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

Diff of /code/trunk/ChangeLog

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

revision 13 by nigel, Sat Feb 24 21:38:21 2007 UTC revision 483 by ph10, Mon Jan 4 16:19:17 2010 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    Version 8.01 11-Dec-09
5    ----------------------
6    
7    1.  If a pattern contained a conditional subpattern with only one branch (in
8        particular, this includes all (DEFINE) patterns), a call to pcre_study()
9        computed the wrong minimum data length (which is of course zero for such
10        subpatterns).
11    
12    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
13        the pattern is reset in the first branch, pcre_compile() failed with
14        "internal error: code overflow at offset...". This happened only when
15        the reset was to the original external option setting. (An optimization
16        abstracts leading options settings into an external setting, which was the
17        cause of this.)
18    
19    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
20        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
21        assertion pattern did not match (meaning that the assertion was true), it
22        was incorrectly treated as false if the SKIP had been reached during the
23        matching. This also applied to assertions used as conditions.
24    
25    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
26        assertion subpattern, including such a pattern used as a condition,
27        unpredictable results occurred, instead of the error return
28        PCRE_ERROR_DFA_UITEM.
29    
30    5.  The C++ GlobalReplace function was not working like Perl for the special
31        situation when an empty string is matched. It now does the fancy magic
32        stuff that is necessary.
33    
34    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
35        removed. (These were left over from very, very early versions of PCRE.)
36    
37    7.  Some cosmetic changes to the code to make life easier when compiling it
38        as part of something else:
39    
40        (a) Change DEBUG to PCRE_DEBUG.
41    
42        (b) In pcre_compile(), rename the member of the "branch_chain" structure
43            called "current" as "current_branch", to prevent a collision with the
44            Linux macro when compiled as a kernel module.
45    
46        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
47            prevent a collision with the Linux macro when compiled as a kernel
48            module.
49    
50    8.  In pcre_compile() there are some checks for integer overflows that used to
51        cast potentially large values to (double). This has been changed to that
52        when building, a check for int64_t is made, and if it is found, it is used
53        instead, thus avoiding the use of floating point arithmetic. (There is no
54        other use of FP in PCRE.) If int64_t is not found, the fallback is to
55        double.
56    
57    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
58        2005 (difference between two addresses compared to an unsigned value).
59    
60    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
61        custom one, because of the following reported problem in Windows:
62    
63          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
64            under Win32.
65          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
66            therefore missing the function definition.
67          - The compiler thus generates a "C" signature for the test function.
68          - The linker fails to find the "C" function.
69          - PCRE fails to configure if asked to do so against libbz2.
70    
71    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
72        messages were output:
73    
74          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
75          rerunning libtoolize, to keep the correct libtool macros in-tree.
76          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
77    
78        I have done both of these things.
79    
80    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
81        most of the time, it *can* run out if it is given a pattern that contains a
82        runaway infinite recursion. I updated the discussion in the pcrestack man
83        page.
84    
85    13. Now that we have gone to the x.xx style of version numbers, the minor
86        version may start with zero. Using 08 or 09 is a bad idea because users
87        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
88        interpreted as invalid octal numbers. I've updated the previous comment in
89        configure.ac, and also added a check that gives an error if 08 or 09 are
90        used.
91    
92    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
93        causing partial matching to fail when the end of the subject matched \W
94        in a UTF-8 pattern where \W was quantified with a minimum of 3.
95    
96    15. There were some discrepancies between the declarations in pcre_internal.h
97        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
98        their definitions. The declarations used "const uschar *" and the
99        definitions used USPTR. Even though USPTR is normally defined as "const
100        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
101        reported that: "This difference in casting confuses some C++ compilers, for
102        example, SunCC recognizes above declarations as different functions and
103        generates broken code for hbpcre." I have changed the declarations to use
104        USPTR.
105    
106    
107    
108    
109    Version 8.00 19-Oct-09
110    ----------------------
111    
112    1.  The table for translating pcre_compile() error codes into POSIX error codes
113        was out-of-date, and there was no check on the pcre_compile() error code
114        being within the table. This could lead to an OK return being given in
115        error.
116    
117    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
118        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
119        in a Windows environment.
120    
121    3.  The pcregrep --count option prints the count for each file even when it is
122        zero, as does GNU grep. However, pcregrep was also printing all files when
123        --files-with-matches was added. Now, when both options are given, it prints
124        counts only for those files that have at least one match. (GNU grep just
125        prints the file name in this circumstance, but including the count seems
126        more useful - otherwise, why use --count?) Also ensured that the
127        combination -clh just lists non-zero counts, with no names.
128    
129    4.  The long form of the pcregrep -F option was incorrectly implemented as
130        --fixed_strings instead of --fixed-strings. This is an incompatible change,
131        but it seems right to fix it, and I didn't think it was worth preserving
132        the old behaviour.
133    
134    5.  The command line items --regex=pattern and --regexp=pattern were not
135        recognized by pcregrep, which required --regex pattern or --regexp pattern
136        (with a space rather than an '='). The man page documented the '=' forms,
137        which are compatible with GNU grep; these now work.
138    
139    6.  No libpcreposix.pc file was created for pkg-config; there was just
140        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
141    
142    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
143        when UCP support is not needed, by modifying the Python script that
144        generates it from Unicode data files. This should not matter if the module
145        is correctly used as a library, but I received one complaint about 50K of
146        unwanted data. My guess is that the person linked everything into his
147        program rather than using a library. Anyway, it does no harm.
148    
149    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
150        was a minimum greater than 1 for a wide character in a possessive
151        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
152        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
153        character. Chaos in the form of incorrect output or a compiling loop could
154        result.
155    
156    9.  The restrictions on what a pattern can contain when partial matching is
157        requested for pcre_exec() have been removed. All patterns can now be
158        partially matched by this function. In addition, if there are at least two
159        slots in the offset vector, the offset of the earliest inspected character
160        for the match and the offset of the end of the subject are set in them when
161        PCRE_ERROR_PARTIAL is returned.
162    
163    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
164        synonymous with PCRE_PARTIAL, for backwards compatibility, and
165        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
166        and may be more useful for multi-segment matching.
167    
168    11. Partial matching with pcre_exec() is now more intuitive. A partial match
169        used to be given if ever the end of the subject was reached; now it is
170        given only if matching could not proceed because another character was
171        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
172        string "Z", which now yields "no match" instead of "partial match". In the
173        case of pcre_dfa_exec(), "no match" is given if every matching path for the
174        final character ended with (*FAIL).
175    
176    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
177        if the pattern had a "must contain" character that was already found in the
178        earlier partial match, unless partial matching was again requested. For
179        example, with the pattern /dog.(body)?/, the "must contain" character is
180        "g". If the first part-match was for the string "dog", restarting with
181        "sbody" failed. This bug has been fixed.
182    
183    13. The string returned by pcre_dfa_exec() after a partial match has been
184        changed so that it starts at the first inspected character rather than the
185        first character of the match. This makes a difference only if the pattern
186        starts with a lookbehind assertion or \b or \B (\K is not supported by
187        pcre_dfa_exec()). It's an incompatible change, but it makes the two
188        matching functions compatible, and I think it's the right thing to do.
189    
190    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
191        so that the demonstration program is easily available in environments where
192        PCRE has not been installed from source.
193    
194    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
195        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
196        library.
197    
198    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
199        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
200        is not the first non-POSIX option to be added. Clearly some people find
201        these options useful.
202    
203    17. If a caller to the POSIX matching function regexec() passes a non-zero
204        value for nmatch with a NULL value for pmatch, the value of
205        nmatch is forced to zero.
206    
207    18. RunGrepTest did not have a test for the availability of the -u option of
208        the diff command, as RunTest does. It now checks in the same way as
209        RunTest, and also checks for the -b option.
210    
211    19. If an odd number of negated classes containing just a single character
212        interposed, within parentheses, between a forward reference to a named
213        subpattern and the definition of the subpattern, compilation crashed with
214        an internal error, complaining that it could not find the referenced
215        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
216        [The bug was that it was starting one character too far in when skipping
217        over the character class, thus treating the ] as data rather than
218        terminating the class. This meant it could skip too much.]
219    
220    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
221        /g option in pcretest when the pattern contains \K, which makes it possible
222        to have an empty string match not at the start, even when the pattern is
223        anchored. Updated pcretest and pcredemo to use this option.
224    
225    21. If the maximum number of capturing subpatterns in a recursion was greater
226        than the maximum at the outer level, the higher number was returned, but
227        with unset values at the outer level. The correct (outer level) value is
228        now given.
229    
230    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
231        PCRE did not set those parentheses (unlike Perl). I have now found a way to
232        make it do so. The string so far is captured, making this feature
233        compatible with Perl.
234    
235    23. The tests have been re-organized, adding tests 11 and 12, to make it
236        possible to check the Perl 5.10 features against Perl 5.10.
237    
238    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
239        pattern matches a fixed length string. PCRE did not allow this; now it
240        does. Neither allows recursion.
241    
242    25. I finally figured out how to implement a request to provide the minimum
243        length of subject string that was needed in order to match a given pattern.
244        (It was back references and recursion that I had previously got hung up
245        on.) This code has now been added to pcre_study(); it finds a lower bound
246        to the length of subject needed. It is not necessarily the greatest lower
247        bound, but using it to avoid searching strings that are too short does give
248        some useful speed-ups. The value is available to calling programs via
249        pcre_fullinfo().
250    
251    26. While implementing 25, I discovered to my embarrassment that pcretest had
252        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
253        study optimizations had never been tested with that matching function.
254        Oops. What is worse, even when it was passed study data, there was a bug in
255        pcre_dfa_exec() that meant it never actually used it. Double oops. There
256        were also very few tests of studied patterns with pcre_dfa_exec().
257    
258    27. If (?| is used to create subpatterns with duplicate numbers, they are now
259        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
260        on the other side of the coin, they are no longer allowed to have different
261        names, because these cannot be distinguished in PCRE, and this has caused
262        confusion. (This is a difference from Perl.)
263    
264    28. When duplicate subpattern names are present (necessarily with different
265        numbers, as required by 27 above), and a test is made by name in a
266        conditional pattern, either for a subpattern having been matched, or for
267        recursion in such a pattern, all the associated numbered subpatterns are
268        tested, and the overall condition is true if the condition is true for any
269        one of them. This is the way Perl works, and is also more like the way
270        testing by number works.
271    
272    
273    Version 7.9 11-Apr-09
274    ---------------------
275    
276    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
277        (pcretest), all targets were linked against these libraries. This included
278        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
279        libraries. This caused unwanted dependencies to be created. This problem
280        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
281        pcretest is linked with readline.
282    
283    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
284        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
285        moved outside it again, because FALSE and TRUE are already defined in AIX,
286        but BOOL is not.
287    
288    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
289        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
290    
291    4.  The pcregrep documentation said spaces were inserted as well as colons (or
292        hyphens) following file names and line numbers when outputting matching
293        lines. This is not true; no spaces are inserted. I have also clarified the
294        wording for the --colour (or --color) option.
295    
296    5.  In pcregrep, when --colour was used with -o, the list of matching strings
297        was not coloured; this is different to GNU grep, so I have changed it to be
298        the same.
299    
300    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
301        each matching line was coloured. Now it goes on to look for further matches
302        of any of the test patterns, which is the same behaviour as GNU grep.
303    
304    7.  A pattern that could match an empty string could cause pcregrep to loop; it
305        doesn't make sense to accept an empty string match in pcregrep, so I have
306        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
307        seems to be how GNU grep behaves.
308    
309    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
310        start or after a newline", because the conditional assertion was not being
311        correctly handled. The rule now is that both the assertion and what follows
312        in the first alternative must satisfy the test.
313    
314    9.  If auto-callout was enabled in a pattern with a conditional group whose
315        condition was an assertion, PCRE could crash during matching, both with
316        pcre_exec() and pcre_dfa_exec().
317    
318    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
319        used for matching.
320    
321    11. Unicode property support in character classes was not working for
322        characters (bytes) greater than 127 when not in UTF-8 mode.
323    
324    12. Added the -M command line option to pcretest.
325    
326    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
327    
328    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
329    
330    16. Added comments and documentation about mis-use of no_arg in the C++
331        wrapper.
332    
333    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
334        from Martin Jerabek that uses macro names for all relevant character and
335        string constants.
336    
337    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
338        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
339        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
340        these, but not everybody uses configure.
341    
342    19. A conditional group that had only one branch was not being correctly
343        recognized as an item that could match an empty string. This meant that an
344        enclosing group might also not be so recognized, causing infinite looping
345        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
346        with the subject "ab", where knowledge that the repeated group can match
347        nothing is needed in order to break the loop.
348    
349    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
350        exec(), but without supplying a callout function, matching went wrong.
351    
352    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
353        leak if the size of the offset vector was greater than 30. When the vector
354        is smaller, the saved offsets during recursion go onto a local stack
355        vector, but for larger vectors malloc() is used. It was failing to free
356        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
357        error, in fact).
358    
359    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
360        heapframe that is used only when UTF-8 support is enabled. This caused no
361        problem, but was untidy.
362    
363    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
364        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
365        included within another project.
366    
367    24. Steven Van Ingelgem's patches to add more options to the CMake support,
368        slightly modified by me:
369    
370          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
371              not building pcregrep.
372    
373          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
374              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
375    
376    25. Forward references, both numeric and by name, in patterns that made use of
377        duplicate group numbers, could behave incorrectly or give incorrect errors,
378        because when scanning forward to find the reference group, PCRE was not
379        taking into account the duplicate group numbers. A pattern such as
380        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
381    
382    26. Changed a few more instances of "const unsigned char *" to USPTR, making
383        the feature of a custom pointer more persuasive (as requested by a user).
384    
385    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
386        pcretest.c, inside #ifndefs, because it seems they are sometimes already
387        pre-defined.
388    
389    28. Added support for (*UTF8) at the start of a pattern.
390    
391    29. Arrange for flags added by the "release type" setting in CMake to be shown
392        in the configuration summary.
393    
394    
395    Version 7.8 05-Sep-08
396    ---------------------
397    
398    1.  Replaced UCP searching code with optimized version as implemented for Ad
399        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
400        stage table and inline lookup instead of a function, giving speed ups of 2
401        to 5 times on some simple patterns that I tested. Permission was given to
402        distribute the MultiStage2.py script that generates the tables (it's not in
403        the tarball, but is in the Subversion repository).
404    
405    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
406        scripts.
407    
408    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
409        a group with a zero qualifier. The result of the study could be incorrect,
410        or the function might crash, depending on the pattern.
411    
412    4.  Caseless matching was not working for non-ASCII characters in back
413        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
414        It now works when Unicode Property Support is available.
415    
416    5.  In pcretest, an escape such as \x{de} in the data was always generating
417        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
418        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
419        truncation.
420    
421    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
422    
423    7.  Added two (int) casts to pcregrep when printing the difference of two
424        pointers, in case they are 64-bit values.
425    
426    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
427        test 2 if it fails.
428    
429    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
430        and a #define of that name to empty if it is not externally set. This is to
431        allow users of MSVC to set it if necessary.
432    
433    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
434        the convenience functions in the pcre_get.c source file.
435    
436    11. An option change at the start of a pattern that had top-level alternatives
437        could cause overwriting and/or a crash. This command provoked a crash in
438        some environments:
439    
440          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
441    
442        This potential security problem was recorded as CVE-2008-2371.
443    
444    12. For a pattern where the match had to start at the beginning or immediately
445        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
446        pcre_dfa_exec() could read past the end of the passed subject if there was
447        no match. To help with detecting such bugs (e.g. with valgrind), I modified
448        pcretest so that it places the subject at the end of its malloc-ed buffer.
449    
450    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
451        exec() might read past the end of the data buffer in UTF-8 mode.
452    
453    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
454        the data contained the byte 0x85 as part of a UTF-8 character within its
455        first line. This applied both to normal and DFA matching.
456    
457    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
458        /^[^d]*?$/8 failed to match "abc".
459    
460    16. Added a missing copyright notice to pcrecpp_internal.h.
461    
462    17. Make it more clear in the documentation that values returned from
463        pcre_exec() in ovector are byte offsets, not character counts.
