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

Diff of /code/trunk/ChangeLog

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

revision 23 by nigel, Sat Feb 24 21:38:41 2007 UTC revision 521 by ph10, Sat May 22 19:03:35 2010 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    Version 8.10 03 May-2010
5    ------------------------
6    
7    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
8        THEN.
9    
10    2.  (*ACCEPT) was not working when inside an atomic group.
11    
12    3.  Inside a character class, \B is treated as a literal by default, but
13        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
14        causes the error). The code is unchanged, but I tidied the documentation.
15    
16    4.  Inside a character class, PCRE always treated \R and \X as literals,
17        whereas Perl faults them if its -w option is set. I have changed PCRE so
18        that it faults them when PCRE_EXTRA is set.
19    
20    5.  Added support for \N, which always matches any character other than
21        newline. (It is the same as "." when PCRE_DOTALL is not set.)
22    
23    6.  When compiling pcregrep with newer versions of gcc which may have
24        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
25        declared with attribute warn_unused_result" were given. Just casting the
26        result to (void) does not stop the warnings; a more elaborate fudge is
27        needed. I've used a macro to implement this.
28    
29    7.  Minor change to pcretest.c to avoid a compiler warning.
30    
31    8.  Added four artifical Unicode properties to help with an option to make
32        \s etc use properties (see next item). The new properties are: Xan
33        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
34    
35    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
36        use Unicode properties. (*UCP) at the start of a pattern can be used to set
37        this option. Modified pcretest to add /W to test this facility. Added
38        REG_UCP to make it available via the POSIX interface.
39    
40    10. Added --line-buffered to pcregrep.
41    
42    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
43        studied, and the match started with a letter with a code point greater than
44        127 whose first byte was different to the first byte of the other case of
45        the letter, the other case of this starting letter was not recognized.
46    
47    12. pcreposix.c included pcre.h before including pcre_internal.h. This caused a
48        conflict in the definition of PCRE_EXP_DECL. I have removed the include of
49        pcre.h as pcre_internal.h includes pcre.h itself. (This may be a bit of
50        historical tidying that never got done.)
51    
52    
53    
54    Version 8.02 19-Mar-2010
55    ------------------------
56    
57    1.  The Unicode data tables have been updated to Unicode 5.2.0.
58    
59    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
60        configured.
61    
62    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
63        original author of that file, following a query about its status.
64    
65    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
66        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
67    
68    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
69        quantifier applied to a forward-referencing subroutine call, could compile
70        incorrect code or give the error "internal error: previously-checked
71        referenced subpattern not found".
72    
73    6.  Both MS Visual Studio and Symbian OS have problems with initializing
74        variables to point to external functions. For these systems, therefore,
75        pcre_malloc etc. are now initialized to local functions that call the
76        relevant global functions.
77    
78    7.  There were two entries missing in the vectors called coptable and poptable
79        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
80        I've fixed the data, and added a kludgy way of testing at compile time that
81        the lengths are correct (equal to the number of opcodes).
82    
83    8.  Following on from 7, I added a similar kludge to check the length of the
84        eint vector in pcreposix.c.
85    
86    9.  Error texts for pcre_compile() are held as one long string to avoid too
87        much relocation at load time. To find a text, the string is searched,
88        counting zeros. There was no check for running off the end of the string,
89        which could happen if a new error number was added without updating the
90        string.
91    
92    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
93    
94    11. \K was not working if it appeared in an atomic group or in a group that
95        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
96        \K is "not well defined" if used in an assertion. PCRE now accepts it if
97        the assertion is positive, but not if it is negative.
98    
99    12. Change 11 fortuitously reduced the size of the stack frame used in the
100        "match()" function of pcre_exec.c by one pointer. Forthcoming
101        implementation of support for (*MARK) will need an extra pointer on the
102        stack; I have reserved it now, so that the stack frame size does not
103        decrease.
104    
105    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
106        item in branch that calls a recursion is a subroutine call - as in the
107        second branch in the above example - was incorrectly given the compile-
108        time error "recursive call could loop indefinitely" because pcre_compile()
109        was not correctly checking the subroutine for matching a non-empty string.
110    
111    14. The checks for overrunning compiling workspace could trigger after an
112        overrun had occurred. This is a "should never occur" error, but it can be
113        triggered by pathological patterns such as hundreds of nested parentheses.
114        The checks now trigger 100 bytes before the end of the workspace.
115    
116    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
117    
118    
119    Version 8.01 19-Jan-2010
120    ------------------------
121    
122    1.  If a pattern contained a conditional subpattern with only one branch (in
123        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
124        computed the wrong minimum data length (which is of course zero for such
125        subpatterns). This could cause incorrect "no match" results.
126    
127    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
128        the pattern is reset in the first branch, pcre_compile() failed with
129        "internal error: code overflow at offset...". This happened only when
130        the reset was to the original external option setting. (An optimization
131        abstracts leading options settings into an external setting, which was the
132        cause of this.)
133    
134    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
135        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
136        assertion pattern did not match (meaning that the assertion was true), it
137        was incorrectly treated as false if the SKIP had been reached during the
138        matching. This also applied to assertions used as conditions.
139    
140    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
141        assertion subpattern, including such a pattern used as a condition,
142        unpredictable results occurred, instead of the error return
143        PCRE_ERROR_DFA_UITEM.
144    
145    5.  The C++ GlobalReplace function was not working like Perl for the special
146        situation when an empty string is matched. It now does the fancy magic
147        stuff that is necessary.
148    
149    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
150        removed. (These were left over from very, very early versions of PCRE.)
151    
152    7.  Some cosmetic changes to the code to make life easier when compiling it
153        as part of something else:
154    
155        (a) Change DEBUG to PCRE_DEBUG.
156    
157        (b) In pcre_compile(), rename the member of the "branch_chain" structure
158            called "current" as "current_branch", to prevent a collision with the
159            Linux macro when compiled as a kernel module.
160    
161        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
162            prevent a collision with the Linux macro when compiled as a kernel
163            module.
164    
165    8.  In pcre_compile() there are some checks for integer overflows that used to
166        cast potentially large values to (double). This has been changed to that
167        when building, a check for int64_t is made, and if it is found, it is used
168        instead, thus avoiding the use of floating point arithmetic. (There is no
169        other use of FP in PCRE.) If int64_t is not found, the fallback is to
170        double.
171    
172    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
173        2005 (difference between two addresses compared to an unsigned value).
174    
175    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
176        custom one, because of the following reported problem in Windows:
177    
178          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
179              under Win32.
180          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
181              therefore missing the function definition.
182          - The compiler thus generates a "C" signature for the test function.
183          - The linker fails to find the "C" function.
184          - PCRE fails to configure if asked to do so against libbz2.
185    
186    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
187        messages were output:
188    
189          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
190          rerunning libtoolize, to keep the correct libtool macros in-tree.
191          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
192    
193        I have done both of these things.
194    
195    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
196        most of the time, it *can* run out if it is given a pattern that contains a
197        runaway infinite recursion. I updated the discussion in the pcrestack man
198        page.
199    
200    13. Now that we have gone to the x.xx style of version numbers, the minor
201        version may start with zero. Using 08 or 09 is a bad idea because users
202        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
203        interpreted as invalid octal numbers. I've updated the previous comment in
204        configure.ac, and also added a check that gives an error if 08 or 09 are
205        used.
206    
207    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
208        causing partial matching to fail when the end of the subject matched \W
209        in a UTF-8 pattern where \W was quantified with a minimum of 3.
210    
211    15. There were some discrepancies between the declarations in pcre_internal.h
212        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
213        their definitions. The declarations used "const uschar *" and the
214        definitions used USPTR. Even though USPTR is normally defined as "const
215        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
216        reported that: "This difference in casting confuses some C++ compilers, for
217        example, SunCC recognizes above declarations as different functions and
218        generates broken code for hbpcre." I have changed the declarations to use
219        USPTR.
220    
221    16. GNU libtool is named differently on some systems. The autogen.sh script now
222        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
223        (FreeBSD).
224    
225    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
226        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
227        comment: "Figure out how to create a longlong from a string: strtoll and
228        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
229        instance, but it only takes 2 args instead of 3!"
230    
231    18. A subtle bug concerned with back references has been fixed by a change of
232        specification, with a corresponding code fix. A pattern such as
233        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
234        refers, was giving matches when it shouldn't. For example, xa=xaaa would
235        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
236        same bug. Such groups have to be quantified to be useful, or contained
237        inside another quantified group. (If there's no repetition, the reference
238        can never match.) The problem arises because, having left the group and
239        moved on to the rest of the pattern, a later failure that backtracks into
240        the group uses the captured value from the final iteration of the group
241        rather than the correct earlier one. I have fixed this in PCRE by forcing
242        any group that contains a reference to itself to be an atomic group; that
243        is, there cannot be any backtracking into it once it has completed. This is
244        similar to recursive and subroutine calls.
245    
246    
247    Version 8.00 19-Oct-09
248    ----------------------
249    
250    1.  The table for translating pcre_compile() error codes into POSIX error codes
251        was out-of-date, and there was no check on the pcre_compile() error code
252        being within the table. This could lead to an OK return being given in
253        error.
254    
255    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
256        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
257        in a Windows environment.
258    
259    3.  The pcregrep --count option prints the count for each file even when it is
260        zero, as does GNU grep. However, pcregrep was also printing all files when
261        --files-with-matches was added. Now, when both options are given, it prints
262        counts only for those files that have at least one match. (GNU grep just
263        prints the file name in this circumstance, but including the count seems
264        more useful - otherwise, why use --count?) Also ensured that the
265        combination -clh just lists non-zero counts, with no names.
266    
267    4.  The long form of the pcregrep -F option was incorrectly implemented as
268        --fixed_strings instead of --fixed-strings. This is an incompatible change,
269        but it seems right to fix it, and I didn't think it was worth preserving
270        the old behaviour.
271    
272    5.  The command line items --regex=pattern and --regexp=pattern were not
273        recognized by pcregrep, which required --regex pattern or --regexp pattern
274        (with a space rather than an '='). The man page documented the '=' forms,
275        which are compatible with GNU grep; these now work.
276    
277    6.  No libpcreposix.pc file was created for pkg-config; there was just
278        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
279    
280    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
281        when UCP support is not needed, by modifying the Python script that
282        generates it from Unicode data files. This should not matter if the module
283        is correctly used as a library, but I received one complaint about 50K of
284        unwanted data. My guess is that the person linked everything into his
285        program rather than using a library. Anyway, it does no harm.
286    
287    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
288        was a minimum greater than 1 for a wide character in a possessive
289        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
290        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
291        character. Chaos in the form of incorrect output or a compiling loop could
292        result.
293    
294    9.  The restrictions on what a pattern can contain when partial matching is
295        requested for pcre_exec() have been removed. All patterns can now be
296        partially matched by this function. In addition, if there are at least two
297        slots in the offset vector, the offset of the earliest inspected character
298        for the match and the offset of the end of the subject are set in them when
299        PCRE_ERROR_PARTIAL is returned.
300    
301    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
302        synonymous with PCRE_PARTIAL, for backwards compatibility, and
303        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
304        and may be more useful for multi-segment matching.
305    
306    11. Partial matching with pcre_exec() is now more intuitive. A partial match
307        used to be given if ever the end of the subject was reached; now it is
308        given only if matching could not proceed because another character was
309        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
310        string "Z", which now yields "no match" instead of "partial match". In the
311        case of pcre_dfa_exec(), "no match" is given if every matching path for the
312        final character ended with (*FAIL).
313    
314    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
315        if the pattern had a "must contain" character that was already found in the
316        earlier partial match, unless partial matching was again requested. For
317        example, with the pattern /dog.(body)?/, the "must contain" character is
318        "g". If the first part-match was for the string "dog", restarting with
319        "sbody" failed. This bug has been fixed.
320    
321    13. The string returned by pcre_dfa_exec() after a partial match has been
322        changed so that it starts at the first inspected character rather than the
323        first character of the match. This makes a difference only if the pattern
324        starts with a lookbehind assertion or \b or \B (\K is not supported by
325        pcre_dfa_exec()). It's an incompatible change, but it makes the two
326        matching functions compatible, and I think it's the right thing to do.
327    
328    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
329        so that the demonstration program is easily available in environments where
330        PCRE has not been installed from source.
331    
332    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
333        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
334        library.
335    
336    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
337        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
338        is not the first non-POSIX option to be added. Clearly some people find
339        these options useful.
340    
341    17. If a caller to the POSIX matching function regexec() passes a non-zero
342        value for nmatch with a NULL value for pmatch, the value of
343        nmatch is forced to zero.
344    
345    18. RunGrepTest did not have a test for the availability of the -u option of
346        the diff command, as RunTest does. It now checks in the same way as
347        RunTest, and also checks for the -b option.
348    
349    19. If an odd number of negated classes containing just a single character
350        interposed, within parentheses, between a forward reference to a named
351        subpattern and the definition of the subpattern, compilation crashed with
352        an internal error, complaining that it could not find the referenced
353        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
354        [The bug was that it was starting one character too far in when skipping
355        over the character class, thus treating the ] as data rather than
356        terminating the class. This meant it could skip too much.]
357    
358    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
359        /g option in pcretest when the pattern contains \K, which makes it possible
360        to have an empty string match not at the start, even when the pattern is
361        anchored. Updated pcretest and pcredemo to use this option.
362    
363    21. If the maximum number of capturing subpatterns in a recursion was greater
364        than the maximum at the outer level, the higher number was returned, but
365        with unset values at the outer level. The correct (outer level) value is
366        now given.
367    
368    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
369        PCRE did not set those parentheses (unlike Perl). I have now found a way to
370        make it do so. The string so far is captured, making this feature
371        compatible with Perl.
372    
373    23. The tests have been re-organized, adding tests 11 and 12, to make it
374        possible to check the Perl 5.10 features against Perl 5.10.
375    
376    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
377        pattern matches a fixed length string. PCRE did not allow this; now it
378        does. Neither allows recursion.
379    
380    25. I finally figured out how to implement a request to provide the minimum
381        length of subject string that was needed in order to match a given pattern.
382        (It was back references and recursion that I had previously got hung up
383        on.) This code has now been added to pcre_study(); it finds a lower bound
384        to the length of subject needed. It is not necessarily the greatest lower
385        bound, but using it to avoid searching strings that are too short does give
386        some useful speed-ups. The value is available to calling programs via
387        pcre_fullinfo().
388    
389    26. While implementing 25, I discovered to my embarrassment that pcretest had
390        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
391        study optimizations had never been tested with that matching function.
392        Oops. What is worse, even when it was passed study data, there was a bug in
393        pcre_dfa_exec() that meant it never actually used it. Double oops. There
394        were also very few tests of studied patterns with pcre_dfa_exec().
395    
396    27. If (?| is used to create subpatterns with duplicate numbers, they are now
397        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
398        on the other side of the coin, they are no longer allowed to have different
399        names, because these cannot be distinguished in PCRE, and this has caused
400        confusion. (This is a difference from Perl.)
401    
402    28. When duplicate subpattern names are present (necessarily with different
403        numbers, as required by 27 above), and a test is made by name in a
404        conditional pattern, either for a subpattern having been matched, or for
405        recursion in such a pattern, all the associated numbered subpatterns are
406        tested, and the overall condition is true if the condition is true for any
407        one of them. This is the way Perl works, and is also more like the way
408        testing by number works.
409    
410    
411    Version 7.9 11-Apr-09
412    ---------------------
413    
414    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
415        (pcretest), all targets were linked against these libraries. This included
416        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
417        libraries. This caused unwanted dependencies to be created. This problem
418        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
419        pcretest is linked with readline.
420    
421    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
422        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
423        moved outside it again, because FALSE and TRUE are already defined in AIX,
424        but BOOL is not.
425    
426    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
427        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
428    
429    4.  The pcregrep documentation said spaces were inserted as well as colons (or
430        hyphens) following file names and line numbers when outputting matching
431        lines. This is not true; no spaces are inserted. I have also clarified the
432        wording for the --colour (or --color) option.
433    
434    5.  In pcregrep, when --colour was used with -o, the list of matching strings
435        was not coloured; this is different to GNU grep, so I have changed it to be
436        the same.
437    
438    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
439        each matching line was coloured. Now it goes on to look for further matches
440        of any of the test patterns, which is the same behaviour as GNU grep.
441    
442    7.  A pattern that could match an empty string could cause pcregrep to loop; it
443        doesn't make sense to accept an empty string match in pcregrep, so I have
444        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
445        seems to be how GNU grep behaves.
