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

Diff of /code/trunk/ChangeLog

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

revision 19 by nigel, Sat Feb 24 21:38:33 2007 UTC revision 524 by ph10, Mon May 24 17:06:28 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    13. If a pattern that was studied started with a repeated Unicode property
53        test, for example, \p{Nd}+, there was the theoretical possibility of
54        setting up an incorrect bitmap of starting bytes, but fortunately it could
55        not have actually happened in practice until change 8 above was made (it
56        added property types that matched character-matching opcodes).
57    
58    14. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
59        possible starting bytes for non-anchored patterns.
60    
61    
62    
63    Version 8.02 19-Mar-2010
64    ------------------------
65    
66    1.  The Unicode data tables have been updated to Unicode 5.2.0.
67    
68    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
69        configured.
70    
71    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
72        original author of that file, following a query about its status.
73    
74    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
75        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
76    
77    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
78        quantifier applied to a forward-referencing subroutine call, could compile
79        incorrect code or give the error "internal error: previously-checked
80        referenced subpattern not found".
81    
82    6.  Both MS Visual Studio and Symbian OS have problems with initializing
83        variables to point to external functions. For these systems, therefore,
84        pcre_malloc etc. are now initialized to local functions that call the
85        relevant global functions.
86    
87    7.  There were two entries missing in the vectors called coptable and poptable
88        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
89        I've fixed the data, and added a kludgy way of testing at compile time that
90        the lengths are correct (equal to the number of opcodes).
91    
92    8.  Following on from 7, I added a similar kludge to check the length of the
93        eint vector in pcreposix.c.
94    
95    9.  Error texts for pcre_compile() are held as one long string to avoid too
96        much relocation at load time. To find a text, the string is searched,
97        counting zeros. There was no check for running off the end of the string,
98        which could happen if a new error number was added without updating the
99        string.
100    
101    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
102    
103    11. \K was not working if it appeared in an atomic group or in a group that
104        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
105        \K is "not well defined" if used in an assertion. PCRE now accepts it if
106        the assertion is positive, but not if it is negative.
107    
108    12. Change 11 fortuitously reduced the size of the stack frame used in the
109        "match()" function of pcre_exec.c by one pointer. Forthcoming
110        implementation of support for (*MARK) will need an extra pointer on the
111        stack; I have reserved it now, so that the stack frame size does not
112        decrease.
113    
114    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
115        item in branch that calls a recursion is a subroutine call - as in the
116        second branch in the above example - was incorrectly given the compile-
117        time error "recursive call could loop indefinitely" because pcre_compile()
118        was not correctly checking the subroutine for matching a non-empty string.
119    
120    14. The checks for overrunning compiling workspace could trigger after an
121        overrun had occurred. This is a "should never occur" error, but it can be
122        triggered by pathological patterns such as hundreds of nested parentheses.
123        The checks now trigger 100 bytes before the end of the workspace.
124    
125    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
126    
127    
128    Version 8.01 19-Jan-2010
129    ------------------------
130    
131    1.  If a pattern contained a conditional subpattern with only one branch (in
132        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
133        computed the wrong minimum data length (which is of course zero for such
134        subpatterns). This could cause incorrect "no match" results.
135    
136    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
137        the pattern is reset in the first branch, pcre_compile() failed with
138        "internal error: code overflow at offset...". This happened only when
139        the reset was to the original external option setting. (An optimization
140        abstracts leading options settings into an external setting, which was the
141        cause of this.)
142    
143    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
144        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
145        assertion pattern did not match (meaning that the assertion was true), it
146        was incorrectly treated as false if the SKIP had been reached during the
147        matching. This also applied to assertions used as conditions.
148    
149    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
150        assertion subpattern, including such a pattern used as a condition,
151        unpredictable results occurred, instead of the error return
152        PCRE_ERROR_DFA_UITEM.
153    
154    5.  The C++ GlobalReplace function was not working like Perl for the special
155        situation when an empty string is matched. It now does the fancy magic
156        stuff that is necessary.
157    
158    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
159        removed. (These were left over from very, very early versions of PCRE.)
160    
161    7.  Some cosmetic changes to the code to make life easier when compiling it
162        as part of something else:
163    
164        (a) Change DEBUG to PCRE_DEBUG.
165    
166        (b) In pcre_compile(), rename the member of the "branch_chain" structure
167            called "current" as "current_branch", to prevent a collision with the
168            Linux macro when compiled as a kernel module.
169    
170        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
171            prevent a collision with the Linux macro when compiled as a kernel
172            module.
173    
174    8.  In pcre_compile() there are some checks for integer overflows that used to
175        cast potentially large values to (double). This has been changed to that
176        when building, a check for int64_t is made, and if it is found, it is used
177        instead, thus avoiding the use of floating point arithmetic. (There is no
178        other use of FP in PCRE.) If int64_t is not found, the fallback is to
179        double.
180    
181    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
182        2005 (difference between two addresses compared to an unsigned value).
183    
184    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
185        custom one, because of the following reported problem in Windows:
186    
187          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
188              under Win32.
189          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
190              therefore missing the function definition.
191          - The compiler thus generates a "C" signature for the test function.
192          - The linker fails to find the "C" function.
193          - PCRE fails to configure if asked to do so against libbz2.
194    
195    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
196        messages were output:
197    
198          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
199          rerunning libtoolize, to keep the correct libtool macros in-tree.
200          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
201    
202        I have done both of these things.
203    
204    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
205        most of the time, it *can* run out if it is given a pattern that contains a
206        runaway infinite recursion. I updated the discussion in the pcrestack man
207        page.
208    
209    13. Now that we have gone to the x.xx style of version numbers, the minor
210        version may start with zero. Using 08 or 09 is a bad idea because users
211        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
212        interpreted as invalid octal numbers. I've updated the previous comment in
213        configure.ac, and also added a check that gives an error if 08 or 09 are
214        used.
215    
216    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
217        causing partial matching to fail when the end of the subject matched \W
218        in a UTF-8 pattern where \W was quantified with a minimum of 3.
219    
220    15. There were some discrepancies between the declarations in pcre_internal.h
221        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
222        their definitions. The declarations used "const uschar *" and the
223        definitions used USPTR. Even though USPTR is normally defined as "const
224        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
225        reported that: "This difference in casting confuses some C++ compilers, for
226        example, SunCC recognizes above declarations as different functions and
227        generates broken code for hbpcre." I have changed the declarations to use
228        USPTR.
229    
230    16. GNU libtool is named differently on some systems. The autogen.sh script now
231        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
232        (FreeBSD).
233    
234    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
235        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
236        comment: "Figure out how to create a longlong from a string: strtoll and
237        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
238        instance, but it only takes 2 args instead of 3!"
239    
240    18. A subtle bug concerned with back references has been fixed by a change of
241        specification, with a corresponding code fix. A pattern such as
242        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
243        refers, was giving matches when it shouldn't. For example, xa=xaaa would
244        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
245        same bug. Such groups have to be quantified to be useful, or contained
246        inside another quantified group. (If there's no repetition, the reference
247        can never match.) The problem arises because, having left the group and
248        moved on to the rest of the pattern, a later failure that backtracks into
249        the group uses the captured value from the final iteration of the group
250        rather than the correct earlier one. I have fixed this in PCRE by forcing
251        any group that contains a reference to itself to be an atomic group; that
252        is, there cannot be any backtracking into it once it has completed. This is
253        similar to recursive and subroutine calls.
254    
255    
256    Version 8.00 19-Oct-09
257    ----------------------
258    
259    1.  The table for translating pcre_compile() error codes into POSIX error codes
260        was out-of-date, and there was no check on the pcre_compile() error code
261        being within the table. This could lead to an OK return being given in
262        error.
263    
264    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
265        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
266        in a Windows environment.
267    
268    3.  The pcregrep --count option prints the count for each file even when it is
269        zero, as does GNU grep. However, pcregrep was also printing all files when
270        --files-with-matches was added. Now, when both options are given, it prints
271        counts only for those files that have at least one match. (GNU grep just
272        prints the file name in this circumstance, but including the count seems
273        more useful - otherwise, why use --count?) Also ensured that the
274        combination -clh just lists non-zero counts, with no names.
275    
276    4.  The long form of the pcregrep -F option was incorrectly implemented as
277        --fixed_strings instead of --fixed-strings. This is an incompatible change,
278        but it seems right to fix it, and I didn't think it was worth preserving
279        the old behaviour.
280    
281    5.  The command line items --regex=pattern and --regexp=pattern were not
282        recognized by pcregrep, which required --regex pattern or --regexp pattern
283        (with a space rather than an '='). The man page documented the '=' forms,
284        which are compatible with GNU grep; these now work.