464    
465    18. Tidied a few places to stop certain compilers from issuing warnings.
466    
467    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
468        supplied by Stefan Weber. I made a further small update for 7.8 because
469        there is a change of source arrangements: the pcre_searchfuncs.c module is
470        replaced by pcre_ucd.c.
471    
472    
473    Version 7.7 07-May-08
474    ---------------------
475    
476    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
477        a string to a long long, pretend we don't even have a long long." This is
478        done by checking for the strtoq, strtoll, and _strtoi64 functions.
479    
480    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
481        pre-7.6 versions, which defined a global no_arg variable instead of putting
482        it in the RE class. (See also #8 below.)
483    
484    3.  Remove a line of dead code, identified by coverity and reported by Nuno
485        Lopes.
486    
487    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
488    
489        (1) The include/exclude patterns were being applied to the whole pathnames
490            of files, instead of just to the final components.
491    
492        (2) If there was more than one level of directory, the subdirectories were
493            skipped unless they satisfied the include/exclude conditions. This is
494            inconsistent with GNU grep (and could even be seen as contrary to the
495            pcregrep specification - which I improved to make it absolutely clear).
496            The action now is always to scan all levels of directory, and just
497            apply the include/exclude patterns to regular files.
498    
499    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
500        --exclude_dir in the tests to avoid scanning .svn directories.
501    
502    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
503        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
504        doesn't support NULs in patterns.
505    
506    7.  Added some missing "const"s to declarations of static tables in
507        pcre_compile.c and pcre_dfa_exec.c.
508    
509    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
510        caused by fix #2  above. (Subsequently also a second patch to fix the
511        first patch. And a third patch - this was a messy problem.)
512    
513    9.  Applied Craig's patch to remove the use of push_back().
514    
515    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
516        matching function regexec().
517    
518    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
519        which, however, unlike Perl's \g{...}, are subroutine calls, not back
520        references. PCRE supports relative numbers with this syntax (I don't think
521        Oniguruma does).
522    
523    12. Previously, a group with a zero repeat such as (...){0} was completely
524        omitted from the compiled regex. However, this means that if the group
525        was called as a subroutine from elsewhere in the pattern, things went wrong
526        (an internal error was given). Such groups are now left in the compiled
527        pattern, with a new opcode that causes them to be skipped at execution
528        time.
529    
530    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
531        to the way PCRE behaves:
532    
533        (a) A lone ] character is dis-allowed (Perl treats it as data).
534    
535        (b) A back reference to an unmatched subpattern matches an empty string
536            (Perl fails the current match path).
537    
538        (c) A data ] in a character class must be notated as \] because if the
539            first data character in a class is ], it defines an empty class. (In
540            Perl it is not possible to have an empty class.) The empty class []
541            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
542            The negative empty class [^] matches any one character, independently
543            of the DOTALL setting.
544    
545    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
546        non-existent subpattern following a character class starting with ']' and
547        containing () gave an internal compiling error instead of "reference to
548        non-existent subpattern". Fortunately, when the pattern did exist, the
549        compiled code was correct. (When scanning forwards to check for the
550        existencd of the subpattern, it was treating the data ']' as terminating
551        the class, so got the count wrong. When actually compiling, the reference
552        was subsequently set up correctly.)
553    
554    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
555        it was being rejected as not supported by pcre_dfa_exec(), even though
556        other assertions are supported. I have made pcre_dfa_exec() support
557        (*FAIL).
558    
559    16. The implementation of 13c above involved the invention of a new opcode,
560        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
561        cannot be changed at match time, I realized I could make a small
562        improvement to matching performance by compiling OP_ALLANY instead of
563        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
564        on the OP_ANY path.
565    
566    17. Compiling pcretest on Windows with readline support failed without the
567        following two fixes: (1) Make the unistd.h include conditional on
568        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
569    
570    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
571        ncurses library to be included for pcretest when ReadLine support is
572        requested, but also to allow for it to be overridden. This patch came from
573        Daniel Bergström.
574    
575    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
576        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
577        any errors with the current Unicode tables. Thanks to Peter Kankowski for
578        spotting this.
579    
580    
581    Version 7.6 28-Jan-08
582    ---------------------
583    
584    1.  A character class containing a very large number of characters with
585        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
586        overflow.
587    
588    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
589        HAVE_LONG_LONG is not defined.
590    
591    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
592        bring them up to date and include new features. This patch includes:
593    
594        - Fixed PH's badly added libz and libbz2 support.
595        - Fixed a problem with static linking.
596        - Added pcredemo. [But later removed - see 7 below.]
597        - Fixed dftables problem and added an option.
598        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
599            HAVE_LONG_LONG.
600        - Added readline support for pcretest.
601        - Added an listing of the option settings after cmake has run.
602    
603    4.  A user submitted a patch to Makefile that makes it easy to create
604        "pcre.dll" under mingw when using Configure/Make. I added stuff to
605        Makefile.am that cause it to include this special target, without
606        affecting anything else. Note that the same mingw target plus all
607        the other distribution libraries and programs are now supported
608        when configuring with CMake (see 6 below) instead of with
609        Configure/Make.
610    
611    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
612        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
613        exported in the Windows port". It has not yet been confirmed that the patch
614        solves the problem, but it does no harm.
615    
616    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
617        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
618        with CMake, and also correct the comment about stack recursion.
619    
620    7.  Remove the automatic building of pcredemo from the ./configure system and
621        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
622        of a program that users should build themselves after PCRE is installed, so
623        building it automatically is not really right. What is more, it gave
624        trouble in some build environments.
625    
626    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
627    
628    
629    Version 7.5 10-Jan-08
630    ---------------------
631    
632    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
633        values in parens when parsing an RE using the C++ wrapper."
634    
635    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
636        Characters greater than 255 were excluded from the class instead of being
637        included.
638    
639    3.  The same bug as (2) above applied to negated POSIX classes such as
640        [:^space:].
641    
642    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
643        defined or documented. It seems to have been a typo for PCRE_STATIC, so
644        I have changed it.
645    
646    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
647        first named subpattern) and a construct such as (?&a) would reference the
648        first named subpattern whose name started with "a" (in other words, the
649        length check was missing). Both these problems are fixed. "Subpattern name
650        expected" is now given for (?&) (a zero-length name), and this patch also
651        makes it give the same error for \k'' (previously it complained that that
652        was a reference to a non-existent subpattern).
653    
654    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
655        this is right because (?- can be followed by option settings as well as by
656        digits. I have, however, made the messages clearer.
657    
658    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
659        than the number used in the conditional) now cause a compile-time error.
660        This is actually not compatible with Perl, which accepts such patterns, but
661        treats the conditional as always being FALSE (as PCRE used to), but it
662        seems to me that giving a diagnostic is better.
663    
664    8.  Change "alphameric" to the more common word "alphanumeric" in comments
665        and messages.
666    
667    9.  Fix two occurrences of "backslash" in comments that should have been
668        "backspace".
669    
670    10. Remove two redundant lines of code that can never be obeyed (their function
671        was moved elsewhere).
672    
673    11. The program that makes PCRE's Unicode character property table had a bug
674        which caused it to generate incorrect table entries for sequences of
675        characters that have the same character type, but are in different scripts.
676        It amalgamated them into a single range, with the script of the first of
677        them. In other words, some characters were in the wrong script. There were
678        thirteen such cases, affecting characters in the following ranges:
679    
680          U+002b0 - U+002c1
681          U+0060c - U+0060d
682          U+0061e - U+00612
683          U+0064b - U+0065e
684          U+0074d - U+0076d
685          U+01800 - U+01805
686          U+01d00 - U+01d77
687          U+01d9b - U+01dbf
688          U+0200b - U+0200f
689          U+030fc - U+030fe
690          U+03260 - U+0327f
691          U+0fb46 - U+0fbb1
692          U+10450 - U+1049d
693    
694    12. The -o option (show only the matching part of a line) for pcregrep was not
695        compatible with GNU grep in that, if there was more than one match in a
696        line, it showed only the first of them. It now behaves in the same way as
697        GNU grep.
698    
699    13. If the -o and -v options were combined for pcregrep, it printed a blank
700        line for every non-matching line. GNU grep prints nothing, and pcregrep now
701        does the same. The return code can be used to tell if there were any
702        non-matching lines.
703    
704    14. Added --file-offsets and --line-offsets to pcregrep.
705    
706    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
707        infinitely looping recursion. The bug was that positive lookaheads were not
708        being skipped when checking for a possible empty match (negative lookaheads
709        and both kinds of lookbehind were skipped).
710    
711    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
712        inclusion of <windows.h> to before rather than after the definition of
713        INVALID_FILE_ATTRIBUTES (patch from David Byron).
714    
715    17. Specifying a possessive quantifier with a specific limit for a Unicode
716        character property caused pcre_compile() to compile bad code, which led at
717        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
718        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
719        caused the error; without that there was no problem.
720    
721    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
722    
723    19. Added --enable-pcretest-libreadline.
724    
725    20. In pcrecpp.cc, the variable 'count' was incremented twice in
726        RE::GlobalReplace(). As a result, the number of replacements returned was
727        double what it should be. I removed one of the increments, but Craig sent a
728        later patch that removed the other one (the right fix) and added unit tests
729        that check the return values (which was not done before).
730    
731    21. Several CMake things:
732    
733        (1) Arranged that, when cmake is used on Unix, the libraries end up with
734            the names libpcre and libpcreposix, not just pcre and pcreposix.
735    
736        (2) The above change means that pcretest and pcregrep are now correctly
737            linked with the newly-built libraries, not previously installed ones.
738    
739        (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
740    
741    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
742        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
743        UTF-8 newline character). The key issue is that the pattern starts .*;
744        this means that the match must be either at the beginning, or after a
745        newline. The bug was in the code for advancing after a failed match and
746        checking that the new position followed a newline. It was not taking
747        account of UTF-8 characters correctly.
748    
749    23. PCRE was behaving differently from Perl in the way it recognized POSIX
750        character classes. PCRE was not treating the sequence [:...:] as a
751        character class unless the ... were all letters. Perl, however, seems to
752        allow any characters between [: and :], though of course it rejects as
753        unknown any "names" that contain non-letters, because all the known class
754        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
755        for example, whereas PCRE did not - it did not recognize a POSIX character
756        class. This seemed a bit dangerous, so the code has been changed to be
757        closer to Perl. The behaviour is not identical to Perl, because PCRE will
758        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
759        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
760        Perl does, and where it didn't before.
761    
762    24. Rewrite so as to remove the single use of %n from pcregrep because in some
763        Windows environments %n is disabled by default.
764    
765    
766    Version 7.4 21-Sep-07
767    ---------------------
768    
769    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
770        means that a class such as [\s] counted as "explicit reference to CR or
771        LF". That isn't really right - the whole point of the change was to try to
772        help when there was an actual mention of one of the two characters. So now
773        the change happens only if \r or \n (or a literal CR or LF) character is
774        encountered.
775    
776    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
777        of both had grown to the point where there were only 3 bits left.
778        Fortunately, there was spare space in the data structure, and so I have
779        moved the internal flags into a new 16-bit field to free up more option
780        bits.
781    
782    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
783        but did not set the internal JCHANGED flag - either of these is enough to
784        control the way the "get" function works - but the PCRE_INFO_JCHANGED
785        facility is supposed to tell if (?J) was ever used, so now (?J) at the
786        start sets both bits.
787    
788    4.  Added options (at build time, compile time, exec time) to change \R from
789        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
790    
791    5.  doc/pcresyntax.html was missing from the distribution.
792    
793    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
794        compatibility, even though it is no longer used.
795    
796    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
797        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
798        windows.h file is present (where different names are used). [This was
799        reversed later after testing - see 16 below.]
800    
801    8.  Changed all #include <config.h> to #include "config.h". There were also
802        some further <pcre.h> cases that I changed to "pcre.h".
803    
804    9.  When pcregrep was used with the --colour option, it missed the line ending
805        sequence off the lines that it output.
806    
807    10. It was pointed out to me that arrays of string pointers cause lots of
808        relocations when a shared library is dynamically loaded. A technique of
809        using a single long string with a table of offsets can drastically reduce
810        these. I have refactored PCRE in four places to do this. The result is
811        dramatic:
812    
813          Originally:                          290
814          After changing UCP table:            187
815          After changing error message table:   43
816          After changing table of "verbs"       36
817          After changing table of Posix names   22
818    
819        Thanks to the folks working on Gregex for glib for this insight.
820    
821    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
822        unicode-properties was also set.
823    
824    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
825    
826    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
827        checked only for CRLF.
828    
829    14. Added casts to pcretest.c to avoid compiler warnings.
830    
831    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
832    
833    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
834        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
835        entirely. This removes changes made in 7 above.
836    
837    17. The CMake files have been updated, and there is now more information about
838        building with CMake in the NON-UNIX-USE document.
839    
840    
841    Version 7.3 28-Aug-07
842    ---------------------
843    
844     1. In the rejigging of the build system that eventually resulted in 7.1, the
845        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
846        brackets there is not right, since it causes compilers to look for an
847        installed pcre.h, not the version that is in the source that is being
848        compiled (which of course may be different). I have changed it back to:
849    
850          #include "pcre.h"
851    
852        I have a vague recollection that the change was concerned with compiling in
853        different directories, but in the new build system, that is taken care of
854        by the VPATH setting the Makefile.
855    
856     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
857        when the subject happened to end in the byte 0x85 (e.g. if the last
858        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
859        characters but of course it shouldn't be taken as a newline when it is part
860        of another character. The bug was that, for an unlimited repeat of . in
861        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
862        characters when looking for a newline.
863    
864     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
865    
866     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
867        in debug output.
868    
869     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
870        long printing in the pcrecpp unittest when running under MinGW.
871    
872     6. ESC_K was left out of the EBCDIC table.
873    
874     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
875        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
876        limit also applies to "virtual nesting" when a pattern is recursive, and in
877        this case 1000 isn't so big. I have been able to remove this limit at the
878        expense of backing off one optimization in certain circumstances. Normally,
879        when pcre_exec() would call its internal match() function recursively and
880        immediately return the result unconditionally, it uses a "tail recursion"
881        feature to save stack. However, when a subpattern that can match an empty
882        string has an unlimited repetition quantifier, it no longer makes this
883        optimization. That gives it a stack frame in which to save the data for
884        checking that an empty string has been matched. Previously this was taken
885        from the 1000-entry workspace that had been reserved. So now there is no
886        explicit limit, but more stack is used.
887    
888     8. Applied Daniel's patches to solve problems with the import/export magic
889        syntax that is required for Windows, and which was going wrong for the
890        pcreposix and pcrecpp parts of the library. These were overlooked when this
891        problem was solved for the main library.
892    
893     9. There were some crude static tests to avoid integer overflow when computing
894        the size of patterns that contain repeated groups with explicit upper
895        limits. As the maximum quantifier is 65535, the maximum group length was
896        set at 30,000 so that the product of these two numbers did not overflow a
897        32-bit integer. However, it turns out that people want to use groups that
898        are longer than 30,000 bytes (though not repeat them that many times).
899        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
900        made it possible to implement the integer overflow checks in a much more
901        dynamic way, which I have now done. The artificial limitation on group
902        length has been removed - we now have only the limit on the total length of
903        the compiled pattern, which depends on the LINK_SIZE setting.
904    
905    10. Fixed a bug in the documentation for get/copy named substring when
906        duplicate names are permitted. If none of the named substrings are set, the
907        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
908        empty string.
909    
910    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
911        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
912        because the ] is interpreted as the first data character and the
913        terminating ] is not found. PCRE has been made compatible with Perl in this
914        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
915        cause memory overwriting.
916    
917    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
918        string has been matched (to stop an infinite loop). It was not recognizing
919        a conditional subpattern that could match an empty string if that
920        subpattern was within another subpattern. For example, it looped when
921        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
922        condition was not nested. This bug has been fixed.
923    
924    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
925        past the start of the subject in the presence of bytes with the top bit
926        set, for example "\x8aBCD".
927    
928    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
929        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
930    
931    14. Optimized (?!) to (*FAIL).
932    
933    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
934        This restricts code points to be within the range 0 to 0x10FFFF, excluding
935        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
936        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
937        does: it's just the validity check that is more restrictive.