446    
447    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
448        start or after a newline", because the conditional assertion was not being
449        correctly handled. The rule now is that both the assertion and what follows
450        in the first alternative must satisfy the test.
451    
452    9.  If auto-callout was enabled in a pattern with a conditional group whose
453        condition was an assertion, PCRE could crash during matching, both with
454        pcre_exec() and pcre_dfa_exec().
455    
456    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
457        used for matching.
458    
459    11. Unicode property support in character classes was not working for
460        characters (bytes) greater than 127 when not in UTF-8 mode.
461    
462    12. Added the -M command line option to pcretest.
463    
464    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
465    
466    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
467    
468    16. Added comments and documentation about mis-use of no_arg in the C++
469        wrapper.
470    
471    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
472        from Martin Jerabek that uses macro names for all relevant character and
473        string constants.
474    
475    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
476        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
477        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
478        these, but not everybody uses configure.
479    
480    19. A conditional group that had only one branch was not being correctly
481        recognized as an item that could match an empty string. This meant that an
482        enclosing group might also not be so recognized, causing infinite looping
483        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
484        with the subject "ab", where knowledge that the repeated group can match
485        nothing is needed in order to break the loop.
486    
487    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
488        exec(), but without supplying a callout function, matching went wrong.
489    
490    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
491        leak if the size of the offset vector was greater than 30. When the vector
492        is smaller, the saved offsets during recursion go onto a local stack
493        vector, but for larger vectors malloc() is used. It was failing to free
494        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
495        error, in fact).
496    
497    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
498        heapframe that is used only when UTF-8 support is enabled. This caused no
499        problem, but was untidy.
500    
501    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
502        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
503        included within another project.
504    
505    24. Steven Van Ingelgem's patches to add more options to the CMake support,
506        slightly modified by me:
507    
508          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
509              not building pcregrep.
510    
511          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
512              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
513    
514    25. Forward references, both numeric and by name, in patterns that made use of
515        duplicate group numbers, could behave incorrectly or give incorrect errors,
516        because when scanning forward to find the reference group, PCRE was not
517        taking into account the duplicate group numbers. A pattern such as
518        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
519    
520    26. Changed a few more instances of "const unsigned char *" to USPTR, making
521        the feature of a custom pointer more persuasive (as requested by a user).
522    
523    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
524        pcretest.c, inside #ifndefs, because it seems they are sometimes already
525        pre-defined.
526    
527    28. Added support for (*UTF8) at the start of a pattern.
528    
529    29. Arrange for flags added by the "release type" setting in CMake to be shown
530        in the configuration summary.
531    
532    
533    Version 7.8 05-Sep-08
534    ---------------------
535    
536    1.  Replaced UCP searching code with optimized version as implemented for Ad
537        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
538        stage table and inline lookup instead of a function, giving speed ups of 2
539        to 5 times on some simple patterns that I tested. Permission was given to
540        distribute the MultiStage2.py script that generates the tables (it's not in
541        the tarball, but is in the Subversion repository).
542    
543    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
544        scripts.
545    
546    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
547        a group with a zero qualifier. The result of the study could be incorrect,
548        or the function might crash, depending on the pattern.
549    
550    4.  Caseless matching was not working for non-ASCII characters in back
551        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
552        It now works when Unicode Property Support is available.
553    
554    5.  In pcretest, an escape such as \x{de} in the data was always generating
555        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
556        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
557        truncation.
558    
559    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
560    
561    7.  Added two (int) casts to pcregrep when printing the difference of two
562        pointers, in case they are 64-bit values.
563    
564    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
565        test 2 if it fails.
566    
567    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
568        and a #define of that name to empty if it is not externally set. This is to
569        allow users of MSVC to set it if necessary.
570    
571    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
572        the convenience functions in the pcre_get.c source file.
573    
574    11. An option change at the start of a pattern that had top-level alternatives
575        could cause overwriting and/or a crash. This command provoked a crash in
576        some environments:
577    
578          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
579    
580        This potential security problem was recorded as CVE-2008-2371.
581    
582    12. For a pattern where the match had to start at the beginning or immediately
583        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
584        pcre_dfa_exec() could read past the end of the passed subject if there was
585        no match. To help with detecting such bugs (e.g. with valgrind), I modified
586        pcretest so that it places the subject at the end of its malloc-ed buffer.
587    
588    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
589        exec() might read past the end of the data buffer in UTF-8 mode.
590    
591    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
592        the data contained the byte 0x85 as part of a UTF-8 character within its
593        first line. This applied both to normal and DFA matching.
594    
595    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
596        /^[^d]*?$/8 failed to match "abc".
597    
598    16. Added a missing copyright notice to pcrecpp_internal.h.
599    
600    17. Make it more clear in the documentation that values returned from
601        pcre_exec() in ovector are byte offsets, not character counts.
602    
603    18. Tidied a few places to stop certain compilers from issuing warnings.
604    
605    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
606        supplied by Stefan Weber. I made a further small update for 7.8 because
607        there is a change of source arrangements: the pcre_searchfuncs.c module is
608        replaced by pcre_ucd.c.
609    
610    
611    Version 7.7 07-May-08
612    ---------------------
613    
614    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
615        a string to a long long, pretend we don't even have a long long." This is
616        done by checking for the strtoq, strtoll, and _strtoi64 functions.
617    
618    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
619        pre-7.6 versions, which defined a global no_arg variable instead of putting
620        it in the RE class. (See also #8 below.)
621    
622    3.  Remove a line of dead code, identified by coverity and reported by Nuno
623        Lopes.
624    
625    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
626    
627        (1) The include/exclude patterns were being applied to the whole pathnames
628            of files, instead of just to the final components.
629    
630        (2) If there was more than one level of directory, the subdirectories were
631            skipped unless they satisfied the include/exclude conditions. This is
632            inconsistent with GNU grep (and could even be seen as contrary to the
633            pcregrep specification - which I improved to make it absolutely clear).
634            The action now is always to scan all levels of directory, and just
635            apply the include/exclude patterns to regular files.
636    
637    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
638        --exclude_dir in the tests to avoid scanning .svn directories.
639    
640    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
641        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
642        doesn't support NULs in patterns.
643    
644    7.  Added some missing "const"s to declarations of static tables in
645        pcre_compile.c and pcre_dfa_exec.c.
646    
647    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
648        caused by fix #2  above. (Subsequently also a second patch to fix the
649        first patch. And a third patch - this was a messy problem.)
650    
651    9.  Applied Craig's patch to remove the use of push_back().
652    
653    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
654        matching function regexec().
655    
656    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
657        which, however, unlike Perl's \g{...}, are subroutine calls, not back
658        references. PCRE supports relative numbers with this syntax (I don't think
659        Oniguruma does).
660    
661    12. Previously, a group with a zero repeat such as (...){0} was completely
662        omitted from the compiled regex. However, this means that if the group
663        was called as a subroutine from elsewhere in the pattern, things went wrong
664        (an internal error was given). Such groups are now left in the compiled
665        pattern, with a new opcode that causes them to be skipped at execution
666        time.
667    
668    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
669        to the way PCRE behaves:
670    
671        (a) A lone ] character is dis-allowed (Perl treats it as data).
672    
673        (b) A back reference to an unmatched subpattern matches an empty string
674            (Perl fails the current match path).
675    
676        (c) A data ] in a character class must be notated as \] because if the
677            first data character in a class is ], it defines an empty class. (In
678            Perl it is not possible to have an empty class.) The empty class []
679            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
680            The negative empty class [^] matches any one character, independently
681            of the DOTALL setting.
682    
683    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
684        non-existent subpattern following a character class starting with ']' and
685        containing () gave an internal compiling error instead of "reference to
686        non-existent subpattern". Fortunately, when the pattern did exist, the
687        compiled code was correct. (When scanning forwards to check for the
688        existencd of the subpattern, it was treating the data ']' as terminating
689        the class, so got the count wrong. When actually compiling, the reference
690        was subsequently set up correctly.)
691    
692    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
693        it was being rejected as not supported by pcre_dfa_exec(), even though
694        other assertions are supported. I have made pcre_dfa_exec() support
695        (*FAIL).
696    
697    16. The implementation of 13c above involved the invention of a new opcode,
698        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
699        cannot be changed at match time, I realized I could make a small
700        improvement to matching performance by compiling OP_ALLANY instead of
701        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
702        on the OP_ANY path.
703    
704    17. Compiling pcretest on Windows with readline support failed without the
705        following two fixes: (1) Make the unistd.h include conditional on
706        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
707    
708    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
709        ncurses library to be included for pcretest when ReadLine support is
710        requested, but also to allow for it to be overridden. This patch came from
711        Daniel Bergström.
712    
713    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
714        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
715        any errors with the current Unicode tables. Thanks to Peter Kankowski for
716        spotting this.
717    
718    
719    Version 7.6 28-Jan-08
720    ---------------------
721    
722    1.  A character class containing a very large number of characters with
723        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
724        overflow.
725    
726    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
727        HAVE_LONG_LONG is not defined.
728    
729    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
730        bring them up to date and include new features. This patch includes:
731    
732        - Fixed PH's badly added libz and libbz2 support.
733        - Fixed a problem with static linking.
734        - Added pcredemo. [But later removed - see 7 below.]
735        - Fixed dftables problem and added an option.
736        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
737            HAVE_LONG_LONG.
738        - Added readline support for pcretest.
739        - Added an listing of the option settings after cmake has run.
740    
741    4.  A user submitted a patch to Makefile that makes it easy to create
742        "pcre.dll" under mingw when using Configure/Make. I added stuff to
743        Makefile.am that cause it to include this special target, without
744        affecting anything else. Note that the same mingw target plus all
745        the other distribution libraries and programs are now supported
746        when configuring with CMake (see 6 below) instead of with
747        Configure/Make.
748    
749    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
750        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
751        exported in the Windows port". It has not yet been confirmed that the patch
752        solves the problem, but it does no harm.
753    
754    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
755        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
756        with CMake, and also correct the comment about stack recursion.
757    
758    7.  Remove the automatic building of pcredemo from the ./configure system and
759        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
760        of a program that users should build themselves after PCRE is installed, so
761        building it automatically is not really right. What is more, it gave
762        trouble in some build environments.
763    
764    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
765    
766    
767    Version 7.5 10-Jan-08
768    ---------------------
769    
770    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
771        values in parens when parsing an RE using the C++ wrapper."
772    
773    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
774        Characters greater than 255 were excluded from the class instead of being
775        included.
776    
777    3.  The same bug as (2) above applied to negated POSIX classes such as
778        [:^space:].
779    
780    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
781        defined or documented. It seems to have been a typo for PCRE_STATIC, so
782        I have changed it.
783    
784    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
785        first named subpattern) and a construct such as (?&a) would reference the
786        first named subpattern whose name started with "a" (in other words, the
787        length check was missing). Both these problems are fixed. "Subpattern name
788        expected" is now given for (?&) (a zero-length name), and this patch also
789        makes it give the same error for \k'' (previously it complained that that
790        was a reference to a non-existent subpattern).
791    
792    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
793        this is right because (?- can be followed by option settings as well as by
794        digits. I have, however, made the messages clearer.
795    
796    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
797        than the number used in the conditional) now cause a compile-time error.
798        This is actually not compatible with Perl, which accepts such patterns, but
799        treats the conditional as always being FALSE (as PCRE used to), but it
800        seems to me that giving a diagnostic is better.
801    
802    8.  Change "alphameric" to the more common word "alphanumeric" in comments
803        and messages.
804    
805    9.  Fix two occurrences of "backslash" in comments that should have been
806        "backspace".
807    
808    10. Remove two redundant lines of code that can never be obeyed (their function
809        was moved elsewhere).
810    
811    11. The program that makes PCRE's Unicode character property table had a bug
812        which caused it to generate incorrect table entries for sequences of
813        characters that have the same character type, but are in different scripts.
814        It amalgamated them into a single range, with the script of the first of
815        them. In other words, some characters were in the wrong script. There were
816        thirteen such cases, affecting characters in the following ranges:
817    
818          U+002b0 - U+002c1
819          U+0060c - U+0060d
820          U+0061e - U+00612
821          U+0064b - U+0065e
822          U+0074d - U+0076d
823          U+01800 - U+01805
824          U+01d00 - U+01d77
825          U+01d9b - U+01dbf
826          U+0200b - U+0200f
827          U+030fc - U+030fe
828          U+03260 - U+0327f
829          U+0fb46 - U+0fbb1
830          U+10450 - U+1049d
831    
832    12. The -o option (show only the matching part of a line) for pcregrep was not
833        compatible with GNU grep in that, if there was more than one match in a
834        line, it showed only the first of them. It now behaves in the same way as
835        GNU grep.
836    
837    13. If the -o and -v options were combined for pcregrep, it printed a blank
838        line for every non-matching line. GNU grep prints nothing, and pcregrep now
839        does the same. The return code can be used to tell if there were any
840        non-matching lines.
841    
842    14. Added --file-offsets and --line-offsets to pcregrep.
843    
844    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
845        infinitely looping recursion. The bug was that positive lookaheads were not
846        being skipped when checking for a possible empty match (negative lookaheads
847        and both kinds of lookbehind were skipped).
848    
849    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
850        inclusion of <windows.h> to before rather than after the definition of
851        INVALID_FILE_ATTRIBUTES (patch from David Byron).
852    
853    17. Specifying a possessive quantifier with a specific limit for a Unicode
854        character property caused pcre_compile() to compile bad code, which led at
855        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
856        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
857        caused the error; without that there was no problem.
858    
859    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
860    
861    19. Added --enable-pcretest-libreadline.
862    
863    20. In pcrecpp.cc, the variable 'count' was incremented twice in
864        RE::GlobalReplace(). As a result, the number of replacements returned was
865        double what it should be. I removed one of the increments, but Craig sent a
866        later patch that removed the other one (the right fix) and added unit tests
867        that check the return values (which was not done before).
868    
869    21. Several CMake things:
870    
871        (1) Arranged that, when cmake is used on Unix, the libraries end up with
872            the names libpcre and libpcreposix, not just pcre and pcreposix.
873    
874        (2) The above change means that pcretest and pcregrep are now correctly
875            linked with the newly-built libraries, not previously installed ones.
876    
877        (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
878    
879    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
880        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
881        UTF-8 newline character). The key issue is that the pattern starts .*;
882        this means that the match must be either at the beginning, or after a
883        newline. The bug was in the code for advancing after a failed match and
884        checking that the new position followed a newline. It was not taking
885        account of UTF-8 characters correctly.
886    
887    23. PCRE was behaving differently from Perl in the way it recognized POSIX
888        character classes. PCRE was not treating the sequence [:...:] as a
889        character class unless the ... were all letters. Perl, however, seems to
890        allow any characters between [: and :], though of course it rejects as
891        unknown any "names" that contain non-letters, because all the known class
892        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
893        for example, whereas PCRE did not - it did not recognize a POSIX character
894        class. This seemed a bit dangerous, so the code has been changed to be
895        closer to Perl. The behaviour is not identical to Perl, because PCRE will
896        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
897        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
898        Perl does, and where it didn't before.
899    
900    24. Rewrite so as to remove the single use of %n from pcregrep because in some
901        Windows environments %n is disabled by default.
902    
903    
904    Version 7.4 21-Sep-07
905    ---------------------
906    
907    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
908        means that a class such as [\s] counted as "explicit reference to CR or
909        LF". That isn't really right - the whole point of the change was to try to
910        help when there was an actual mention of one of the two characters. So now
911        the change happens only if \r or \n (or a literal CR or LF) character is
912        encountered.
913    
914    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
915        of both had grown to the point where there were only 3 bits left.
916        Fortunately, there was spare space in the data structure, and so I have
917        moved the internal flags into a new 16-bit field to free up more option
918        bits.
919    
920    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
921        but did not set the internal JCHANGED flag - either of these is enough to
922        control the way the "get" function works - but the PCRE_INFO_JCHANGED
923        facility is supposed to tell if (?J) was ever used, so now (?J) at the
924        start sets both bits.
925    
926    4.  Added options (at build time, compile time, exec time) to change \R from
927        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
928    
929    5.  doc/pcresyntax.html was missing from the distribution.
930    
931    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
932        compatibility, even though it is no longer used.
933    
934    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
935        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
936        windows.h file is present (where different names are used). [This was
937        reversed later after testing - see 16 below.]
938    
939    8.  Changed all #include <config.h> to #include "config.h". There were also
940        some further <pcre.h> cases that I changed to "pcre.h".
941    
942    9.  When pcregrep was used with the --colour option, it missed the line ending
943        sequence off the lines that it output.
