2066        (k) Re-factored the options decoding logic because we are going to add
2067            two more options that take data. Such options can now be given in four
2068            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
2069    
2070        (l) Added the -A, -B, and -C options for requesting that lines of context
2071            around matches be printed.
2072    
2073        (m) Added the -L option to print the names of files that do not contain
2074            any matching lines, that is, the complement of -l.
2075    
2076        (n) The return code is 2 if any file cannot be opened, but pcregrep does
2077            continue to scan other files.
2078    
2079        (o) The -s option was incorrectly implemented. For compatibility with other
2080            greps, it now suppresses the error message for a non-existent or non-
2081            accessible file (but not the return code). There is a new option called
2082            -q that suppresses the output of matching lines, which was what -s was
2083            previously doing.
2084    
2085        (p) Added --include and --exclude options to specify files for inclusion
2086            and exclusion when recursing.
2087    
2088    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
2089        Hopefully, it now does.
2090    
2091    12. Missing cast in pcre_study().
2092    
2093    13. Added an "uninstall" target to the makefile.
2094    
2095    14. Replaced "extern" in the function prototypes in Makefile.in with
2096        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
2097        world, but is set differently for Windows.
2098    
2099    15. Added a second compiling function called pcre_compile2(). The only
2100        difference is that it has an extra argument, which is a pointer to an
2101        integer error code. When there is a compile-time failure, this is set
2102        non-zero, in addition to the error test pointer being set to point to an
2103        error message. The new argument may be NULL if no error number is required
2104        (but then you may as well call pcre_compile(), which is now just a
2105        wrapper). This facility is provided because some applications need a
2106        numeric error indication, but it has also enabled me to tidy up the way
2107        compile-time errors are handled in the POSIX wrapper.
2108    
2109    16. Added VPATH=.libs to the makefile; this should help when building with one
2110        prefix path and installing with another. (Or so I'm told by someone who
2111        knows more about this stuff than I do.)
2112    
2113    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
2114        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
2115        match everything, including newlines. This is not POSIX-compatible, but
2116        somebody wanted the feature. From pcretest it can be activated by using
2117        both the P and the s flags.
2118    
2119    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
2120    
2121    19. libpcre.pc was being incorrectly installed as executable.
2122    
2123    20. A couple of places in pcretest check for end-of-line by looking for '\n';
2124        it now also looks for '\r' so that it will work unmodified on Windows.
2125    
2126    21. Added Google's contributed C++ wrapper to the distribution.
2127    
2128    22. Added some untidy missing memory free() calls in pcretest, to keep
2129        Electric Fence happy when testing.
2130    
2131    
2132    
2133    Version 5.0 13-Sep-04
2134    ---------------------
2135    
2136     1. Internal change: literal characters are no longer packed up into items
2137        containing multiple characters in a single byte-string. Each character
2138        is now matched using a separate opcode. However, there may be more than one
2139        byte in the character in UTF-8 mode.
2140    
2141     2. The pcre_callout_block structure has two new fields: pattern_position and
2142        next_item_length. These contain the offset in the pattern to the next match
2143        item, and its length, respectively.
2144    
2145     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
2146        insertion of callouts before each pattern item. Added the /C option to
2147        pcretest to make use of this.
2148    
2149     4. On the advice of a Windows user, the lines
2150    
2151          #if defined(_WIN32) || defined(WIN32)
2152          _setmode( _fileno( stdout ), 0x8000 );
2153          #endif  /* defined(_WIN32) || defined(WIN32) */
2154    
2155        have been added to the source of pcretest. This apparently does useful
2156        magic in relation to line terminators.
2157    
2158     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
2159        for the benefit of those environments where the "b" makes a difference.
2160    
2161     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
2162        to know about it. I have put a hack into configure.in that adds in code
2163        to set GCC=yes if CC=icc. This seems to end up at a point in the
2164        generated configure script that is early enough to affect the setting of
2165        compiler options, which is what is needed, but I have no means of testing
2166        whether it really works. (The user who reported this had patched the
2167        generated configure script, which of course I cannot do.)
2168    
2169        LATER: After change 22 below (new libtool files), the configure script
2170        seems to know about icc (and also ecc). Therefore, I have commented out
2171        this hack in configure.in.
2172    
2173     7. Added support for pkg-config (2 patches were sent in).
2174    
2175     8. Negated POSIX character classes that used a combination of internal tables
2176        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2177        [[:^ascii]]. Typically, they would match almost any characters. The other
2178        POSIX classes were not broken in this way.
2179    
2180     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2181        to find the match, as PCRE was deluded into thinking that the match had to
2182        start at the start point or following a newline. The same bug applied to
2183        patterns with negative forward assertions or any backward assertions
2184        preceding ".*" at the start, unless the pattern required a fixed first
2185        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2186    
2187    10. In UTF-8 mode, when moving forwards in the subject after a failed match
2188        starting at the last subject character, bytes beyond the end of the subject
2189        string were read.