938    
939    16. Inserted checks for integer overflows during escape sequence (backslash)
940        processing, and also fixed erroneous offset values for syntax errors during
941        backslash processing.
942    
943    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
944        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
945    
946    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
947        caused an overrun.
948    
949    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
950        something other than just ASCII characters) inside a group that had an
951        unlimited repeat caused a loop at compile time (while checking to see
952        whether the group could match an empty string).
953    
954    20. Debugging a pattern containing \p or \P could cause a crash. For example,
955        [\P{Any}] did so. (Error in the code for printing property names.)
956    
957    21. An orphan \E inside a character class could cause a crash.
958    
959    22. A repeated capturing bracket such as (A)? could cause a wild memory
960        reference during compilation.
961    
962    23. There are several functions in pcre_compile() that scan along a compiled
963        expression for various reasons (e.g. to see if it's fixed length for look
964        behind). There were bugs in these functions when a repeated \p or \P was
965        present in the pattern. These operators have additional parameters compared
966        with \d, etc, and these were not being taken into account when moving along
967        the compiled data. Specifically:
968    
969        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
970            length.
971    
972        (b) An item such as \pL+ within a repeated group could cause crashes or
973            loops.
974    
975        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
976            "reference to non-existent subpattern" error.
977    
978        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
979    
980    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
981        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
982    
983    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
984    
985    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
986        character were causing crashes (broken optimization).
987    
988    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
989        \p or \P) caused a compile-time loop.
990    
991    28. More problems have arisen in unanchored patterns when CRLF is a valid line
992        break. For example, the unstudied pattern [\r\n]A does not match the string
993        "\r\nA" because change 7.0/46 below moves the current point on by two
994        characters after failing to match at the start. However, the pattern \nA
995        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
996        the same is true. There doesn't seem any very clean way out of this, but
997        what I have chosen to do makes the common cases work: PCRE now takes note
998        of whether there can be an explicit match for \r or \n anywhere in the
999        pattern, and if so, 7.0/46 no longer applies. As part of this change,
1000        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1001        pattern has explicit CR or LF references.
1002    
1003    29. Added (*CR) etc for changing newline setting at start of pattern.
1004    
1005    
1006    Version 7.2 19-Jun-07
1007    ---------------------
1008    
1009     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
1010        which is apparently normally available under Windows.
1011    
1012     2. Re-jig the pcregrep tests with different newline settings in an attempt
1013        to make them independent of the local environment's newline setting.
1014    
1015     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
1016    
1017     4. Some of the "internals" tests were previously cut out when the link size
1018        was not 2, because the output contained actual offsets. The recent new
1019        "Z" feature of pcretest means that these can be cut out, making the tests
1020        usable with all link sizes.
1021    
1022     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
1023        stack recursion. This gives a massive performance boost under BSD, but just
1024        a small improvement under Linux. However, it saves one field in the frame
1025        in all cases.
1026    
1027     6. Added more features from the forthcoming Perl 5.10:
1028    
1029        (a) (?-n) (where n is a string of digits) is a relative subroutine or
1030            recursion call. It refers to the nth most recently opened parentheses.
1031    
1032        (b) (?+n) is also a relative subroutine call; it refers to the nth next
1033            to be opened parentheses.
1034    
1035        (c) Conditions that refer to capturing parentheses can be specified
1036            relatively, for example, (?(-2)... or (?(+3)...
1037    
1038        (d) \K resets the start of the current match so that everything before
1039            is not part of it.
1040    
1041        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1042    
1043        (f) \g{name} is another synonym - part of Perl 5.10's unification of
1044            reference syntax.
1045    
1046        (g) (?| introduces a group in which the numbering of parentheses in each
1047            alternative starts with the same number.
1048    
1049        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
1050    
1051     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
1052        PCRE_INFO_JCHANGED.
1053    
1054     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
1055        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
1056        for detecting groups that can match an empty string.
1057    
1058     9. A pattern with a very large number of alternatives (more than several
1059        hundred) was running out of internal workspace during the pre-compile
1060        phase, where pcre_compile() figures out how much memory will be needed. A
1061        bit of new cunning has reduced the workspace needed for groups with
1062        alternatives. The 1000-alternative test pattern now uses 12 bytes of
1063        workspace instead of running out of the 4096 that are available.
1064    
1065    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
1066    
1067    11. Applied patch from Google to remove an optimization that didn't quite work.
1068        The report of the bug said:
1069    
1070          pcrecpp::RE("a*").FullMatch("aaa") matches, while
1071          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
1072          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
1073    
1074    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
1075        it matched the wrong number of bytes.
1076    
1077    
1078    Version 7.1 24-Apr-07
1079    ---------------------
1080    
1081     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
1082        that is more "standard", making use of automake and other Autotools. There
1083        is some re-arrangement of the files and adjustment of comments consequent
1084        on this.
1085    
1086     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
1087        for recursive directory scanning broke on some systems because the files
1088        are not scanned in any specific order and on different systems the order
1089        was different. A call to "sort" has been inserted into RunGrepTest for the
1090        approprate test as a short-term fix. In the longer term there may be an
1091        alternative.
1092    
1093     3. I had an email from Eric Raymond about problems translating some of PCRE's
1094        man pages to HTML (despite the fact that I distribute HTML pages, some
1095        people do their own conversions for various reasons). The problems
1096        concerned the use of low-level troff macros .br and .in. I have therefore
1097        removed all such uses from the man pages (some were redundant, some could
1098        be replaced by .nf/.fi pairs). The 132html script that I use to generate
1099        HTML has been updated to handle .nf/.fi and to complain if it encounters
1100        .br or .in.
1101    
1102     4. Updated comments in configure.ac that get placed in config.h.in and also
1103        arranged for config.h to be included in the distribution, with the name
1104        config.h.generic, for the benefit of those who have to compile without
1105        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
1106    
1107     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
1108        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
1109        makevp.bat for the current PCRE, using the additional files
1110        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
1111    
1112     6. A Windows user reported a minor discrepancy with test 2, which turned out
1113        to be caused by a trailing space on an input line that had got lost in his
1114        copy. The trailing space was an accident, so I've just removed it.
1115    
1116     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
1117        that is needed.
1118    
1119     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
1120        as "const" (a) because they are and (b) because it helps the PHP
1121        maintainers who have recently made a script to detect big data structures
1122        in the php code that should be moved to the .rodata section. I remembered
1123        to update Builducptable as well, so it won't revert if ucptable.h is ever
1124        re-created.
1125    
1126     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
1127        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
1128        order to be able to cut out the UTF-8 tables in the latter when UTF-8
1129        support is not required. This saves 1.5-2K of code, which is important in
1130        some applications.
1131    
1132        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
1133        so as not to refer to the tables, even though these functions will never be
1134        called when UTF-8 support is disabled. Otherwise there are problems with a
1135        shared library.
1136    
1137    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
1138    
1139        (a) It was defining its arguments as char * instead of void *.
1140    
1141        (b) It was assuming that all moves were upwards in memory; this was true
1142            a long time ago when I wrote it, but is no longer the case.
1143    
1144        The emulated memove() is provided for those environments that have neither
1145        memmove() nor bcopy(). I didn't think anyone used it these days, but that
1146        is clearly not the case, as these two bugs were recently reported.
1147    
1148    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
1149        and Detrail to create the HTML documentation, the .txt form of the man
1150        pages, and it removes trailing spaces from listed files. It also creates
1151        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
1152        case, it wraps all the #defines with #ifndefs. This script should be run
1153        before "make dist".
1154    
1155    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1156        with Unicode property support.
1157    
1158        (a) For a maximizing quantifier, if the two different cases of the
1159            character were of different lengths in their UTF-8 codings (there are
1160            some cases like this - I found 11), and the matching function had to
1161            back up over a mixture of the two cases, it incorrectly assumed they
1162            were both the same length.
1163    
1164        (b) When PCRE was configured to use the heap rather than the stack for
1165            recursion during matching, it was not correctly preserving the data for
1166            the other case of a UTF-8 character when checking ahead for a match
1167            while processing a minimizing repeat. If the check also involved
1168            matching a wide character, but failed, corruption could cause an
1169            erroneous result when trying to check for a repeat of the original
1170            character.
1171    
1172    13. Some tidying changes to the testing mechanism:
1173    
1174        (a) The RunTest script now detects the internal link size and whether there
1175            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1176            values substituted by "configure". (The RunGrepTest script already did
1177            this for UTF-8.) The configure.ac script no longer substitutes the
1178            relevant variables.
1179    
1180        (b) The debugging options /B and /D in pcretest show the compiled bytecode
1181            with length and offset values. This means that the output is different
1182            for different internal link sizes. Test 2 is skipped for link sizes
1183            other than 2 because of this, bypassing the problem. Unfortunately,
1184            there was also a test in test 3 (the locale tests) that used /B and
1185            failed for link sizes other than 2. Rather than cut the whole test out,
1186            I have added a new /Z option to pcretest that replaces the length and
1187            offset values with spaces. This is now used to make test 3 independent
1188            of link size. (Test 2 will be tidied up later.)
1189    
1190    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1191        segmentation fault instead of returning the appropriate error message.
1192    
1193    15. In multiline mode when the newline sequence was set to "any", the pattern
1194        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1195        This doesn't seem right; it now treats the CRLF combination as the line
1196        ending, and so does not match in that case. It's only a pattern such as ^$
1197        that would hit this one: something like ^ABC$ would have failed after \r
1198        and then tried again after \r\n.
1199    
1200    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1201        in an attempt to make files that differ only in their line terminators
1202        compare equal. This works on Linux.
1203    
1204    17. Under certain error circumstances pcregrep might try to free random memory
1205        as it exited. This is now fixed, thanks to valgrind.
1206    
1207    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1208        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1209        was because its rules for how to advance for /g after matching an empty
1210        string at the end of a line did not allow for this case. They now check for
1211        it specially.
1212    
1213    20. pcretest is supposed to handle patterns and data of any length, by
1214        extending its buffers when necessary. It was getting this wrong when the
1215        buffer for a data line had to be extended.
1216    
1217    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1218        CRLF as a newline sequence.
1219    
1220    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1221        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1222        I have nevertheless tidied it up.
1223    
1224    23. Added some casts to kill warnings from HP-UX ia64 compiler.
1225    
1226    24. Added a man page for pcre-config.
1227    
1228    
1229    Version 7.0 19-Dec-06
1230    ---------------------
1231    
1232     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1233        moving to gcc 4.1.1.
1234    
1235     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1236        sys/time.h, which is documented as needed for this function. It doesn't
1237        seem to matter on Linux, but it showed up on some releases of OS X.
1238    
1239     3. It seems that there are systems where bytes whose values are greater than
1240        127 match isprint() in the "C" locale. The "C" locale should be the
1241        default when a C program starts up. In most systems, only ASCII printing
1242        characters match isprint(). This difference caused the output from pcretest
1243        to vary, making some of the tests fail. I have changed pcretest so that:
1244    
1245        (a) When it is outputting text in the compiled version of a pattern, bytes
1246            other than 32-126 are always shown as hex escapes.
1247    
1248        (b) When it is outputting text that is a matched part of a subject string,
1249            it does the same, unless a different locale has been set for the match
1250            (using the /L modifier). In this case, it uses isprint() to decide.
1251    
1252     4. Fixed a major bug that caused incorrect computation of the amount of memory
1253        required for a compiled pattern when options that changed within the
1254        pattern affected the logic of the preliminary scan that determines the
1255        length. The relevant options are -x, and -i in UTF-8 mode. The result was
1256        that the computed length was too small. The symptoms of this bug were
1257        either the PCRE error "internal error: code overflow" from pcre_compile(),
1258        or a glibc crash with a message such as "pcretest: free(): invalid next
1259        size (fast)". Examples of patterns that provoked this bug (shown in
1260        pcretest format) are:
1261    
1262          /(?-x: )/x
1263          /(?x)(?-x: \s*#\s*)/
1264          /((?i)[\x{c0}])/8
1265          /(?i:[\x{c0}])/8
1266    
1267        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1268        is now done differently.
1269    
1270     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1271        wrapper classes; (b) implement a new function in the C++ scanner that is
1272        more efficient than the old way of doing things because it avoids levels of
1273        recursion in the regex matching; (c) add a paragraph to the documentation
1274        for the FullMatch() function.
1275    
1276     6. The escape sequence \n was being treated as whatever was defined as
1277        "newline". Not only was this contrary to the documentation, which states
1278        that \n is character 10 (hex 0A), but it also went horribly wrong when
1279        "newline" was defined as CRLF. This has been fixed.
1280    
1281     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1282        was being set to -1 for the "end of line" case (supposedly a value that no
1283        character can have). Though this value is never used (the check for end of
1284        line is "zero bytes in current character"), it caused compiler complaints.
1285        I've changed it to 0xffffffff.
1286    
1287     8. In pcre_version.c, the version string was being built by a sequence of
1288        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1289        string (as it is for production releases) called a macro with an empty
1290        argument. The C standard says the result of this is undefined. The gcc
1291        compiler treats it as an empty string (which was what was wanted) but it is
1292        reported that Visual C gives an error. The source has been hacked around to
1293        avoid this problem.
1294    
1295     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1296        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1297        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1298        of them did).
1299    
1300    10. Originally, pcretest opened its input and output without "b"; then I was
1301        told that "b" was needed in some environments, so it was added for release
1302        5.0 to both the input and output. (It makes no difference on Unix-like
1303        systems.) Later I was told that it is wrong for the input on Windows. I've
1304        now abstracted the modes into two macros, to make it easier to fiddle with
1305        them, and removed "b" from the input mode under Windows.
1306    
1307    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1308    
1309    12. Added -help and --help to pcretest as an official way of being reminded
1310        of the options.
1311    
1312    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1313        and pcrecpp.cc because they annoy compilers at high warning levels.
1314    
1315    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1316    
1317    15. Fixed an occurrence of == in configure.ac that should have been = (shell
1318        scripts are not C programs :-) and which was not noticed because it works
1319        on Linux.
1320    
1321    16. pcretest is supposed to handle any length of pattern and data line (as one
1322        line or as a continued sequence of lines) by extending its input buffer if
1323        necessary. This feature was broken for very long pattern lines, leading to
1324        a string of junk being passed to pcre_compile() if the pattern was longer
1325        than about 50K.
1326    
1327    17. I have done a major re-factoring of the way pcre_compile() computes the
1328        amount of memory needed for a compiled pattern. Previously, there was code
1329        that made a preliminary scan of the pattern in order to do this. That was
1330        OK when PCRE was new, but as the facilities have expanded, it has become
1331        harder and harder to keep it in step with the real compile phase, and there
1332        have been a number of bugs (see for example, 4 above). I have now found a
1333        cunning way of running the real compile function in a "fake" mode that
1334        enables it to compute how much memory it would need, while actually only
1335        ever using a few hundred bytes of working memory and without too many
1336        tests of the mode. This should make future maintenance and development
1337        easier. A side effect of this work is that the limit of 200 on the nesting
1338        depth of parentheses has been removed (though this was never a serious
1339        limitation, I suspect). However, there is a downside: pcre_compile() now
1340        runs more slowly than before (30% or more, depending on the pattern). I
1341        hope this isn't a big issue. There is no effect on runtime performance.
1342    
1343    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1344        newline (only possible for the last line of a file) and it was a
1345        pattern that set a locale (followed by /Lsomething), pcretest crashed.
1346    
1347    19. Added additional timing features to pcretest. (1) The -tm option now times
1348        matching only, not compiling. (2) Both -t and -tm can be followed, as a
1349        separate command line item, by a number that specifies the number of
1350        repeats to use when timing. The default is 50000; this gives better
1351        precision, but takes uncomfortably long for very large patterns.
1352    
1353    20. Extended pcre_study() to be more clever in cases where a branch of a
1354        subpattern has no definite first character. For example, (a*|b*)[cd] would
1355        previously give no result from pcre_study(). Now it recognizes that the
1356        first character must be a, b, c, or d.
1357    
1358    21. There was an incorrect error "recursive call could loop indefinitely" if
1359        a subpattern (or the entire pattern) that was being tested for matching an
1360        empty string contained only one non-empty item after a nested subpattern.