944    
945    10. It was pointed out to me that arrays of string pointers cause lots of
946        relocations when a shared library is dynamically loaded. A technique of
947        using a single long string with a table of offsets can drastically reduce
948        these. I have refactored PCRE in four places to do this. The result is
949        dramatic:
950    
951          Originally:                          290
952          After changing UCP table:            187
953          After changing error message table:   43
954          After changing table of "verbs"       36
955          After changing table of Posix names   22
956    
957        Thanks to the folks working on Gregex for glib for this insight.
958    
959    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
960        unicode-properties was also set.
961    
962    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
963    
964    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
965        checked only for CRLF.
966    
967    14. Added casts to pcretest.c to avoid compiler warnings.
968    
969    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
970    
971    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
972        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
973        entirely. This removes changes made in 7 above.
974    
975    17. The CMake files have been updated, and there is now more information about
976        building with CMake in the NON-UNIX-USE document.
977    
978    
979    Version 7.3 28-Aug-07
980    ---------------------
981    
982     1. In the rejigging of the build system that eventually resulted in 7.1, the
983        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
984        brackets there is not right, since it causes compilers to look for an
985        installed pcre.h, not the version that is in the source that is being
986        compiled (which of course may be different). I have changed it back to:
987    
988          #include "pcre.h"
989    
990        I have a vague recollection that the change was concerned with compiling in
991        different directories, but in the new build system, that is taken care of
992        by the VPATH setting the Makefile.
993    
994     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
995        when the subject happened to end in the byte 0x85 (e.g. if the last
996        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
997        characters but of course it shouldn't be taken as a newline when it is part
998        of another character. The bug was that, for an unlimited repeat of . in
999        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
1000        characters when looking for a newline.
1001    
1002     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
1003    
1004     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
1005        in debug output.
1006    
1007     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
1008        long printing in the pcrecpp unittest when running under MinGW.
1009    
1010     6. ESC_K was left out of the EBCDIC table.
1011    
1012     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
1013        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
1014        limit also applies to "virtual nesting" when a pattern is recursive, and in
1015        this case 1000 isn't so big. I have been able to remove this limit at the
1016        expense of backing off one optimization in certain circumstances. Normally,
1017        when pcre_exec() would call its internal match() function recursively and
1018        immediately return the result unconditionally, it uses a "tail recursion"
1019        feature to save stack. However, when a subpattern that can match an empty
1020        string has an unlimited repetition quantifier, it no longer makes this
1021        optimization. That gives it a stack frame in which to save the data for
1022        checking that an empty string has been matched. Previously this was taken
1023        from the 1000-entry workspace that had been reserved. So now there is no
1024        explicit limit, but more stack is used.
1025    
1026     8. Applied Daniel's patches to solve problems with the import/export magic
1027        syntax that is required for Windows, and which was going wrong for the
1028        pcreposix and pcrecpp parts of the library. These were overlooked when this
1029        problem was solved for the main library.
1030    
1031     9. There were some crude static tests to avoid integer overflow when computing
1032        the size of patterns that contain repeated groups with explicit upper
1033        limits. As the maximum quantifier is 65535, the maximum group length was
1034        set at 30,000 so that the product of these two numbers did not overflow a
1035        32-bit integer. However, it turns out that people want to use groups that
1036        are longer than 30,000 bytes (though not repeat them that many times).
1037        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1038        made it possible to implement the integer overflow checks in a much more
1039        dynamic way, which I have now done. The artificial limitation on group
1040        length has been removed - we now have only the limit on the total length of
1041        the compiled pattern, which depends on the LINK_SIZE setting.
1042    
1043    10. Fixed a bug in the documentation for get/copy named substring when
1044        duplicate names are permitted. If none of the named substrings are set, the
1045        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1046        empty string.
1047    
1048    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1049        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1050        because the ] is interpreted as the first data character and the
1051        terminating ] is not found. PCRE has been made compatible with Perl in this
1052        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1053        cause memory overwriting.
1054    
1055    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1056        string has been matched (to stop an infinite loop). It was not recognizing
1057        a conditional subpattern that could match an empty string if that
1058        subpattern was within another subpattern. For example, it looped when
1059        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
1060        condition was not nested. This bug has been fixed.
1061    
1062    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1063        past the start of the subject in the presence of bytes with the top bit
1064        set, for example "\x8aBCD".
1065    
1066    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1067        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1068    
1069    14. Optimized (?!) to (*FAIL).
1070    
1071    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1072        This restricts code points to be within the range 0 to 0x10FFFF, excluding
1073        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1074        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1075        does: it's just the validity check that is more restrictive.
1076    
1077    16. Inserted checks for integer overflows during escape sequence (backslash)
1078        processing, and also fixed erroneous offset values for syntax errors during
1079        backslash processing.
1080    
1081    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1082        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1083    
1084    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1085        caused an overrun.
1086    
1087    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1088        something other than just ASCII characters) inside a group that had an
1089        unlimited repeat caused a loop at compile time (while checking to see
1090        whether the group could match an empty string).
1091    
1092    20. Debugging a pattern containing \p or \P could cause a crash. For example,
1093        [\P{Any}] did so. (Error in the code for printing property names.)
1094    
1095    21. An orphan \E inside a character class could cause a crash.
1096    
1097    22. A repeated capturing bracket such as (A)? could cause a wild memory
1098        reference during compilation.
1099    
1100    23. There are several functions in pcre_compile() that scan along a compiled
1101        expression for various reasons (e.g. to see if it's fixed length for look
1102        behind). There were bugs in these functions when a repeated \p or \P was
1103        present in the pattern. These operators have additional parameters compared
1104        with \d, etc, and these were not being taken into account when moving along
1105        the compiled data. Specifically:
1106    
1107        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1108            length.
1109    
1110        (b) An item such as \pL+ within a repeated group could cause crashes or
1111            loops.
1112    
1113        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1114            "reference to non-existent subpattern" error.
1115    
1116        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1117    
1118    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1119        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1120    
1121    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1122    
1123    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1124        character were causing crashes (broken optimization).
1125    
1126    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1127        \p or \P) caused a compile-time loop.
1128    
1129    28. More problems have arisen in unanchored patterns when CRLF is a valid line
1130        break. For example, the unstudied pattern [\r\n]A does not match the string
1131        "\r\nA" because change 7.0/46 below moves the current point on by two
1132        characters after failing to match at the start. However, the pattern \nA
1133        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
1134        the same is true. There doesn't seem any very clean way out of this, but
1135        what I have chosen to do makes the common cases work: PCRE now takes note
1136        of whether there can be an explicit match for \r or \n anywhere in the
1137        pattern, and if so, 7.0/46 no longer applies. As part of this change,
1138        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1139        pattern has explicit CR or LF references.
1140    
1141    29. Added (*CR) etc for changing newline setting at start of pattern.
1142    
1143    
1144    Version 7.2 19-Jun-07
1145    ---------------------
1146    
1147     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
1148        which is apparently normally available under Windows.
1149    
1150     2. Re-jig the pcregrep tests with different newline settings in an attempt
1151        to make them independent of the local environment's newline setting.
1152    
1153     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
1154    
1155     4. Some of the "internals" tests were previously cut out when the link size
1156        was not 2, because the output contained actual offsets. The recent new
1157        "Z" feature of pcretest means that these can be cut out, making the tests
1158        usable with all link sizes.
1159    
1160     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
1161        stack recursion. This gives a massive performance boost under BSD, but just
1162        a small improvement under Linux. However, it saves one field in the frame
1163        in all cases.
1164    
1165     6. Added more features from the forthcoming Perl 5.10:
1166    
1167        (a) (?-n) (where n is a string of digits) is a relative subroutine or
1168            recursion call. It refers to the nth most recently opened parentheses.
1169    
1170        (b) (?+n) is also a relative subroutine call; it refers to the nth next
1171            to be opened parentheses.
1172    
1173        (c) Conditions that refer to capturing parentheses can be specified
1174            relatively, for example, (?(-2)... or (?(+3)...
1175    
1176        (d) \K resets the start of the current match so that everything before
1177            is not part of it.
1178    
1179        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1180    
1181        (f) \g{name} is another synonym - part of Perl 5.10's unification of
1182            reference syntax.
1183    
1184        (g) (?| introduces a group in which the numbering of parentheses in each
1185            alternative starts with the same number.
1186    
1187        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
1188    
1189     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
1190        PCRE_INFO_JCHANGED.
1191    
1192     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
1193        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
1194        for detecting groups that can match an empty string.
1195    
1196     9. A pattern with a very large number of alternatives (more than several
1197        hundred) was running out of internal workspace during the pre-compile
1198        phase, where pcre_compile() figures out how much memory will be needed. A
1199        bit of new cunning has reduced the workspace needed for groups with
1200        alternatives. The 1000-alternative test pattern now uses 12 bytes of
1201        workspace instead of running out of the 4096 that are available.
1202    
1203    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
1204    
1205    11. Applied patch from Google to remove an optimization that didn't quite work.
1206        The report of the bug said:
1207    
1208          pcrecpp::RE("a*").FullMatch("aaa") matches, while
1209          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
1210          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
1211    
1212    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
1213        it matched the wrong number of bytes.
1214    
1215    
1216    Version 7.1 24-Apr-07
1217    ---------------------
1218    
1219     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
1220        that is more "standard", making use of automake and other Autotools. There
1221        is some re-arrangement of the files and adjustment of comments consequent
1222        on this.
1223    
1224     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
1225        for recursive directory scanning broke on some systems because the files
1226        are not scanned in any specific order and on different systems the order
1227        was different. A call to "sort" has been inserted into RunGrepTest for the
1228        approprate test as a short-term fix. In the longer term there may be an
1229        alternative.
1230    
1231     3. I had an email from Eric Raymond about problems translating some of PCRE's
1232        man pages to HTML (despite the fact that I distribute HTML pages, some
1233        people do their own conversions for various reasons). The problems
1234        concerned the use of low-level troff macros .br and .in. I have therefore
1235        removed all such uses from the man pages (some were redundant, some could
1236        be replaced by .nf/.fi pairs). The 132html script that I use to generate
1237        HTML has been updated to handle .nf/.fi and to complain if it encounters
1238        .br or .in.
1239    
1240     4. Updated comments in configure.ac that get placed in config.h.in and also
1241        arranged for config.h to be included in the distribution, with the name
1242        config.h.generic, for the benefit of those who have to compile without
1243        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
1244    
1245     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
1246        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
1247        makevp.bat for the current PCRE, using the additional files
1248        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
1249    
1250     6. A Windows user reported a minor discrepancy with test 2, which turned out
1251        to be caused by a trailing space on an input line that had got lost in his
1252        copy. The trailing space was an accident, so I've just removed it.
1253    
1254     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
1255        that is needed.
1256    
1257     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
1258        as "const" (a) because they are and (b) because it helps the PHP
1259        maintainers who have recently made a script to detect big data structures
1260        in the php code that should be moved to the .rodata section. I remembered
1261        to update Builducptable as well, so it won't revert if ucptable.h is ever
1262        re-created.
1263    
1264     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
1265        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
1266        order to be able to cut out the UTF-8 tables in the latter when UTF-8
1267        support is not required. This saves 1.5-2K of code, which is important in
1268        some applications.
1269    
1270        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
1271        so as not to refer to the tables, even though these functions will never be
1272        called when UTF-8 support is disabled. Otherwise there are problems with a
1273        shared library.
1274    
1275    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
1276    
1277        (a) It was defining its arguments as char * instead of void *.
1278    
1279        (b) It was assuming that all moves were upwards in memory; this was true
1280            a long time ago when I wrote it, but is no longer the case.
1281    
1282        The emulated memove() is provided for those environments that have neither
1283        memmove() nor bcopy(). I didn't think anyone used it these days, but that
1284        is clearly not the case, as these two bugs were recently reported.
1285    
1286    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
1287        and Detrail to create the HTML documentation, the .txt form of the man
1288        pages, and it removes trailing spaces from listed files. It also creates
1289        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
1290        case, it wraps all the #defines with #ifndefs. This script should be run
1291        before "make dist".
1292    
1293    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1294        with Unicode property support.
1295    
1296        (a) For a maximizing quantifier, if the two different cases of the
1297            character were of different lengths in their UTF-8 codings (there are
1298            some cases like this - I found 11), and the matching function had to
1299            back up over a mixture of the two cases, it incorrectly assumed they
1300            were both the same length.
1301    
1302        (b) When PCRE was configured to use the heap rather than the stack for
1303            recursion during matching, it was not correctly preserving the data for
1304            the other case of a UTF-8 character when checking ahead for a match
1305            while processing a minimizing repeat. If the check also involved
1306            matching a wide character, but failed, corruption could cause an
1307            erroneous result when trying to check for a repeat of the original
1308            character.
1309    
1310    13. Some tidying changes to the testing mechanism:
1311    
1312        (a) The RunTest script now detects the internal link size and whether there
1313            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1314            values substituted by "configure". (The RunGrepTest script already did
1315            this for UTF-8.) The configure.ac script no longer substitutes the
1316            relevant variables.
1317    
1318        (b) The debugging options /B and /D in pcretest show the compiled bytecode
1319            with length and offset values. This means that the output is different
1320            for different internal link sizes. Test 2 is skipped for link sizes
1321            other than 2 because of this, bypassing the problem. Unfortunately,
1322            there was also a test in test 3 (the locale tests) that used /B and
1323            failed for link sizes other than 2. Rather than cut the whole test out,
1324            I have added a new /Z option to pcretest that replaces the length and
1325            offset values with spaces. This is now used to make test 3 independent
1326            of link size. (Test 2 will be tidied up later.)
1327    
1328    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1329        segmentation fault instead of returning the appropriate error message.
1330    
1331    15. In multiline mode when the newline sequence was set to "any", the pattern
1332        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1333        This doesn't seem right; it now treats the CRLF combination as the line
1334        ending, and so does not match in that case. It's only a pattern such as ^$
1335        that would hit this one: something like ^ABC$ would have failed after \r
1336        and then tried again after \r\n.
1337    
1338    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1339        in an attempt to make files that differ only in their line terminators
1340        compare equal. This works on Linux.
1341    
1342    17. Under certain error circumstances pcregrep might try to free random memory
1343        as it exited. This is now fixed, thanks to valgrind.
1344    
1345    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1346        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1347        was because its rules for how to advance for /g after matching an empty
1348        string at the end of a line did not allow for this case. They now check for
1349        it specially.
1350    
1351    20. pcretest is supposed to handle patterns and data of any length, by
1352        extending its buffers when necessary. It was getting this wrong when the
1353        buffer for a data line had to be extended.
1354    
1355    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1356        CRLF as a newline sequence.
1357    
1358    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1359        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1360        I have nevertheless tidied it up.
1361    
1362    23. Added some casts to kill warnings from HP-UX ia64 compiler.
1363    
1364    24. Added a man page for pcre-config.
1365    
1366    
1367    Version 7.0 19-Dec-06
1368    ---------------------
1369    
1370     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1371        moving to gcc 4.1.1.
1372    
1373     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1374        sys/time.h, which is documented as needed for this function. It doesn't
1375        seem to matter on Linux, but it showed up on some releases of OS X.
1376    
1377     3. It seems that there are systems where bytes whose values are greater than
1378        127 match isprint() in the "C" locale. The "C" locale should be the
1379        default when a C program starts up. In most systems, only ASCII printing
1380        characters match isprint(). This difference caused the output from pcretest
1381        to vary, making some of the tests fail. I have changed pcretest so that:
1382    
1383        (a) When it is outputting text in the compiled version of a pattern, bytes
1384            other than 32-126 are always shown as hex escapes.
1385    
1386        (b) When it is outputting text that is a matched part of a subject string,
1387            it does the same, unless a different locale has been set for the match
1388            (using the /L modifier). In this case, it uses isprint() to decide.
1389    
1390     4. Fixed a major bug that caused incorrect computation of the amount of memory
1391        required for a compiled pattern when options that changed within the
1392        pattern affected the logic of the preliminary scan that determines the
1393        length. The relevant options are -x, and -i in UTF-8 mode. The result was
1394        that the computed length was too small. The symptoms of this bug were
1395        either the PCRE error "internal error: code overflow" from pcre_compile(),
1396        or a glibc crash with a message such as "pcretest: free(): invalid next
1397        size (fast)". Examples of patterns that provoked this bug (shown in
1398        pcretest format) are:
1399    
1400          /(?-x: )/x
1401          /(?x)(?-x: \s*#\s*)/
1402          /((?i)[\x{c0}])/8
1403          /(?i:[\x{c0}])/8
1404    
1405        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1406        is now done differently.