2190    
2191    11. Renamed the variable "class" as "classbits" to make life easier for C++
2192        users. (Previously there was a macro definition, but it apparently wasn't
2193        enough.)
2194    
2195    12. Added the new field "tables" to the extra data so that tables can be passed
2196        in at exec time, or the internal tables can be re-selected. This allows
2197        a compiled regex to be saved and re-used at a later time by a different
2198        program that might have everything at different addresses.
2199    
2200    13. Modified the pcre-config script so that, when run on Solaris, it shows a
2201        -R library as well as a -L library.
2202    
2203    14. The debugging options of pcretest (-d on the command line or D on a
2204        pattern) showed incorrect output for anything following an extended class
2205        that contained multibyte characters and which was followed by a quantifier.
2206    
2207    15. Added optional support for general category Unicode character properties
2208        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2209        support. It adds about 90K to the size of the library. The meanings of the
2210        inbuilt class escapes such as \d and \s have NOT been changed.
2211    
2212    16. Updated pcredemo.c to include calls to free() to release the memory for the
2213        compiled pattern.
2214    
2215    17. The generated file chartables.c was being created in the source directory
2216        instead of in the building directory. This caused the build to fail if the
2217        source directory was different from the building directory, and was
2218        read-only.
2219    
2220    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2221        file. No doubt somebody will tell me if they don't make sense... Also added
2222        Dan Mooney's comments about building on OpenVMS.
2223    
2224    19. Added support for partial matching via the PCRE_PARTIAL option for
2225        pcre_exec() and the \P data escape in pcretest.
2226    
2227    20. Extended pcretest with 3 new pattern features:
2228    
2229        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
2230              write the compiled pattern to the file whose name is "rest-of-line".
2231              This is a straight binary dump of the data, with the saved pointer to
2232              the character tables forced to be NULL. The study data, if any, is
2233              written too. After writing, pcretest reads a new pattern.
2234    
2235        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2236              compiled pattern from the given file. There must not be any
2237              occurrences of "<" in the file name (pretty unlikely); if there are,
2238              pcretest will instead treat the initial "<" as a pattern delimiter.
2239              After reading in the pattern, pcretest goes on to read data lines as
2240              usual.
2241    
2242        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2243              and 16-bit fields in a compiled pattern, to simulate a pattern that
2244              was compiled on a host of opposite endianness.
2245    
2246    21. The pcre-exec() function can now cope with patterns that were compiled on
2247        hosts of opposite endianness, with this restriction:
2248    
2249          As for any compiled expression that is saved and used later, the tables
2250          pointer field cannot be preserved; the extra_data field in the arguments
2251          to pcre_exec() should be used to pass in a tables address if a value
2252          other than the default internal tables were used at compile time.
2253    
2254    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2255        now diagnosed as an error. Previously, most of the time, a negative number
2256        would have been treated as zero, but if in addition "ovector" was passed as
2257        NULL, a crash could occur.
2258    
2259    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2260        new versions from the libtool 1.5 distribution (the last one is a copy of
2261        a file called libtool.m4). This seems to have fixed the need to patch
2262        "configure" to support Darwin 1.3 (which I used to do). However, I still
2263        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2264        workstation).
2265    
2266    24. Changed the PCRE licence to be the more standard "BSD" licence.
2267    
2268    
2269    Version 4.5 01-Dec-03
2270    ---------------------
2271    
2272     1. There has been some re-arrangement of the code for the match() function so
2273        that it can be compiled in a version that does not call itself recursively.
2274        Instead, it keeps those local variables that need separate instances for
2275        each "recursion" in a frame on the heap, and gets/frees frames whenever it
2276        needs to "recurse". Keeping track of where control must go is done by means
2277        of setjmp/longjmp. The whole thing is implemented by a set of macros that
2278        hide most of the details from the main code, and operates only if
2279        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2280        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2281        operating.
2282    
2283        To make it easier for callers to provide specially tailored get/free
2284        functions for this usage, two new functions, pcre_stack_malloc, and
2285        pcre_stack_free, are used. They are always called in strict stacking order,
2286        and the size of block requested is always the same.
2287    
2288        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2289        PCRE has been compiled to use the stack or the heap for recursion. The
2290        -C option of pcretest uses this to show which version is compiled.
2291    
2292        A new data escape \S, is added to pcretest; it causes the amounts of store
2293        obtained and freed by both kinds of malloc/free at match time to be added
2294        to the output.
2295    
2296     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2297        what's available on my current Linux desktop machine.
2298    
2299     3. When matching a UTF-8 string, the test for a valid string at the start has
2300        been extended. If start_offset is not zero, PCRE now checks that it points
2301        to a byte that is the start of a UTF-8 character. If not, it returns
2302        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2303        this is necessary because there may be backward assertions in the pattern.
2304        When matching the same subject several times, it may save resources to use
2305        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2306    
2307     4. The code for checking the validity of UTF-8 strings has been tightened so
2308        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2309        containing "overlong sequences".
2310    
2311     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2312        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2313        should have been "&", but it just so happened that all the cases this let
2314        through by mistake were picked up later in the function.