285    
286    6.  No libpcreposix.pc file was created for pkg-config; there was just
287        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
288    
289    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
290        when UCP support is not needed, by modifying the Python script that
291        generates it from Unicode data files. This should not matter if the module
292        is correctly used as a library, but I received one complaint about 50K of
293        unwanted data. My guess is that the person linked everything into his
294        program rather than using a library. Anyway, it does no harm.
295    
296    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
297        was a minimum greater than 1 for a wide character in a possessive
298        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
299        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
300        character. Chaos in the form of incorrect output or a compiling loop could
301        result.
302    
303    9.  The restrictions on what a pattern can contain when partial matching is
304        requested for pcre_exec() have been removed. All patterns can now be
305        partially matched by this function. In addition, if there are at least two
306        slots in the offset vector, the offset of the earliest inspected character
307        for the match and the offset of the end of the subject are set in them when
308        PCRE_ERROR_PARTIAL is returned.
309    
310    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
311        synonymous with PCRE_PARTIAL, for backwards compatibility, and
312        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
313        and may be more useful for multi-segment matching.
314    
315    11. Partial matching with pcre_exec() is now more intuitive. A partial match
316        used to be given if ever the end of the subject was reached; now it is
317        given only if matching could not proceed because another character was
318        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
319        string "Z", which now yields "no match" instead of "partial match". In the
320        case of pcre_dfa_exec(), "no match" is given if every matching path for the
321        final character ended with (*FAIL).
322    
323    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
324        if the pattern had a "must contain" character that was already found in the
325        earlier partial match, unless partial matching was again requested. For
326        example, with the pattern /dog.(body)?/, the "must contain" character is
327        "g". If the first part-match was for the string "dog", restarting with
328        "sbody" failed. This bug has been fixed.
329    
330    13. The string returned by pcre_dfa_exec() after a partial match has been
331        changed so that it starts at the first inspected character rather than the
332        first character of the match. This makes a difference only if the pattern
333        starts with a lookbehind assertion or \b or \B (\K is not supported by
334        pcre_dfa_exec()). It's an incompatible change, but it makes the two
335        matching functions compatible, and I think it's the right thing to do.
336    
337    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
338        so that the demonstration program is easily available in environments where
339        PCRE has not been installed from source.
340    
341    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
342        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
343        library.
344    
345    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
346        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
347        is not the first non-POSIX option to be added. Clearly some people find
348        these options useful.
349    
350    17. If a caller to the POSIX matching function regexec() passes a non-zero
351        value for nmatch with a NULL value for pmatch, the value of
352        nmatch is forced to zero.
353    
354    18. RunGrepTest did not have a test for the availability of the -u option of
355        the diff command, as RunTest does. It now checks in the same way as
356        RunTest, and also checks for the -b option.
357    
358    19. If an odd number of negated classes containing just a single character
359        interposed, within parentheses, between a forward reference to a named
360        subpattern and the definition of the subpattern, compilation crashed with
361        an internal error, complaining that it could not find the referenced
362        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
363        [The bug was that it was starting one character too far in when skipping
364        over the character class, thus treating the ] as data rather than
365        terminating the class. This meant it could skip too much.]
366    
367    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
368        /g option in pcretest when the pattern contains \K, which makes it possible
369        to have an empty string match not at the start, even when the pattern is
370        anchored. Updated pcretest and pcredemo to use this option.
371    
372    21. If the maximum number of capturing subpatterns in a recursion was greater
373        than the maximum at the outer level, the higher number was returned, but
374        with unset values at the outer level. The correct (outer level) value is
375        now given.
376    
377    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
378        PCRE did not set those parentheses (unlike Perl). I have now found a way to
379        make it do so. The string so far is captured, making this feature
380        compatible with Perl.
381    
382    23. The tests have been re-organized, adding tests 11 and 12, to make it
383        possible to check the Perl 5.10 features against Perl 5.10.
384    
385    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
386        pattern matches a fixed length string. PCRE did not allow this; now it
387        does. Neither allows recursion.
388    
389    25. I finally figured out how to implement a request to provide the minimum
390        length of subject string that was needed in order to match a given pattern.
391        (It was back references and recursion that I had previously got hung up
392        on.) This code has now been added to pcre_study(); it finds a lower bound
393        to the length of subject needed. It is not necessarily the greatest lower
394        bound, but using it to avoid searching strings that are too short does give
395        some useful speed-ups. The value is available to calling programs via
396        pcre_fullinfo().
397    
398    26. While implementing 25, I discovered to my embarrassment that pcretest had
399        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
400        study optimizations had never been tested with that matching function.
401        Oops. What is worse, even when it was passed study data, there was a bug in
402        pcre_dfa_exec() that meant it never actually used it. Double oops. There
403        were also very few tests of studied patterns with pcre_dfa_exec().
404    
405    27. If (?| is used to create subpatterns with duplicate numbers, they are now
406        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
407        on the other side of the coin, they are no longer allowed to have different
408        names, because these cannot be distinguished in PCRE, and this has caused
409        confusion. (This is a difference from Perl.)
410    
411    28. When duplicate subpattern names are present (necessarily with different
412        numbers, as required by 27 above), and a test is made by name in a
413        conditional pattern, either for a subpattern having been matched, or for
414        recursion in such a pattern, all the associated numbered subpatterns are
415        tested, and the overall condition is true if the condition is true for any
416        one of them. This is the way Perl works, and is also more like the way
417        testing by number works.
418    
419    
420    Version 7.9 11-Apr-09
421    ---------------------
422    
423    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
424        (pcretest), all targets were linked against these libraries. This included
425        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
426        libraries. This caused unwanted dependencies to be created. This problem
427        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
428        pcretest is linked with readline.
429    
430    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
431        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
432        moved outside it again, because FALSE and TRUE are already defined in AIX,
433        but BOOL is not.
434    
435    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
436        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
437    
438    4.  The pcregrep documentation said spaces were inserted as well as colons (or
439        hyphens) following file names and line numbers when outputting matching
440        lines. This is not true; no spaces are inserted. I have also clarified the
441        wording for the --colour (or --color) option.
442    
443    5.  In pcregrep, when --colour was used with -o, the list of matching strings
444        was not coloured; this is different to GNU grep, so I have changed it to be
445        the same.
446    
447    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
448        each matching line was coloured. Now it goes on to look for further matches
449        of any of the test patterns, which is the same behaviour as GNU grep.
450    
451    7.  A pattern that could match an empty string could cause pcregrep to loop; it
452        doesn't make sense to accept an empty string match in pcregrep, so I have
453        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
454        seems to be how GNU grep behaves.
455    
456    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
457        start or after a newline", because the conditional assertion was not being
458        correctly handled. The rule now is that both the assertion and what follows
459        in the first alternative must satisfy the test.
460    
461    9.  If auto-callout was enabled in a pattern with a conditional group whose
462        condition was an assertion, PCRE could crash during matching, both with
463        pcre_exec() and pcre_dfa_exec().
464    
465    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
466        used for matching.
467    
468    11. Unicode property support in character classes was not working for
469        characters (bytes) greater than 127 when not in UTF-8 mode.
470    
471    12. Added the -M command line option to pcretest.
472    
473    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
474    
475    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
476    
477    16. Added comments and documentation about mis-use of no_arg in the C++
478        wrapper.
479    
480    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
481        from Martin Jerabek that uses macro names for all relevant character and
482        string constants.
483    
484    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
485        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
486        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
487        these, but not everybody uses configure.
488    
489    19. A conditional group that had only one branch was not being correctly
490        recognized as an item that could match an empty string. This meant that an
491        enclosing group might also not be so recognized, causing infinite looping
492        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
493        with the subject "ab", where knowledge that the repeated group can match
494        nothing is needed in order to break the loop.
495    
496    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
497        exec(), but without supplying a callout function, matching went wrong.
498    
499    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
500        leak if the size of the offset vector was greater than 30. When the vector
501        is smaller, the saved offsets during recursion go onto a local stack
502        vector, but for larger vectors malloc() is used. It was failing to free
503        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
504        error, in fact).
505    
506    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
507        heapframe that is used only when UTF-8 support is enabled. This caused no
508        problem, but was untidy.
509    
510    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
511        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
512        included within another project.
513    
514    24. Steven Van Ingelgem's patches to add more options to the CMake support,
515        slightly modified by me:
516    
517          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
518              not building pcregrep.
519    
520          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
521              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
522    
523    25. Forward references, both numeric and by name, in patterns that made use of
524        duplicate group numbers, could behave incorrectly or give incorrect errors,
525        because when scanning forward to find the reference group, PCRE was not
526        taking into account the duplicate group numbers. A pattern such as
527        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
528    
529    26. Changed a few more instances of "const unsigned char *" to USPTR, making
530        the feature of a custom pointer more persuasive (as requested by a user).