1361        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1362        incorrectly, because the \d was being skipped in the check.
1363    
1364    22. The pcretest program now has a new pattern option /B and a command line
1365        option -b, which is equivalent to adding /B to every pattern. This causes
1366        it to show the compiled bytecode, without the additional information that
1367        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1368        is the same as /B/I).
1369    
1370    23. A new optimization is now able automatically to treat some sequences such
1371        as a*b as a*+b. More specifically, if something simple (such as a character
1372        or a simple class like \d) has an unlimited quantifier, and is followed by
1373        something that cannot possibly match the quantified thing, the quantifier
1374        is automatically "possessified".
1375    
1376    24. A recursive reference to a subpattern whose number was greater than 39
1377        went wrong under certain circumstances in UTF-8 mode. This bug could also
1378        have affected the operation of pcre_study().
1379    
1380    25. Realized that a little bit of performance could be had by replacing
1381        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1382    
1383    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1384    
1385    27. Possessive quantifiers such as a++ were previously implemented by turning
1386        them into atomic groups such as ($>a+). Now they have their own opcodes,
1387        which improves performance. This includes the automatically created ones
1388        from 23 above.
1389    
1390    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1391        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1392        the first matched character to be a colon. This applied both to named and
1393        numbered groups.
1394    
1395    29. The ucpinternal.h header file was missing its idempotency #ifdef.
1396    
1397    30. I was sent a "project" file called libpcre.a.dev which I understand makes
1398        building PCRE on Windows easier, so I have included it in the distribution.
1399    
1400    31. There is now a check in pcretest against a ridiculously large number being
1401        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1402        loop, the loop is abandoned.
1403    
1404    32. Forward references to subpatterns in conditions such as (?(2)...) where
1405        subpattern 2 is defined later cause pcre_compile() to search forwards in
1406        the pattern for the relevant set of parentheses. This search went wrong
1407        when there were unescaped parentheses in a character class, parentheses
1408        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1409    
1410    33. "Subroutine" calls and backreferences were previously restricted to
1411        referencing subpatterns earlier in the regex. This restriction has now
1412        been removed.
1413    
1414    34. Added a number of extra features that are going to be in Perl 5.10. On the
1415        whole, these are just syntactic alternatives for features that PCRE had
1416        previously implemented using the Python syntax or my own invention. The
1417        other formats are all retained for compatibility.
1418    
1419        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1420            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1421            also .NET compatible.
1422    
1423        (b) A recursion or subroutine call to a named group can now be defined as
1424            (?&name) as well as (?P>name).
1425    
1426        (c) A backreference to a named group can now be defined as \k<name> or
1427            \k'name' as well as (?P=name). The new forms, as well as being in Perl
1428            5.10, are also .NET compatible.
1429    
1430        (d) A conditional reference to a named group can now use the syntax
1431            (?(<name>) or (?('name') as well as (?(name).
1432    
1433        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1434            groups (named and numbered) that are never evaluated inline, but can be
1435            called as "subroutines" from elsewhere. In effect, the DEFINE condition
1436            is always false. There may be only one alternative in such a group.
1437    
1438        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1439            as the simple (?(R). The condition is true only if the most recent
1440            recursion is that of the given number or name. It does not search out
1441            through the entire recursion stack.
1442    
1443        (g) The escape \gN or \g{N} has been added, where N is a positive or
1444            negative number, specifying an absolute or relative reference.
1445    
1446    35. Tidied to get rid of some further signed/unsigned compiler warnings and
1447        some "unreachable code" warnings.
1448    
1449    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1450        things, this adds five new scripts.
1451    
1452    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1453        There were also incompatibilities regarding the handling of \Q..\E inside
1454        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1455        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1456    
1457    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1458        matches an empty string, and forcibly breaks the loop. There were bugs in
1459        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
1460        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
1461        separate and independent bugs (that affected different cases) have been
1462        fixed.
1463    
1464    39. Refactored the code to abolish the use of different opcodes for small
1465        capturing bracket numbers. This is a tidy that I avoided doing when I
1466        removed the limit on the number of capturing brackets for 3.5 back in 2001.
1467        The new approach is not only tidier, it makes it possible to reduce the
1468        memory needed to fix the previous bug (38).
1469    
1470    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1471        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1472        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1473        mode.
1474    
1475    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1476        report.
1477    
1478    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1479        copy construction and assignment in the C++ wrapper.
1480    
1481    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1482        couple of bugs that could have given wrong results in the "--newline=crlf"
1483        case.
1484    
1485    44. Added a number of casts and did some reorganization of signed/unsigned int
1486        variables following suggestions from Dair Grant. Also renamed the variable
1487        "this" as "item" because it is a C++ keyword.
1488    
1489    45. Arranged for dftables to add
1490    
1491          #include "pcre_internal.h"
1492    
1493        to pcre_chartables.c because without it, gcc 4.x may remove the array
1494        definition from the final binary if PCRE is built into a static library and
1495        dead code stripping is activated.
1496    
1497    46. For an unanchored pattern, if a match attempt fails at the start of a
1498        newline sequence, and the newline setting is CRLF or ANY, and the next two
1499        characters are CRLF, advance by two characters instead of one.
1500    
1501    
1502    Version 6.7 04-Jul-06
1503    ---------------------
1504    
1505     1. In order to handle tests when input lines are enormously long, pcretest has
1506        been re-factored so that it automatically extends its buffers when
1507        necessary. The code is crude, but this _is_ just a test program. The
1508        default size has been increased from 32K to 50K.
1509    
1510     2. The code in pcre_study() was using the value of the re argument before
1511        testing it for NULL. (Of course, in any sensible call of the function, it
1512        won't be NULL.)
1513    
1514     3. The memmove() emulation function in pcre_internal.h, which is used on
1515        systems that lack both memmove() and bcopy() - that is, hardly ever -
1516        was missing a "static" storage class specifier.
1517    
1518     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1519        containing an extended class (one that cannot be represented by a bitmap
1520        because it contains high-valued characters or Unicode property items, e.g.
1521        [\pZ]). Almost always one would set UTF-8 mode when processing such a
1522        pattern, but PCRE should not loop if you do not (it no longer does).
1523        [Detail: two cases were found: (a) a repeated subpattern containing an
1524        extended class; (b) a recursive reference to a subpattern that followed a
1525        previous extended class. It wasn't skipping over the extended class
1526        correctly when UTF-8 mode was not set.]
1527    
1528     5. A negated single-character class was not being recognized as fixed-length
1529        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1530        compile error "lookbehind assertion is not fixed length".
1531    
1532     6. The RunPerlTest auxiliary script was showing an unexpected difference
1533        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1534        write a Perl script that can interpret lines of an input file either as
1535        byte characters or as UTF-8, which is what "perltest" was being required to
1536        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1537        can't do is switch easily at run time between having the "use utf8;" pragma
1538        or not. In the end, I fudged it by using the RunPerlTest script to insert
1539        "use utf8;" explicitly for the UTF-8 tests.
1540    
1541     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1542        the end of the subject string, contrary to the documentation and to what
1543        Perl does. This was true of both matching functions. Now it matches only at
1544        the start of the subject and immediately after *internal* newlines.
1545    
1546     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1547        a pointer to an int instead of a pointer to an unsigned long int. This
1548        caused problems on 64-bit systems.
1549    
1550     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1551        instance of the 'standard' template library not being so standard".
1552    
1553    10. There was no check on the number of named subpatterns nor the maximum
1554        length of a subpattern name. The product of these values is used to compute
1555        the size of the memory block for a compiled pattern. By supplying a very
1556        long subpattern name and a large number of named subpatterns, the size
1557        computation could be caused to overflow. This is now prevented by limiting
1558        the length of names to 32 characters, and the number of named subpatterns
1559        to 10,000.
1560    
1561    11. Subpatterns that are repeated with specific counts have to be replicated in
1562        the compiled pattern. The size of memory for this was computed from the
1563        length of the subpattern and the repeat count. The latter is limited to
1564        65535, but there was no limit on the former, meaning that integer overflow
1565        could in principle occur. The compiled length of a repeated subpattern is
1566        now limited to 30,000 bytes in order to prevent this.
1567    
1568    12. Added the optional facility to have named substrings with the same name.
1569    
1570    13. Added the ability to use a named substring as a condition, using the
1571        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1572        are numbers (not recommended). Forward references are permitted.
1573    
1574    14. Added forward references in named backreferences (if you see what I mean).
1575    
1576    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1577        pattern could run off the end of the subject. For example, the pattern
1578        "(?s)(.{1,5})"8 did this with the subject "ab".
1579    
1580    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1581        PCRE_CASELESS was set when matching characters that were quantified with ?
1582        or *.
1583    
1584    17. A character class other than a single negated character that had a minimum
1585        but no maximum quantifier - for example [ab]{6,} - was not handled
1586        correctly by pce_dfa_exec(). It would match only one character.
1587    
1588    18. A valid (though odd) pattern that looked like a POSIX character
1589        class but used an invalid character after [ (for example [[,abc,]]) caused
1590        pcre_compile() to give the error "Failed: internal error: code overflow" or
1591        in some cases to crash with a glibc free() error. This could even happen if
1592        the pattern terminated after [[ but there just happened to be a sequence of
1593        letters, a binary zero, and a closing ] in the memory that followed.
1594    
1595    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1596        over the years. Originally (before any Unicode support), just the bottom 8
1597        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1598        output from "man perlunicode" includes this:
1599    
1600          The regular expression compiler produces polymorphic opcodes.  That
1601          is, the pattern adapts to the data and automatically switches to
1602          the Unicode character scheme when presented with Unicode data--or
1603          instead uses a traditional byte scheme when presented with byte
1604          data.
1605    
1606        Sadly, a wide octal escape does not cause a switch, and in a string with
1607        no other multibyte characters, these octal escapes are treated as before.
1608        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
1609        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1610        Unicode string.
1611    
1612        I have not perpetrated such confusion in PCRE. Up till now, it took just
1613        the bottom 8 bits, as in old Perl. I have now made octal escapes with
1614        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1615        translate to the appropriate multibyte character.
1616    
1617    29. Applied some refactoring to reduce the number of warnings from Microsoft
1618        and Borland compilers. This has included removing the fudge introduced
1619        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1620        a warning about an unused variable.
1621    
1622    21. PCRE has not included VT (character 0x0b) in the set of whitespace
1623        characters since release 4.0, because Perl (from release 5.004) does not.
1624        [Or at least, is documented not to: some releases seem to be in conflict
1625        with the documentation.] However, when a pattern was studied with
1626        pcre_study() and all its branches started with \s, PCRE still included VT
1627        as a possible starting character. Of course, this did no harm; it just
1628        caused an unnecessary match attempt.
1629    
1630    22. Removed a now-redundant internal flag bit that recorded the fact that case
1631        dependency changed within the pattern. This was once needed for "required
1632        byte" processing, but is no longer used. This recovers a now-scarce options
1633        bit. Also moved the least significant internal flag bit to the most-
1634        significant bit of the word, which was not previously used (hangover from
1635        the days when it was an int rather than a uint) to free up another bit for
1636        the future.
1637    
1638    23. Added support for CRLF line endings as well as CR and LF. As well as the
1639        default being selectable at build time, it can now be changed at runtime
1640        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1641        specify that it is scanning data with non-default line endings.
1642    
1643    24. Changed the definition of CXXLINK to make it agree with the definition of
1644        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1645    
1646    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1647        recursions. This makes a big different to stack usage for some patterns.
1648    
1649    26. If a subpattern containing a named recursion or subroutine reference such
1650        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1651        the space required for the compiled pattern went wrong and gave too small a
1652        value. Depending on the environment, this could lead to "Failed: internal
1653        error: code overflow at offset 49" or "glibc detected double free or
1654        corruption" errors.
1655    
1656    27. Applied patches from Google (a) to support the new newline modes and (b) to
1657        advance over multibyte UTF-8 characters in GlobalReplace.
1658    
1659    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1660        difference for some implementation of PCRE in some Windows version.
1661    
1662    29. Added some extra testing facilities to pcretest:
1663    
1664        \q<number>   in a data line sets the "match limit" value
1665        \Q<number>   in a data line sets the "match recursion limt" value
1666        -S <number>  sets the stack size, where <number> is in megabytes
1667    
1668        The -S option isn't available for Windows.
1669    
1670    
1671    Version 6.6 06-Feb-06
1672    ---------------------
1673    
1674     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1675        in pcreposix.h. I have copied the definition from pcre.h.
1676    
1677     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1678        because pcre.h is no longer a built file.
1679    
1680     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1681        not normally included in the compiled code.
1682    
1683    
1684    Version 6.5 01-Feb-06
1685    ---------------------
1686    
1687     1. When using the partial match feature with pcre_dfa_exec(), it was not
1688        anchoring the second and subsequent partial matches at the new starting
1689        point. This could lead to incorrect results. For example, with the pattern
1690        /1234/, partially matching against "123" and then "a4" gave a match.
1691    
1692     2. Changes to pcregrep:
1693    
1694        (a) All non-match returns from pcre_exec() were being treated as failures
1695            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1696            error message is output. Some extra information is given for the
1697            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
1698            probably the only errors that are likely to be caused by users (by
1699            specifying a regex that has nested indefinite repeats, for instance).
1700            If there are more than 20 of these errors, pcregrep is abandoned.
1701    
1702        (b) A binary zero was treated as data while matching, but terminated the
1703            output line if it was written out. This has been fixed: binary zeroes
1704            are now no different to any other data bytes.
1705    
1706        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1707            used to set a locale for matching. The --locale=xxxx long option has
1708            been added (no short equivalent) to specify a locale explicitly on the
1709            pcregrep command, overriding the environment variables.
1710    
1711        (d) When -B was used with -n, some line numbers in the output were one less
1712            than they should have been.
1713    
1714        (e) Added the -o (--only-matching) option.
1715    
1716        (f) If -A or -C was used with -c (count only), some lines of context were
1717            accidentally printed for the final match.
1718    
1719        (g) Added the -H (--with-filename) option.
1720    
1721        (h) The combination of options -rh failed to suppress file names for files
1722            that were found from directory arguments.
1723    
1724        (i) Added the -D (--devices) and -d (--directories) options.
1725    
1726        (j) Added the -F (--fixed-strings) option.
1727    
1728        (k) Allow "-" to be used as a file name for -f as well as for a data file.
1729    
1730        (l) Added the --colo(u)r option.
1731    
1732        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1733            is not present by default.
1734    
1735     3. A nasty bug was discovered in the handling of recursive patterns, that is,
1736        items such as (?R) or (?1), when the recursion could match a number of
1737        alternatives. If it matched one of the alternatives, but subsequently,
1738        outside the recursion, there was a failure, the code tried to back up into
1739        the recursion. However, because of the way PCRE is implemented, this is not
1740        possible, and the result was an incorrect result from the match.
1741    
1742        In order to prevent this happening, the specification of recursion has
1743        been changed so that all such subpatterns are automatically treated as
1744        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1745    
1746     4. I had overlooked the fact that, in some locales, there are characters for
1747        which isalpha() is true but neither isupper() nor islower() are true. In
1748        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1749        and ordfeminine) are like this. This affected the treatment of \w and \W
1750        when they appeared in character classes, but not when they appeared outside
1751        a character class. The bit map for "word" characters is now created
1752        separately from the results of isalnum() instead of just taking it from the
1753        upper, lower, and digit maps. (Plus the underscore character, of course.)
1754    
1755     5. The above bug also affected the handling of POSIX character classes such as
1756        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1757        permanent tables. Instead, the bit maps for such a class were previously
1758        created as the appropriate unions of the upper, lower, and digit bitmaps.
1759        Now they are created by subtraction from the [[:word:]] class, which has
1760        its own bitmap.
1761    
1762     6. The [[:blank:]] character class matches horizontal, but not vertical space.
1763        It is created by subtracting the vertical space characters (\x09, \x0a,
1764        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1765        subtraction was done in the overall bitmap for a character class, meaning
1766        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1767        be recognized. This bug has been fixed.
1768    
1769     7. Patches from the folks at Google:
1770    
1771          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1772          real life, but is still worth protecting against".
1773    
1774          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1775          regular expressions".