1407    
1408     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1409        wrapper classes; (b) implement a new function in the C++ scanner that is
1410        more efficient than the old way of doing things because it avoids levels of
1411        recursion in the regex matching; (c) add a paragraph to the documentation
1412        for the FullMatch() function.
1413    
1414     6. The escape sequence \n was being treated as whatever was defined as
1415        "newline". Not only was this contrary to the documentation, which states
1416        that \n is character 10 (hex 0A), but it also went horribly wrong when
1417        "newline" was defined as CRLF. This has been fixed.
1418    
1419     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1420        was being set to -1 for the "end of line" case (supposedly a value that no
1421        character can have). Though this value is never used (the check for end of
1422        line is "zero bytes in current character"), it caused compiler complaints.
1423        I've changed it to 0xffffffff.
1424    
1425     8. In pcre_version.c, the version string was being built by a sequence of
1426        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1427        string (as it is for production releases) called a macro with an empty
1428        argument. The C standard says the result of this is undefined. The gcc
1429        compiler treats it as an empty string (which was what was wanted) but it is
1430        reported that Visual C gives an error. The source has been hacked around to
1431        avoid this problem.
1432    
1433     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1434        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1435        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1436        of them did).
1437    
1438    10. Originally, pcretest opened its input and output without "b"; then I was
1439        told that "b" was needed in some environments, so it was added for release
1440        5.0 to both the input and output. (It makes no difference on Unix-like
1441        systems.) Later I was told that it is wrong for the input on Windows. I've
1442        now abstracted the modes into two macros, to make it easier to fiddle with
1443        them, and removed "b" from the input mode under Windows.
1444    
1445    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1446    
1447    12. Added -help and --help to pcretest as an official way of being reminded
1448        of the options.
1449    
1450    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1451        and pcrecpp.cc because they annoy compilers at high warning levels.
1452    
1453    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1454    
1455    15. Fixed an occurrence of == in configure.ac that should have been = (shell
1456        scripts are not C programs :-) and which was not noticed because it works
1457        on Linux.
1458    
1459    16. pcretest is supposed to handle any length of pattern and data line (as one
1460        line or as a continued sequence of lines) by extending its input buffer if
1461        necessary. This feature was broken for very long pattern lines, leading to
1462        a string of junk being passed to pcre_compile() if the pattern was longer
1463        than about 50K.
1464    
1465    17. I have done a major re-factoring of the way pcre_compile() computes the
1466        amount of memory needed for a compiled pattern. Previously, there was code
1467        that made a preliminary scan of the pattern in order to do this. That was
1468        OK when PCRE was new, but as the facilities have expanded, it has become
1469        harder and harder to keep it in step with the real compile phase, and there
1470        have been a number of bugs (see for example, 4 above). I have now found a
1471        cunning way of running the real compile function in a "fake" mode that
1472        enables it to compute how much memory it would need, while actually only
1473        ever using a few hundred bytes of working memory and without too many
1474        tests of the mode. This should make future maintenance and development
1475        easier. A side effect of this work is that the limit of 200 on the nesting
1476        depth of parentheses has been removed (though this was never a serious
1477        limitation, I suspect). However, there is a downside: pcre_compile() now
1478        runs more slowly than before (30% or more, depending on the pattern). I
1479        hope this isn't a big issue. There is no effect on runtime performance.
1480    
1481    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1482        newline (only possible for the last line of a file) and it was a
1483        pattern that set a locale (followed by /Lsomething), pcretest crashed.
1484    
1485    19. Added additional timing features to pcretest. (1) The -tm option now times
1486        matching only, not compiling. (2) Both -t and -tm can be followed, as a
1487        separate command line item, by a number that specifies the number of
1488        repeats to use when timing. The default is 50000; this gives better
1489        precision, but takes uncomfortably long for very large patterns.
1490    
1491    20. Extended pcre_study() to be more clever in cases where a branch of a
1492        subpattern has no definite first character. For example, (a*|b*)[cd] would
1493        previously give no result from pcre_study(). Now it recognizes that the
1494        first character must be a, b, c, or d.
1495    
1496    21. There was an incorrect error "recursive call could loop indefinitely" if
1497        a subpattern (or the entire pattern) that was being tested for matching an
1498        empty string contained only one non-empty item after a nested subpattern.
1499        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1500        incorrectly, because the \d was being skipped in the check.
1501    
1502    22. The pcretest program now has a new pattern option /B and a command line
1503        option -b, which is equivalent to adding /B to every pattern. This causes
1504        it to show the compiled bytecode, without the additional information that
1505        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1506        is the same as /B/I).
1507    
1508    23. A new optimization is now able automatically to treat some sequences such
1509        as a*b as a*+b. More specifically, if something simple (such as a character
1510        or a simple class like \d) has an unlimited quantifier, and is followed by
1511        something that cannot possibly match the quantified thing, the quantifier
1512        is automatically "possessified".
1513    
1514    24. A recursive reference to a subpattern whose number was greater than 39
1515        went wrong under certain circumstances in UTF-8 mode. This bug could also
1516        have affected the operation of pcre_study().
1517    
1518    25. Realized that a little bit of performance could be had by replacing
1519        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1520    
1521    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1522    
1523    27. Possessive quantifiers such as a++ were previously implemented by turning
1524        them into atomic groups such as ($>a+). Now they have their own opcodes,
1525        which improves performance. This includes the automatically created ones
1526        from 23 above.
1527    
1528    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1529        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1530        the first matched character to be a colon. This applied both to named and
1531        numbered groups.
1532    
1533    29. The ucpinternal.h header file was missing its idempotency #ifdef.
1534    
1535    30. I was sent a "project" file called libpcre.a.dev which I understand makes
1536        building PCRE on Windows easier, so I have included it in the distribution.
1537    
1538    31. There is now a check in pcretest against a ridiculously large number being
1539        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1540        loop, the loop is abandoned.
1541    
1542    32. Forward references to subpatterns in conditions such as (?(2)...) where
1543        subpattern 2 is defined later cause pcre_compile() to search forwards in
1544        the pattern for the relevant set of parentheses. This search went wrong
1545        when there were unescaped parentheses in a character class, parentheses
1546        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1547    
1548    33. "Subroutine" calls and backreferences were previously restricted to
1549        referencing subpatterns earlier in the regex. This restriction has now
1550        been removed.
1551    
1552    34. Added a number of extra features that are going to be in Perl 5.10. On the
1553        whole, these are just syntactic alternatives for features that PCRE had
1554        previously implemented using the Python syntax or my own invention. The
1555        other formats are all retained for compatibility.
1556    
1557        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1558            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1559            also .NET compatible.
1560    
1561        (b) A recursion or subroutine call to a named group can now be defined as
1562            (?&name) as well as (?P>name).
1563    
1564        (c) A backreference to a named group can now be defined as \k<name> or
1565            \k'name' as well as (?P=name). The new forms, as well as being in Perl
1566            5.10, are also .NET compatible.
1567    
1568        (d) A conditional reference to a named group can now use the syntax
1569            (?(<name>) or (?('name') as well as (?(name).
1570    
1571        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1572            groups (named and numbered) that are never evaluated inline, but can be
1573            called as "subroutines" from elsewhere. In effect, the DEFINE condition
1574            is always false. There may be only one alternative in such a group.
1575    
1576        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1577            as the simple (?(R). The condition is true only if the most recent
1578            recursion is that of the given number or name. It does not search out
1579            through the entire recursion stack.
1580    
1581        (g) The escape \gN or \g{N} has been added, where N is a positive or
1582            negative number, specifying an absolute or relative reference.
1583    
1584    35. Tidied to get rid of some further signed/unsigned compiler warnings and
1585        some "unreachable code" warnings.
1586    
1587    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1588        things, this adds five new scripts.
1589    
1590    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1591        There were also incompatibilities regarding the handling of \Q..\E inside
1592        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1593        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1594    
1595    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1596        matches an empty string, and forcibly breaks the loop. There were bugs in
1597        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
1598        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
1599        separate and independent bugs (that affected different cases) have been
1600        fixed.
1601    
1602    39. Refactored the code to abolish the use of different opcodes for small
1603        capturing bracket numbers. This is a tidy that I avoided doing when I
1604        removed the limit on the number of capturing brackets for 3.5 back in 2001.
1605        The new approach is not only tidier, it makes it possible to reduce the
1606        memory needed to fix the previous bug (38).
1607    
1608    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1609        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1610        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1611        mode.
1612    
1613    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1614        report.
1615    
1616    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1617        copy construction and assignment in the C++ wrapper.
1618    
1619    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1620        couple of bugs that could have given wrong results in the "--newline=crlf"
1621        case.
1622    
1623    44. Added a number of casts and did some reorganization of signed/unsigned int
1624        variables following suggestions from Dair Grant. Also renamed the variable
1625        "this" as "item" because it is a C++ keyword.
1626    
1627    45. Arranged for dftables to add
1628    
1629          #include "pcre_internal.h"
1630    
1631        to pcre_chartables.c because without it, gcc 4.x may remove the array
1632        definition from the final binary if PCRE is built into a static library and
1633        dead code stripping is activated.
1634    
1635    46. For an unanchored pattern, if a match attempt fails at the start of a
1636        newline sequence, and the newline setting is CRLF or ANY, and the next two
1637        characters are CRLF, advance by two characters instead of one.
1638    
1639    
1640    Version 6.7 04-Jul-06
1641    ---------------------
1642    
1643     1. In order to handle tests when input lines are enormously long, pcretest has
1644        been re-factored so that it automatically extends its buffers when
1645        necessary. The code is crude, but this _is_ just a test program. The
1646        default size has been increased from 32K to 50K.
1647    
1648     2. The code in pcre_study() was using the value of the re argument before
1649        testing it for NULL. (Of course, in any sensible call of the function, it
1650        won't be NULL.)
1651    
1652     3. The memmove() emulation function in pcre_internal.h, which is used on
1653        systems that lack both memmove() and bcopy() - that is, hardly ever -
1654        was missing a "static" storage class specifier.
1655    
1656     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1657        containing an extended class (one that cannot be represented by a bitmap
1658        because it contains high-valued characters or Unicode property items, e.g.
1659        [\pZ]). Almost always one would set UTF-8 mode when processing such a
1660        pattern, but PCRE should not loop if you do not (it no longer does).
1661        [Detail: two cases were found: (a) a repeated subpattern containing an
1662        extended class; (b) a recursive reference to a subpattern that followed a
1663        previous extended class. It wasn't skipping over the extended class
1664        correctly when UTF-8 mode was not set.]
1665    
1666     5. A negated single-character class was not being recognized as fixed-length
1667        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1668        compile error "lookbehind assertion is not fixed length".
1669    
1670     6. The RunPerlTest auxiliary script was showing an unexpected difference
1671        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1672        write a Perl script that can interpret lines of an input file either as
1673        byte characters or as UTF-8, which is what "perltest" was being required to
1674        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1675        can't do is switch easily at run time between having the "use utf8;" pragma
1676        or not. In the end, I fudged it by using the RunPerlTest script to insert
1677        "use utf8;" explicitly for the UTF-8 tests.
1678    
1679     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1680        the end of the subject string, contrary to the documentation and to what
1681        Perl does. This was true of both matching functions. Now it matches only at
1682        the start of the subject and immediately after *internal* newlines.
1683    
1684     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1685        a pointer to an int instead of a pointer to an unsigned long int. This
1686        caused problems on 64-bit systems.
1687    
1688     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1689        instance of the 'standard' template library not being so standard".
1690    
1691    10. There was no check on the number of named subpatterns nor the maximum
1692        length of a subpattern name. The product of these values is used to compute
1693        the size of the memory block for a compiled pattern. By supplying a very
1694        long subpattern name and a large number of named subpatterns, the size
1695        computation could be caused to overflow. This is now prevented by limiting
1696        the length of names to 32 characters, and the number of named subpatterns
1697        to 10,000.
1698    
1699    11. Subpatterns that are repeated with specific counts have to be replicated in
1700        the compiled pattern. The size of memory for this was computed from the
1701        length of the subpattern and the repeat count. The latter is limited to
1702        65535, but there was no limit on the former, meaning that integer overflow
1703        could in principle occur. The compiled length of a repeated subpattern is
1704        now limited to 30,000 bytes in order to prevent this.
1705    
1706    12. Added the optional facility to have named substrings with the same name.
1707    
1708    13. Added the ability to use a named substring as a condition, using the
1709        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1710        are numbers (not recommended). Forward references are permitted.
1711    
1712    14. Added forward references in named backreferences (if you see what I mean).
1713    
1714    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1715        pattern could run off the end of the subject. For example, the pattern
1716        "(?s)(.{1,5})"8 did this with the subject "ab".
1717    
1718    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1719        PCRE_CASELESS was set when matching characters that were quantified with ?
1720        or *.
1721    
1722    17. A character class other than a single negated character that had a minimum
1723        but no maximum quantifier - for example [ab]{6,} - was not handled
1724        correctly by pce_dfa_exec(). It would match only one character.
1725    
1726    18. A valid (though odd) pattern that looked like a POSIX character
1727        class but used an invalid character after [ (for example [[,abc,]]) caused
1728        pcre_compile() to give the error "Failed: internal error: code overflow" or
1729        in some cases to crash with a glibc free() error. This could even happen if
1730        the pattern terminated after [[ but there just happened to be a sequence of
1731        letters, a binary zero, and a closing ] in the memory that followed.
1732    
1733    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1734        over the years. Originally (before any Unicode support), just the bottom 8
1735        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1736        output from "man perlunicode" includes this:
1737    
1738          The regular expression compiler produces polymorphic opcodes.  That
1739          is, the pattern adapts to the data and automatically switches to
1740          the Unicode character scheme when presented with Unicode data--or
1741          instead uses a traditional byte scheme when presented with byte
1742          data.
1743    
1744        Sadly, a wide octal escape does not cause a switch, and in a string with
1745        no other multibyte characters, these octal escapes are treated as before.
1746        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
1747        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1748        Unicode string.
1749    
1750        I have not perpetrated such confusion in PCRE. Up till now, it took just
1751        the bottom 8 bits, as in old Perl. I have now made octal escapes with
1752        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1753        translate to the appropriate multibyte character.
1754    
1755    29. Applied some refactoring to reduce the number of warnings from Microsoft
1756        and Borland compilers. This has included removing the fudge introduced
1757        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1758        a warning about an unused variable.
1759    
1760    21. PCRE has not included VT (character 0x0b) in the set of whitespace
1761        characters since release 4.0, because Perl (from release 5.004) does not.
1762        [Or at least, is documented not to: some releases seem to be in conflict
1763        with the documentation.] However, when a pattern was studied with
1764        pcre_study() and all its branches started with \s, PCRE still included VT
1765        as a possible starting character. Of course, this did no harm; it just
1766        caused an unnecessary match attempt.
1767    
1768    22. Removed a now-redundant internal flag bit that recorded the fact that case
1769        dependency changed within the pattern. This was once needed for "required
1770        byte" processing, but is no longer used. This recovers a now-scarce options
1771        bit. Also moved the least significant internal flag bit to the most-
1772        significant bit of the word, which was not previously used (hangover from
1773        the days when it was an int rather than a uint) to free up another bit for
1774        the future.
1775    
1776    23. Added support for CRLF line endings as well as CR and LF. As well as the
1777        default being selectable at build time, it can now be changed at runtime
1778        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1779        specify that it is scanning data with non-default line endings.
1780    
1781    24. Changed the definition of CXXLINK to make it agree with the definition of
1782        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1783    
1784    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1785        recursions. This makes a big different to stack usage for some patterns.
1786    
1787    26. If a subpattern containing a named recursion or subroutine reference such
1788        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1789        the space required for the compiled pattern went wrong and gave too small a
1790        value. Depending on the environment, this could lead to "Failed: internal
1791        error: code overflow at offset 49" or "glibc detected double free or
1792        corruption" errors.
1793    
1794    27. Applied patches from Google (a) to support the new newline modes and (b) to
1795        advance over multibyte UTF-8 characters in GlobalReplace.
1796    
1797    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1798        difference for some implementation of PCRE in some Windows version.
1799    
1800    29. Added some extra testing facilities to pcretest:
1801    
1802        \q<number>   in a data line sets the "match limit" value
1803        \Q<number>   in a data line sets the "match recursion limt" value
1804        -S <number>  sets the stack size, where <number> is in megabytes
1805    
1806        The -S option isn't available for Windows.
1807    
1808    
1809    Version 6.6 06-Feb-06
1810    ---------------------
1811    
1812     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1813        in pcreposix.h. I have copied the definition from pcre.h.