531    
532    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
533        pcretest.c, inside #ifndefs, because it seems they are sometimes already
534        pre-defined.
535    
536    28. Added support for (*UTF8) at the start of a pattern.
537    
538    29. Arrange for flags added by the "release type" setting in CMake to be shown
539        in the configuration summary.
540    
541    
542    Version 7.8 05-Sep-08
543    ---------------------
544    
545    1.  Replaced UCP searching code with optimized version as implemented for Ad
546        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
547        stage table and inline lookup instead of a function, giving speed ups of 2
548        to 5 times on some simple patterns that I tested. Permission was given to
549        distribute the MultiStage2.py script that generates the tables (it's not in
550        the tarball, but is in the Subversion repository).
551    
552    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
553        scripts.
554    
555    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
556        a group with a zero qualifier. The result of the study could be incorrect,
557        or the function might crash, depending on the pattern.
558    
559    4.  Caseless matching was not working for non-ASCII characters in back
560        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
561        It now works when Unicode Property Support is available.
562    
563    5.  In pcretest, an escape such as \x{de} in the data was always generating
564        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
565        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
566        truncation.
567    
568    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
569    
570    7.  Added two (int) casts to pcregrep when printing the difference of two
571        pointers, in case they are 64-bit values.
572    
573    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
574        test 2 if it fails.
575    
576    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
577        and a #define of that name to empty if it is not externally set. This is to
578        allow users of MSVC to set it if necessary.
579    
580    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
581        the convenience functions in the pcre_get.c source file.
582    
583    11. An option change at the start of a pattern that had top-level alternatives
584        could cause overwriting and/or a crash. This command provoked a crash in
585        some environments:
586    
587          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
588    
589        This potential security problem was recorded as CVE-2008-2371.
590    
591    12. For a pattern where the match had to start at the beginning or immediately
592        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
593        pcre_dfa_exec() could read past the end of the passed subject if there was
594        no match. To help with detecting such bugs (e.g. with valgrind), I modified
595        pcretest so that it places the subject at the end of its malloc-ed buffer.
596    
597    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
598        exec() might read past the end of the data buffer in UTF-8 mode.
599    
600    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
601        the data contained the byte 0x85 as part of a UTF-8 character within its
602        first line. This applied both to normal and DFA matching.
603    
604    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
605        /^[^d]*?$/8 failed to match "abc".
606    
607    16. Added a missing copyright notice to pcrecpp_internal.h.
608    
609    17. Make it more clear in the documentation that values returned from
610        pcre_exec() in ovector are byte offsets, not character counts.
611    
612    18. Tidied a few places to stop certain compilers from issuing warnings.
613    
614    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
615        supplied by Stefan Weber. I made a further small update for 7.8 because
616        there is a change of source arrangements: the pcre_searchfuncs.c module is
617        replaced by pcre_ucd.c.
618    
619    
620    Version 7.7 07-May-08
621    ---------------------
622    
623    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
624        a string to a long long, pretend we don't even have a long long." This is
625        done by checking for the strtoq, strtoll, and _strtoi64 functions.
626    
627    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
628        pre-7.6 versions, which defined a global no_arg variable instead of putting
629        it in the RE class. (See also #8 below.)
630    
631    3.  Remove a line of dead code, identified by coverity and reported by Nuno
632        Lopes.
633    
634    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
635    
636        (1) The include/exclude patterns were being applied to the whole pathnames
637            of files, instead of just to the final components.
638    
639        (2) If there was more than one level of directory, the subdirectories were
640            skipped unless they satisfied the include/exclude conditions. This is
641            inconsistent with GNU grep (and could even be seen as contrary to the
642            pcregrep specification - which I improved to make it absolutely clear).
643            The action now is always to scan all levels of directory, and just
644            apply the include/exclude patterns to regular files.
645    
646    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
647        --exclude_dir in the tests to avoid scanning .svn directories.
648    
649    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
650        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
651        doesn't support NULs in patterns.
652    
653    7.  Added some missing "const"s to declarations of static tables in
654        pcre_compile.c and pcre_dfa_exec.c.
655    
656    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
657        caused by fix #2  above. (Subsequently also a second patch to fix the
658        first patch. And a third patch - this was a messy problem.)
659    
660    9.  Applied Craig's patch to remove the use of push_back().
661    
662    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
663        matching function regexec().
664    
665    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
666        which, however, unlike Perl's \g{...}, are subroutine calls, not back
667        references. PCRE supports relative numbers with this syntax (I don't think
668        Oniguruma does).
669    
670    12. Previously, a group with a zero repeat such as (...){0} was completely
671        omitted from the compiled regex. However, this means that if the group
672        was called as a subroutine from elsewhere in the pattern, things went wrong
673        (an internal error was given). Such groups are now left in the compiled
674        pattern, with a new opcode that causes them to be skipped at execution
675        time.
676    
677    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
678        to the way PCRE behaves:
679    
680        (a) A lone ] character is dis-allowed (Perl treats it as data).
681    
682        (b) A back reference to an unmatched subpattern matches an empty string
683            (Perl fails the current match path).
684    
685        (c) A data ] in a character class must be notated as \] because if the
686            first data character in a class is ], it defines an empty class. (In
687            Perl it is not possible to have an empty class.) The empty class []
688            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
689            The negative empty class [^] matches any one character, independently
690            of the DOTALL setting.
691    
692    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
693        non-existent subpattern following a character class starting with ']' and
694        containing () gave an internal compiling error instead of "reference to
695        non-existent subpattern". Fortunately, when the pattern did exist, the
696        compiled code was correct. (When scanning forwards to check for the
697        existencd of the subpattern, it was treating the data ']' as terminating
698        the class, so got the count wrong. When actually compiling, the reference
699        was subsequently set up correctly.)
700    
701    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
702        it was being rejected as not supported by pcre_dfa_exec(), even though
703        other assertions are supported. I have made pcre_dfa_exec() support
704        (*FAIL).
705    
706    16. The implementation of 13c above involved the invention of a new opcode,
707        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
708        cannot be changed at match time, I realized I could make a small
709        improvement to matching performance by compiling OP_ALLANY instead of
710        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
711        on the OP_ANY path.
712    
713    17. Compiling pcretest on Windows with readline support failed without the
714        following two fixes: (1) Make the unistd.h include conditional on
715        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
716    
717    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
718        ncurses library to be included for pcretest when ReadLine support is
719        requested, but also to allow for it to be overridden. This patch came from
720        Daniel Bergström.
721    
722    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
723        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
724        any errors with the current Unicode tables. Thanks to Peter Kankowski for
725        spotting this.
726    
727    
728    Version 7.6 28-Jan-08
729    ---------------------
730    
731    1.  A character class containing a very large number of characters with
732        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
733        overflow.
734    
735    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
736        HAVE_LONG_LONG is not defined.
737    
738    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
739        bring them up to date and include new features. This patch includes:
740    
741        - Fixed PH's badly added libz and libbz2 support.
742        - Fixed a problem with static linking.
743        - Added pcredemo. [But later removed - see 7 below.]
744        - Fixed dftables problem and added an option.
745        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
746            HAVE_LONG_LONG.
747        - Added readline support for pcretest.
748        - Added an listing of the option settings after cmake has run.
749    
750    4.  A user submitted a patch to Makefile that makes it easy to create
751        "pcre.dll" under mingw when using Configure/Make. I added stuff to
752        Makefile.am that cause it to include this special target, without
753        affecting anything else. Note that the same mingw target plus all
754        the other distribution libraries and programs are now supported
755        when configuring with CMake (see 6 below) instead of with
756        Configure/Make.
757    
758    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
759        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
760        exported in the Windows port". It has not yet been confirmed that the patch
761        solves the problem, but it does no harm.
762    
763    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
764        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
765        with CMake, and also correct the comment about stack recursion.
766    
767    7.  Remove the automatic building of pcredemo from the ./configure system and
768        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
769        of a program that users should build themselves after PCRE is installed, so
770        building it automatically is not really right. What is more, it gave
771        trouble in some build environments.
772    
773    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
774    
775    
776    Version 7.5 10-Jan-08
777    ---------------------
778    
779    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
780        values in parens when parsing an RE using the C++ wrapper."
781    
782    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
783        Characters greater than 255 were excluded from the class instead of being
784        included.
785    
786    3.  The same bug as (2) above applied to negated POSIX classes such as
787        [:^space:].