2315    
2316     6. I had used a variable called "isblank" - this is a C99 function, causing
2317        some compilers to warn. To avoid this, I renamed it (as "blankclass").
2318    
2319     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2320        prompting; (b) run "./pcretest /dev/null" at the start of the test script
2321        so the version is shown; (c) stop "make test" echoing "./RunTest".
2322    
2323     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2324    
2325     9. The prototype for memmove() for systems that don't have it was using
2326        size_t, but the inclusion of the header that defines size_t was later. I've
2327        moved the #includes for the C headers earlier to avoid this.
2328    
2329    10. Added some adjustments to the code to make it easier to compiler on certain
2330        special systems:
2331    
2332          (a) Some "const" qualifiers were missing.
2333          (b) Added the macro EXPORT before all exported functions; by default this
2334              is defined to be empty.
2335          (c) Changed the dftables auxiliary program (that builds chartables.c) so
2336              that it reads its output file name as an argument instead of writing
2337              to the standard output and assuming this can be redirected.
2338    
2339    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2340        class containing characters with values greater than 255, PCRE compilation
2341        went into a loop.
2342    
2343    12. A recursive reference to a subpattern that was within another subpattern
2344        that had a minimum quantifier of zero caused PCRE to crash. For example,
2345        (x(y(?2))z)? provoked this bug with a subject that got as far as the
2346        recursion. If the recursively-called subpattern itself had a zero repeat,
2347        that was OK.
2348    
2349    13. In pcretest, the buffer for reading a data line was set at 30K, but the
2350        buffer into which it was copied (for escape processing) was still set at
2351        1024, so long lines caused crashes.
2352    
2353    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2354        "internal error: code overflow...". This applied to any character class
2355        that was followed by a possessive quantifier.
2356    
2357    15. Modified the Makefile to add libpcre.la as a prerequisite for
2358        libpcreposix.la because I was told this is needed for a parallel build to
2359        work.
2360    
2361    16. If a pattern that contained .* following optional items at the start was
2362        studied, the wrong optimizing data was generated, leading to matching
2363        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2364        matching string must start with a or b or c. The correct conclusion for
2365        this pattern is that a match can start with any character.
2366    
2367    
2368    Version 4.4 13-Aug-03
2369    ---------------------
2370    
2371     1. In UTF-8 mode, a character class containing characters with values between
2372        127 and 255 was not handled correctly if the compiled pattern was studied.
2373        In fixing this, I have also improved the studying algorithm for such
2374        classes (slightly).
2375    
2376     2. Three internal functions had redundant arguments passed to them. Removal
2377        might give a very teeny performance improvement.
2378    
2379     3. Documentation bug: the value of the capture_top field in a callout is *one
2380        more than* the number of the hightest numbered captured substring.
2381    
2382     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2383        in incorrectly linking with a previously installed version. They now link
2384        explicitly with libpcre.la.
2385    
2386     5. configure.in no longer needs to recognize Cygwin specially.
2387    
2388     6. A problem in pcre.in for Windows platforms is fixed.
2389    
2390     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2391        pcretest, it used to include the size of the study block as part of its
2392        output. Unfortunately, the structure contains a field that has a different
2393        size on different hardware architectures. This meant that the tests that
2394        showed this size failed. As the block is currently always of a fixed size,
2395        this information isn't actually particularly useful in pcretest output, so
2396        I have just removed it.
2397    
2398     8. Three pre-processor statements accidentally did not start in column 1.
2399        Sadly, there are *still* compilers around that complain, even though
2400        standard C has not required this for well over a decade. Sigh.
2401    
2402     9. In pcretest, the code for checking callouts passed small integers in the
2403        callout_data field, which is a void * field. However, some picky compilers
2404        complained about the casts involved for this on 64-bit systems. Now
2405        pcretest passes the address of the small integer instead, which should get
2406        rid of the warnings.
2407    
2408    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2409        both compile and run time, and gives an error if an invalid UTF-8 sequence
2410        is found. There is a option for disabling this check in cases where the
2411        string is known to be correct and/or the maximum performance is wanted.
2412    
2413    11. In response to a bug report, I changed one line in Makefile.in from
2414    
2415            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2416        to
2417            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2418    
2419        to look similar to other lines, but I have no way of telling whether this
2420        is the right thing to do, as I do not use Windows. No doubt I'll get told
2421        if it's wrong...
2422    
2423    
2424    Version 4.3 21-May-03
2425    ---------------------
2426    
2427    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2428       Makefile.
2429    
2430    2. Some refactoring to improve the quality of the code:
2431    
2432       (i)   The utf8_table... variables are now declared "const".
2433    
2434       (ii)  The code for \cx, which used the "case flipping" table to upper case
2435             lower case letters, now just substracts 32. This is ASCII-specific,
2436             but the whole concept of \cx is ASCII-specific, so it seems
2437             reasonable.