1776    
1777          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1778          have it.
1779    
1780          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1781          "configure" and the latter not, in order to fix a problem somebody had
1782          with compiling the Arg class on HP-UX.
1783    
1784          (e) Improve the error-handling of the C++ wrapper a little bit.
1785    
1786          (f) New tests for checking recursion limiting.
1787    
1788     8. The pcre_memmove() function, which is used only if the environment does not
1789        have a standard memmove() function (and is therefore rarely compiled),
1790        contained two bugs: (a) use of int instead of size_t, and (b) it was not
1791        returning a result (though PCRE never actually uses the result).
1792    
1793     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1794        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1795        returned instead of calling malloc() with an overflowing number that would
1796        most likely cause subsequent chaos.
1797    
1798    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1799    
1800    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1801        with this option is matched, the nmatch and pmatch options of regexec() are
1802        ignored.
1803    
1804    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1805        provided in case anyone wants to the the POSIX interface with UTF-8
1806        strings.
1807    
1808    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1809        C++ linking (needed for some HP-UX environments).
1810    
1811    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1812        (unused parameter) and in the pcre_printint() function (omitted "default"
1813        switch label when the default is to do nothing).
1814    
1815    15. Added some code to make it possible, when PCRE is compiled as a C++
1816        library, to replace subject pointers for pcre_exec() with a smart pointer
1817        class, thus making it possible to process discontinuous strings.
1818    
1819    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1820        much the same function. They were added by different people who were trying
1821        to make PCRE easy to compile on non-Unix systems. It has been suggested
1822        that PCRE_EXPORT be abolished now that there is more automatic apparatus
1823        for compiling on Windows systems. I have therefore replaced it with
1824        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1825        defaults to "extern" for C or "extern C" for C++, which works fine on
1826        Unix-like systems. It is now possible to override the value of PCRE_DATA_
1827        SCOPE with something explicit in config.h. In addition:
1828    
1829        (a) pcreposix.h still had just "extern" instead of either of these macros;
1830            I have replaced it with PCRE_DATA_SCOPE.
1831    
1832        (b) Functions such as _pcre_xclass(), which are internal to the library,
1833            but external in the C sense, all had PCRE_EXPORT in their definitions.
1834            This is apparently wrong for the Windows case, so I have removed it.
1835            (It makes no difference on Unix-like systems.)
1836    
1837    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1838        of recursive calls to match(). This is different to MATCH_LIMIT because
1839        that limits the total number of calls to match(), not all of which increase
1840        the depth of recursion. Limiting the recursion depth limits the amount of
1841        stack (or heap if NO_RECURSE is set) that is used. The default can be set
1842        when PCRE is compiled, and changed at run time. A patch from Google adds
1843        this functionality to the C++ interface.
1844    
1845    18. Changes to the handling of Unicode character properties:
1846    
1847        (a) Updated the table to Unicode 4.1.0.
1848    
1849        (b) Recognize characters that are not in the table as "Cn" (undefined).
1850    
1851        (c) I revised the way the table is implemented to a much improved format
1852            which includes recognition of ranges. It now supports the ranges that
1853            are defined in UnicodeData.txt, and it also amalgamates other
1854            characters into ranges. This has reduced the number of entries in the
1855            table from around 16,000 to around 3,000, thus reducing its size
1856            considerably. I realized I did not need to use a tree structure after
1857            all - a binary chop search is just as efficient. Having reduced the
1858            number of entries, I extended their size from 6 bytes to 8 bytes to
1859            allow for more data.
1860    
1861        (d) Added support for Unicode script names via properties such as \p{Han}.
1862    
1863    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1864        matching that character.
1865    
1866    20. When matching a repeated Unicode property with a minimum greater than zero,
1867        (for example \pL{2,}), PCRE could look past the end of the subject if it
1868        reached it while seeking the minimum number of characters. This could
1869        happen only if some of the characters were more than one byte long, because
1870        there is a check for at least the minimum number of bytes.
1871    
1872    21. Refactored the implementation of \p and \P so as to be more general, to
1873        allow for more different types of property in future. This has changed the
1874        compiled form incompatibly. Anybody with saved compiled patterns that use
1875        \p or \P will have to recompile them.
1876    
1877    22. Added "Any" and "L&" to the supported property types.
1878    
1879    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1880        but give a compile time error if the value is greater than 0xff.
1881    
1882    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1883        accidentally not being installed or uninstalled.
1884    
1885    25. The pcre.h file was built from pcre.h.in, but the only changes that were
1886        made were to insert the current release number. This seemed silly, because
1887        it made things harder for people building PCRE on systems that don't run
1888        "configure". I have turned pcre.h into a distributed file, no longer built
1889        by "configure", with the version identification directly included. There is
1890        no longer a pcre.h.in file.
1891    
1892        However, this change necessitated a change to the pcre-config script as
1893        well. It is built from pcre-config.in, and one of the substitutions was the
1894        release number. I have updated configure.ac so that ./configure now finds
1895        the release number by grepping pcre.h.
1896    
1897    26. Added the ability to run the tests under valgrind.
1898    
1899    
1900    Version 6.4 05-Sep-05
1901    ---------------------
1902    
1903     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1904        "--" to be printed when multiple files were scanned, even when none of the
1905        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1906        consider it to be a bug, and have restored the previous behaviour.
1907    
1908     2. A couple of code tidies to get rid of compiler warnings.
1909    
1910     3. The pcretest program used to cheat by referring to symbols in the library
1911        whose names begin with _pcre_. These are internal symbols that are not
1912        really supposed to be visible externally, and in some environments it is
1913        possible to suppress them. The cheating is now confined to including
1914        certain files from the library's source, which is a bit cleaner.
1915    
1916     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1917        file's purpose clearer.
1918    
1919     5. Reorganized pcre_ucp_findchar().
1920    
1921    
1922    Version 6.3 15-Aug-05
1923    ---------------------
1924    
1925     1. The file libpcre.pc.in did not have general read permission in the tarball.
1926    
1927     2. There were some problems when building without C++ support:
1928    
1929        (a) If C++ support was not built, "make install" and "make test" still
1930            tried to test it.
1931    
1932        (b) There were problems when the value of CXX was explicitly set. Some
1933            changes have been made to try to fix these, and ...
1934    
1935        (c) --disable-cpp can now be used to explicitly disable C++ support.
1936    
1937        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1938            backslash in a target when C++ was disabled. This confuses some
1939            versions of "make", apparently. Using an intermediate variable solves
1940            this. (Same for CPP_LOBJ.)
1941    
1942     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1943        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1944        necessary on certain architectures.
1945    
1946     4. Added a setting of -export-symbols-regex to the link command to remove
1947        those symbols that are exported in the C sense, but actually are local
1948        within the library, and not documented. Their names all begin with
1949        "_pcre_". This is not a perfect job, because (a) we have to except some
1950        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1951        available (and never for static libraries). I have made a note to try to
1952        find a way round (a) in the future.
1953    
1954    
1955    Version 6.2 01-Aug-05
1956    ---------------------
1957    
1958     1. There was no test for integer overflow of quantifier values. A construction
1959        such as {1111111111111111} would give undefined results. What is worse, if
1960        a minimum quantifier for a parenthesized subpattern overflowed and became
1961        negative, the calculation of the memory size went wrong. This could have
1962        led to memory overwriting.
1963    
1964     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1965    
1966     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1967        operating environments where this matters.
1968    
1969     4. Applied Giuseppe Maxia's patch to add additional features for controlling
1970        PCRE options from within the C++ wrapper.
1971    
1972     5. Named capturing subpatterns were not being correctly counted when a pattern
1973        was compiled. This caused two problems: (a) If there were more than 100
1974        such subpatterns, the calculation of the memory needed for the whole
1975        compiled pattern went wrong, leading to an overflow error. (b) Numerical
1976        back references of the form \12, where the number was greater than 9, were
1977        not recognized as back references, even though there were sufficient
1978        previous subpatterns.
1979    
1980     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1981        versions of gcc, e.g. 2.95.4.
1982    
1983    
1984    Version 6.1 21-Jun-05
1985    ---------------------
1986    
1987     1. There was one reference to the variable "posix" in pcretest.c that was not
1988        surrounded by "#if !defined NOPOSIX".
1989    
1990     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1991        the cross-check on the old pcre_info() function, for the benefit of the
1992        cut-down version of PCRE that is currently imported into Exim.
1993    
1994     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1995        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1996        patterns that start (?i) but I don't think that matters. The use of (?i) is
1997        just an example; this all applies to the other options as well.
1998    
1999     4. Since libtool seems to echo the compile commands it is issuing, the output
2000        from "make" can be reduced a bit by putting "@" in front of each libtool
2001        compile command.
2002    
2003     5. Patch from the folks at Google for configure.in to be a bit more thorough
2004        in checking for a suitable C++ installation before trying to compile the
2005        C++ stuff. This should fix a reported problem when a compiler was present,
2006        but no suitable headers.
2007    
2008     6. The man pages all had just "PCRE" as their title. I have changed them to
2009        be the relevant file name. I have also arranged that these names are
2010        retained in the file doc/pcre.txt, which is a concatenation in text format
2011        of all the man pages except the little individual ones for each function.
2012    
2013     7. The NON-UNIX-USE file had not been updated for the different set of source
2014        files that come with release 6. I also added a few comments about the C++
2015        wrapper.
2016    
2017    
2018    Version 6.0 07-Jun-05
2019    ---------------------
2020    
2021     1. Some minor internal re-organization to help with my DFA experiments.
2022    
2023     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
2024        didn't matter for the library itself when fully configured, but did matter
2025        when compiling without UCP support, or within Exim, where the ucp files are
2026        not imported.
2027    
2028     3. Refactoring of the library code to split up the various functions into
2029        different source modules. The addition of the new DFA matching code (see
2030        below) to a single monolithic source would have made it really too
2031        unwieldy, quite apart from causing all the code to be include in a
2032        statically linked application, when only some functions are used. This is
2033        relevant even without the DFA addition now that patterns can be compiled in
2034        one application and matched in another.
2035    
2036        The downside of splitting up is that there have to be some external
2037        functions and data tables that are used internally in different modules of
2038        the library but which are not part of the API. These have all had their
2039        names changed to start with "_pcre_" so that they are unlikely to clash
2040        with other external names.
2041    
2042     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
2043        a different (DFA) algorithm. Although it is slower than the original
2044        function, it does have some advantages for certain types of matching
2045        problem.
2046    
2047     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
2048        including restarting after a partial match.
2049    
2050     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
2051        defined when compiling for Windows was sent to me. I have put it into the
2052        code, though I have no means of testing or verifying it.
2053    
2054     7. Added the pcre_refcount() auxiliary function.
2055    
2056     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
2057        match before or at the first newline in the subject string. In pcretest,
2058        the /f option on a pattern can be used to set this.
2059    
2060     9. A repeated \w when used in UTF-8 mode with characters greater than 256
2061        would behave wrongly. This has been present in PCRE since release 4.0.
2062    
2063    10. A number of changes to the pcregrep command:
2064    
2065        (a) Refactored how -x works; insert ^(...)$ instead of setting
2066            PCRE_ANCHORED and checking the length, in preparation for adding
2067            something similar for -w.
2068    
2069        (b) Added the -w (match as a word) option.
2070    
2071        (c) Refactored the way lines are read and buffered so as to have more
2072            than one at a time available.
2073    
2074        (d) Implemented a pcregrep test script.
2075    
2076        (e) Added the -M (multiline match) option. This allows patterns to match
2077            over several lines of the subject. The buffering ensures that at least
2078            8K, or the rest of the document (whichever is the shorter) is available
2079            for matching (and similarly the previous 8K for lookbehind assertions).
2080    
2081        (f) Changed the --help output so that it now says
2082    
2083              -w, --word-regex(p)
2084    
2085            instead of two lines, one with "regex" and the other with "regexp"
2086            because that confused at least one person since the short forms are the
2087            same. (This required a bit of code, as the output is generated
2088            automatically from a table. It wasn't just a text change.)
2089    
2090        (g) -- can be used to terminate pcregrep options if the next thing isn't an
2091            option but starts with a hyphen. Could be a pattern or a path name
2092            starting with a hyphen, for instance.
2093    
2094        (h) "-" can be given as a file name to represent stdin.
2095    
2096        (i) When file names are being printed, "(standard input)" is used for
2097            the standard input, for compatibility with GNU grep. Previously
2098            "<stdin>" was used.
2099    
2100        (j) The option --label=xxx can be used to supply a name to be used for
2101            stdin when file names are being printed. There is no short form.
2102    
2103        (k) Re-factored the options decoding logic because we are going to add
2104            two more options that take data. Such options can now be given in four
2105            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
2106    
2107        (l) Added the -A, -B, and -C options for requesting that lines of context
2108            around matches be printed.
2109    
2110        (m) Added the -L option to print the names of files that do not contain
2111            any matching lines, that is, the complement of -l.
2112    
2113        (n) The return code is 2 if any file cannot be opened, but pcregrep does
2114            continue to scan other files.
2115    
2116        (o) The -s option was incorrectly implemented. For compatibility with other
2117            greps, it now suppresses the error message for a non-existent or non-
2118            accessible file (but not the return code). There is a new option called
2119            -q that suppresses the output of matching lines, which was what -s was
2120            previously doing.
2121    
2122        (p) Added --include and --exclude options to specify files for inclusion
2123            and exclusion when recursing.
2124    
2125    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
2126        Hopefully, it now does.
2127    
2128    12. Missing cast in pcre_study().
2129    
2130    13. Added an "uninstall" target to the makefile.
2131    
2132    14. Replaced "extern" in the function prototypes in Makefile.in with
2133        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
2134        world, but is set differently for Windows.
2135    
2136    15. Added a second compiling function called pcre_compile2(). The only
2137        difference is that it has an extra argument, which is a pointer to an
2138        integer error code. When there is a compile-time failure, this is set
2139        non-zero, in addition to the error test pointer being set to point to an
2140        error message. The new argument may be NULL if no error number is required
2141        (but then you may as well call pcre_compile(), which is now just a
2142        wrapper). This facility is provided because some applications need a
2143        numeric error indication, but it has also enabled me to tidy up the way
2144        compile-time errors are handled in the POSIX wrapper.
2145    
2146    16. Added VPATH=.libs to the makefile; this should help when building with one
2147        prefix path and installing with another. (Or so I'm told by someone who
2148        knows more about this stuff than I do.)
2149    
2150    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
2151        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
2152        match everything, including newlines. This is not POSIX-compatible, but
2153        somebody wanted the feature. From pcretest it can be activated by using
2154        both the P and the s flags.
2155    
2156    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
2157    
2158    19. libpcre.pc was being incorrectly installed as executable.
2159    
2160    20. A couple of places in pcretest check for end-of-line by looking for '\n';
2161        it now also looks for '\r' so that it will work unmodified on Windows.
2162    
2163    21. Added Google's contributed C++ wrapper to the distribution.
2164    
2165    22. Added some untidy missing memory free() calls in pcretest, to keep
2166        Electric Fence happy when testing.
2167    
2168    
2169    
2170    Version 5.0 13-Sep-04
2171    ---------------------
2172    
2173     1. Internal change: literal characters are no longer packed up into items
2174        containing multiple characters in a single byte-string. Each character
2175        is now matched using a separate opcode. However, there may be more than one
2176        byte in the character in UTF-8 mode.
2177    
2178     2. The pcre_callout_block structure has two new fields: pattern_position and
2179        next_item_length. These contain the offset in the pattern to the next match
2180        item, and its length, respectively.
2181    
2182     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
2183        insertion of callouts before each pattern item. Added the /C option to
2184        pcretest to make use of this.
2185    
2186     4. On the advice of a Windows user, the lines
2187    
2188          #if defined(_WIN32) || defined(WIN32)
2189          _setmode( _fileno( stdout ), 0x8000 );
2190          #endif  /* defined(_WIN32) || defined(WIN32) */
2191    
2192        have been added to the source of pcretest. This apparently does useful
2193        magic in relation to line terminators.
2194    
2195     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
2196        for the benefit of those environments where the "b" makes a difference.