788    
789    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
790        defined or documented. It seems to have been a typo for PCRE_STATIC, so
791        I have changed it.
792    
793    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
794        first named subpattern) and a construct such as (?&a) would reference the
795        first named subpattern whose name started with "a" (in other words, the
796        length check was missing). Both these problems are fixed. "Subpattern name
797        expected" is now given for (?&) (a zero-length name), and this patch also
798        makes it give the same error for \k'' (previously it complained that that
799        was a reference to a non-existent subpattern).
800    
801    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
802        this is right because (?- can be followed by option settings as well as by
803        digits. I have, however, made the messages clearer.
804    
805    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
806        than the number used in the conditional) now cause a compile-time error.
807        This is actually not compatible with Perl, which accepts such patterns, but
808        treats the conditional as always being FALSE (as PCRE used to), but it
809        seems to me that giving a diagnostic is better.
810    
811    8.  Change "alphameric" to the more common word "alphanumeric" in comments
812        and messages.
813    
814    9.  Fix two occurrences of "backslash" in comments that should have been
815        "backspace".
816    
817    10. Remove two redundant lines of code that can never be obeyed (their function
818        was moved elsewhere).
819    
820    11. The program that makes PCRE's Unicode character property table had a bug
821        which caused it to generate incorrect table entries for sequences of
822        characters that have the same character type, but are in different scripts.
823        It amalgamated them into a single range, with the script of the first of
824        them. In other words, some characters were in the wrong script. There were
825        thirteen such cases, affecting characters in the following ranges:
826    
827          U+002b0 - U+002c1
828          U+0060c - U+0060d
829          U+0061e - U+00612
830          U+0064b - U+0065e
831          U+0074d - U+0076d
832          U+01800 - U+01805
833          U+01d00 - U+01d77
834          U+01d9b - U+01dbf
835          U+0200b - U+0200f
836          U+030fc - U+030fe
837          U+03260 - U+0327f
838          U+0fb46 - U+0fbb1
839          U+10450 - U+1049d
840    
841    12. The -o option (show only the matching part of a line) for pcregrep was not
842        compatible with GNU grep in that, if there was more than one match in a
843        line, it showed only the first of them. It now behaves in the same way as
844        GNU grep.
845    
846    13. If the -o and -v options were combined for pcregrep, it printed a blank
847        line for every non-matching line. GNU grep prints nothing, and pcregrep now
848        does the same. The return code can be used to tell if there were any
849        non-matching lines.
850    
851    14. Added --file-offsets and --line-offsets to pcregrep.
852    
853    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
854        infinitely looping recursion. The bug was that positive lookaheads were not
855        being skipped when checking for a possible empty match (negative lookaheads
856        and both kinds of lookbehind were skipped).
857    
858    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
859        inclusion of <windows.h> to before rather than after the definition of
860        INVALID_FILE_ATTRIBUTES (patch from David Byron).
861    
862    17. Specifying a possessive quantifier with a specific limit for a Unicode
863        character property caused pcre_compile() to compile bad code, which led at
864        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
865        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
866        caused the error; without that there was no problem.
867    
868    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
869    
870    19. Added --enable-pcretest-libreadline.
871    
872    20. In pcrecpp.cc, the variable 'count' was incremented twice in
873        RE::GlobalReplace(). As a result, the number of replacements returned was
874        double what it should be. I removed one of the increments, but Craig sent a
875        later patch that removed the other one (the right fix) and added unit tests
876        that check the return values (which was not done before).
877    
878    21. Several CMake things:
879    
880        (1) Arranged that, when cmake is used on Unix, the libraries end up with
881            the names libpcre and libpcreposix, not just pcre and pcreposix.
882    
883        (2) The above change means that pcretest and pcregrep are now correctly
884            linked with the newly-built libraries, not previously installed ones.
885    
886        (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
887    
888    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
889        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
890        UTF-8 newline character). The key issue is that the pattern starts .*;
891        this means that the match must be either at the beginning, or after a
892        newline. The bug was in the code for advancing after a failed match and
893        checking that the new position followed a newline. It was not taking
894        account of UTF-8 characters correctly.
895    
896    23. PCRE was behaving differently from Perl in the way it recognized POSIX
897        character classes. PCRE was not treating the sequence [:...:] as a
898        character class unless the ... were all letters. Perl, however, seems to
899        allow any characters between [: and :], though of course it rejects as
900        unknown any "names" that contain non-letters, because all the known class
901        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
902        for example, whereas PCRE did not - it did not recognize a POSIX character
903        class. This seemed a bit dangerous, so the code has been changed to be
904        closer to Perl. The behaviour is not identical to Perl, because PCRE will
905        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
906        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
907        Perl does, and where it didn't before.
908    
909    24. Rewrite so as to remove the single use of %n from pcregrep because in some
910        Windows environments %n is disabled by default.
911    
912    
913    Version 7.4 21-Sep-07
914    ---------------------
915    
916    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
917        means that a class such as [\s] counted as "explicit reference to CR or
918        LF". That isn't really right - the whole point of the change was to try to
919        help when there was an actual mention of one of the two characters. So now
920        the change happens only if \r or \n (or a literal CR or LF) character is
921        encountered.
922    
923    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
924        of both had grown to the point where there were only 3 bits left.
925        Fortunately, there was spare space in the data structure, and so I have
926        moved the internal flags into a new 16-bit field to free up more option
927        bits.
928    
929    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
930        but did not set the internal JCHANGED flag - either of these is enough to
931        control the way the "get" function works - but the PCRE_INFO_JCHANGED
932        facility is supposed to tell if (?J) was ever used, so now (?J) at the
933        start sets both bits.
934    
935    4.  Added options (at build time, compile time, exec time) to change \R from
936        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
937    
938    5.  doc/pcresyntax.html was missing from the distribution.
939    
940    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
941        compatibility, even though it is no longer used.
942    
943    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
944        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
945        windows.h file is present (where different names are used). [This was
946        reversed later after testing - see 16 below.]
947    
948    8.  Changed all #include <config.h> to #include "config.h". There were also
949        some further <pcre.h> cases that I changed to "pcre.h".
950    
951    9.  When pcregrep was used with the --colour option, it missed the line ending
952        sequence off the lines that it output.
953    
954    10. It was pointed out to me that arrays of string pointers cause lots of
955        relocations when a shared library is dynamically loaded. A technique of
956        using a single long string with a table of offsets can drastically reduce
957        these. I have refactored PCRE in four places to do this. The result is
958        dramatic:
959    
960          Originally:                          290
961          After changing UCP table:            187
962          After changing error message table:   43
963          After changing table of "verbs"       36
964          After changing table of Posix names   22
965    
966        Thanks to the folks working on Gregex for glib for this insight.
967    
968    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
969        unicode-properties was also set.
970    
971    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
972    
973    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
974        checked only for CRLF.
975    
976    14. Added casts to pcretest.c to avoid compiler warnings.
977    
978    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
979    
980    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
981        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
982        entirely. This removes changes made in 7 above.
983    
984    17. The CMake files have been updated, and there is now more information about
985        building with CMake in the NON-UNIX-USE document.
986    
987    
988    Version 7.3 28-Aug-07
989    ---------------------
990    
991     1. In the rejigging of the build system that eventually resulted in 7.1, the
992        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
993        brackets there is not right, since it causes compilers to look for an
994        installed pcre.h, not the version that is in the source that is being
995        compiled (which of course may be different). I have changed it back to:
996    
997          #include "pcre.h"
998    
999        I have a vague recollection that the change was concerned with compiling in
1000        different directories, but in the new build system, that is taken care of
1001        by the VPATH setting the Makefile.
1002    
1003     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
1004        when the subject happened to end in the byte 0x85 (e.g. if the last
1005        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
1006        characters but of course it shouldn't be taken as a newline when it is part
1007        of another character. The bug was that, for an unlimited repeat of . in
1008        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
1009        characters when looking for a newline.
1010    
1011     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
1012    
1013     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
1014        in debug output.
1015    
1016     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
1017        long printing in the pcrecpp unittest when running under MinGW.
1018    
1019     6. ESC_K was left out of the EBCDIC table.
1020    
1021     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
1022        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
1023        limit also applies to "virtual nesting" when a pattern is recursive, and in
1024        this case 1000 isn't so big. I have been able to remove this limit at the
1025        expense of backing off one optimization in certain circumstances. Normally,
1026        when pcre_exec() would call its internal match() function recursively and
1027        immediately return the result unconditionally, it uses a "tail recursion"
1028        feature to save stack. However, when a subpattern that can match an empty
1029        string has an unlimited repetition quantifier, it no longer makes this
1030        optimization. That gives it a stack frame in which to save the data for
1031        checking that an empty string has been matched. Previously this was taken
1032        from the 1000-entry workspace that had been reserved. So now there is no
1033        explicit limit, but more stack is used.