2438    
2439       (iii) PCRE was using its character types table to recognize decimal and
2440             hexadecimal digits in the pattern. This is silly, because it handles
2441             only 0-9, a-f, and A-F, but the character types table is locale-
2442             specific, which means strange things might happen. A private
2443             table is now used for this - though it costs 256 bytes, a table is
2444             much faster than multiple explicit tests. Of course, the standard
2445             character types table is still used for matching digits in subject
2446             strings against \d.
2447    
2448       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2449             ending in _t are). So I've renamed it as ESC_tee.
2450    
2451    3. The first argument for regexec() in the POSIX wrapper should have been
2452       defined as "const".
2453    
2454    4. Changed pcretest to use malloc() for its buffers so that they can be
2455       Electric Fenced for debugging.
2456    
2457    5. There were several places in the code where, in UTF-8 mode, PCRE would try
2458       to read one or more bytes before the start of the subject string. Often this
2459       had no effect on PCRE's behaviour, but in some circumstances it could
2460       provoke a segmentation fault.
2461    
2462    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2463       to try to read one or more bytes before the start of the subject string.
2464    
2465    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
2466       UTF-8 support could misbehave in various ways if the subject string
2467       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
2468       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
2469       back over UTF-8 characters.)
2470    
2471    
2472    Version 4.2 14-Apr-03
2473    ---------------------
2474    
2475    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
2476    
2477    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
2478         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
2479         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
2480         [WIN_PREFIX]: new variable, "cyg" for Cygwin
2481         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
2482           and BUILD_EXEEXT
2483         Note: automatic setting of the BUILD variables is not yet working
2484         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
2485           compile-time but not at link-time
2486         [LINK]: use for linking executables only
2487         make different versions for Windows and non-Windows
2488         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
2489           libraries
2490         [LINK_FOR_BUILD]: new variable
2491         [OBJEXT]: use throughout
2492         [EXEEXT]: use throughout
2493         <winshared>: new target
2494         <wininstall>: new target
2495         <dftables.o>: use native compiler
2496         <dftables>: use native linker
2497         <install>: handle Windows platform correctly
2498         <clean>: ditto
2499         <check>: ditto
2500         copy DLL to top builddir before testing
2501    
2502       As part of these changes, -no-undefined was removed again. This was reported
2503       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
2504       in any case.
2505    
2506    3. Some tidies to get rid of compiler warnings:
2507    
2508       . In the match_data structure, match_limit was an unsigned long int, whereas
2509         match_call_count was an int. I've made them both unsigned long ints.
2510    
2511       . In pcretest the fact that a const uschar * doesn't automatically cast to
2512         a void * provoked a warning.
2513    
2514       . Turning on some more compiler warnings threw up some "shadow" variables
2515         and a few more missing casts.
2516    
2517    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
2518       option, a class that contained a single character with a value between 128
2519       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
2520    
2521    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
2522       option, a class that contained several characters, but with at least one
2523       whose value was between 128 and 255 caused PCRE to crash.
2524    
2525    
2526    Version 4.1 12-Mar-03
2527    ---------------------
2528    
2529    1. Compiling with gcc -pedantic found a couple of places where casts were
2530    needed, and a string in dftables.c that was longer than standard compilers are
2531    required to support.
2532    
2533    2. Compiling with Sun's compiler found a few more places where the code could
2534    be tidied up in order to avoid warnings.
2535    
2536    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
2537    first of these names is deprecated in the latest Autoconf in favour of the name
2538    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
2539    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2540    analogy I have changed it to CFLAGS_FOR_BUILD.
2541    
2542    4. Added -no-undefined to the linking command in the Makefile, because this is
2543    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2544    linking step for the pcreposix library.
2545    
2546    5. PCRE was failing to diagnose the case of two named groups with the same
2547    name.
2548    
2549    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2550    literal character that is needed in the subject for a match, and scans along to
2551    ensure that it is present before embarking on the full matching process. This
2552    saves time in cases of nested unlimited repeats that are never going to match.
2553    Problem: the scan can take a lot of time if the subject is very long (e.g.
2554    megabytes), thus penalizing straightforward matches. It is now done only if the
2555    amount of subject to be scanned is less than 1000 bytes.
2556    
2557    7. A lesser problem with the same optimization is that it was recording the
2558    first character of an anchored pattern as "needed", thus provoking a search
2559    right along the subject, even when the first match of the pattern was going to
2560    fail. The "needed" character is now not set for anchored patterns, unless it
2561    follows something in the pattern that is of non-fixed length. Thus, it still
2562    fulfils its original purpose of finding quick non-matches in cases of nested
2563    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
2564    
2565    
2566    Version 4.0 17-Feb-03
2567    ---------------------
2568    
2569    1. If a comment in an extended regex that started immediately after a meta-item
2570    extended to the end of string, PCRE compiled incorrect data. This could lead to
2571    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
2572    
2573    2. Moved to autoconf 2.53 and libtool 1.4.2.
2574    
2575    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
2576    the special perltest8 script is no longer needed - all the tests can be run
2577    from a single perltest script.
