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

Diff of /code/trunk/ChangeLog

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

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35        was out-of-date, and there was no check on the pcre_compile() error code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

Legend:
Removed from v.67  
changed lines
  Added in v.473

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12