1034    
1035     8. Applied Daniel's patches to solve problems with the import/export magic
1036        syntax that is required for Windows, and which was going wrong for the
1037        pcreposix and pcrecpp parts of the library. These were overlooked when this
1038        problem was solved for the main library.
1039    
1040     9. There were some crude static tests to avoid integer overflow when computing
1041        the size of patterns that contain repeated groups with explicit upper
1042        limits. As the maximum quantifier is 65535, the maximum group length was
1043        set at 30,000 so that the product of these two numbers did not overflow a
1044        32-bit integer. However, it turns out that people want to use groups that
1045        are longer than 30,000 bytes (though not repeat them that many times).
1046        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1047        made it possible to implement the integer overflow checks in a much more
1048        dynamic way, which I have now done. The artificial limitation on group
1049        length has been removed - we now have only the limit on the total length of
1050        the compiled pattern, which depends on the LINK_SIZE setting.
1051    
1052    10. Fixed a bug in the documentation for get/copy named substring when
1053        duplicate names are permitted. If none of the named substrings are set, the
1054        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1055        empty string.
1056    
1057    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1058        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1059        because the ] is interpreted as the first data character and the
1060        terminating ] is not found. PCRE has been made compatible with Perl in this
1061        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1062        cause memory overwriting.
1063    
1064    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1065        string has been matched (to stop an infinite loop). It was not recognizing
1066        a conditional subpattern that could match an empty string if that
1067        subpattern was within another subpattern. For example, it looped when
1068        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
1069        condition was not nested. This bug has been fixed.
1070    
1071    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1072        past the start of the subject in the presence of bytes with the top bit
1073        set, for example "\x8aBCD".
1074    
1075    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1076        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1077    
1078    14. Optimized (?!) to (*FAIL).
1079    
1080    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1081        This restricts code points to be within the range 0 to 0x10FFFF, excluding
1082        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1083        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1084        does: it's just the validity check that is more restrictive.
1085    
1086    16. Inserted checks for integer overflows during escape sequence (backslash)
1087        processing, and also fixed erroneous offset values for syntax errors during
1088        backslash processing.
1089    
1090    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1091        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1092    
1093    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1094        caused an overrun.
1095    
1096    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1097        something other than just ASCII characters) inside a group that had an
1098        unlimited repeat caused a loop at compile time (while checking to see
1099        whether the group could match an empty string).
1100    
1101    20. Debugging a pattern containing \p or \P could cause a crash. For example,
1102        [\P{Any}] did so. (Error in the code for printing property names.)
1103    
1104    21. An orphan \E inside a character class could cause a crash.
1105    
1106    22. A repeated capturing bracket such as (A)? could cause a wild memory
1107        reference during compilation.
1108    
1109    23. There are several functions in pcre_compile() that scan along a compiled
1110        expression for various reasons (e.g. to see if it's fixed length for look
1111        behind). There were bugs in these functions when a repeated \p or \P was
1112        present in the pattern. These operators have additional parameters compared
1113        with \d, etc, and these were not being taken into account when moving along
1114        the compiled data. Specifically:
1115    
1116        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1117            length.
1118    
1119        (b) An item such as \pL+ within a repeated group could cause crashes or
1120            loops.
1121    
1122        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1123            "reference to non-existent subpattern" error.
1124    
1125        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1126    
1127    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1128        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1129    
1130    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1131    
1132    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1133        character were causing crashes (broken optimization).
1134    
1135    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1136        \p or \P) caused a compile-time loop.
1137    
1138    28. More problems have arisen in unanchored patterns when CRLF is a valid line
1139        break. For example, the unstudied pattern [\r\n]A does not match the string
1140        "\r\nA" because change 7.0/46 below moves the current point on by two
1141        characters after failing to match at the start. However, the pattern \nA
1142        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
1143        the same is true. There doesn't seem any very clean way out of this, but
1144        what I have chosen to do makes the common cases work: PCRE now takes note
1145        of whether there can be an explicit match for \r or \n anywhere in the
1146        pattern, and if so, 7.0/46 no longer applies. As part of this change,
1147        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1148        pattern has explicit CR or LF references.
1149    
1150    29. Added (*CR) etc for changing newline setting at start of pattern.
1151    
1152    
1153    Version 7.2 19-Jun-07
1154    ---------------------
1155    
1156     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
1157        which is apparently normally available under Windows.
1158    
1159     2. Re-jig the pcregrep tests with different newline settings in an attempt
1160        to make them independent of the local environment's newline setting.
1161    
1162     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
1163    
1164     4. Some of the "internals" tests were previously cut out when the link size
1165        was not 2, because the output contained actual offsets. The recent new
1166        "Z" feature of pcretest means that these can be cut out, making the tests
1167        usable with all link sizes.
1168    
1169     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
1170        stack recursion. This gives a massive performance boost under BSD, but just
1171        a small improvement under Linux. However, it saves one field in the frame
1172        in all cases.
1173    
1174     6. Added more features from the forthcoming Perl 5.10:
1175    
1176        (a) (?-n) (where n is a string of digits) is a relative subroutine or
1177            recursion call. It refers to the nth most recently opened parentheses.
1178    
1179        (b) (?+n) is also a relative subroutine call; it refers to the nth next
1180            to be opened parentheses.
1181    
1182        (c) Conditions that refer to capturing parentheses can be specified
1183            relatively, for example, (?(-2)... or (?(+3)...
1184    
1185        (d) \K resets the start of the current match so that everything before
1186            is not part of it.
1187    
1188        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1189    
1190        (f) \g{name} is another synonym - part of Perl 5.10's unification of
1191            reference syntax.
1192    
1193        (g) (?| introduces a group in which the numbering of parentheses in each
1194            alternative starts with the same number.
1195    
1196        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
1197    
1198     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
1199        PCRE_INFO_JCHANGED.
1200    
1201     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
1202        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
1203        for detecting groups that can match an empty string.
1204    
1205     9. A pattern with a very large number of alternatives (more than several
1206        hundred) was running out of internal workspace during the pre-compile
1207        phase, where pcre_compile() figures out how much memory will be needed. A
1208        bit of new cunning has reduced the workspace needed for groups with
1209        alternatives. The 1000-alternative test pattern now uses 12 bytes of
1210        workspace instead of running out of the 4096 that are available.
1211    
1212    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
1213    
1214    11. Applied patch from Google to remove an optimization that didn't quite work.
1215        The report of the bug said:
1216    
1217          pcrecpp::RE("a*").FullMatch("aaa") matches, while
1218          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
1219          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
1220    
1221    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
1222        it matched the wrong number of bytes.
1223    
1224    
1225    Version 7.1 24-Apr-07
1226    ---------------------
1227    
1228     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
1229        that is more "standard", making use of automake and other Autotools. There
1230        is some re-arrangement of the files and adjustment of comments consequent
1231        on this.
1232    
1233     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
1234        for recursive directory scanning broke on some systems because the files
1235        are not scanned in any specific order and on different systems the order
1236        was different. A call to "sort" has been inserted into RunGrepTest for the
1237        approprate test as a short-term fix. In the longer term there may be an
1238        alternative.
1239    
1240     3. I had an email from Eric Raymond about problems translating some of PCRE's
1241        man pages to HTML (despite the fact that I distribute HTML pages, some
1242        people do their own conversions for various reasons). The problems
1243        concerned the use of low-level troff macros .br and .in. I have therefore
1244        removed all such uses from the man pages (some were redundant, some could
1245        be replaced by .nf/.fi pairs). The 132html script that I use to generate
1246        HTML has been updated to handle .nf/.fi and to complain if it encounters
1247        .br or .in.
1248    
1249     4. Updated comments in configure.ac that get placed in config.h.in and also
1250        arranged for config.h to be included in the distribution, with the name
1251        config.h.generic, for the benefit of those who have to compile without
1252        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
1253    
1254     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
1255        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
1256        makevp.bat for the current PCRE, using the additional files
1257        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
1258    
1259     6. A Windows user reported a minor discrepancy with test 2, which turned out
1260        to be caused by a trailing space on an input line that had got lost in his
1261        copy. The trailing space was an accident, so I've just removed it.
1262    
1263     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
1264        that is needed.
1265    
1266     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
1267        as "const" (a) because they are and (b) because it helps the PHP
1268        maintainers who have recently made a script to detect big data structures
1269        in the php code that should be moved to the .rodata section. I remembered
1270        to update Builducptable as well, so it won't revert if ucptable.h is ever
1271        re-created.
1272    
1273     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
1274        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
1275        order to be able to cut out the UTF-8 tables in the latter when UTF-8
1276        support is not required. This saves 1.5-2K of code, which is important in
1277        some applications.