2578    
2579    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
2580    by \s. It has now been removed in PCRE. This means it isn't recognized as
2581    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
2582    class [:space:] *does* include VT, thereby creating a mess.
2583    
2584    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
2585    space and tab.
2586    
2587    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
2588    its new features into the main test script, reducing the number of scripts.
2589    
2590    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
2591    were backward compatible, and made the (?i) apply to the whole pattern, as if
2592    /i were given. Now it behaves more logically, and applies the option setting
2593    only to what follows. PCRE has been changed to follow suit. However, if it
2594    finds options settings right at the start of the pattern, it extracts them into
2595    the global options, as before. Thus, they show up in the info data.
2596    
2597    8. Added support for the \Q...\E escape sequence. Characters in between are
2598    treated as literals. This is slightly different from Perl in that $ and @ are
2599    also handled as literals inside the quotes. In Perl, they will cause variable
2600    interpolation. Note the following examples:
2601    
2602        Pattern            PCRE matches      Perl matches
2603    
2604        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
2605        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
2606        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
2607    
2608    For compatibility with Perl, \Q...\E sequences are recognized inside character
2609    classes as well as outside them.
2610    
2611    9. Re-organized 3 code statements in pcretest to avoid "overflow in
2612    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2613    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2614    signed/unsigned warnings.
2615    
2616    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2617    option for pcretest, so I've replaced it by a simple function that does just
2618    that job.
2619    
2620    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2621    "pcregrep -".
2622    
2623    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2624    Java package. This provides some syntactic sugar for simple cases of what my
2625    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2626    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2627    item, you can use this simplified notation. Note that only makes sense with
2628    greedy quantifiers. Consequently, the use of the possessive quantifier forces
2629    greediness, whatever the setting of the PCRE_UNGREEDY option.
2630    
2631    13. A change of greediness default within a pattern was not taking effect at
2632    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2633    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2634    was abstracted outside.
2635    
2636    14. PCRE now supports the \G assertion. It is true when the current matching
2637    position is at the start point of the match. This differs from \A when the
2638    starting offset is non-zero. Used with the /g option of pcretest (or similar
2639    code), it works in the same way as it does for Perl's /g option. If all
2640    alternatives of a regex begin with \G, the expression is anchored to the start
2641    match position, and the "anchored" flag is set in the compiled expression.
2642    
2643    15. Some bugs concerning the handling of certain option changes within patterns
2644    have been fixed. These applied to options other than (?ims). For example,
2645    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2646    been the other way round. Some of this was related to change 7 above.
2647    
2648    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2649    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2650    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2651    POSIX classes only within a class (e.g. /[[:alpha:]]/).
2652    
2653    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2654    mode. Unlike ".", it always matches newline, whatever the setting of
2655    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2656    assertions. Perl allows it, but it doesn't (in general) work because it can't
2657    calculate the length of the lookbehind. At least, that's the case for Perl
2658    5.8.0 - I've been told they are going to document that it doesn't work in
2659    future.
2660    
2661    18. Added an error diagnosis for escapes that PCRE does not support: these are
2662    \L, \l, \N, \P, \p, \U, \u, and \X.
2663    
2664    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2665    reading past the end of the pattern in cases such as /[abcd/.
2666    
2667    20. PCRE was getting more memory than necessary for patterns with classes that
2668    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2669    
2670    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2671    compiling PCRE for use with Virtual Pascal.
2672    
2673    22. Small fix to the Makefile to make it work properly if the build is done
2674    outside the source tree.
2675    
2676    23. Added a new extension: a condition to go with recursion. If a conditional
2677    subpattern starts with (?(R) the "true" branch is used if recursion has
2678    happened, whereas the "false" branch is used only at the top level.
2679    
2680    24. When there was a very long string of literal characters (over 255 bytes
2681    without UTF support, over 250 bytes with UTF support), the computation of how
2682    much memory was required could be incorrect, leading to segfaults or other
2683    strange effects.
2684    
2685    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2686    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2687    there was a subsequent back reference to those brackets. This meant that, for
2688    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2689    possible to check for precisely this case. All we can do is abandon the
2690    optimization if .* occurs inside capturing brackets when there are any back
2691    references whatsoever. (See below for a better fix that came later.)
2692    
2693    26. The handling of the optimization for finding the first character of a
2694    non-anchored pattern, and for finding a character that is required later in the
2695    match were failing in some cases. This didn't break the matching; it just
2696    failed to optimize when it could. The way this is done has been re-implemented.
2697    
2698    27. Fixed typo in error message for invalid (?R item (it said "(?p").
2699    
2700    28. Added a new feature that provides some of the functionality that Perl
2701    provides with (?{...}). The facility is termed a "callout". The way it is done
2702    in PCRE is for the caller to provide an optional function, by setting
2703    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2704    global variable. By default it is unset, which disables all calling out. To get
2705    the function called, the regex must include (?C) at appropriate points. This
2706    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2707    This provides a means of identifying different callout points. When PCRE
2708    reaches such a point in the regex, if pcre_callout has been set, the external
2709    function is called. It is provided with data in a structure called
2710    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2711    matching continues; if it returns a non-zero value, the match at the current
2712    point fails. However, backtracking will occur if possible. [This was changed
2713    later and other features added - see item 49 below.]