2197    
2198     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
2199        to know about it. I have put a hack into configure.in that adds in code
2200        to set GCC=yes if CC=icc. This seems to end up at a point in the
2201        generated configure script that is early enough to affect the setting of
2202        compiler options, which is what is needed, but I have no means of testing
2203        whether it really works. (The user who reported this had patched the
2204        generated configure script, which of course I cannot do.)
2205    
2206        LATER: After change 22 below (new libtool files), the configure script
2207        seems to know about icc (and also ecc). Therefore, I have commented out
2208        this hack in configure.in.
2209    
2210     7. Added support for pkg-config (2 patches were sent in).
2211    
2212     8. Negated POSIX character classes that used a combination of internal tables
2213        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2214        [[:^ascii]]. Typically, they would match almost any characters. The other
2215        POSIX classes were not broken in this way.
2216    
2217     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2218        to find the match, as PCRE was deluded into thinking that the match had to
2219        start at the start point or following a newline. The same bug applied to
2220        patterns with negative forward assertions or any backward assertions
2221        preceding ".*" at the start, unless the pattern required a fixed first
2222        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2223    
2224    10. In UTF-8 mode, when moving forwards in the subject after a failed match
2225        starting at the last subject character, bytes beyond the end of the subject
2226        string were read.
2227    
2228    11. Renamed the variable "class" as "classbits" to make life easier for C++
2229        users. (Previously there was a macro definition, but it apparently wasn't
2230        enough.)
2231    
2232    12. Added the new field "tables" to the extra data so that tables can be passed
2233        in at exec time, or the internal tables can be re-selected. This allows
2234        a compiled regex to be saved and re-used at a later time by a different
2235        program that might have everything at different addresses.
2236    
2237    13. Modified the pcre-config script so that, when run on Solaris, it shows a
2238        -R library as well as a -L library.
2239    
2240    14. The debugging options of pcretest (-d on the command line or D on a
2241        pattern) showed incorrect output for anything following an extended class
2242        that contained multibyte characters and which was followed by a quantifier.
2243    
2244    15. Added optional support for general category Unicode character properties
2245        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2246        support. It adds about 90K to the size of the library. The meanings of the
2247        inbuilt class escapes such as \d and \s have NOT been changed.
2248    
2249    16. Updated pcredemo.c to include calls to free() to release the memory for the
2250        compiled pattern.
2251    
2252    17. The generated file chartables.c was being created in the source directory
2253        instead of in the building directory. This caused the build to fail if the
2254        source directory was different from the building directory, and was
2255        read-only.
2256    
2257    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2258        file. No doubt somebody will tell me if they don't make sense... Also added
2259        Dan Mooney's comments about building on OpenVMS.
2260    
2261    19. Added support for partial matching via the PCRE_PARTIAL option for
2262        pcre_exec() and the \P data escape in pcretest.
2263    
2264    20. Extended pcretest with 3 new pattern features:
2265    
2266        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
2267              write the compiled pattern to the file whose name is "rest-of-line".
2268              This is a straight binary dump of the data, with the saved pointer to
2269              the character tables forced to be NULL. The study data, if any, is
2270              written too. After writing, pcretest reads a new pattern.
2271    
2272        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2273              compiled pattern from the given file. There must not be any
2274              occurrences of "<" in the file name (pretty unlikely); if there are,
2275              pcretest will instead treat the initial "<" as a pattern delimiter.
2276              After reading in the pattern, pcretest goes on to read data lines as
2277              usual.
2278    
2279        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2280              and 16-bit fields in a compiled pattern, to simulate a pattern that
2281              was compiled on a host of opposite endianness.
2282    
2283    21. The pcre-exec() function can now cope with patterns that were compiled on
2284        hosts of opposite endianness, with this restriction:
2285    
2286          As for any compiled expression that is saved and used later, the tables
2287          pointer field cannot be preserved; the extra_data field in the arguments
2288          to pcre_exec() should be used to pass in a tables address if a value
2289          other than the default internal tables were used at compile time.
2290    
2291    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2292        now diagnosed as an error. Previously, most of the time, a negative number
2293        would have been treated as zero, but if in addition "ovector" was passed as
2294        NULL, a crash could occur.
2295    
2296    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2297        new versions from the libtool 1.5 distribution (the last one is a copy of
2298        a file called libtool.m4). This seems to have fixed the need to patch
2299        "configure" to support Darwin 1.3 (which I used to do). However, I still
2300        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2301        workstation).
2302    
2303    24. Changed the PCRE licence to be the more standard "BSD" licence.
2304    
2305    
2306    Version 4.5 01-Dec-03
2307    ---------------------
2308    
2309     1. There has been some re-arrangement of the code for the match() function so
2310        that it can be compiled in a version that does not call itself recursively.
2311        Instead, it keeps those local variables that need separate instances for
2312        each "recursion" in a frame on the heap, and gets/frees frames whenever it
2313        needs to "recurse". Keeping track of where control must go is done by means
2314        of setjmp/longjmp. The whole thing is implemented by a set of macros that
2315        hide most of the details from the main code, and operates only if
2316        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2317        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2318        operating.
2319    
2320        To make it easier for callers to provide specially tailored get/free
2321        functions for this usage, two new functions, pcre_stack_malloc, and
2322        pcre_stack_free, are used. They are always called in strict stacking order,
2323        and the size of block requested is always the same.
2324    
2325        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2326        PCRE has been compiled to use the stack or the heap for recursion. The
2327        -C option of pcretest uses this to show which version is compiled.
2328    
2329        A new data escape \S, is added to pcretest; it causes the amounts of store
2330        obtained and freed by both kinds of malloc/free at match time to be added
2331        to the output.
2332    
2333     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2334        what's available on my current Linux desktop machine.
2335    
2336     3. When matching a UTF-8 string, the test for a valid string at the start has
2337        been extended. If start_offset is not zero, PCRE now checks that it points
2338        to a byte that is the start of a UTF-8 character. If not, it returns
2339        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2340        this is necessary because there may be backward assertions in the pattern.
2341        When matching the same subject several times, it may save resources to use
2342        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2343    
2344     4. The code for checking the validity of UTF-8 strings has been tightened so
2345        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2346        containing "overlong sequences".
2347    
2348     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2349        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2350        should have been "&", but it just so happened that all the cases this let
2351        through by mistake were picked up later in the function.
2352    
2353     6. I had used a variable called "isblank" - this is a C99 function, causing
2354        some compilers to warn. To avoid this, I renamed it (as "blankclass").
2355    
2356     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2357        prompting; (b) run "./pcretest /dev/null" at the start of the test script
2358        so the version is shown; (c) stop "make test" echoing "./RunTest".
2359    
2360     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2361    
2362     9. The prototype for memmove() for systems that don't have it was using
2363        size_t, but the inclusion of the header that defines size_t was later. I've
2364        moved the #includes for the C headers earlier to avoid this.
2365    
2366    10. Added some adjustments to the code to make it easier to compiler on certain
2367        special systems:
2368    
2369          (a) Some "const" qualifiers were missing.
2370          (b) Added the macro EXPORT before all exported functions; by default this
2371              is defined to be empty.
2372          (c) Changed the dftables auxiliary program (that builds chartables.c) so
2373              that it reads its output file name as an argument instead of writing
2374              to the standard output and assuming this can be redirected.
2375    
2376    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2377        class containing characters with values greater than 255, PCRE compilation
2378        went into a loop.
2379    
2380    12. A recursive reference to a subpattern that was within another subpattern
2381        that had a minimum quantifier of zero caused PCRE to crash. For example,
2382        (x(y(?2))z)? provoked this bug with a subject that got as far as the
2383        recursion. If the recursively-called subpattern itself had a zero repeat,
2384        that was OK.
2385    
2386    13. In pcretest, the buffer for reading a data line was set at 30K, but the
2387        buffer into which it was copied (for escape processing) was still set at
2388        1024, so long lines caused crashes.
2389    
2390    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2391        "internal error: code overflow...". This applied to any character class
2392        that was followed by a possessive quantifier.
2393    
2394    15. Modified the Makefile to add libpcre.la as a prerequisite for
2395        libpcreposix.la because I was told this is needed for a parallel build to
2396        work.
2397    
2398    16. If a pattern that contained .* following optional items at the start was
2399        studied, the wrong optimizing data was generated, leading to matching
2400        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2401        matching string must start with a or b or c. The correct conclusion for
2402        this pattern is that a match can start with any character.
2403    
2404    
2405    Version 4.4 13-Aug-03
2406    ---------------------
2407    
2408     1. In UTF-8 mode, a character class containing characters with values between
2409        127 and 255 was not handled correctly if the compiled pattern was studied.
2410        In fixing this, I have also improved the studying algorithm for such
2411        classes (slightly).
2412    
2413     2. Three internal functions had redundant arguments passed to them. Removal
2414        might give a very teeny performance improvement.
2415    
2416     3. Documentation bug: the value of the capture_top field in a callout is *one
2417        more than* the number of the hightest numbered captured substring.
2418    
2419     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2420        in incorrectly linking with a previously installed version. They now link
2421        explicitly with libpcre.la.
2422    
2423     5. configure.in no longer needs to recognize Cygwin specially.
2424    
2425     6. A problem in pcre.in for Windows platforms is fixed.
2426    
2427     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2428        pcretest, it used to include the size of the study block as part of its
2429        output. Unfortunately, the structure contains a field that has a different
2430        size on different hardware architectures. This meant that the tests that
2431        showed this size failed. As the block is currently always of a fixed size,
2432        this information isn't actually particularly useful in pcretest output, so
2433        I have just removed it.
2434    
2435     8. Three pre-processor statements accidentally did not start in column 1.
2436        Sadly, there are *still* compilers around that complain, even though
2437        standard C has not required this for well over a decade. Sigh.
2438    
2439     9. In pcretest, the code for checking callouts passed small integers in the
2440        callout_data field, which is a void * field. However, some picky compilers
2441        complained about the casts involved for this on 64-bit systems. Now
2442        pcretest passes the address of the small integer instead, which should get
2443        rid of the warnings.
2444    
2445    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2446        both compile and run time, and gives an error if an invalid UTF-8 sequence
2447        is found. There is a option for disabling this check in cases where the
2448        string is known to be correct and/or the maximum performance is wanted.
2449    
2450    11. In response to a bug report, I changed one line in Makefile.in from
2451    
2452            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2453        to
2454            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2455    
2456        to look similar to other lines, but I have no way of telling whether this
2457        is the right thing to do, as I do not use Windows. No doubt I'll get told
2458        if it's wrong...
2459    
2460    
2461    Version 4.3 21-May-03
2462    ---------------------
2463    
2464    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2465       Makefile.
2466    
2467    2. Some refactoring to improve the quality of the code:
2468    
2469       (i)   The utf8_table... variables are now declared "const".
2470    
2471       (ii)  The code for \cx, which used the "case flipping" table to upper case
2472             lower case letters, now just substracts 32. This is ASCII-specific,
2473             but the whole concept of \cx is ASCII-specific, so it seems
2474             reasonable.
2475    
2476       (iii) PCRE was using its character types table to recognize decimal and
2477             hexadecimal digits in the pattern. This is silly, because it handles
2478             only 0-9, a-f, and A-F, but the character types table is locale-
2479             specific, which means strange things might happen. A private
2480             table is now used for this - though it costs 256 bytes, a table is
2481             much faster than multiple explicit tests. Of course, the standard
2482             character types table is still used for matching digits in subject
2483             strings against \d.
2484    
2485       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2486             ending in _t are). So I've renamed it as ESC_tee.
2487    
2488    3. The first argument for regexec() in the POSIX wrapper should have been
2489       defined as "const".
2490    
2491    4. Changed pcretest to use malloc() for its buffers so that they can be
2492       Electric Fenced for debugging.
2493    
2494    5. There were several places in the code where, in UTF-8 mode, PCRE would try
2495       to read one or more bytes before the start of the subject string. Often this
2496       had no effect on PCRE's behaviour, but in some circumstances it could
2497       provoke a segmentation fault.
2498    
2499    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2500       to try to read one or more bytes before the start of the subject string.
2501    
2502    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
2503       UTF-8 support could misbehave in various ways if the subject string
2504       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
2505       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
2506       back over UTF-8 characters.)
2507    
2508    
2509    Version 4.2 14-Apr-03
2510    ---------------------
2511    
2512    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
2513    
2514    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
2515         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
2516         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
2517         [WIN_PREFIX]: new variable, "cyg" for Cygwin
2518         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
2519           and BUILD_EXEEXT
2520         Note: automatic setting of the BUILD variables is not yet working
2521         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
2522           compile-time but not at link-time
2523         [LINK]: use for linking executables only
2524         make different versions for Windows and non-Windows
2525         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
2526           libraries
2527         [LINK_FOR_BUILD]: new variable
2528         [OBJEXT]: use throughout
2529         [EXEEXT]: use throughout
2530         <winshared>: new target
2531         <wininstall>: new target
2532         <dftables.o>: use native compiler
2533         <dftables>: use native linker
2534         <install>: handle Windows platform correctly
2535         <clean>: ditto
2536         <check>: ditto
2537         copy DLL to top builddir before testing
2538    
2539       As part of these changes, -no-undefined was removed again. This was reported
2540       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
2541       in any case.
2542    
2543    3. Some tidies to get rid of compiler warnings:
2544    
2545       . In the match_data structure, match_limit was an unsigned long int, whereas
2546         match_call_count was an int. I've made them both unsigned long ints.
2547    
2548       . In pcretest the fact that a const uschar * doesn't automatically cast to
2549         a void * provoked a warning.
2550    
2551       . Turning on some more compiler warnings threw up some "shadow" variables
2552         and a few more missing casts.
2553    
2554    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
2555       option, a class that contained a single character with a value between 128
2556       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
2557    
2558    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
2559       option, a class that contained several characters, but with at least one
2560       whose value was between 128 and 255 caused PCRE to crash.
2561    
2562    
2563    Version 4.1 12-Mar-03
2564    ---------------------
2565    
2566    1. Compiling with gcc -pedantic found a couple of places where casts were
2567    needed, and a string in dftables.c that was longer than standard compilers are
2568    required to support.
2569    
2570    2. Compiling with Sun's compiler found a few more places where the code could
2571    be tidied up in order to avoid warnings.
2572    
2573    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
2574    first of these names is deprecated in the latest Autoconf in favour of the name
2575    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
2576    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2577    analogy I have changed it to CFLAGS_FOR_BUILD.
2578    
2579    4. Added -no-undefined to the linking command in the Makefile, because this is
2580    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2581    linking step for the pcreposix library.
2582    
2583    5. PCRE was failing to diagnose the case of two named groups with the same
2584    name.
2585    
2586    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2587    literal character that is needed in the subject for a match, and scans along to
2588    ensure that it is present before embarking on the full matching process. This
2589    saves time in cases of nested unlimited repeats that are never going to match.
2590    Problem: the scan can take a lot of time if the subject is very long (e.g.
2591    megabytes), thus penalizing straightforward matches. It is now done only if the
2592    amount of subject to be scanned is less than 1000 bytes.
2593    
2594    7. A lesser problem with the same optimization is that it was recording the
2595    first character of an anchored pattern as "needed", thus provoking a search
2596    right along the subject, even when the first match of the pattern was going to
2597    fail. The "needed" character is now not set for anchored patterns, unless it
2598    follows something in the pattern that is of non-fixed length. Thus, it still
2599    fulfils its original purpose of finding quick non-matches in cases of nested
2600    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
2601    
2602    
2603    Version 4.0 17-Feb-03
2604    ---------------------
2605    
2606    1. If a comment in an extended regex that started immediately after a meta-item
2607    extended to the end of string, PCRE compiled incorrect data. This could lead to
2608    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
2609    
2610    2. Moved to autoconf 2.53 and libtool 1.4.2.
2611    
2612    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
2613    the special perltest8 script is no longer needed - all the tests can be run
2614    from a single perltest script.
2615    
2616    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
2617    by \s. It has now been removed in PCRE. This means it isn't recognized as
2618    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
2619    class [:space:] *does* include VT, thereby creating a mess.
2620    
2621    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
2622    space and tab.
2623    
2624    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
2625    its new features into the main test script, reducing the number of scripts.