1278    
1279        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
1280        so as not to refer to the tables, even though these functions will never be
1281        called when UTF-8 support is disabled. Otherwise there are problems with a
1282        shared library.
1283    
1284    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
1285    
1286        (a) It was defining its arguments as char * instead of void *.
1287    
1288        (b) It was assuming that all moves were upwards in memory; this was true
1289            a long time ago when I wrote it, but is no longer the case.
1290    
1291        The emulated memove() is provided for those environments that have neither
1292        memmove() nor bcopy(). I didn't think anyone used it these days, but that
1293        is clearly not the case, as these two bugs were recently reported.
1294    
1295    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
1296        and Detrail to create the HTML documentation, the .txt form of the man
1297        pages, and it removes trailing spaces from listed files. It also creates
1298        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
1299        case, it wraps all the #defines with #ifndefs. This script should be run
1300        before "make dist".
1301    
1302    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1303        with Unicode property support.
1304    
1305        (a) For a maximizing quantifier, if the two different cases of the
1306            character were of different lengths in their UTF-8 codings (there are
1307            some cases like this - I found 11), and the matching function had to
1308            back up over a mixture of the two cases, it incorrectly assumed they
1309            were both the same length.
1310    
1311        (b) When PCRE was configured to use the heap rather than the stack for
1312            recursion during matching, it was not correctly preserving the data for
1313            the other case of a UTF-8 character when checking ahead for a match
1314            while processing a minimizing repeat. If the check also involved
1315            matching a wide character, but failed, corruption could cause an
1316            erroneous result when trying to check for a repeat of the original
1317            character.
1318    
1319    13. Some tidying changes to the testing mechanism:
1320    
1321        (a) The RunTest script now detects the internal link size and whether there
1322            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1323            values substituted by "configure". (The RunGrepTest script already did
1324            this for UTF-8.) The configure.ac script no longer substitutes the
1325            relevant variables.
1326    
1327        (b) The debugging options /B and /D in pcretest show the compiled bytecode
1328            with length and offset values. This means that the output is different
1329            for different internal link sizes. Test 2 is skipped for link sizes
1330            other than 2 because of this, bypassing the problem. Unfortunately,
1331            there was also a test in test 3 (the locale tests) that used /B and
1332            failed for link sizes other than 2. Rather than cut the whole test out,
1333            I have added a new /Z option to pcretest that replaces the length and
1334            offset values with spaces. This is now used to make test 3 independent
1335            of link size. (Test 2 will be tidied up later.)
1336    
1337    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1338        segmentation fault instead of returning the appropriate error message.
1339    
1340    15. In multiline mode when the newline sequence was set to "any", the pattern
1341        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1342        This doesn't seem right; it now treats the CRLF combination as the line
1343        ending, and so does not match in that case. It's only a pattern such as ^$
1344        that would hit this one: something like ^ABC$ would have failed after \r
1345        and then tried again after \r\n.
1346    
1347    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1348        in an attempt to make files that differ only in their line terminators
1349        compare equal. This works on Linux.
1350    
1351    17. Under certain error circumstances pcregrep might try to free random memory
1352        as it exited. This is now fixed, thanks to valgrind.
1353    
1354    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1355        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1356        was because its rules for how to advance for /g after matching an empty
1357        string at the end of a line did not allow for this case. They now check for
1358        it specially.
1359    
1360    20. pcretest is supposed to handle patterns and data of any length, by
1361        extending its buffers when necessary. It was getting this wrong when the
1362        buffer for a data line had to be extended.
1363    
1364    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1365        CRLF as a newline sequence.
1366    
1367    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1368        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1369        I have nevertheless tidied it up.
1370    
1371    23. Added some casts to kill warnings from HP-UX ia64 compiler.
1372    
1373    24. Added a man page for pcre-config.
1374    
1375    
1376    Version 7.0 19-Dec-06
1377    ---------------------
1378    
1379     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1380        moving to gcc 4.1.1.
1381    
1382     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1383        sys/time.h, which is documented as needed for this function. It doesn't
1384        seem to matter on Linux, but it showed up on some releases of OS X.
1385    
1386     3. It seems that there are systems where bytes whose values are greater than
1387        127 match isprint() in the "C" locale. The "C" locale should be the
1388        default when a C program starts up. In most systems, only ASCII printing
1389        characters match isprint(). This difference caused the output from pcretest
1390        to vary, making some of the tests fail. I have changed pcretest so that:
1391    
1392        (a) When it is outputting text in the compiled version of a pattern, bytes
1393            other than 32-126 are always shown as hex escapes.
1394    
1395        (b) When it is outputting text that is a matched part of a subject string,
1396            it does the same, unless a different locale has been set for the match
1397            (using the /L modifier). In this case, it uses isprint() to decide.
1398    
1399     4. Fixed a major bug that caused incorrect computation of the amount of memory
1400        required for a compiled pattern when options that changed within the
1401        pattern affected the logic of the preliminary scan that determines the
1402        length. The relevant options are -x, and -i in UTF-8 mode. The result was
1403        that the computed length was too small. The symptoms of this bug were
1404        either the PCRE error "internal error: code overflow" from pcre_compile(),
1405        or a glibc crash with a message such as "pcretest: free(): invalid next
1406        size (fast)". Examples of patterns that provoked this bug (shown in
1407        pcretest format) are:
1408    
1409          /(?-x: )/x
1410          /(?x)(?-x: \s*#\s*)/
1411          /((?i)[\x{c0}])/8
1412          /(?i:[\x{c0}])/8
1413    
1414        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1415        is now done differently.
1416    
1417     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1418        wrapper classes; (b) implement a new function in the C++ scanner that is
1419        more efficient than the old way of doing things because it avoids levels of
1420        recursion in the regex matching; (c) add a paragraph to the documentation
1421        for the FullMatch() function.
1422    
1423     6. The escape sequence \n was being treated as whatever was defined as
1424        "newline". Not only was this contrary to the documentation, which states
1425        that \n is character 10 (hex 0A), but it also went horribly wrong when
1426        "newline" was defined as CRLF. This has been fixed.
1427    
1428     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1429        was being set to -1 for the "end of line" case (supposedly a value that no
1430        character can have). Though this value is never used (the check for end of
1431        line is "zero bytes in current character"), it caused compiler complaints.
1432        I've changed it to 0xffffffff.
1433    
1434     8. In pcre_version.c, the version string was being built by a sequence of
1435        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1436        string (as it is for production releases) called a macro with an empty
1437        argument. The C standard says the result of this is undefined. The gcc
1438        compiler treats it as an empty string (which was what was wanted) but it is
1439        reported that Visual C gives an error. The source has been hacked around to
1440        avoid this problem.
1441    
1442     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1443        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1444        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1445        of them did).
1446    
1447    10. Originally, pcretest opened its input and output without "b"; then I was
1448        told that "b" was needed in some environments, so it was added for release
1449        5.0 to both the input and output. (It makes no difference on Unix-like
1450        systems.) Later I was told that it is wrong for the input on Windows. I've
1451        now abstracted the modes into two macros, to make it easier to fiddle with
1452        them, and removed "b" from the input mode under Windows.
1453    
1454    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1455    
1456    12. Added -help and --help to pcretest as an official way of being reminded
1457        of the options.
1458    
1459    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1460        and pcrecpp.cc because they annoy compilers at high warning levels.
1461    
1462    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1463    
1464    15. Fixed an occurrence of == in configure.ac that should have been = (shell
1465        scripts are not C programs :-) and which was not noticed because it works
1466        on Linux.
1467    
1468    16. pcretest is supposed to handle any length of pattern and data line (as one
1469        line or as a continued sequence of lines) by extending its input buffer if
1470        necessary. This feature was broken for very long pattern lines, leading to
1471        a string of junk being passed to pcre_compile() if the pattern was longer
1472        than about 50K.
1473    
1474    17. I have done a major re-factoring of the way pcre_compile() computes the
1475        amount of memory needed for a compiled pattern. Previously, there was code
1476        that made a preliminary scan of the pattern in order to do this. That was
1477        OK when PCRE was new, but as the facilities have expanded, it has become
1478        harder and harder to keep it in step with the real compile phase, and there
1479        have been a number of bugs (see for example, 4 above). I have now found a
1480        cunning way of running the real compile function in a "fake" mode that
1481        enables it to compute how much memory it would need, while actually only
1482        ever using a few hundred bytes of working memory and without too many
1483        tests of the mode. This should make future maintenance and development
1484        easier. A side effect of this work is that the limit of 200 on the nesting
1485        depth of parentheses has been removed (though this was never a serious
1486        limitation, I suspect). However, there is a downside: pcre_compile() now
1487        runs more slowly than before (30% or more, depending on the pattern). I
1488        hope this isn't a big issue. There is no effect on runtime performance.