2714    
2715    29. pcretest is upgraded to test the callout functionality. It provides a
2716    callout function that displays information. By default, it shows the start of
2717    the match and the current position in the text. There are some new data escapes
2718    to vary what happens:
2719    
2720        \C+         in addition, show current contents of captured substrings
2721        \C-         do not supply a callout function
2722        \C!n        return 1 when callout number n is reached
2723        \C!n!m      return 1 when callout number n is reached for the mth time
2724    
2725    30. If pcregrep was called with the -l option and just a single file name, it
2726    output "<stdin>" if a match was found, instead of the file name.
2727    
2728    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2729    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2730    pcre_exec(). This saves a malloc/free per call. The default value of
2731    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2732    when configuring.
2733    
2734    32. The default maximum size of a compiled pattern is 64K. There have been a
2735    few cases of people hitting this limit. The code now uses macros to handle the
2736    storing of links as offsets within the compiled pattern. It defaults to 2-byte
2737    links, but this can be changed to 3 or 4 bytes by --with-link-size when
2738    configuring. Tests 2 and 5 work only with 2-byte links because they output
2739    debugging information about compiled patterns.
2740    
2741    33. Internal code re-arrangements:
2742    
2743    (a) Moved the debugging function for printing out a compiled regex into
2744        its own source file (printint.c) and used #include to pull it into
2745        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2746        separate copies.
2747    
2748    (b) Defined the list of op-code names for debugging as a macro in
2749        internal.h so that it is next to the definition of the opcodes.
2750    
2751    (c) Defined a table of op-code lengths for simpler skipping along compiled
2752        code. This is again a macro in internal.h so that it is next to the
2753        definition of the opcodes.
2754    
2755    34. Added support for recursive calls to individual subpatterns, along the
2756    lines of Robin Houston's patch (but implemented somewhat differently).
2757    
2758    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2759    allow it to read and process whole directories in Win32. This code was
2760    contributed by Lionel Fourquaux; it has not been tested by me.
2761    
2762    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2763    used to name a group. Names consist of alphanumerics and underscores, and must
2764    be unique. Back references use the syntax (?P=name) and recursive calls use
2765    (?P>name) which is a PCRE extension to the Python extension. Groups still have
2766    numbers. The function pcre_fullinfo() can be used after compilation to extract
2767    a name/number map. There are three relevant calls:
2768    
2769      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
2770      PCRE_INFO_NAMECOUNT            yields the number of entries
2771      PCRE_INFO_NAMETABLE            yields a pointer to the map.
2772    
2773    The map is a vector of fixed-size entries. The size of each entry depends on
2774    the length of the longest name used. The first two bytes of each entry are the
2775    group number, most significant byte first. There follows the corresponding
2776    name, zero terminated. The names are in alphabetical order.
2777    
2778    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2779    case instead of 255. Making it the same both with and without UTF-8 support
2780    means that the same test output works with both.
2781    
2782    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2783    calling malloc() with a zero argument.
2784    
2785    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2786    optimization. I've improved things by keeping a bitmap of backreferences with
2787    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2788    fact referenced, the optimization can be applied. It is unlikely that a
2789    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2790    the match to follow \n) will appear inside brackets with a number greater than
2791    31, but if it does, any back reference > 31 suppresses the optimization.
2792    
2793    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2794    of disabling numbered capturing parentheses. Any opening parenthesis that is
2795    not followed by ? behaves as if it were followed by ?: but named parentheses
2796    can still be used for capturing (and they will acquire numbers in the usual
2797    way).
2798    
2799    41. Redesigned the return codes from the match() function into yes/no/error so
2800    that errors can be passed back from deep inside the nested calls. A malloc
2801    failure while inside a recursive subpattern call now causes the
2802    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2803    
2804    42. It is now possible to set a limit on the number of times the match()
2805    function is called in a call to pcre_exec(). This facility makes it possible to
2806    limit the amount of recursion and backtracking, though not in a directly
2807    obvious way, because the match() function is used in a number of different
2808    circumstances. The count starts from zero for each position in the subject
2809    string (for non-anchored patterns). The default limit is, for compatibility, a
2810    large number, namely 10 000 000. You can change this in two ways:
2811    
2812    (a) When configuring PCRE before making, you can use --with-match-limit=n
2813        to set a default value for the compiled library.
2814    
2815    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2816        a different value is set. See 45 below.
2817    
2818    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2819    
2820    43. Added a new function pcre_config(int, void *) to enable run-time extraction
2821    of things that can be changed at compile time. The first argument specifies
2822    what is wanted and the second points to where the information is to be placed.
2823    The current list of available information is:
2824    
2825      PCRE_CONFIG_UTF8
2826    
2827    The output is an integer that is set to one if UTF-8 support is available;
2828    otherwise it is set to zero.