2626    
2627    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
2628    were backward compatible, and made the (?i) apply to the whole pattern, as if
2629    /i were given. Now it behaves more logically, and applies the option setting
2630    only to what follows. PCRE has been changed to follow suit. However, if it
2631    finds options settings right at the start of the pattern, it extracts them into
2632    the global options, as before. Thus, they show up in the info data.
2633    
2634    8. Added support for the \Q...\E escape sequence. Characters in between are
2635    treated as literals. This is slightly different from Perl in that $ and @ are
2636    also handled as literals inside the quotes. In Perl, they will cause variable
2637    interpolation. Note the following examples:
2638    
2639        Pattern            PCRE matches      Perl matches
2640    
2641        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
2642        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
2643        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
2644    
2645    For compatibility with Perl, \Q...\E sequences are recognized inside character
2646    classes as well as outside them.
2647    
2648    9. Re-organized 3 code statements in pcretest to avoid "overflow in
2649    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2650    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2651    signed/unsigned warnings.
2652    
2653    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2654    option for pcretest, so I've replaced it by a simple function that does just
2655    that job.
2656    
2657    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2658    "pcregrep -".
2659    
2660    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2661    Java package. This provides some syntactic sugar for simple cases of what my
2662    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2663    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2664    item, you can use this simplified notation. Note that only makes sense with
2665    greedy quantifiers. Consequently, the use of the possessive quantifier forces
2666    greediness, whatever the setting of the PCRE_UNGREEDY option.
2667    
2668    13. A change of greediness default within a pattern was not taking effect at
2669    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2670    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2671    was abstracted outside.
2672    
2673    14. PCRE now supports the \G assertion. It is true when the current matching
2674    position is at the start point of the match. This differs from \A when the
2675    starting offset is non-zero. Used with the /g option of pcretest (or similar
2676    code), it works in the same way as it does for Perl's /g option. If all
2677    alternatives of a regex begin with \G, the expression is anchored to the start
2678    match position, and the "anchored" flag is set in the compiled expression.
2679    
2680    15. Some bugs concerning the handling of certain option changes within patterns
2681    have been fixed. These applied to options other than (?ims). For example,
2682    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2683    been the other way round. Some of this was related to change 7 above.
2684    
2685    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2686    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2687    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2688    POSIX classes only within a class (e.g. /[[:alpha:]]/).
2689    
2690    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2691    mode. Unlike ".", it always matches newline, whatever the setting of
2692    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2693    assertions. Perl allows it, but it doesn't (in general) work because it can't
2694    calculate the length of the lookbehind. At least, that's the case for Perl
2695    5.8.0 - I've been told they are going to document that it doesn't work in
2696    future.
2697    
2698    18. Added an error diagnosis for escapes that PCRE does not support: these are
2699    \L, \l, \N, \P, \p, \U, \u, and \X.
2700    
2701    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2702    reading past the end of the pattern in cases such as /[abcd/.
2703    
2704    20. PCRE was getting more memory than necessary for patterns with classes that
2705    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2706    
2707    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2708    compiling PCRE for use with Virtual Pascal.
2709    
2710    22. Small fix to the Makefile to make it work properly if the build is done
2711    outside the source tree.
2712    
2713    23. Added a new extension: a condition to go with recursion. If a conditional
2714    subpattern starts with (?(R) the "true" branch is used if recursion has
2715    happened, whereas the "false" branch is used only at the top level.
2716    
2717    24. When there was a very long string of literal characters (over 255 bytes
2718    without UTF support, over 250 bytes with UTF support), the computation of how
2719    much memory was required could be incorrect, leading to segfaults or other
2720    strange effects.
2721    
2722    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2723    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2724    there was a subsequent back reference to those brackets. This meant that, for
2725    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2726    possible to check for precisely this case. All we can do is abandon the
2727    optimization if .* occurs inside capturing brackets when there are any back
2728    references whatsoever. (See below for a better fix that came later.)
2729    
2730    26. The handling of the optimization for finding the first character of a
2731    non-anchored pattern, and for finding a character that is required later in the
2732    match were failing in some cases. This didn't break the matching; it just
2733    failed to optimize when it could. The way this is done has been re-implemented.
2734    
2735    27. Fixed typo in error message for invalid (?R item (it said "(?p").
2736    
2737    28. Added a new feature that provides some of the functionality that Perl
2738    provides with (?{...}). The facility is termed a "callout". The way it is done
2739    in PCRE is for the caller to provide an optional function, by setting
2740    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2741    global variable. By default it is unset, which disables all calling out. To get
2742    the function called, the regex must include (?C) at appropriate points. This
2743    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2744    This provides a means of identifying different callout points. When PCRE
2745    reaches such a point in the regex, if pcre_callout has been set, the external
2746    function is called. It is provided with data in a structure called
2747    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2748    matching continues; if it returns a non-zero value, the match at the current
2749    point fails. However, backtracking will occur if possible. [This was changed
2750    later and other features added - see item 49 below.]
2751    
2752    29. pcretest is upgraded to test the callout functionality. It provides a
2753    callout function that displays information. By default, it shows the start of
2754    the match and the current position in the text. There are some new data escapes
2755    to vary what happens:
2756    
2757        \C+         in addition, show current contents of captured substrings
2758        \C-         do not supply a callout function
2759        \C!n        return 1 when callout number n is reached
2760        \C!n!m      return 1 when callout number n is reached for the mth time
2761    
2762    30. If pcregrep was called with the -l option and just a single file name, it
2763    output "<stdin>" if a match was found, instead of the file name.
2764    
2765    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2766    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2767    pcre_exec(). This saves a malloc/free per call. The default value of
2768    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2769    when configuring.
2770    
2771    32. The default maximum size of a compiled pattern is 64K. There have been a
2772    few cases of people hitting this limit. The code now uses macros to handle the
2773    storing of links as offsets within the compiled pattern. It defaults to 2-byte
2774    links, but this can be changed to 3 or 4 bytes by --with-link-size when
2775    configuring. Tests 2 and 5 work only with 2-byte links because they output
2776    debugging information about compiled patterns.
2777    
2778    33. Internal code re-arrangements:
2779    
2780    (a) Moved the debugging function for printing out a compiled regex into
2781        its own source file (printint.c) and used #include to pull it into
2782        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2783        separate copies.
2784    
2785    (b) Defined the list of op-code names for debugging as a macro in
2786        internal.h so that it is next to the definition of the opcodes.
2787    
2788    (c) Defined a table of op-code lengths for simpler skipping along compiled
2789        code. This is again a macro in internal.h so that it is next to the
2790        definition of the opcodes.
2791    
2792    34. Added support for recursive calls to individual subpatterns, along the
2793    lines of Robin Houston's patch (but implemented somewhat differently).
2794    
2795    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2796    allow it to read and process whole directories in Win32. This code was
2797    contributed by Lionel Fourquaux; it has not been tested by me.
2798    
2799    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2800    used to name a group. Names consist of alphanumerics and underscores, and must
2801    be unique. Back references use the syntax (?P=name) and recursive calls use
2802    (?P>name) which is a PCRE extension to the Python extension. Groups still have
2803    numbers. The function pcre_fullinfo() can be used after compilation to extract
2804    a name/number map. There are three relevant calls:
2805    
2806      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
2807      PCRE_INFO_NAMECOUNT            yields the number of entries
2808      PCRE_INFO_NAMETABLE            yields a pointer to the map.
2809    
2810    The map is a vector of fixed-size entries. The size of each entry depends on
2811    the length of the longest name used. The first two bytes of each entry are the
2812    group number, most significant byte first. There follows the corresponding
2813    name, zero terminated. The names are in alphabetical order.
2814    
2815    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2816    case instead of 255. Making it the same both with and without UTF-8 support
2817    means that the same test output works with both.
2818    
2819    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2820    calling malloc() with a zero argument.
2821    
2822    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2823    optimization. I've improved things by keeping a bitmap of backreferences with
2824    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2825    fact referenced, the optimization can be applied. It is unlikely that a
2826    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2827    the match to follow \n) will appear inside brackets with a number greater than
2828    31, but if it does, any back reference > 31 suppresses the optimization.
2829    
2830    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2831    of disabling numbered capturing parentheses. Any opening parenthesis that is
2832    not followed by ? behaves as if it were followed by ?: but named parentheses
2833    can still be used for capturing (and they will acquire numbers in the usual
2834    way).
2835    
2836    41. Redesigned the return codes from the match() function into yes/no/error so
2837    that errors can be passed back from deep inside the nested calls. A malloc
2838    failure while inside a recursive subpattern call now causes the
2839    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2840    
2841    42. It is now possible to set a limit on the number of times the match()
2842    function is called in a call to pcre_exec(). This facility makes it possible to
2843    limit the amount of recursion and backtracking, though not in a directly
2844    obvious way, because the match() function is used in a number of different
2845    circumstances. The count starts from zero for each position in the subject
2846    string (for non-anchored patterns). The default limit is, for compatibility, a
2847    large number, namely 10 000 000. You can change this in two ways:
2848    
2849    (a) When configuring PCRE before making, you can use --with-match-limit=n
2850        to set a default value for the compiled library.
2851    
2852    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2853        a different value is set. See 45 below.
2854    
2855    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2856    
2857    43. Added a new function pcre_config(int, void *) to enable run-time extraction
2858    of things that can be changed at compile time. The first argument specifies
2859    what is wanted and the second points to where the information is to be placed.
2860    The current list of available information is:
2861    
2862      PCRE_CONFIG_UTF8
2863    
2864    The output is an integer that is set to one if UTF-8 support is available;
2865    otherwise it is set to zero.
2866    
2867      PCRE_CONFIG_NEWLINE
2868    
2869    The output is an integer that it set to the value of the code that is used for
2870    newline. It is either LF (10) or CR (13).
2871    
2872      PCRE_CONFIG_LINK_SIZE
2873    
2874    The output is an integer that contains the number of bytes used for internal
2875    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2876    
2877      PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
2878    
2879    The output is an integer that contains the threshold above which the POSIX
2880    interface uses malloc() for output vectors. See item 31 above.
2881    
2882      PCRE_CONFIG_MATCH_LIMIT
2883    
2884    The output is an unsigned integer that contains the default limit of the number
2885    of match() calls in a pcre_exec() execution. See 42 above.
2886    
2887    44. pcretest has been upgraded by the addition of the -C option. This causes it
2888    to extract all the available output from the new pcre_config() function, and to
2889    output it. The program then exits immediately.
2890    
2891    45. A need has arisen to pass over additional data with calls to pcre_exec() in
2892    order to support additional features. One way would have been to define
2893    pcre_exec2() (for example) with extra arguments, but this would not have been
2894    extensible, and would also have required all calls to the original function to
2895    be mapped to the new one. Instead, I have chosen to extend the mechanism that
2896    is used for passing in "extra" data from pcre_study().
2897    
2898    The pcre_extra structure is now exposed and defined in pcre.h. It currently
2899    contains the following fields:
2900    
2901      flags         a bitmap indicating which of the following fields are set
2902      study_data    opaque data from pcre_study()
2903      match_limit   a way of specifying a limit on match() calls for a specific
2904                      call to pcre_exec()
2905      callout_data  data for callouts (see 49 below)
2906    
2907    The flag bits are also defined in pcre.h, and are
2908    
2909      PCRE_EXTRA_STUDY_DATA
2910      PCRE_EXTRA_MATCH_LIMIT
2911      PCRE_EXTRA_CALLOUT_DATA
2912    
2913    The pcre_study() function now returns one of these new pcre_extra blocks, with
2914    the actual study data pointed to by the study_data field, and the
2915    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2916    before. That is, this change is entirely upwards-compatible and requires no
2917    change to existing code.
2918    
2919    If you want to pass in additional data to pcre_exec(), you can either place it
2920    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2921    block.
2922    
2923    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2924    data string contains the escape sequence \M, pcretest calls pcre_exec() several
2925    times with different match limits, until it finds the minimum value needed for
2926    pcre_exec() to complete. The value is then output. This can be instructive; for
2927    most simple matches the number is quite small, but for pathological cases it
2928    gets very large very quickly.
2929    
2930    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2931    returns the size of the data block pointed to by the study_data field in a
2932    pcre_extra block, that is, the value that was passed as the argument to
2933    pcre_malloc() when PCRE was getting memory in which to place the information
2934    created by pcre_study(). The fourth argument should point to a size_t variable.
2935    pcretest has been extended so that this information is shown after a successful
2936    pcre_study() call when information about the compiled regex is being displayed.
2937    
2938    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2939    because what follows is always an absolute path. (Later: it turns out that this
2940    is more than cosmetic for MinGW, because it doesn't like empty path
2941    components.)
2942    
2943    49. Some changes have been made to the callout feature (see 28 above):
2944    
2945    (i)  A callout function now has three choices for what it returns:
2946    
2947           0  =>  success, carry on matching
2948         > 0  =>  failure at this point, but backtrack if possible
2949         < 0  =>  serious error, return this value from pcre_exec()
2950    
2951         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2952         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2953         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2954         use by callout functions. It will never be used by PCRE itself.
2955    
2956    (ii) The pcre_extra structure (see 45 above) has a void * field called
2957         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2958         pcre_callout_block structure has a field of the same name. The contents of
2959         the field passed in the pcre_extra structure are passed to the callout
2960         function in the corresponding field in the callout block. This makes it
2961         easier to use the same callout-containing regex from multiple threads. For
2962         testing, the pcretest program has a new data escape
2963    
2964           \C*n        pass the number n (may be negative) as callout_data
2965    
2966         If the callout function in pcretest receives a non-zero value as
2967         callout_data, it returns that value.
2968    
2969    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2970    there were some redundant $(CFLAGS) in commands that are now specified as
2971    $(LINK), which already includes $(CFLAGS).
2972    
2973    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2974    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2975    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2976    one-byte characters throughout. Note that case-insensitive matching applies
2977    only to characters whose values are less than 256. PCRE doesn't support the
2978    notion of cases for higher-valued characters.
2979    
2980    (i)   A character class whose characters are all within 0-255 is handled as
2981          a bit map, and the map is inverted for negative classes. Previously, a
2982          character > 255 always failed to match such a class; however it should
2983          match if the class was a negative one (e.g. [^ab]). This has been fixed.
2984    
2985    (ii)  A negated character class with a single character < 255 is coded as
2986          "not this character" (OP_NOT). This wasn't working properly when the test
2987          character was multibyte, either singly or repeated.
2988    
2989    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2990          mode, for example: \x{100}{2,3}.
2991    
2992    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2993          singly or repeated) now correctly test multibyte characters. However,
2994          PCRE doesn't recognize any characters with values greater than 255 as
2995          digits, spaces, or word characters. Such characters always match \D, \S,
2996          and \W, and never match \d, \s, or \w.
2997    
2998    (v)   Classes may now contain characters and character ranges with values
2999          greater than 255. For example: [ab\x{100}-\x{400}].
3000    
3001    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
3002          PCRE in UTF-8 mode.
3003    
3004    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
3005    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
3006    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
3007    value.)
3008    
3009    53. The single man page has become too large. I have therefore split it up into
3010    a number of separate man pages. These also give rise to individual HTML pages;
3011    these are now put in a separate directory, and there is an index.html page that
3012    lists them all. Some hyperlinking between the pages has been installed.
3013    
3014    54. Added convenience functions for handling named capturing parentheses.
3015    
3016    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
3017    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
3018    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
3019    are faulted.
3020    
3021    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
3022    calling configure. These values are used when compiling the dftables.c program
3023    which is run to generate the source of the default character tables. They
3024    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
3025    you will need to set these values.
3026    
3027    57. Updated the building process for Windows DLL, as provided by Fred Cox.
3028    
3029    
3030    Version 3.9 02-Jan-02
3031    ---------------------
3032    
3033    1. A bit of extraneous text had somehow crept into the pcregrep documentation.
3034    
3035    2. If --disable-static was given, the building process failed when trying to
3036    build pcretest and pcregrep. (For some reason it was using libtool to compile
3037    them, which is not right, as they aren't part of the library.)
3038    
3039    
3040    Version 3.8 18-Dec-01
3041    ---------------------
3042    
3043    1. The experimental UTF-8 code was completely screwed up. It was packing the
3044    bytes in the wrong order. How dumb can you get?