1814    
1815     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1816        because pcre.h is no longer a built file.
1817    
1818     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1819        not normally included in the compiled code.
1820    
1821    
1822    Version 6.5 01-Feb-06
1823    ---------------------
1824    
1825     1. When using the partial match feature with pcre_dfa_exec(), it was not
1826        anchoring the second and subsequent partial matches at the new starting
1827        point. This could lead to incorrect results. For example, with the pattern
1828        /1234/, partially matching against "123" and then "a4" gave a match.
1829    
1830     2. Changes to pcregrep:
1831    
1832        (a) All non-match returns from pcre_exec() were being treated as failures
1833            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1834            error message is output. Some extra information is given for the
1835            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
1836            probably the only errors that are likely to be caused by users (by
1837            specifying a regex that has nested indefinite repeats, for instance).
1838            If there are more than 20 of these errors, pcregrep is abandoned.
1839    
1840        (b) A binary zero was treated as data while matching, but terminated the
1841            output line if it was written out. This has been fixed: binary zeroes
1842            are now no different to any other data bytes.
1843    
1844        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1845            used to set a locale for matching. The --locale=xxxx long option has
1846            been added (no short equivalent) to specify a locale explicitly on the
1847            pcregrep command, overriding the environment variables.
1848    
1849        (d) When -B was used with -n, some line numbers in the output were one less
1850            than they should have been.
1851    
1852        (e) Added the -o (--only-matching) option.
1853    
1854        (f) If -A or -C was used with -c (count only), some lines of context were
1855            accidentally printed for the final match.
1856    
1857        (g) Added the -H (--with-filename) option.
1858    
1859        (h) The combination of options -rh failed to suppress file names for files
1860            that were found from directory arguments.
1861    
1862        (i) Added the -D (--devices) and -d (--directories) options.
1863    
1864        (j) Added the -F (--fixed-strings) option.
1865    
1866        (k) Allow "-" to be used as a file name for -f as well as for a data file.
1867    
1868        (l) Added the --colo(u)r option.
1869    
1870        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1871            is not present by default.
1872    
1873     3. A nasty bug was discovered in the handling of recursive patterns, that is,
1874        items such as (?R) or (?1), when the recursion could match a number of
1875        alternatives. If it matched one of the alternatives, but subsequently,
1876        outside the recursion, there was a failure, the code tried to back up into
1877        the recursion. However, because of the way PCRE is implemented, this is not
1878        possible, and the result was an incorrect result from the match.
1879    
1880        In order to prevent this happening, the specification of recursion has
1881        been changed so that all such subpatterns are automatically treated as
1882        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1883    
1884     4. I had overlooked the fact that, in some locales, there are characters for
1885        which isalpha() is true but neither isupper() nor islower() are true. In
1886        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1887        and ordfeminine) are like this. This affected the treatment of \w and \W
1888        when they appeared in character classes, but not when they appeared outside
1889        a character class. The bit map for "word" characters is now created
1890        separately from the results of isalnum() instead of just taking it from the
1891        upper, lower, and digit maps. (Plus the underscore character, of course.)
1892    
1893     5. The above bug also affected the handling of POSIX character classes such as
1894        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1895        permanent tables. Instead, the bit maps for such a class were previously
1896        created as the appropriate unions of the upper, lower, and digit bitmaps.
1897        Now they are created by subtraction from the [[:word:]] class, which has
1898        its own bitmap.
1899    
1900     6. The [[:blank:]] character class matches horizontal, but not vertical space.
1901        It is created by subtracting the vertical space characters (\x09, \x0a,
1902        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1903        subtraction was done in the overall bitmap for a character class, meaning
1904        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1905        be recognized. This bug has been fixed.
1906    
1907     7. Patches from the folks at Google:
1908    
1909          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1910          real life, but is still worth protecting against".
1911    
1912          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1913          regular expressions".
1914    
1915          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1916          have it.
1917    
1918          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1919          "configure" and the latter not, in order to fix a problem somebody had
1920          with compiling the Arg class on HP-UX.
1921    
1922          (e) Improve the error-handling of the C++ wrapper a little bit.
1923    
1924          (f) New tests for checking recursion limiting.
1925    
1926     8. The pcre_memmove() function, which is used only if the environment does not
1927        have a standard memmove() function (and is therefore rarely compiled),
1928        contained two bugs: (a) use of int instead of size_t, and (b) it was not
1929        returning a result (though PCRE never actually uses the result).
1930    
1931     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1932        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1933        returned instead of calling malloc() with an overflowing number that would
1934        most likely cause subsequent chaos.
1935    
1936    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1937    
1938    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1939        with this option is matched, the nmatch and pmatch options of regexec() are
1940        ignored.
1941    
1942    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1943        provided in case anyone wants to the the POSIX interface with UTF-8
1944        strings.
1945    
1946    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1947        C++ linking (needed for some HP-UX environments).
1948    
1949    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1950        (unused parameter) and in the pcre_printint() function (omitted "default"
1951        switch label when the default is to do nothing).
1952    
1953    15. Added some code to make it possible, when PCRE is compiled as a C++
1954        library, to replace subject pointers for pcre_exec() with a smart pointer
1955        class, thus making it possible to process discontinuous strings.
1956    
1957    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1958        much the same function. They were added by different people who were trying
1959        to make PCRE easy to compile on non-Unix systems. It has been suggested
1960        that PCRE_EXPORT be abolished now that there is more automatic apparatus
1961        for compiling on Windows systems. I have therefore replaced it with
1962        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1963        defaults to "extern" for C or "extern C" for C++, which works fine on
1964        Unix-like systems. It is now possible to override the value of PCRE_DATA_
1965        SCOPE with something explicit in config.h. In addition:
1966    
1967        (a) pcreposix.h still had just "extern" instead of either of these macros;
1968            I have replaced it with PCRE_DATA_SCOPE.
1969    
1970        (b) Functions such as _pcre_xclass(), which are internal to the library,
1971            but external in the C sense, all had PCRE_EXPORT in their definitions.
1972            This is apparently wrong for the Windows case, so I have removed it.
1973            (It makes no difference on Unix-like systems.)
1974    
1975    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1976        of recursive calls to match(). This is different to MATCH_LIMIT because
1977        that limits the total number of calls to match(), not all of which increase
1978        the depth of recursion. Limiting the recursion depth limits the amount of
1979        stack (or heap if NO_RECURSE is set) that is used. The default can be set
1980        when PCRE is compiled, and changed at run time. A patch from Google adds
1981        this functionality to the C++ interface.
1982    
1983    18. Changes to the handling of Unicode character properties:
1984    
1985        (a) Updated the table to Unicode 4.1.0.
1986    
1987        (b) Recognize characters that are not in the table as "Cn" (undefined).
1988    
1989        (c) I revised the way the table is implemented to a much improved format
1990            which includes recognition of ranges. It now supports the ranges that
1991            are defined in UnicodeData.txt, and it also amalgamates other
1992            characters into ranges. This has reduced the number of entries in the
1993            table from around 16,000 to around 3,000, thus reducing its size
1994            considerably. I realized I did not need to use a tree structure after
1995            all - a binary chop search is just as efficient. Having reduced the
1996            number of entries, I extended their size from 6 bytes to 8 bytes to
1997            allow for more data.
1998    
1999        (d) Added support for Unicode script names via properties such as \p{Han}.
2000    
2001    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
2002        matching that character.
2003    
2004    20. When matching a repeated Unicode property with a minimum greater than zero,
2005        (for example \pL{2,}), PCRE could look past the end of the subject if it
2006        reached it while seeking the minimum number of characters. This could
2007        happen only if some of the characters were more than one byte long, because
2008        there is a check for at least the minimum number of bytes.
2009    
2010    21. Refactored the implementation of \p and \P so as to be more general, to
2011        allow for more different types of property in future. This has changed the
2012        compiled form incompatibly. Anybody with saved compiled patterns that use
2013        \p or \P will have to recompile them.
2014    
2015    22. Added "Any" and "L&" to the supported property types.
2016    
2017    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
2018        but give a compile time error if the value is greater than 0xff.
2019    
2020    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
2021        accidentally not being installed or uninstalled.
2022    
2023    25. The pcre.h file was built from pcre.h.in, but the only changes that were
2024        made were to insert the current release number. This seemed silly, because
2025        it made things harder for people building PCRE on systems that don't run
2026        "configure". I have turned pcre.h into a distributed file, no longer built
2027        by "configure", with the version identification directly included. There is
2028        no longer a pcre.h.in file.
2029    
2030        However, this change necessitated a change to the pcre-config script as
2031        well. It is built from pcre-config.in, and one of the substitutions was the
2032        release number. I have updated configure.ac so that ./configure now finds
2033        the release number by grepping pcre.h.
2034    
2035    26. Added the ability to run the tests under valgrind.
2036    
2037    
2038    Version 6.4 05-Sep-05
2039    ---------------------
2040    
2041     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
2042        "--" to be printed when multiple files were scanned, even when none of the
2043        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
2044        consider it to be a bug, and have restored the previous behaviour.
2045    
2046     2. A couple of code tidies to get rid of compiler warnings.
2047    
2048     3. The pcretest program used to cheat by referring to symbols in the library
2049        whose names begin with _pcre_. These are internal symbols that are not
2050        really supposed to be visible externally, and in some environments it is
2051        possible to suppress them. The cheating is now confined to including
2052        certain files from the library's source, which is a bit cleaner.
2053    
2054     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
2055        file's purpose clearer.
2056    
2057     5. Reorganized pcre_ucp_findchar().
2058    
2059    
2060    Version 6.3 15-Aug-05
2061    ---------------------
2062    
2063     1. The file libpcre.pc.in did not have general read permission in the tarball.
2064    
2065     2. There were some problems when building without C++ support:
2066    
2067        (a) If C++ support was not built, "make install" and "make test" still
2068            tried to test it.
2069    
2070        (b) There were problems when the value of CXX was explicitly set. Some
2071            changes have been made to try to fix these, and ...
2072    
2073        (c) --disable-cpp can now be used to explicitly disable C++ support.
2074    
2075        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
2076            backslash in a target when C++ was disabled. This confuses some
2077            versions of "make", apparently. Using an intermediate variable solves
2078            this. (Same for CPP_LOBJ.)
2079    
2080     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
2081        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
2082        necessary on certain architectures.
2083    
2084     4. Added a setting of -export-symbols-regex to the link command to remove
2085        those symbols that are exported in the C sense, but actually are local
2086        within the library, and not documented. Their names all begin with
2087        "_pcre_". This is not a perfect job, because (a) we have to except some
2088        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
2089        available (and never for static libraries). I have made a note to try to
2090        find a way round (a) in the future.
2091    
2092    
2093    Version 6.2 01-Aug-05
2094    ---------------------
2095    
2096     1. There was no test for integer overflow of quantifier values. A construction
2097        such as {1111111111111111} would give undefined results. What is worse, if
2098        a minimum quantifier for a parenthesized subpattern overflowed and became
2099        negative, the calculation of the memory size went wrong. This could have
2100        led to memory overwriting.
2101    
2102     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
2103    
2104     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
2105        operating environments where this matters.
2106    
2107     4. Applied Giuseppe Maxia's patch to add additional features for controlling
2108        PCRE options from within the C++ wrapper.
2109    
2110     5. Named capturing subpatterns were not being correctly counted when a pattern
2111        was compiled. This caused two problems: (a) If there were more than 100
2112        such subpatterns, the calculation of the memory needed for the whole
2113        compiled pattern went wrong, leading to an overflow error. (b) Numerical
2114        back references of the form \12, where the number was greater than 9, were
2115        not recognized as back references, even though there were sufficient
2116        previous subpatterns.
2117    
2118     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
2119        versions of gcc, e.g. 2.95.4.
2120    
2121    
2122    Version 6.1 21-Jun-05
2123    ---------------------
2124    
2125     1. There was one reference to the variable "posix" in pcretest.c that was not
2126        surrounded by "#if !defined NOPOSIX".
2127    
2128     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
2129        the cross-check on the old pcre_info() function, for the benefit of the
2130        cut-down version of PCRE that is currently imported into Exim.
2131    
2132     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
2133        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
2134        patterns that start (?i) but I don't think that matters. The use of (?i) is
2135        just an example; this all applies to the other options as well.
2136    
2137     4. Since libtool seems to echo the compile commands it is issuing, the output
2138        from "make" can be reduced a bit by putting "@" in front of each libtool
2139        compile command.
2140    
2141     5. Patch from the folks at Google for configure.in to be a bit more thorough
2142        in checking for a suitable C++ installation before trying to compile the
2143        C++ stuff. This should fix a reported problem when a compiler was present,
2144        but no suitable headers.
2145    
2146     6. The man pages all had just "PCRE" as their title. I have changed them to
2147        be the relevant file name. I have also arranged that these names are
2148        retained in the file doc/pcre.txt, which is a concatenation in text format
2149        of all the man pages except the little individual ones for each function.
2150    
2151     7. The NON-UNIX-USE file had not been updated for the different set of source
2152        files that come with release 6. I also added a few comments about the C++
2153        wrapper.
2154    
2155    
2156    Version 6.0 07-Jun-05
2157    ---------------------
2158    
2159     1. Some minor internal re-organization to help with my DFA experiments.
2160    
2161     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
2162        didn't matter for the library itself when fully configured, but did matter
2163        when compiling without UCP support, or within Exim, where the ucp files are
2164        not imported.
2165    
2166     3. Refactoring of the library code to split up the various functions into
2167        different source modules. The addition of the new DFA matching code (see
2168        below) to a single monolithic source would have made it really too
2169        unwieldy, quite apart from causing all the code to be include in a
2170        statically linked application, when only some functions are used. This is
2171        relevant even without the DFA addition now that patterns can be compiled in
2172        one application and matched in another.
2173    
2174        The downside of splitting up is that there have to be some external
2175        functions and data tables that are used internally in different modules of
2176        the library but which are not part of the API. These have all had their
2177        names changed to start with "_pcre_" so that they are unlikely to clash
2178        with other external names.
2179    
2180     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
2181        a different (DFA) algorithm. Although it is slower than the original
2182        function, it does have some advantages for certain types of matching
2183        problem.
2184    
2185     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
2186        including restarting after a partial match.
2187    
2188     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
2189        defined when compiling for Windows was sent to me. I have put it into the
2190        code, though I have no means of testing or verifying it.
2191    
2192     7. Added the pcre_refcount() auxiliary function.
2193    
2194     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
2195        match before or at the first newline in the subject string. In pcretest,
2196        the /f option on a pattern can be used to set this.
2197    
2198     9. A repeated \w when used in UTF-8 mode with characters greater than 256
2199        would behave wrongly. This has been present in PCRE since release 4.0.
2200    
2201    10. A number of changes to the pcregrep command:
2202    
2203        (a) Refactored how -x works; insert ^(...)$ instead of setting
2204            PCRE_ANCHORED and checking the length, in preparation for adding
2205            something similar for -w.
2206    
2207        (b) Added the -w (match as a word) option.
2208    
2209        (c) Refactored the way lines are read and buffered so as to have more
2210            than one at a time available.
2211    
2212        (d) Implemented a pcregrep test script.
2213    
2214        (e) Added the -M (multiline match) option. This allows patterns to match
2215            over several lines of the subject. The buffering ensures that at least
2216            8K, or the rest of the document (whichever is the shorter) is available
2217            for matching (and similarly the previous 8K for lookbehind assertions).
2218    
2219        (f) Changed the --help output so that it now says
2220    
2221              -w, --word-regex(p)
2222    
2223            instead of two lines, one with "regex" and the other with "regexp"
2224            because that confused at least one person since the short forms are the
2225            same. (This required a bit of code, as the output is generated
2226            automatically from a table. It wasn't just a text change.)
2227    
2228        (g) -- can be used to terminate pcregrep options if the next thing isn't an
2229            option but starts with a hyphen. Could be a pattern or a path name
2230            starting with a hyphen, for instance.
2231    
2232        (h) "-" can be given as a file name to represent stdin.
2233    
2234        (i) When file names are being printed, "(standard input)" is used for
2235            the standard input, for compatibility with GNU grep. Previously
2236            "<stdin>" was used.
2237    
2238        (j) The option --label=xxx can be used to supply a name to be used for
2239            stdin when file names are being printed. There is no short form.