2829    
2830      PCRE_CONFIG_NEWLINE
2831    
2832    The output is an integer that it set to the value of the code that is used for
2833    newline. It is either LF (10) or CR (13).
2834    
2835      PCRE_CONFIG_LINK_SIZE
2836    
2837    The output is an integer that contains the number of bytes used for internal
2838    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2839    
2840      PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
2841    
2842    The output is an integer that contains the threshold above which the POSIX
2843    interface uses malloc() for output vectors. See item 31 above.
2844    
2845      PCRE_CONFIG_MATCH_LIMIT
2846    
2847    The output is an unsigned integer that contains the default limit of the number
2848    of match() calls in a pcre_exec() execution. See 42 above.
2849    
2850    44. pcretest has been upgraded by the addition of the -C option. This causes it
2851    to extract all the available output from the new pcre_config() function, and to
2852    output it. The program then exits immediately.
2853    
2854    45. A need has arisen to pass over additional data with calls to pcre_exec() in
2855    order to support additional features. One way would have been to define
2856    pcre_exec2() (for example) with extra arguments, but this would not have been
2857    extensible, and would also have required all calls to the original function to
2858    be mapped to the new one. Instead, I have chosen to extend the mechanism that
2859    is used for passing in "extra" data from pcre_study().
2860    
2861    The pcre_extra structure is now exposed and defined in pcre.h. It currently
2862    contains the following fields:
2863    
2864      flags         a bitmap indicating which of the following fields are set
2865      study_data    opaque data from pcre_study()
2866      match_limit   a way of specifying a limit on match() calls for a specific
2867                      call to pcre_exec()
2868      callout_data  data for callouts (see 49 below)
2869    
2870    The flag bits are also defined in pcre.h, and are
2871    
2872      PCRE_EXTRA_STUDY_DATA
2873      PCRE_EXTRA_MATCH_LIMIT
2874      PCRE_EXTRA_CALLOUT_DATA
2875    
2876    The pcre_study() function now returns one of these new pcre_extra blocks, with
2877    the actual study data pointed to by the study_data field, and the
2878    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2879    before. That is, this change is entirely upwards-compatible and requires no
2880    change to existing code.
2881    
2882    If you want to pass in additional data to pcre_exec(), you can either place it
2883    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2884    block.
2885    
2886    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2887    data string contains the escape sequence \M, pcretest calls pcre_exec() several
2888    times with different match limits, until it finds the minimum value needed for
2889    pcre_exec() to complete. The value is then output. This can be instructive; for
2890    most simple matches the number is quite small, but for pathological cases it
2891    gets very large very quickly.
2892    
2893    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2894    returns the size of the data block pointed to by the study_data field in a
2895    pcre_extra block, that is, the value that was passed as the argument to
2896    pcre_malloc() when PCRE was getting memory in which to place the information
2897    created by pcre_study(). The fourth argument should point to a size_t variable.
2898    pcretest has been extended so that this information is shown after a successful
2899    pcre_study() call when information about the compiled regex is being displayed.
2900    
2901    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2902    because what follows is always an absolute path. (Later: it turns out that this
2903    is more than cosmetic for MinGW, because it doesn't like empty path
2904    components.)
2905    
2906    49. Some changes have been made to the callout feature (see 28 above):
2907    
2908    (i)  A callout function now has three choices for what it returns:
2909    
2910           0  =>  success, carry on matching
2911         > 0  =>  failure at this point, but backtrack if possible
2912         < 0  =>  serious error, return this value from pcre_exec()
2913    
2914         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2915         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2916         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2917         use by callout functions. It will never be used by PCRE itself.
2918    
2919    (ii) The pcre_extra structure (see 45 above) has a void * field called
2920         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2921         pcre_callout_block structure has a field of the same name. The contents of
2922         the field passed in the pcre_extra structure are passed to the callout
2923         function in the corresponding field in the callout block. This makes it
2924         easier to use the same callout-containing regex from multiple threads. For
2925         testing, the pcretest program has a new data escape
2926    
2927           \C*n        pass the number n (may be negative) as callout_data
2928    
2929         If the callout function in pcretest receives a non-zero value as
2930         callout_data, it returns that value.
2931    
2932    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2933    there were some redundant $(CFLAGS) in commands that are now specified as
2934    $(LINK), which already includes $(CFLAGS).
2935    
2936    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2937    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2938    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2939    one-byte characters throughout. Note that case-insensitive matching applies
2940    only to characters whose values are less than 256. PCRE doesn't support the
2941    notion of cases for higher-valued characters.
2942    
2943    (i)   A character class whose characters are all within 0-255 is handled as
2944          a bit map, and the map is inverted for negative classes. Previously, a
2945          character > 255 always failed to match such a class; however it should
2946          match if the class was a negative one (e.g. [^ab]). This has been fixed.
2947    
2948    (ii)  A negated character class with a single character < 255 is coded as
2949          "not this character" (OP_NOT). This wasn't working properly when the test
2950          character was multibyte, either singly or repeated.