1489    
1490    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1491        newline (only possible for the last line of a file) and it was a
1492        pattern that set a locale (followed by /Lsomething), pcretest crashed.
1493    
1494    19. Added additional timing features to pcretest. (1) The -tm option now times
1495        matching only, not compiling. (2) Both -t and -tm can be followed, as a
1496        separate command line item, by a number that specifies the number of
1497        repeats to use when timing. The default is 50000; this gives better
1498        precision, but takes uncomfortably long for very large patterns.
1499    
1500    20. Extended pcre_study() to be more clever in cases where a branch of a
1501        subpattern has no definite first character. For example, (a*|b*)[cd] would
1502        previously give no result from pcre_study(). Now it recognizes that the
1503        first character must be a, b, c, or d.
1504    
1505    21. There was an incorrect error "recursive call could loop indefinitely" if
1506        a subpattern (or the entire pattern) that was being tested for matching an
1507        empty string contained only one non-empty item after a nested subpattern.
1508        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1509        incorrectly, because the \d was being skipped in the check.
1510    
1511    22. The pcretest program now has a new pattern option /B and a command line
1512        option -b, which is equivalent to adding /B to every pattern. This causes
1513        it to show the compiled bytecode, without the additional information that
1514        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1515        is the same as /B/I).
1516    
1517    23. A new optimization is now able automatically to treat some sequences such
1518        as a*b as a*+b. More specifically, if something simple (such as a character
1519        or a simple class like \d) has an unlimited quantifier, and is followed by
1520        something that cannot possibly match the quantified thing, the quantifier
1521        is automatically "possessified".
1522    
1523    24. A recursive reference to a subpattern whose number was greater than 39
1524        went wrong under certain circumstances in UTF-8 mode. This bug could also
1525        have affected the operation of pcre_study().
1526    
1527    25. Realized that a little bit of performance could be had by replacing
1528        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1529    
1530    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1531    
1532    27. Possessive quantifiers such as a++ were previously implemented by turning
1533        them into atomic groups such as ($>a+). Now they have their own opcodes,
1534        which improves performance. This includes the automatically created ones
1535        from 23 above.
1536    
1537    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1538        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1539        the first matched character to be a colon. This applied both to named and
1540        numbered groups.
1541    
1542    29. The ucpinternal.h header file was missing its idempotency #ifdef.
1543    
1544    30. I was sent a "project" file called libpcre.a.dev which I understand makes
1545        building PCRE on Windows easier, so I have included it in the distribution.
1546    
1547    31. There is now a check in pcretest against a ridiculously large number being
1548        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1549        loop, the loop is abandoned.
1550    
1551    32. Forward references to subpatterns in conditions such as (?(2)...) where
1552        subpattern 2 is defined later cause pcre_compile() to search forwards in
1553        the pattern for the relevant set of parentheses. This search went wrong
1554        when there were unescaped parentheses in a character class, parentheses
1555        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1556    
1557    33. "Subroutine" calls and backreferences were previously restricted to
1558        referencing subpatterns earlier in the regex. This restriction has now
1559        been removed.
1560    
1561    34. Added a number of extra features that are going to be in Perl 5.10. On the
1562        whole, these are just syntactic alternatives for features that PCRE had
1563        previously implemented using the Python syntax or my own invention. The
1564        other formats are all retained for compatibility.
1565    
1566        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1567            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1568            also .NET compatible.
1569    
1570        (b) A recursion or subroutine call to a named group can now be defined as
1571            (?&name) as well as (?P>name).
1572    
1573        (c) A backreference to a named group can now be defined as \k<name> or
1574            \k'name' as well as (?P=name). The new forms, as well as being in Perl
1575            5.10, are also .NET compatible.
1576    
1577        (d) A conditional reference to a named group can now use the syntax
1578            (?(<name>) or (?('name') as well as (?(name).
1579    
1580        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1581            groups (named and numbered) that are never evaluated inline, but can be
1582            called as "subroutines" from elsewhere. In effect, the DEFINE condition
1583            is always false. There may be only one alternative in such a group.
1584    
1585        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1586            as the simple (?(R). The condition is true only if the most recent
1587            recursion is that of the given number or name. It does not search out
1588            through the entire recursion stack.
1589    
1590        (g) The escape \gN or \g{N} has been added, where N is a positive or
1591            negative number, specifying an absolute or relative reference.
1592    
1593    35. Tidied to get rid of some further signed/unsigned compiler warnings and
1594        some "unreachable code" warnings.
1595    
1596    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1597        things, this adds five new scripts.
1598    
1599    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1600        There were also incompatibilities regarding the handling of \Q..\E inside
1601        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1602        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1603    
1604    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1605        matches an empty string, and forcibly breaks the loop. There were bugs in
1606        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
1607        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
1608        separate and independent bugs (that affected different cases) have been
1609        fixed.
1610    
1611    39. Refactored the code to abolish the use of different opcodes for small
1612        capturing bracket numbers. This is a tidy that I avoided doing when I
1613        removed the limit on the number of capturing brackets for 3.5 back in 2001.
1614        The new approach is not only tidier, it makes it possible to reduce the
1615        memory needed to fix the previous bug (38).
1616    
1617    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1618        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1619        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1620        mode.
1621    
1622    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1623        report.
1624    
1625    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1626        copy construction and assignment in the C++ wrapper.
1627    
1628    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1629        couple of bugs that could have given wrong results in the "--newline=crlf"
1630        case.
1631    
1632    44. Added a number of casts and did some reorganization of signed/unsigned int
1633        variables following suggestions from Dair Grant. Also renamed the variable
1634        "this" as "item" because it is a C++ keyword.
1635    
1636    45. Arranged for dftables to add
1637    
1638          #include "pcre_internal.h"
1639    
1640        to pcre_chartables.c because without it, gcc 4.x may remove the array
1641        definition from the final binary if PCRE is built into a static library and
1642        dead code stripping is activated.
1643    
1644    46. For an unanchored pattern, if a match attempt fails at the start of a
1645        newline sequence, and the newline setting is CRLF or ANY, and the next two
1646        characters are CRLF, advance by two characters instead of one.
1647    
1648    
1649    Version 6.7 04-Jul-06
1650    ---------------------
1651    
1652     1. In order to handle tests when input lines are enormously long, pcretest has
1653        been re-factored so that it automatically extends its buffers when
1654        necessary. The code is crude, but this _is_ just a test program. The
1655        default size has been increased from 32K to 50K.
1656    
1657     2. The code in pcre_study() was using the value of the re argument before
1658        testing it for NULL. (Of course, in any sensible call of the function, it
1659        won't be NULL.)
1660    
1661     3. The memmove() emulation function in pcre_internal.h, which is used on
1662        systems that lack both memmove() and bcopy() - that is, hardly ever -
1663        was missing a "static" storage class specifier.
1664    
1665     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1666        containing an extended class (one that cannot be represented by a bitmap
1667        because it contains high-valued characters or Unicode property items, e.g.
1668        [\pZ]). Almost always one would set UTF-8 mode when processing such a
1669        pattern, but PCRE should not loop if you do not (it no longer does).
1670        [Detail: two cases were found: (a) a repeated subpattern containing an
1671        extended class; (b) a recursive reference to a subpattern that followed a
1672        previous extended class. It wasn't skipping over the extended class
1673        correctly when UTF-8 mode was not set.]
1674    
1675     5. A negated single-character class was not being recognized as fixed-length
1676        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1677        compile error "lookbehind assertion is not fixed length".
1678    
1679     6. The RunPerlTest auxiliary script was showing an unexpected difference
1680        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1681        write a Perl script that can interpret lines of an input file either as
1682        byte characters or as UTF-8, which is what "perltest" was being required to
1683        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1684        can't do is switch easily at run time between having the "use utf8;" pragma
1685        or not. In the end, I fudged it by using the RunPerlTest script to insert
1686        "use utf8;" explicitly for the UTF-8 tests.
1687    
1688     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1689        the end of the subject string, contrary to the documentation and to what
1690        Perl does. This was true of both matching functions. Now it matches only at
1691        the start of the subject and immediately after *internal* newlines.
1692    
1693     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1694        a pointer to an int instead of a pointer to an unsigned long int. This
1695        caused problems on 64-bit systems.
1696    
1697     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1698        instance of the 'standard' template library not being so standard".