2240    
2241        (k) Re-factored the options decoding logic because we are going to add
2242            two more options that take data. Such options can now be given in four
2243            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
2244    
2245        (l) Added the -A, -B, and -C options for requesting that lines of context
2246            around matches be printed.
2247    
2248        (m) Added the -L option to print the names of files that do not contain
2249            any matching lines, that is, the complement of -l.
2250    
2251        (n) The return code is 2 if any file cannot be opened, but pcregrep does
2252            continue to scan other files.
2253    
2254        (o) The -s option was incorrectly implemented. For compatibility with other
2255            greps, it now suppresses the error message for a non-existent or non-
2256            accessible file (but not the return code). There is a new option called
2257            -q that suppresses the output of matching lines, which was what -s was
2258            previously doing.
2259    
2260        (p) Added --include and --exclude options to specify files for inclusion
2261            and exclusion when recursing.
2262    
2263    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
2264        Hopefully, it now does.
2265    
2266    12. Missing cast in pcre_study().
2267    
2268    13. Added an "uninstall" target to the makefile.
2269    
2270    14. Replaced "extern" in the function prototypes in Makefile.in with
2271        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
2272        world, but is set differently for Windows.
2273    
2274    15. Added a second compiling function called pcre_compile2(). The only
2275        difference is that it has an extra argument, which is a pointer to an
2276        integer error code. When there is a compile-time failure, this is set
2277        non-zero, in addition to the error test pointer being set to point to an
2278        error message. The new argument may be NULL if no error number is required
2279        (but then you may as well call pcre_compile(), which is now just a
2280        wrapper). This facility is provided because some applications need a
2281        numeric error indication, but it has also enabled me to tidy up the way
2282        compile-time errors are handled in the POSIX wrapper.
2283    
2284    16. Added VPATH=.libs to the makefile; this should help when building with one
2285        prefix path and installing with another. (Or so I'm told by someone who
2286        knows more about this stuff than I do.)
2287    
2288    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
2289        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
2290        match everything, including newlines. This is not POSIX-compatible, but
2291        somebody wanted the feature. From pcretest it can be activated by using
2292        both the P and the s flags.
2293    
2294    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
2295    
2296    19. libpcre.pc was being incorrectly installed as executable.
2297    
2298    20. A couple of places in pcretest check for end-of-line by looking for '\n';
2299        it now also looks for '\r' so that it will work unmodified on Windows.
2300    
2301    21. Added Google's contributed C++ wrapper to the distribution.
2302    
2303    22. Added some untidy missing memory free() calls in pcretest, to keep
2304        Electric Fence happy when testing.
2305    
2306    
2307    
2308    Version 5.0 13-Sep-04
2309    ---------------------
2310    
2311     1. Internal change: literal characters are no longer packed up into items
2312        containing multiple characters in a single byte-string. Each character
2313        is now matched using a separate opcode. However, there may be more than one
2314        byte in the character in UTF-8 mode.
2315    
2316     2. The pcre_callout_block structure has two new fields: pattern_position and
2317        next_item_length. These contain the offset in the pattern to the next match
2318        item, and its length, respectively.
2319    
2320     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
2321        insertion of callouts before each pattern item. Added the /C option to
2322        pcretest to make use of this.
2323    
2324     4. On the advice of a Windows user, the lines
2325    
2326          #if defined(_WIN32) || defined(WIN32)
2327          _setmode( _fileno( stdout ), 0x8000 );
2328          #endif  /* defined(_WIN32) || defined(WIN32) */
2329    
2330        have been added to the source of pcretest. This apparently does useful
2331        magic in relation to line terminators.
2332    
2333     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
2334        for the benefit of those environments where the "b" makes a difference.
2335    
2336     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
2337        to know about it. I have put a hack into configure.in that adds in code
2338        to set GCC=yes if CC=icc. This seems to end up at a point in the
2339        generated configure script that is early enough to affect the setting of
2340        compiler options, which is what is needed, but I have no means of testing
2341        whether it really works. (The user who reported this had patched the
2342        generated configure script, which of course I cannot do.)
2343    
2344        LATER: After change 22 below (new libtool files), the configure script
2345        seems to know about icc (and also ecc). Therefore, I have commented out
2346        this hack in configure.in.
2347    
2348     7. Added support for pkg-config (2 patches were sent in).
2349    
2350     8. Negated POSIX character classes that used a combination of internal tables
2351        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2352        [[:^ascii]]. Typically, they would match almost any characters. The other
2353        POSIX classes were not broken in this way.
2354    
2355     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2356        to find the match, as PCRE was deluded into thinking that the match had to
2357        start at the start point or following a newline. The same bug applied to
2358        patterns with negative forward assertions or any backward assertions
2359        preceding ".*" at the start, unless the pattern required a fixed first
2360        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2361    
2362    10. In UTF-8 mode, when moving forwards in the subject after a failed match
2363        starting at the last subject character, bytes beyond the end of the subject
2364        string were read.
2365    
2366    11. Renamed the variable "class" as "classbits" to make life easier for C++
2367        users. (Previously there was a macro definition, but it apparently wasn't
2368        enough.)
2369    
2370    12. Added the new field "tables" to the extra data so that tables can be passed
2371        in at exec time, or the internal tables can be re-selected. This allows
2372        a compiled regex to be saved and re-used at a later time by a different
2373        program that might have everything at different addresses.
2374    
2375    13. Modified the pcre-config script so that, when run on Solaris, it shows a
2376        -R library as well as a -L library.
2377    
2378    14. The debugging options of pcretest (-d on the command line or D on a
2379        pattern) showed incorrect output for anything following an extended class
2380        that contained multibyte characters and which was followed by a quantifier.
2381    
2382    15. Added optional support for general category Unicode character properties
2383        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2384        support. It adds about 90K to the size of the library. The meanings of the
2385        inbuilt class escapes such as \d and \s have NOT been changed.
2386    
2387    16. Updated pcredemo.c to include calls to free() to release the memory for the
2388        compiled pattern.
2389    
2390    17. The generated file chartables.c was being created in the source directory
2391        instead of in the building directory. This caused the build to fail if the
2392        source directory was different from the building directory, and was
2393        read-only.
2394    
2395    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2396        file. No doubt somebody will tell me if they don't make sense... Also added
2397        Dan Mooney's comments about building on OpenVMS.
2398    
2399    19. Added support for partial matching via the PCRE_PARTIAL option for
2400        pcre_exec() and the \P data escape in pcretest.
2401    
2402    20. Extended pcretest with 3 new pattern features:
2403    
2404        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
2405              write the compiled pattern to the file whose name is "rest-of-line".
2406              This is a straight binary dump of the data, with the saved pointer to
2407              the character tables forced to be NULL. The study data, if any, is
2408              written too. After writing, pcretest reads a new pattern.
2409    
2410        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2411              compiled pattern from the given file. There must not be any
2412              occurrences of "<" in the file name (pretty unlikely); if there are,
2413              pcretest will instead treat the initial "<" as a pattern delimiter.
2414              After reading in the pattern, pcretest goes on to read data lines as
2415              usual.
2416    
2417        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2418              and 16-bit fields in a compiled pattern, to simulate a pattern that
2419              was compiled on a host of opposite endianness.
2420    
2421    21. The pcre-exec() function can now cope with patterns that were compiled on
2422        hosts of opposite endianness, with this restriction:
2423    
2424          As for any compiled expression that is saved and used later, the tables
2425          pointer field cannot be preserved; the extra_data field in the arguments
2426          to pcre_exec() should be used to pass in a tables address if a value
2427          other than the default internal tables were used at compile time.
2428    
2429    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2430        now diagnosed as an error. Previously, most of the time, a negative number
2431        would have been treated as zero, but if in addition "ovector" was passed as
2432        NULL, a crash could occur.
2433    
2434    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2435        new versions from the libtool 1.5 distribution (the last one is a copy of
2436        a file called libtool.m4). This seems to have fixed the need to patch
2437        "configure" to support Darwin 1.3 (which I used to do). However, I still
2438        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2439        workstation).
2440    
2441    24. Changed the PCRE licence to be the more standard "BSD" licence.
2442    
2443    
2444    Version 4.5 01-Dec-03
2445    ---------------------
2446    
2447     1. There has been some re-arrangement of the code for the match() function so
2448        that it can be compiled in a version that does not call itself recursively.
2449        Instead, it keeps those local variables that need separate instances for
2450        each "recursion" in a frame on the heap, and gets/frees frames whenever it
2451        needs to "recurse". Keeping track of where control must go is done by means
2452        of setjmp/longjmp. The whole thing is implemented by a set of macros that
2453        hide most of the details from the main code, and operates only if
2454        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2455        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2456        operating.
2457    
2458        To make it easier for callers to provide specially tailored get/free
2459        functions for this usage, two new functions, pcre_stack_malloc, and
2460        pcre_stack_free, are used. They are always called in strict stacking order,
2461        and the size of block requested is always the same.
2462    
2463        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2464        PCRE has been compiled to use the stack or the heap for recursion. The
2465        -C option of pcretest uses this to show which version is compiled.
2466    
2467        A new data escape \S, is added to pcretest; it causes the amounts of store
2468        obtained and freed by both kinds of malloc/free at match time to be added
2469        to the output.
2470    
2471     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2472        what's available on my current Linux desktop machine.
2473    
2474     3. When matching a UTF-8 string, the test for a valid string at the start has
2475        been extended. If start_offset is not zero, PCRE now checks that it points
2476        to a byte that is the start of a UTF-8 character. If not, it returns
2477        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2478        this is necessary because there may be backward assertions in the pattern.
2479        When matching the same subject several times, it may save resources to use
2480        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2481    
2482     4. The code for checking the validity of UTF-8 strings has been tightened so
2483        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2484        containing "overlong sequences".
2485    
2486     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2487        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2488        should have been "&", but it just so happened that all the cases this let
2489        through by mistake were picked up later in the function.
2490    
2491     6. I had used a variable called "isblank" - this is a C99 function, causing
2492        some compilers to warn. To avoid this, I renamed it (as "blankclass").
2493    
2494     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2495        prompting; (b) run "./pcretest /dev/null" at the start of the test script
2496        so the version is shown; (c) stop "make test" echoing "./RunTest".
2497    
2498     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2499    
2500     9. The prototype for memmove() for systems that don't have it was using
2501        size_t, but the inclusion of the header that defines size_t was later. I've
2502        moved the #includes for the C headers earlier to avoid this.
2503    
2504    10. Added some adjustments to the code to make it easier to compiler on certain
2505        special systems:
2506    
2507          (a) Some "const" qualifiers were missing.
2508          (b) Added the macro EXPORT before all exported functions; by default this
2509              is defined to be empty.
2510          (c) Changed the dftables auxiliary program (that builds chartables.c) so
2511              that it reads its output file name as an argument instead of writing
2512              to the standard output and assuming this can be redirected.
2513    
2514    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2515        class containing characters with values greater than 255, PCRE compilation
2516        went into a loop.
2517    
2518    12. A recursive reference to a subpattern that was within another subpattern
2519        that had a minimum quantifier of zero caused PCRE to crash. For example,
2520        (x(y(?2))z)? provoked this bug with a subject that got as far as the
2521        recursion. If the recursively-called subpattern itself had a zero repeat,
2522        that was OK.
2523    
2524    13. In pcretest, the buffer for reading a data line was set at 30K, but the
2525        buffer into which it was copied (for escape processing) was still set at
2526        1024, so long lines caused crashes.
2527    
2528    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2529        "internal error: code overflow...". This applied to any character class
2530        that was followed by a possessive quantifier.
2531    
2532    15. Modified the Makefile to add libpcre.la as a prerequisite for
2533        libpcreposix.la because I was told this is needed for a parallel build to
2534        work.
2535    
2536    16. If a pattern that contained .* following optional items at the start was
2537        studied, the wrong optimizing data was generated, leading to matching
2538        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2539        matching string must start with a or b or c. The correct conclusion for
2540        this pattern is that a match can start with any character.
2541    
2542    
2543    Version 4.4 13-Aug-03
2544    ---------------------
2545    
2546     1. In UTF-8 mode, a character class containing characters with values between
2547        127 and 255 was not handled correctly if the compiled pattern was studied.
2548        In fixing this, I have also improved the studying algorithm for such
2549        classes (slightly).
2550    
2551     2. Three internal functions had redundant arguments passed to them. Removal
2552        might give a very teeny performance improvement.
2553    
2554     3. Documentation bug: the value of the capture_top field in a callout is *one
2555        more than* the number of the hightest numbered captured substring.
2556    
2557     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2558        in incorrectly linking with a previously installed version. They now link
2559        explicitly with libpcre.la.
2560    
2561     5. configure.in no longer needs to recognize Cygwin specially.
2562    
2563     6. A problem in pcre.in for Windows platforms is fixed.
2564    
2565     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2566        pcretest, it used to include the size of the study block as part of its
2567        output. Unfortunately, the structure contains a field that has a different
2568        size on different hardware architectures. This meant that the tests that
2569        showed this size failed. As the block is currently always of a fixed size,
2570        this information isn't actually particularly useful in pcretest output, so
2571        I have just removed it.
2572    
2573     8. Three pre-processor statements accidentally did not start in column 1.
2574        Sadly, there are *still* compilers around that complain, even though
2575        standard C has not required this for well over a decade. Sigh.
2576    
2577     9. In pcretest, the code for checking callouts passed small integers in the
2578        callout_data field, which is a void * field. However, some picky compilers
2579        complained about the casts involved for this on 64-bit systems. Now
2580        pcretest passes the address of the small integer instead, which should get
2581        rid of the warnings.
2582    
2583    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2584        both compile and run time, and gives an error if an invalid UTF-8 sequence
2585        is found. There is a option for disabling this check in cases where the
2586        string is known to be correct and/or the maximum performance is wanted.
2587    
2588    11. In response to a bug report, I changed one line in Makefile.in from
2589    
2590            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2591        to
2592            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2593    
2594        to look similar to other lines, but I have no way of telling whether this
2595        is the right thing to do, as I do not use Windows. No doubt I'll get told
2596        if it's wrong...
2597    
2598    
2599    Version 4.3 21-May-03
2600    ---------------------
2601    
2602    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2603       Makefile.
2604    
2605    2. Some refactoring to improve the quality of the code:
2606    
2607       (i)   The utf8_table... variables are now declared "const".
2608    
2609       (ii)  The code for \cx, which used the "case flipping" table to upper case
2610             lower case letters, now just substracts 32. This is ASCII-specific,
2611             but the whole concept of \cx is ASCII-specific, so it seems
2612             reasonable.
2613    
2614       (iii) PCRE was using its character types table to recognize decimal and
2615             hexadecimal digits in the pattern. This is silly, because it handles
2616             only 0-9, a-f, and A-F, but the character types table is locale-
2617             specific, which means strange things might happen. A private
2618             table is now used for this - though it costs 256 bytes, a table is
2619             much faster than multiple explicit tests. Of course, the standard
2620             character types table is still used for matching digits in subject
2621             strings against \d.
2622    
2623       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2624             ending in _t are). So I've renamed it as ESC_tee.
2625    
2626    3. The first argument for regexec() in the POSIX wrapper should have been
2627       defined as "const".
2628    
2629    4. Changed pcretest to use malloc() for its buffers so that they can be
2630       Electric Fenced for debugging.
2631    
2632    5. There were several places in the code where, in UTF-8 mode, PCRE would try
2633       to read one or more bytes before the start of the subject string. Often this
2634       had no effect on PCRE's behaviour, but in some circumstances it could
2635       provoke a segmentation fault.
2636    
2637    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2638       to try to read one or more bytes before the start of the subject string.
2639    
2640    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
2641       UTF-8 support could misbehave in various ways if the subject string
2642       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
2643       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
2644       back over UTF-8 characters.)
2645    
2646    
2647    Version 4.2 14-Apr-03
2648    ---------------------
2649    
2650    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
2651    
2652    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
2653         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
2654         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
2655         [WIN_PREFIX]: new variable, "cyg" for Cygwin
2656         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
2657           and BUILD_EXEEXT
2658         Note: automatic setting of the BUILD variables is not yet working
2659         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
2660           compile-time but not at link-time
2661         [LINK]: use for linking executables only
2662         make different versions for Windows and non-Windows
2663         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
2664           libraries
2665         [LINK_FOR_BUILD]: new variable
2666         [OBJEXT]: use throughout
2667         [EXEEXT]: use throughout
2668         <winshared>: new target
2669         <wininstall>: new target
2670         <dftables.o>: use native compiler
2671         <dftables>: use native linker
2672         <install>: handle Windows platform correctly
2673         <clean>: ditto
2674         <check>: ditto
2675         copy DLL to top builddir before testing
2676    
2677       As part of these changes, -no-undefined was removed again. This was reported
2678       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
2679       in any case.