2951    
2952    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2953          mode, for example: \x{100}{2,3}.
2954    
2955    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2956          singly or repeated) now correctly test multibyte characters. However,
2957          PCRE doesn't recognize any characters with values greater than 255 as
2958          digits, spaces, or word characters. Such characters always match \D, \S,
2959          and \W, and never match \d, \s, or \w.
2960    
2961    (v)   Classes may now contain characters and character ranges with values
2962          greater than 255. For example: [ab\x{100}-\x{400}].
2963    
2964    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
2965          PCRE in UTF-8 mode.
2966    
2967    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2968    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2969    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2970    value.)
2971    
2972    53. The single man page has become too large. I have therefore split it up into
2973    a number of separate man pages. These also give rise to individual HTML pages;
2974    these are now put in a separate directory, and there is an index.html page that
2975    lists them all. Some hyperlinking between the pages has been installed.
2976    
2977    54. Added convenience functions for handling named capturing parentheses.
2978    
2979    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2980    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2981    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2982    are faulted.
2983    
2984    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2985    calling configure. These values are used when compiling the dftables.c program
2986    which is run to generate the source of the default character tables. They
2987    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2988    you will need to set these values.
2989    
2990    57. Updated the building process for Windows DLL, as provided by Fred Cox.
2991    
2992    
2993    Version 3.9 02-Jan-02
2994    ---------------------
2995    
2996    1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2997    
2998    2. If --disable-static was given, the building process failed when trying to
2999    build pcretest and pcregrep. (For some reason it was using libtool to compile
3000    them, which is not right, as they aren't part of the library.)
3001    
3002    
3003    Version 3.8 18-Dec-01
3004    ---------------------
3005    
3006    1. The experimental UTF-8 code was completely screwed up. It was packing the
3007    bytes in the wrong order. How dumb can you get?
3008    
3009    
3010    Version 3.7 29-Oct-01
3011    ---------------------
3012    
3013    1. In updating pcretest to check change 1 of version 3.6, I screwed up.
3014    This caused pcretest, when used on the test data, to segfault. Unfortunately,
3015    this didn't happen under Solaris 8, where I normally test things.
3016    
3017    2. The Makefile had to be changed to make it work on BSD systems, where 'make'
3018    doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
3019    isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
3020    this fix an hour or so after the initial 3.7 release.)
3021    
3022    
3023    Version 3.6 23-Oct-01
3024    ---------------------
3025    
3026    1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
3027    offsets passed as NULL with zero offset count.
3028    
3029    2. The config.guess and config.sub files had not been updated when I moved to
3030    the latest autoconf.
3031    
3032    
3033    Version 3.5 15-Aug-01
3034    ---------------------
3035    
3036    1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
3037    had been forgotten.
3038    
3039    2. By using declared but undefined structures, we can avoid using "void"
3040    definitions in pcre.h while keeping the internal definitions of the structures
3041    private.
3042    
3043    3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
3044    user point of view, this means that both static and shared libraries are built
3045    by default, but this can be individually controlled. More of the work of
3046    handling this static/shared cases is now inside libtool instead of PCRE's make
3047    file.
3048    
3049    4. The pcretest utility is now installed along with pcregrep because it is
3050    useful for users (to test regexs) and by doing this, it automatically gets
3051    relinked by libtool. The documentation has been turned into a man page, so
3052    there are now .1, .txt, and .html versions in /doc.
3053    
3054    5. Upgrades to pcregrep:
3055       (i)   Added long-form option names like gnu grep.
3056       (ii)  Added --help to list all options with an explanatory phrase.
3057       (iii) Added -r, --recursive to recurse into sub-directories.
3058       (iv)  Added -f, --file to read patterns from a file.
3059    
3060    6. pcre_exec() was referring to its "code" argument before testing that
3061    argument for NULL (and giving an error if it was NULL).
3062    
3063    7. Upgraded Makefile.in to allow for compiling in a different directory from
3064    the source directory.
3065    
3066    8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
3067    options bits, the pointer it was passed was to an int instead of to an unsigned
3068    long int. This mattered only on 64-bit systems.
3069    
3070    9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
3071    generated) instead of pcre.in, which it its source. Also made the same change
3072    in several of the .c files.
3073    
3074    10. A new release of gcc defines printf() as a macro, which broke pcretest
3075    because it had an ifdef in the middle of a string argument for printf(). Fixed
3076    by using separate calls to printf().
3077    
3078    11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
3079    script, to force use of CR or LF instead of \n in the source. On non-Unix
3080    systems, the value can be set in config.h.
3081    
3082    12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
3083    absolute limit. Changed the text of the error message to make this clear, and
3084    likewise updated the man page.
3085    
3086    13. The limit of 99 on the number of capturing subpatterns has been removed.
3087    The new limit is 65535, which I hope will not be a "real" limit.
3088    
3089    
3090  Version 3.4 22-Aug-00  Version 3.4 22-Aug-00
3091  ---------------------  ---------------------

Legend:
Removed from v.51  
changed lines
  Added in v.478

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12