3045    
3046    
3047    Version 3.7 29-Oct-01
3048    ---------------------
3049    
3050    1. In updating pcretest to check change 1 of version 3.6, I screwed up.
3051    This caused pcretest, when used on the test data, to segfault. Unfortunately,
3052    this didn't happen under Solaris 8, where I normally test things.
3053    
3054    2. The Makefile had to be changed to make it work on BSD systems, where 'make'
3055    doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
3056    isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
3057    this fix an hour or so after the initial 3.7 release.)
3058    
3059    
3060    Version 3.6 23-Oct-01
3061    ---------------------
3062    
3063    1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
3064    offsets passed as NULL with zero offset count.
3065    
3066    2. The config.guess and config.sub files had not been updated when I moved to
3067    the latest autoconf.
3068    
3069    
3070    Version 3.5 15-Aug-01
3071    ---------------------
3072    
3073    1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
3074    had been forgotten.
3075    
3076    2. By using declared but undefined structures, we can avoid using "void"
3077    definitions in pcre.h while keeping the internal definitions of the structures
3078    private.
3079    
3080    3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
3081    user point of view, this means that both static and shared libraries are built
3082    by default, but this can be individually controlled. More of the work of
3083    handling this static/shared cases is now inside libtool instead of PCRE's make
3084    file.
3085    
3086    4. The pcretest utility is now installed along with pcregrep because it is
3087    useful for users (to test regexs) and by doing this, it automatically gets
3088    relinked by libtool. The documentation has been turned into a man page, so
3089    there are now .1, .txt, and .html versions in /doc.
3090    
3091    5. Upgrades to pcregrep:
3092       (i)   Added long-form option names like gnu grep.
3093       (ii)  Added --help to list all options with an explanatory phrase.
3094       (iii) Added -r, --recursive to recurse into sub-directories.
3095       (iv)  Added -f, --file to read patterns from a file.
3096    
3097    6. pcre_exec() was referring to its "code" argument before testing that
3098    argument for NULL (and giving an error if it was NULL).
3099    
3100    7. Upgraded Makefile.in to allow for compiling in a different directory from
3101    the source directory.
3102    
3103    8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
3104    options bits, the pointer it was passed was to an int instead of to an unsigned
3105    long int. This mattered only on 64-bit systems.
3106    
3107    9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
3108    generated) instead of pcre.in, which it its source. Also made the same change
3109    in several of the .c files.
3110    
3111    10. A new release of gcc defines printf() as a macro, which broke pcretest
3112    because it had an ifdef in the middle of a string argument for printf(). Fixed
3113    by using separate calls to printf().
3114    
3115    11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
3116    script, to force use of CR or LF instead of \n in the source. On non-Unix
3117    systems, the value can be set in config.h.
3118    
3119    12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
3120    absolute limit. Changed the text of the error message to make this clear, and
3121    likewise updated the man page.
3122    
3123    13. The limit of 99 on the number of capturing subpatterns has been removed.
3124    The new limit is 65535, which I hope will not be a "real" limit.
3125    
3126    
3127    Version 3.4 22-Aug-00
3128    ---------------------
3129    
3130    1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
3131    
3132    2. Diagnose condition (?(0) as an error instead of crashing on matching.
3133    
3134    
3135    Version 3.3 01-Aug-00
3136    ---------------------
3137    
3138    1. If an octal character was given, but the value was greater than \377, it
3139    was not getting masked to the least significant bits, as documented. This could
3140    lead to crashes in some systems.
3141    
3142    2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
3143    the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
3144    
3145    3. Added the functions pcre_free_substring() and pcre_free_substring_list().
3146    These just pass their arguments on to (pcre_free)(), but they are provided
3147    because some uses of PCRE bind it to non-C systems that can call its functions,
3148    but cannot call free() or pcre_free() directly.
3149    
3150    4. Add "make test" as a synonym for "make check". Corrected some comments in
3151    the Makefile.
3152    
3153    5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
3154    Makefile.
3155    
3156    6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
3157    command called pgrep for grepping around the active processes.
3158    
3159    7. Added the beginnings of support for UTF-8 character strings.
3160    
3161    8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
3162    RANLIB to ./ltconfig so that they are used by libtool. I think these are all
3163    the relevant ones. (AR is not passed because ./ltconfig does its own figuring
3164    out for the ar command.)
3165    
3166    
3167    Version 3.2 12-May-00
3168    ---------------------
3169    
3170    This is purely a bug fixing release.
3171    
3172    1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
3173    of ZA. This was just one example of several cases that could provoke this bug,
3174    which was introduced by change 9 of version 2.00. The code for breaking
3175    infinite loops after an iteration that matches an empty string was't working
3176    correctly.
3177    
3178    2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
3179    when matched against abbab (for example). After matching an empty string, it
3180    wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
3181    caused it to match further down the string than it should.
3182    
3183    3. The code contained an inclusion of sys/types.h. It isn't clear why this
3184    was there because it doesn't seem to be needed, and it causes trouble on some
3185    systems, as it is not a Standard C header. It has been removed.
3186    
3187    4. Made 4 silly changes to the source to avoid stupid compiler warnings that
3188    were reported on the Macintosh. The changes were from
3189    
3190      while ((c = *(++ptr)) != 0 && c != '\n');
3191    to
3192      while ((c = *(++ptr)) != 0 && c != '\n') ;
3193    
3194    Totally extraordinary, but if that's what it takes...
3195    
3196    5. PCRE is being used in one environment where neither memmove() nor bcopy() is
3197    available. Added HAVE_BCOPY and an autoconf test for it; if neither
3198    HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
3199    assumes the way PCRE uses memmove() (always moving upwards).
3200    
3201    6. PCRE is being used in one environment where strchr() is not available. There
3202    was only one use in pcre.c, and writing it out to avoid strchr() probably gives
3203    faster code anyway.
3204    
3205    
3206    Version 3.1 09-Feb-00
3207    ---------------------
3208    
3209    The only change in this release is the fixing of some bugs in Makefile.in for
3210    the "install" target:
3211    
3212    (1) It was failing to install pcreposix.h.
3213    
3214    (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
3215    
3216    
3217    Version 3.0 01-Feb-00
3218    ---------------------
3219    
3220    1. Add support for the /+ modifier to perltest (to output $` like it does in
3221    pcretest).
3222    
3223    2. Add support for the /g modifier to perltest.
3224    
3225    3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
3226    matches null strings.
3227    
3228    4. Fix perltest so that it doesn't do unwanted things when fed an empty
3229    pattern. Perl treats empty patterns specially - it reuses the most recent
3230    pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
3231    effect.
3232    
3233    5. The POSIX interface was broken in that it was just handing over the POSIX
3234    captured string vector to pcre_exec(), but (since release 2.00) PCRE has
3235    required a bigger vector, with some working space on the end. This means that
3236    the POSIX wrapper now has to get and free some memory, and copy the results.
3237    
3238    6. Added some simple autoconf support, placing the test data and the
3239    documentation in separate directories, re-organizing some of the
3240    information files, and making it build pcre-config (a GNU standard). Also added
3241    libtool support for building PCRE as a shared library, which is now the
3242    default.
3243    
3244    7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
3245    09 are not valid octal constants. Single digits will be used for minor values
3246    less than 10.
3247    
3248    8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
3249    existing programs that set these in the POSIX interface can use PCRE without
3250    modification.
3251    
3252    9. Added a new function, pcre_fullinfo() with an extensible interface. It can
3253    return all that pcre_info() returns, plus additional data. The pcre_info()
3254    function is retained for compatibility, but is considered to be obsolete.
3255    
3256    10. Added experimental recursion feature (?R) to handle one common case that
3257    Perl 5.6 will be able to do with (?p{...}).
3258    
3259    11. Added support for POSIX character classes like [:alpha:], which Perl is
3260    adopting.
3261    
3262    
3263    Version 2.08 31-Aug-99
3264    ----------------------
3265    
3266    1. When startoffset was not zero and the pattern began with ".*", PCRE was not
3267    trying to match at the startoffset position, but instead was moving forward to
3268    the next newline as if a previous match had failed.
3269    
3270    2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
3271    and could get into a loop if a null string was matched other than at the start
3272    of the subject.
3273    
3274    3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
3275    be distinguished at compile time, and for completeness also added PCRE_DATE.
3276    
3277    5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
3278    in GnuWin32 environments.
3279    
3280    
3281    Version 2.07 29-Jul-99
3282    ----------------------
3283    
3284    1. The documentation is now supplied in plain text form and HTML as well as in
3285    the form of man page sources.
3286    
3287    2. C++ compilers don't like assigning (void *) values to other pointer types.
3288    In particular this affects malloc(). Although there is no problem in Standard
3289    C, I've put in casts to keep C++ compilers happy.
3290    
3291    3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
3292    should be (const char *).
3293    
3294    4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
3295    be useful for non-Unix systems who don't want to bother with the POSIX stuff.
3296    However, I haven't made this a standard facility. The documentation doesn't
3297    mention it, and the Makefile doesn't support it.
3298    
3299    5. The Makefile now contains an "install" target, with editable destinations at
3300    the top of the file. The pcretest program is not installed.
3301    
3302    6. pgrep -V now gives the PCRE version number and date.
3303    
3304    7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
3305    causing the entire string to be ignored, instead of just the last character.
3306    
3307    8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
3308    non-matching string, it can take a very, very long time, even for strings of
3309    quite modest length, because of the nested recursion. PCRE now does better in
3310    some of these cases. It does this by remembering the last required literal
3311    character in the pattern, and pre-searching the subject to ensure it is present
3312    before running the real match. In other words, it applies a heuristic to detect
3313    some types of certain failure quickly, and in the above example, if presented
3314    with a string that has no trailing " it gives "no match" very quickly.
3315    
3316    9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
3317    other alternatives are tried instead.
3318    
3319    
3320    Version 2.06 09-Jun-99
3321    ----------------------
3322    
3323    1. Change pcretest's output for amount of store used to show just the code
3324    space, because the remainder (the data block) varies in size between 32-bit and
3325    64-bit systems.
3326    
3327    2. Added an extra argument to pcre_exec() to supply an offset in the subject to
3328    start matching at. This allows lookbehinds to work when searching for multiple
3329    occurrences in a string.
3330    
3331    3. Added additional options to pcretest for testing multiple occurrences:
3332    
3333       /+   outputs the rest of the string that follows a match
3334       /g   loops for multiple occurrences, using the new startoffset argument
3335       /G   loops for multiple occurrences by passing an incremented pointer
3336    
3337    4. PCRE wasn't doing the "first character" optimization for patterns starting
3338    with \b or \B, though it was doing it for other lookbehind assertions. That is,
3339    it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
3340    the letter 'x'. On long subject strings, this gives a significant speed-up.
3341    
3342    
3343    Version 2.05 21-Apr-99
3344    ----------------------
3345    
3346    1. Changed the type of magic_number from int to long int so that it works
3347    properly on 16-bit systems.
3348    
3349    2. Fixed a bug which caused patterns starting with .* not to work correctly
3350    when the subject string contained newline characters. PCRE was assuming
3351    anchoring for such patterns in all cases, which is not correct because .* will
3352    not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
3353    DOTALL is set at top level; otherwise it knows that patterns starting with .*
3354    must be retried after every newline in the subject.
3355    
3356    
3357    Version 2.04 18-Feb-99
3358    ----------------------
3359    
3360    1. For parenthesized subpatterns with repeats whose minimum was zero, the
3361    computation of the store needed to hold the pattern was incorrect (too large).
3362    If such patterns were nested a few deep, this could multiply and become a real
3363    problem.
3364    
3365    2. Added /M option to pcretest to show the memory requirement of a specific
3366    pattern. Made -m a synonym of -s (which does this globally) for compatibility.
3367    
3368    3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
3369    compiled in such a way that the backtracking after subsequent failure was
3370    pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
3371    ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
3372    
3373    
3374    Version 2.03 02-Feb-99
3375    ----------------------
3376    
3377    1. Fixed typo and small mistake in man page.
3378    
3379    2. Added 4th condition (GPL supersedes if conflict) and created separate
3380    LICENCE file containing the conditions.
3381    
3382    3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
3383    Perl, that is the internal \ allows the delimiter to be included in the
3384    pattern. Locked out the use of \ as a delimiter. If \ immediately follows
3385    the final delimiter, add \ to the end of the pattern (to test the error).
3386    
3387    4. Added the convenience functions for extracting substrings after a successful
3388    match. Updated pcretest to make it able to test these functions.
3389    
3390    
3391    Version 2.02 14-Jan-99
3392    ----------------------
3393    
3394    1. Initialized the working variables associated with each extraction so that
3395    their saving and restoring doesn't refer to uninitialized store.
3396    
3397    2. Put dummy code into study.c in order to trick the optimizer of the IBM C
3398    compiler for OS/2 into generating correct code. Apparently IBM isn't going to
3399    fix the problem.
3400    
3401    3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
3402    calls, and wasn't printing the correct value for compiling calls. Increased the
3403    default value of LOOPREPEAT, and the number of significant figures in the
3404    times.
3405    
3406    4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
3407    
3408    5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
3409    a building problem on Windows NT with a FAT file system.
3410    
3411    
3412    Version 2.01 21-Oct-98
3413    ----------------------
3414    
3415    1. Changed the API for pcre_compile() to allow for the provision of a pointer
3416    to character tables built by pcre_maketables() in the current locale. If NULL
3417    is passed, the default tables are used.
3418    
3419    
3420    Version 2.00 24-Sep-98
3421    ----------------------
3422    
3423    1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
3424    it any more.
3425    
3426    2. Allow quantification of (?>) groups, and make it work correctly.
3427    
3428    3. The first character computation wasn't working for (?>) groups.
3429    
3430    4. Correct the implementation of \Z (it is permitted to match on the \n at the
3431    end of the subject) and add 5.005's \z, which really does match only at the
3432    very end of the subject.
3433    
3434    5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
3435    
3436    6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
3437    DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
3438    localized options. All options to pcre_study() were also removed.
3439    
3440    7. Add other new features from 5.005:
3441    
3442       $(?<=           positive lookbehind
3443       $(?<!           negative lookbehind
3444       (?imsx-imsx)    added the unsetting capability
3445                       such a setting is global if at outer level; local otherwise
3446       (?imsx-imsx:)   non-capturing groups with option setting
3447       (?(cond)re|re)  conditional pattern matching
3448    
3449       A backreference to itself in a repeated group matches the previous
3450       captured string.
3451    
3452    8. General tidying up of studying (both automatic and via "study")
3453    consequential on the addition of new assertions.
3454    
3455    9. As in 5.005, unlimited repeated groups that could match an empty substring
3456    are no longer faulted at compile time. Instead, the loop is forcibly broken at
3457    runtime if any iteration does actually match an empty substring.
3458    
3459    10. Include the RunTest script in the distribution.
3460    
3461    11. Added tests from the Perl 5.005_02 distribution. This showed up a few
3462    discrepancies, some of which were old and were also with respect to 5.004. They
3463    have now been fixed.
3464    
3465    
3466    Version 1.09 28-Apr-98
3467    ----------------------
3468    
3469    1. A negated single character class followed by a quantifier with a minimum
3470    value of one (e.g.  [^x]{1,6}  ) was not compiled correctly. This could lead to
3471    program crashes, or just wrong answers. This did not apply to negated classes
3472    containing more than one character, or to minima other than one.
3473    
3474    
3475    Version 1.08 27-Mar-98
3476    ----------------------
3477    
3478    1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
3479    
3480    2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
3481    latter must appear before anything that relies on it in the pattern.
3482    
3483    
3484    Version 1.07 16-Feb-98
3485    ----------------------
3486    
3487    1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
3488    repeat of a potentially empty string).
3489    
3490    
3491    Version 1.06 23-Jan-98
3492    ----------------------
3493    
3494    1. Added Markus Oberhumer's little patches for C++.
3495    
3496    2. Literal strings longer than 255 characters were broken.
3497    
3498    
3499  Version 1.05 23-Dec-97  Version 1.05 23-Dec-97
3500  ----------------------  ----------------------

Legend:
Removed from v.13  
changed lines
  Added in v.483

  ViewVC Help
Powered by ViewVC 1.1.5