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

Diff of /code/trunk/ChangeLog

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

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98        was out-of-date, and there was no check on the pcre_compile() error code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

Legend:
Removed from v.55  
changed lines
  Added in v.481

  ViewVC Help
Powered by ViewVC 1.1.5