2680    
2681    3. Some tidies to get rid of compiler warnings:
2682    
2683       . In the match_data structure, match_limit was an unsigned long int, whereas
2684         match_call_count was an int. I've made them both unsigned long ints.
2685    
2686       . In pcretest the fact that a const uschar * doesn't automatically cast to
2687         a void * provoked a warning.
2688    
2689       . Turning on some more compiler warnings threw up some "shadow" variables
2690         and a few more missing casts.
2691    
2692    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
2693       option, a class that contained a single character with a value between 128
2694       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
2695    
2696    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
2697       option, a class that contained several characters, but with at least one
2698       whose value was between 128 and 255 caused PCRE to crash.
2699    
2700    
2701    Version 4.1 12-Mar-03
2702    ---------------------
2703    
2704    1. Compiling with gcc -pedantic found a couple of places where casts were
2705    needed, and a string in dftables.c that was longer than standard compilers are
2706    required to support.
2707    
2708    2. Compiling with Sun's compiler found a few more places where the code could
2709    be tidied up in order to avoid warnings.
2710    
2711    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
2712    first of these names is deprecated in the latest Autoconf in favour of the name
2713    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
2714    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2715    analogy I have changed it to CFLAGS_FOR_BUILD.
2716    
2717    4. Added -no-undefined to the linking command in the Makefile, because this is
2718    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2719    linking step for the pcreposix library.
2720    
2721    5. PCRE was failing to diagnose the case of two named groups with the same
2722    name.
2723    
2724    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2725    literal character that is needed in the subject for a match, and scans along to
2726    ensure that it is present before embarking on the full matching process. This
2727    saves time in cases of nested unlimited repeats that are never going to match.
2728    Problem: the scan can take a lot of time if the subject is very long (e.g.
2729    megabytes), thus penalizing straightforward matches. It is now done only if the
2730    amount of subject to be scanned is less than 1000 bytes.
2731    
2732    7. A lesser problem with the same optimization is that it was recording the
2733    first character of an anchored pattern as "needed", thus provoking a search
2734    right along the subject, even when the first match of the pattern was going to
2735    fail. The "needed" character is now not set for anchored patterns, unless it
2736    follows something in the pattern that is of non-fixed length. Thus, it still
2737    fulfils its original purpose of finding quick non-matches in cases of nested
2738    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
2739    
2740    
2741    Version 4.0 17-Feb-03
2742    ---------------------
2743    
2744    1. If a comment in an extended regex that started immediately after a meta-item
2745    extended to the end of string, PCRE compiled incorrect data. This could lead to
2746    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
2747    
2748    2. Moved to autoconf 2.53 and libtool 1.4.2.
2749    
2750    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
2751    the special perltest8 script is no longer needed - all the tests can be run
2752    from a single perltest script.
2753    
2754    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
2755    by \s. It has now been removed in PCRE. This means it isn't recognized as
2756    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
2757    class [:space:] *does* include VT, thereby creating a mess.
2758    
2759    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
2760    space and tab.
2761    
2762    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
2763    its new features into the main test script, reducing the number of scripts.
2764    
2765    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
2766    were backward compatible, and made the (?i) apply to the whole pattern, as if
2767    /i were given. Now it behaves more logically, and applies the option setting
2768    only to what follows. PCRE has been changed to follow suit. However, if it
2769    finds options settings right at the start of the pattern, it extracts them into
2770    the global options, as before. Thus, they show up in the info data.
2771    
2772    8. Added support for the \Q...\E escape sequence. Characters in between are
2773    treated as literals. This is slightly different from Perl in that $ and @ are
2774    also handled as literals inside the quotes. In Perl, they will cause variable
2775    interpolation. Note the following examples:
2776    
2777        Pattern            PCRE matches      Perl matches
2778    
2779        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
2780        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
2781        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
2782    
2783    For compatibility with Perl, \Q...\E sequences are recognized inside character
2784    classes as well as outside them.
2785    
2786    9. Re-organized 3 code statements in pcretest to avoid "overflow in
2787    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2788    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2789    signed/unsigned warnings.
2790    
2791    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2792    option for pcretest, so I've replaced it by a simple function that does just
2793    that job.
2794    
2795    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2796    "pcregrep -".
2797    
2798    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2799    Java package. This provides some syntactic sugar for simple cases of what my
2800    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2801    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2802    item, you can use this simplified notation. Note that only makes sense with
2803    greedy quantifiers. Consequently, the use of the possessive quantifier forces
2804    greediness, whatever the setting of the PCRE_UNGREEDY option.
2805    
2806    13. A change of greediness default within a pattern was not taking effect at
2807    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2808    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2809    was abstracted outside.
2810    
2811    14. PCRE now supports the \G assertion. It is true when the current matching
2812    position is at the start point of the match. This differs from \A when the
2813    starting offset is non-zero. Used with the /g option of pcretest (or similar
2814    code), it works in the same way as it does for Perl's /g option. If all
2815    alternatives of a regex begin with \G, the expression is anchored to the start
2816    match position, and the "anchored" flag is set in the compiled expression.
2817    
2818    15. Some bugs concerning the handling of certain option changes within patterns
2819    have been fixed. These applied to options other than (?ims). For example,
2820    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2821    been the other way round. Some of this was related to change 7 above.
2822    
2823    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2824    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2825    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2826    POSIX classes only within a class (e.g. /[[:alpha:]]/).
2827    
2828    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2829    mode. Unlike ".", it always matches newline, whatever the setting of
2830    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2831    assertions. Perl allows it, but it doesn't (in general) work because it can't
2832    calculate the length of the lookbehind. At least, that's the case for Perl
2833    5.8.0 - I've been told they are going to document that it doesn't work in
2834    future.
2835    
2836    18. Added an error diagnosis for escapes that PCRE does not support: these are
2837    \L, \l, \N, \P, \p, \U, \u, and \X.
2838    
2839    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2840    reading past the end of the pattern in cases such as /[abcd/.
2841    
2842    20. PCRE was getting more memory than necessary for patterns with classes that
2843    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2844    
2845    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2846    compiling PCRE for use with Virtual Pascal.
2847    
2848    22. Small fix to the Makefile to make it work properly if the build is done
2849    outside the source tree.
2850    
2851    23. Added a new extension: a condition to go with recursion. If a conditional
2852    subpattern starts with (?(R) the "true" branch is used if recursion has
2853    happened, whereas the "false" branch is used only at the top level.
2854    
2855    24. When there was a very long string of literal characters (over 255 bytes
2856    without UTF support, over 250 bytes with UTF support), the computation of how
2857    much memory was required could be incorrect, leading to segfaults or other
2858    strange effects.
2859    
2860    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2861    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2862    there was a subsequent back reference to those brackets. This meant that, for
2863    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2864    possible to check for precisely this case. All we can do is abandon the
2865    optimization if .* occurs inside capturing brackets when there are any back
2866    references whatsoever. (See below for a better fix that came later.)
2867    
2868    26. The handling of the optimization for finding the first character of a
2869    non-anchored pattern, and for finding a character that is required later in the
2870    match were failing in some cases. This didn't break the matching; it just
2871    failed to optimize when it could. The way this is done has been re-implemented.
2872    
2873    27. Fixed typo in error message for invalid (?R item (it said "(?p").
2874    
2875    28. Added a new feature that provides some of the functionality that Perl
2876    provides with (?{...}). The facility is termed a "callout". The way it is done
2877    in PCRE is for the caller to provide an optional function, by setting
2878    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2879    global variable. By default it is unset, which disables all calling out. To get
2880    the function called, the regex must include (?C) at appropriate points. This
2881    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2882    This provides a means of identifying different callout points. When PCRE
2883    reaches such a point in the regex, if pcre_callout has been set, the external
2884    function is called. It is provided with data in a structure called
2885    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2886    matching continues; if it returns a non-zero value, the match at the current
2887    point fails. However, backtracking will occur if possible. [This was changed
2888    later and other features added - see item 49 below.]
2889    
2890    29. pcretest is upgraded to test the callout functionality. It provides a
2891    callout function that displays information. By default, it shows the start of
2892    the match and the current position in the text. There are some new data escapes
2893    to vary what happens:
2894    
2895        \C+         in addition, show current contents of captured substrings
2896        \C-         do not supply a callout function
2897        \C!n        return 1 when callout number n is reached
2898        \C!n!m      return 1 when callout number n is reached for the mth time
2899    
2900    30. If pcregrep was called with the -l option and just a single file name, it
2901    output "<stdin>" if a match was found, instead of the file name.
2902    
2903    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2904    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2905    pcre_exec(). This saves a malloc/free per call. The default value of
2906    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2907    when configuring.
2908    
2909    32. The default maximum size of a compiled pattern is 64K. There have been a
2910    few cases of people hitting this limit. The code now uses macros to handle the
2911    storing of links as offsets within the compiled pattern. It defaults to 2-byte
2912    links, but this can be changed to 3 or 4 bytes by --with-link-size when
2913    configuring. Tests 2 and 5 work only with 2-byte links because they output
2914    debugging information about compiled patterns.
2915    
2916    33. Internal code re-arrangements:
2917    
2918    (a) Moved the debugging function for printing out a compiled regex into
2919        its own source file (printint.c) and used #include to pull it into
2920        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2921        separate copies.
2922    
2923    (b) Defined the list of op-code names for debugging as a macro in
2924        internal.h so that it is next to the definition of the opcodes.
2925    
2926    (c) Defined a table of op-code lengths for simpler skipping along compiled
2927        code. This is again a macro in internal.h so that it is next to the
2928        definition of the opcodes.
2929    
2930    34. Added support for recursive calls to individual subpatterns, along the
2931    lines of Robin Houston's patch (but implemented somewhat differently).
2932    
2933    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2934    allow it to read and process whole directories in Win32. This code was
2935    contributed by Lionel Fourquaux; it has not been tested by me.
2936    
2937    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2938    used to name a group. Names consist of alphanumerics and underscores, and must
2939    be unique. Back references use the syntax (?P=name) and recursive calls use
2940    (?P>name) which is a PCRE extension to the Python extension. Groups still have
2941    numbers. The function pcre_fullinfo() can be used after compilation to extract
2942    a name/number map. There are three relevant calls:
2943    
2944      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
2945      PCRE_INFO_NAMECOUNT            yields the number of entries
2946      PCRE_INFO_NAMETABLE            yields a pointer to the map.
2947    
2948    The map is a vector of fixed-size entries. The size of each entry depends on
2949    the length of the longest name used. The first two bytes of each entry are the
2950    group number, most significant byte first. There follows the corresponding
2951    name, zero terminated. The names are in alphabetical order.
2952    
2953    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2954    case instead of 255. Making it the same both with and without UTF-8 support
2955    means that the same test output works with both.
2956    
2957    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2958    calling malloc() with a zero argument.
2959    
2960    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2961    optimization. I've improved things by keeping a bitmap of backreferences with
2962    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2963    fact referenced, the optimization can be applied. It is unlikely that a
2964    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2965    the match to follow \n) will appear inside brackets with a number greater than
2966    31, but if it does, any back reference > 31 suppresses the optimization.
2967    
2968    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2969    of disabling numbered capturing parentheses. Any opening parenthesis that is
2970    not followed by ? behaves as if it were followed by ?: but named parentheses
2971    can still be used for capturing (and they will acquire numbers in the usual
2972    way).
2973    
2974    41. Redesigned the return codes from the match() function into yes/no/error so
2975    that errors can be passed back from deep inside the nested calls. A malloc
2976    failure while inside a recursive subpattern call now causes the
2977    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2978    
2979    42. It is now possible to set a limit on the number of times the match()
2980    function is called in a call to pcre_exec(). This facility makes it possible to
2981    limit the amount of recursion and backtracking, though not in a directly
2982    obvious way, because the match() function is used in a number of different
2983    circumstances. The count starts from zero for each position in the subject
2984    string (for non-anchored patterns). The default limit is, for compatibility, a
2985    large number, namely 10 000 000. You can change this in two ways:
2986    
2987    (a) When configuring PCRE before making, you can use --with-match-limit=n
2988        to set a default value for the compiled library.
2989    
2990    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2991        a different value is set. See 45 below.
2992    
2993    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2994    
2995    43. Added a new function pcre_config(int, void *) to enable run-time extraction
2996    of things that can be changed at compile time. The first argument specifies
2997    what is wanted and the second points to where the information is to be placed.
2998    The current list of available information is:
2999    
3000      PCRE_CONFIG_UTF8
3001    
3002    The output is an integer that is set to one if UTF-8 support is available;
3003    otherwise it is set to zero.
3004    
3005      PCRE_CONFIG_NEWLINE
3006    
3007    The output is an integer that it set to the value of the code that is used for
3008    newline. It is either LF (10) or CR (13).
3009    
3010      PCRE_CONFIG_LINK_SIZE
3011    
3012    The output is an integer that contains the number of bytes used for internal
3013    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
3014    
3015      PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
3016    
3017    The output is an integer that contains the threshold above which the POSIX
3018    interface uses malloc() for output vectors. See item 31 above.
3019    
3020      PCRE_CONFIG_MATCH_LIMIT
3021    
3022    The output is an unsigned integer that contains the default limit of the number
3023    of match() calls in a pcre_exec() execution. See 42 above.
3024    
3025    44. pcretest has been upgraded by the addition of the -C option. This causes it
3026    to extract all the available output from the new pcre_config() function, and to
3027    output it. The program then exits immediately.
3028    
3029    45. A need has arisen to pass over additional data with calls to pcre_exec() in
3030    order to support additional features. One way would have been to define
3031    pcre_exec2() (for example) with extra arguments, but this would not have been
3032    extensible, and would also have required all calls to the original function to
3033    be mapped to the new one. Instead, I have chosen to extend the mechanism that
3034    is used for passing in "extra" data from pcre_study().
3035    
3036    The pcre_extra structure is now exposed and defined in pcre.h. It currently
3037    contains the following fields:
3038    
3039      flags         a bitmap indicating which of the following fields are set
3040      study_data    opaque data from pcre_study()
3041      match_limit   a way of specifying a limit on match() calls for a specific
3042                      call to pcre_exec()
3043      callout_data  data for callouts (see 49 below)
3044    
3045    The flag bits are also defined in pcre.h, and are
3046    
3047      PCRE_EXTRA_STUDY_DATA
3048      PCRE_EXTRA_MATCH_LIMIT
3049      PCRE_EXTRA_CALLOUT_DATA
3050    
3051    The pcre_study() function now returns one of these new pcre_extra blocks, with
3052    the actual study data pointed to by the study_data field, and the
3053    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
3054    before. That is, this change is entirely upwards-compatible and requires no
3055    change to existing code.
3056    
3057    If you want to pass in additional data to pcre_exec(), you can either place it
3058    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
3059    block.
3060    
3061    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
3062    data string contains the escape sequence \M, pcretest calls pcre_exec() several
3063    times with different match limits, until it finds the minimum value needed for
3064    pcre_exec() to complete. The value is then output. This can be instructive; for
3065    most simple matches the number is quite small, but for pathological cases it
3066    gets very large very quickly.
3067    
3068    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
3069    returns the size of the data block pointed to by the study_data field in a
3070    pcre_extra block, that is, the value that was passed as the argument to
3071    pcre_malloc() when PCRE was getting memory in which to place the information
3072    created by pcre_study(). The fourth argument should point to a size_t variable.
3073    pcretest has been extended so that this information is shown after a successful
3074    pcre_study() call when information about the compiled regex is being displayed.
3075    
3076    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
3077    because what follows is always an absolute path. (Later: it turns out that this
3078    is more than cosmetic for MinGW, because it doesn't like empty path
3079    components.)
3080    
3081    49. Some changes have been made to the callout feature (see 28 above):
3082    
3083    (i)  A callout function now has three choices for what it returns:
3084    
3085           0  =>  success, carry on matching
3086         > 0  =>  failure at this point, but backtrack if possible
3087         < 0  =>  serious error, return this value from pcre_exec()
3088    
3089         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
3090         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
3091         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
3092         use by callout functions. It will never be used by PCRE itself.
3093    
3094    (ii) The pcre_extra structure (see 45 above) has a void * field called
3095         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
3096         pcre_callout_block structure has a field of the same name. The contents of
3097         the field passed in the pcre_extra structure are passed to the callout
3098         function in the corresponding field in the callout block. This makes it
3099         easier to use the same callout-containing regex from multiple threads. For
3100         testing, the pcretest program has a new data escape
3101    
3102           \C*n        pass the number n (may be negative) as callout_data
3103    
3104         If the callout function in pcretest receives a non-zero value as
3105         callout_data, it returns that value.
3106    
3107    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
3108    there were some redundant $(CFLAGS) in commands that are now specified as
3109    $(LINK), which already includes $(CFLAGS).
3110    
3111    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
3112    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
3113    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
3114    one-byte characters throughout. Note that case-insensitive matching applies
3115    only to characters whose values are less than 256. PCRE doesn't support the
3116    notion of cases for higher-valued characters.
3117    
3118    (i)   A character class whose characters are all within 0-255 is handled as
3119          a bit map, and the map is inverted for negative classes. Previously, a
3120          character > 255 always failed to match such a class; however it should
3121          match if the class was a negative one (e.g. [^ab]). This has been fixed.
3122    
3123    (ii)  A negated character class with a single character < 255 is coded as
3124          "not this character" (OP_NOT). This wasn't working properly when the test
3125          character was multibyte, either singly or repeated.
3126    
3127    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
3128          mode, for example: \x{100}{2,3}.
3129    
3130    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
3131          singly or repeated) now correctly test multibyte characters. However,
3132          PCRE doesn't recognize any characters with values greater than 255 as
3133          digits, spaces, or word characters. Such characters always match \D, \S,
3134          and \W, and never match \d, \s, or \w.
3135    
3136    (v)   Classes may now contain characters and character ranges with values
3137          greater than 255. For example: [ab\x{100}-\x{400}].
3138    
3139    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
3140          PCRE in UTF-8 mode.
3141    
3142    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
3143    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
3144    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
3145    value.)
3146    
3147    53. The single man page has become too large. I have therefore split it up into
3148    a number of separate man pages. These also give rise to individual HTML pages;
3149    these are now put in a separate directory, and there is an index.html page that
3150    lists them all. Some hyperlinking between the pages has been installed.
3151    
3152    54. Added convenience functions for handling named capturing parentheses.
3153    
3154    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
3155    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
3156    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
3157    are faulted.
3158    
3159    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
3160    calling configure. These values are used when compiling the dftables.c program
3161    which is run to generate the source of the default character tables. They
3162    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
3163    you will need to set these values.
3164    
3165    57. Updated the building process for Windows DLL, as provided by Fred Cox.
3166    
3167    
3168    Version 3.9 02-Jan-02
3169    ---------------------
3170    
3171    1. A bit of extraneous text had somehow crept into the pcregrep documentation.
3172    
3173    2. If --disable-static was given, the building process failed when trying to
3174    build pcretest and pcregrep. (For some reason it was using libtool to compile
3175    them, which is not right, as they aren't part of the library.)
3176    
3177    
3178    Version 3.8 18-Dec-01
3179    ---------------------
3180    
3181    1. The experimental UTF-8 code was completely screwed up. It was packing the
3182    bytes in the wrong order. How dumb can you get?
3183    
3184    
3185    Version 3.7 29-Oct-01
3186    ---------------------
3187    
3188    1. In updating pcretest to check change 1 of version 3.6, I screwed up.
3189    This caused pcretest, when used on the test data, to segfault. Unfortunately,
3190    this didn't happen under Solaris 8, where I normally test things.
3191    
3192    2. The Makefile had to be changed to make it work on BSD systems, where 'make'
3193    doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
3194    isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
3195    this fix an hour or so after the initial 3.7 release.)
3196    
3197    
3198    Version 3.6 23-Oct-01
3199    ---------------------
3200    
3201    1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
3202    offsets passed as NULL with zero offset count.
3203    
3204    2. The config.guess and config.sub files had not been updated when I moved to
3205    the latest autoconf.
3206    
3207    
3208    Version 3.5 15-Aug-01
3209    ---------------------
3210    
3211    1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
3212    had been forgotten.
3213    
3214    2. By using declared but undefined structures, we can avoid using "void"
3215    definitions in pcre.h while keeping the internal definitions of the structures
3216    private.
3217    
3218    3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
3219    user point of view, this means that both static and shared libraries are built
3220    by default, but this can be individually controlled. More of the work of
3221    handling this static/shared cases is now inside libtool instead of PCRE's make
3222    file.
3223    
3224    4. The pcretest utility is now installed along with pcregrep because it is
3225    useful for users (to test regexs) and by doing this, it automatically gets
3226    relinked by libtool. The documentation has been turned into a man page, so
3227    there are now .1, .txt, and .html versions in /doc.
3228    
3229    5. Upgrades to pcregrep:
3230       (i)   Added long-form option names like gnu grep.
3231       (ii)  Added --help to list all options with an explanatory phrase.
3232       (iii) Added -r, --recursive to recurse into sub-directories.
3233       (iv)  Added -f, --file to read patterns from a file.
3234    
3235    6. pcre_exec() was referring to its "code" argument before testing that
3236    argument for NULL (and giving an error if it was NULL).
3237    
3238    7. Upgraded Makefile.in to allow for compiling in a different directory from
3239    the source directory.
3240    
3241    8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
3242    options bits, the pointer it was passed was to an int instead of to an unsigned
3243    long int. This mattered only on 64-bit systems.
3244    
3245    9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
3246    generated) instead of pcre.in, which it its source. Also made the same change
3247    in several of the .c files.
3248    
3249    10. A new release of gcc defines printf() as a macro, which broke pcretest
3250    because it had an ifdef in the middle of a string argument for printf(). Fixed
3251    by using separate calls to printf().
3252    
3253    11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
3254    script, to force use of CR or LF instead of \n in the source. On non-Unix
3255    systems, the value can be set in config.h.
3256    
3257    12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
3258    absolute limit. Changed the text of the error message to make this clear, and
3259    likewise updated the man page.
3260    
3261    13. The limit of 99 on the number of capturing subpatterns has been removed.
3262    The new limit is 65535, which I hope will not be a "real" limit.
3263    
3264    
3265    Version 3.4 22-Aug-00
3266    ---------------------
3267    
3268    1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
3269    
3270    2. Diagnose condition (?(0) as an error instead of crashing on matching.
3271    
3272    
3273    Version 3.3 01-Aug-00
3274    ---------------------
3275    
3276    1. If an octal character was given, but the value was greater than \377, it
3277    was not getting masked to the least significant bits, as documented. This could
3278    lead to crashes in some systems.
3279    
3280    2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
3281    the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
3282    
3283    3. Added the functions pcre_free_substring() and pcre_free_substring_list().
3284    These just pass their arguments on to (pcre_free)(), but they are provided
3285    because some uses of PCRE bind it to non-C systems that can call its functions,
3286    but cannot call free() or pcre_free() directly.
3287    
3288    4. Add "make test" as a synonym for "make check". Corrected some comments in
3289    the Makefile.
3290    
3291    5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
3292    Makefile.
3293    
3294    6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
3295    command called pgrep for grepping around the active processes.
3296    
3297    7. Added the beginnings of support for UTF-8 character strings.
3298    
3299    8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
3300    RANLIB to ./ltconfig so that they are used by libtool. I think these are all
3301    the relevant ones. (AR is not passed because ./ltconfig does its own figuring
3302    out for the ar command.)
3303    
3304    
3305    Version 3.2 12-May-00
3306    ---------------------
3307    
3308    This is purely a bug fixing release.
3309    
3310    1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
3311    of ZA. This was just one example of several cases that could provoke this bug,
3312    which was introduced by change 9 of version 2.00. The code for breaking
3313    infinite loops after an iteration that matches an empty string was't working
3314    correctly.
3315    
3316    2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
3317    when matched against abbab (for example). After matching an empty string, it
3318    wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
3319    caused it to match further down the string than it should.
3320    
3321    3. The code contained an inclusion of sys/types.h. It isn't clear why this
3322    was there because it doesn't seem to be needed, and it causes trouble on some
3323    systems, as it is not a Standard C header. It has been removed.
3324    
3325    4. Made 4 silly changes to the source to avoid stupid compiler warnings that
3326    were reported on the Macintosh. The changes were from
3327    
3328      while ((c = *(++ptr)) != 0 && c != '\n');
3329    to
3330      while ((c = *(++ptr)) != 0 && c != '\n') ;
3331    
3332    Totally extraordinary, but if that's what it takes...
3333    
3334    5. PCRE is being used in one environment where neither memmove() nor bcopy() is
3335    available. Added HAVE_BCOPY and an autoconf test for it; if neither
3336    HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
3337    assumes the way PCRE uses memmove() (always moving upwards).
3338    
3339    6. PCRE is being used in one environment where strchr() is not available. There
3340    was only one use in pcre.c, and writing it out to avoid strchr() probably gives
3341    faster code anyway.
3342    
3343    
3344    Version 3.1 09-Feb-00
3345    ---------------------
3346    
3347    The only change in this release is the fixing of some bugs in Makefile.in for
3348    the "install" target:
3349    
3350    (1) It was failing to install pcreposix.h.
3351    
3352    (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
3353    
3354    
3355    Version 3.0 01-Feb-00
3356    ---------------------
3357    
3358    1. Add support for the /+ modifier to perltest (to output $` like it does in
3359    pcretest).
3360    
3361    2. Add support for the /g modifier to perltest.
3362    
3363    3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
3364    matches null strings.
3365    
3366    4. Fix perltest so that it doesn't do unwanted things when fed an empty
3367    pattern. Perl treats empty patterns specially - it reuses the most recent
3368    pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
3369    effect.
3370    
3371    5. The POSIX interface was broken in that it was just handing over the POSIX
3372    captured string vector to pcre_exec(), but (since release 2.00) PCRE has
3373    required a bigger vector, with some working space on the end. This means that
3374    the POSIX wrapper now has to get and free some memory, and copy the results.
3375    
3376    6. Added some simple autoconf support, placing the test data and the
3377    documentation in separate directories, re-organizing some of the
3378    information files, and making it build pcre-config (a GNU standard). Also added
3379    libtool support for building PCRE as a shared library, which is now the
3380    default.
3381    
3382    7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
3383    09 are not valid octal constants. Single digits will be used for minor values
3384    less than 10.
3385    
3386    8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
3387    existing programs that set these in the POSIX interface can use PCRE without
3388    modification.
3389    
3390    9. Added a new function, pcre_fullinfo() with an extensible interface. It can
3391    return all that pcre_info() returns, plus additional data. The pcre_info()
3392    function is retained for compatibility, but is considered to be obsolete.
3393    
3394    10. Added experimental recursion feature (?R) to handle one common case that
3395    Perl 5.6 will be able to do with (?p{...}).
3396    
3397    11. Added support for POSIX character classes like [:alpha:], which Perl is
3398    adopting.
3399    
3400    
3401    Version 2.08 31-Aug-99
3402    ----------------------
3403    
3404    1. When startoffset was not zero and the pattern began with ".*", PCRE was not
3405    trying to match at the startoffset position, but instead was moving forward to
3406    the next newline as if a previous match had failed.
3407    
3408    2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
3409    and could get into a loop if a null string was matched other than at the start
3410    of the subject.
3411    
3412    3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
3413    be distinguished at compile time, and for completeness also added PCRE_DATE.
3414    
3415    5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
3416    in GnuWin32 environments.
3417    
3418    
3419    Version 2.07 29-Jul-99
3420    ----------------------
3421    
3422    1. The documentation is now supplied in plain text form and HTML as well as in
3423    the form of man page sources.
3424    
3425    2. C++ compilers don't like assigning (void *) values to other pointer types.
3426    In particular this affects malloc(). Although there is no problem in Standard
3427    C, I've put in casts to keep C++ compilers happy.
3428    
3429    3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
3430    should be (const char *).
3431    
3432    4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
3433    be useful for non-Unix systems who don't want to bother with the POSIX stuff.
3434    However, I haven't made this a standard facility. The documentation doesn't
3435    mention it, and the Makefile doesn't support it.
3436    
3437    5. The Makefile now contains an "install" target, with editable destinations at
3438    the top of the file. The pcretest program is not installed.
3439    
3440    6. pgrep -V now gives the PCRE version number and date.
3441    
3442    7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
3443    causing the entire string to be ignored, instead of just the last character.
3444    
3445    8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
3446    non-matching string, it can take a very, very long time, even for strings of
3447    quite modest length, because of the nested recursion. PCRE now does better in
3448    some of these cases. It does this by remembering the last required literal
3449    character in the pattern, and pre-searching the subject to ensure it is present
3450    before running the real match. In other words, it applies a heuristic to detect
3451    some types of certain failure quickly, and in the above example, if presented
3452    with a string that has no trailing " it gives "no match" very quickly.
3453    
3454    9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
3455    other alternatives are tried instead.
3456    
3457    
3458    Version 2.06 09-Jun-99
3459    ----------------------
3460    
3461    1. Change pcretest's output for amount of store used to show just the code
3462    space, because the remainder (the data block) varies in size between 32-bit and
3463    64-bit systems.
3464    
3465    2. Added an extra argument to pcre_exec() to supply an offset in the subject to
3466    start matching at. This allows lookbehinds to work when searching for multiple
3467    occurrences in a string.
3468    
3469    3. Added additional options to pcretest for testing multiple occurrences:
3470    
3471       /+   outputs the rest of the string that follows a match
3472       /g   loops for multiple occurrences, using the new startoffset argument
3473       /G   loops for multiple occurrences by passing an incremented pointer
3474    
3475    4. PCRE wasn't doing the "first character" optimization for patterns starting
3476    with \b or \B, though it was doing it for other lookbehind assertions. That is,
3477    it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
3478    the letter 'x'. On long subject strings, this gives a significant speed-up.
3479    
3480    
3481    Version 2.05 21-Apr-99
3482    ----------------------
3483    
3484    1. Changed the type of magic_number from int to long int so that it works
3485    properly on 16-bit systems.
3486    
3487    2. Fixed a bug which caused patterns starting with .* not to work correctly
3488    when the subject string contained newline characters. PCRE was assuming
3489    anchoring for such patterns in all cases, which is not correct because .* will
3490    not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
3491    DOTALL is set at top level; otherwise it knows that patterns starting with .*
3492    must be retried after every newline in the subject.
3493    
3494    
3495    Version 2.04 18-Feb-99
3496    ----------------------
3497    
3498    1. For parenthesized subpatterns with repeats whose minimum was zero, the
3499    computation of the store needed to hold the pattern was incorrect (too large).
3500    If such patterns were nested a few deep, this could multiply and become a real
3501    problem.
3502    
3503    2. Added /M option to pcretest to show the memory requirement of a specific
3504    pattern. Made -m a synonym of -s (which does this globally) for compatibility.
3505    
3506    3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
3507    compiled in such a way that the backtracking after subsequent failure was
3508    pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
3509    ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
3510    
3511    
3512    Version 2.03 02-Feb-99
3513    ----------------------
3514    
3515    1. Fixed typo and small mistake in man page.
3516    
3517    2. Added 4th condition (GPL supersedes if conflict) and created separate
3518    LICENCE file containing the conditions.
3519    
3520    3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
3521    Perl, that is the internal \ allows the delimiter to be included in the
3522    pattern. Locked out the use of \ as a delimiter. If \ immediately follows
3523    the final delimiter, add \ to the end of the pattern (to test the error).
3524    
3525    4. Added the convenience functions for extracting substrings after a successful
3526    match. Updated pcretest to make it able to test these functions.
3527    
3528    
3529    Version 2.02 14-Jan-99
3530    ----------------------
3531    
3532    1. Initialized the working variables associated with each extraction so that
3533    their saving and restoring doesn't refer to uninitialized store.
3534    
3535    2. Put dummy code into study.c in order to trick the optimizer of the IBM C
3536    compiler for OS/2 into generating correct code. Apparently IBM isn't going to
3537    fix the problem.
3538    
3539    3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
3540    calls, and wasn't printing the correct value for compiling calls. Increased the
3541    default value of LOOPREPEAT, and the number of significant figures in the
3542    times.
3543    
3544    4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
3545    
3546    5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
3547    a building problem on Windows NT with a FAT file system.
3548    
3549    
3550    Version 2.01 21-Oct-98
3551    ----------------------
3552    
3553    1. Changed the API for pcre_compile() to allow for the provision of a pointer
3554    to character tables built by pcre_maketables() in the current locale. If NULL
3555    is passed, the default tables are used.
3556    
3557    
3558  Version 2.00 24-Sep-98  Version 2.00 24-Sep-98
3559  ----------------------  ----------------------

Legend:
Removed from v.23  
changed lines
  Added in v.521

  ViewVC Help
Powered by ViewVC 1.1.5