1699    
1700    10. There was no check on the number of named subpatterns nor the maximum
1701        length of a subpattern name. The product of these values is used to compute
1702        the size of the memory block for a compiled pattern. By supplying a very
1703        long subpattern name and a large number of named subpatterns, the size
1704        computation could be caused to overflow. This is now prevented by limiting
1705        the length of names to 32 characters, and the number of named subpatterns
1706        to 10,000.
1707    
1708    11. Subpatterns that are repeated with specific counts have to be replicated in
1709        the compiled pattern. The size of memory for this was computed from the
1710        length of the subpattern and the repeat count. The latter is limited to
1711        65535, but there was no limit on the former, meaning that integer overflow
1712        could in principle occur. The compiled length of a repeated subpattern is
1713        now limited to 30,000 bytes in order to prevent this.
1714    
1715    12. Added the optional facility to have named substrings with the same name.
1716    
1717    13. Added the ability to use a named substring as a condition, using the
1718        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1719        are numbers (not recommended). Forward references are permitted.
1720    
1721    14. Added forward references in named backreferences (if you see what I mean).
1722    
1723    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1724        pattern could run off the end of the subject. For example, the pattern
1725        "(?s)(.{1,5})"8 did this with the subject "ab".
1726    
1727    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1728        PCRE_CASELESS was set when matching characters that were quantified with ?
1729        or *.
1730    
1731    17. A character class other than a single negated character that had a minimum
1732        but no maximum quantifier - for example [ab]{6,} - was not handled
1733        correctly by pce_dfa_exec(). It would match only one character.
1734    
1735    18. A valid (though odd) pattern that looked like a POSIX character
1736        class but used an invalid character after [ (for example [[,abc,]]) caused
1737        pcre_compile() to give the error "Failed: internal error: code overflow" or
1738        in some cases to crash with a glibc free() error. This could even happen if
1739        the pattern terminated after [[ but there just happened to be a sequence of
1740        letters, a binary zero, and a closing ] in the memory that followed.
1741    
1742    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1743        over the years. Originally (before any Unicode support), just the bottom 8
1744        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1745        output from "man perlunicode" includes this:
1746    
1747          The regular expression compiler produces polymorphic opcodes.  That
1748          is, the pattern adapts to the data and automatically switches to
1749          the Unicode character scheme when presented with Unicode data--or
1750          instead uses a traditional byte scheme when presented with byte
1751          data.
1752    
1753        Sadly, a wide octal escape does not cause a switch, and in a string with
1754        no other multibyte characters, these octal escapes are treated as before.
1755        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
1756        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1757        Unicode string.
1758    
1759        I have not perpetrated such confusion in PCRE. Up till now, it took just
1760        the bottom 8 bits, as in old Perl. I have now made octal escapes with
1761        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1762        translate to the appropriate multibyte character.
1763    
1764    29. Applied some refactoring to reduce the number of warnings from Microsoft
1765        and Borland compilers. This has included removing the fudge introduced
1766        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1767        a warning about an unused variable.
1768    
1769    21. PCRE has not included VT (character 0x0b) in the set of whitespace
1770        characters since release 4.0, because Perl (from release 5.004) does not.
1771        [Or at least, is documented not to: some releases seem to be in conflict
1772        with the documentation.] However, when a pattern was studied with
1773        pcre_study() and all its branches started with \s, PCRE still included VT
1774        as a possible starting character. Of course, this did no harm; it just
1775        caused an unnecessary match attempt.
1776    
1777    22. Removed a now-redundant internal flag bit that recorded the fact that case
1778        dependency changed within the pattern. This was once needed for "required
1779        byte" processing, but is no longer used. This recovers a now-scarce options
1780        bit. Also moved the least significant internal flag bit to the most-
1781        significant bit of the word, which was not previously used (hangover from
1782        the days when it was an int rather than a uint) to free up another bit for
1783        the future.
1784    
1785    23. Added support for CRLF line endings as well as CR and LF. As well as the
1786        default being selectable at build time, it can now be changed at runtime
1787        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1788        specify that it is scanning data with non-default line endings.
1789    
1790    24. Changed the definition of CXXLINK to make it agree with the definition of
1791        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1792    
1793    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1794        recursions. This makes a big different to stack usage for some patterns.
1795    
1796    26. If a subpattern containing a named recursion or subroutine reference such
1797        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1798        the space required for the compiled pattern went wrong and gave too small a
1799        value. Depending on the environment, this could lead to "Failed: internal
1800        error: code overflow at offset 49" or "glibc detected double free or
1801        corruption" errors.
1802    
1803    27. Applied patches from Google (a) to support the new newline modes and (b) to
1804        advance over multibyte UTF-8 characters in GlobalReplace.
1805    
1806    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1807        difference for some implementation of PCRE in some Windows version.
1808    
1809    29. Added some extra testing facilities to pcretest:
1810    
1811        \q<number>   in a data line sets the "match limit" value
1812        \Q<number>   in a data line sets the "match recursion limt" value
1813        -S <number>  sets the stack size, where <number> is in megabytes
1814    
1815        The -S option isn't available for Windows.
1816    
1817    
1818    Version 6.6 06-Feb-06
1819    ---------------------
1820    
1821     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1822        in pcreposix.h. I have copied the definition from pcre.h.
1823    
1824     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1825        because pcre.h is no longer a built file.
1826    
1827     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1828        not normally included in the compiled code.
1829    
1830    
1831    Version 6.5 01-Feb-06
1832    ---------------------
1833    
1834     1. When using the partial match feature with pcre_dfa_exec(), it was not
1835        anchoring the second and subsequent partial matches at the new starting
1836        point. This could lead to incorrect results. For example, with the pattern
1837        /1234/, partially matching against "123" and then "a4" gave a match.
1838    
1839     2. Changes to pcregrep:
1840    
1841        (a) All non-match returns from pcre_exec() were being treated as failures
1842            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1843            error message is output. Some extra information is given for the
1844            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
1845            probably the only errors that are likely to be caused by users (by
1846            specifying a regex that has nested indefinite repeats, for instance).
1847            If there are more than 20 of these errors, pcregrep is abandoned.
1848    
1849        (b) A binary zero was treated as data while matching, but terminated the
1850            output line if it was written out. This has been fixed: binary zeroes
1851            are now no different to any other data bytes.
1852    
1853        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1854            used to set a locale for matching. The --locale=xxxx long option has
1855            been added (no short equivalent) to specify a locale explicitly on the
1856            pcregrep command, overriding the environment variables.
1857    
1858        (d) When -B was used with -n, some line numbers in the output were one less
1859            than they should have been.
1860    
1861        (e) Added the -o (--only-matching) option.
1862    
1863        (f) If -A or -C was used with -c (count only), some lines of context were
1864            accidentally printed for the final match.
1865    
1866        (g) Added the -H (--with-filename) option.
1867    
1868        (h) The combination of options -rh failed to suppress file names for files
1869            that were found from directory arguments.
1870    
1871        (i) Added the -D (--devices) and -d (--directories) options.
1872    
1873        (j) Added the -F (--fixed-strings) option.
1874    
1875        (k) Allow "-" to be used as a file name for -f as well as for a data file.
1876    
1877        (l) Added the --colo(u)r option.
1878    
1879        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1880            is not present by default.
1881    
1882     3. A nasty bug was discovered in the handling of recursive patterns, that is,
1883        items such as (?R) or (?1), when the recursion could match a number of
1884        alternatives. If it matched one of the alternatives, but subsequently,
1885        outside the recursion, there was a failure, the code tried to back up into
1886        the recursion. However, because of the way PCRE is implemented, this is not
1887        possible, and the result was an incorrect result from the match.
1888    
1889        In order to prevent this happening, the specification of recursion has
1890        been changed so that all such subpatterns are automatically treated as
1891        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1892    
1893     4. I had overlooked the fact that, in some locales, there are characters for
1894        which isalpha() is true but neither isupper() nor islower() are true. In
1895        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1896        and ordfeminine) are like this. This affected the treatment of \w and \W
1897        when they appeared in character classes, but not when they appeared outside
1898        a character class. The bit map for "word" characters is now created
1899        separately from the results of isalnum() instead of just taking it from the
1900        upper, lower, and digit maps. (Plus the underscore character, of course.)
1901    
1902     5. The above bug also affected the handling of POSIX character classes such as
1903        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1904        permanent tables. Instead, the bit maps for such a class were previously
1905        created as the appropriate unions of the upper, lower, and digit bitmaps.
1906        Now they are created by subtraction from the [[:word:]] class, which has
1907        its own bitmap.
1908    
1909     6. The [[:blank:]] character class matches horizontal, but not vertical space.
1910        It is created by subtracting the vertical space characters (\x09, \x0a,
1911        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1912        subtraction was done in the overall bitmap for a character class, meaning
1913        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1914        be recognized. This bug has been fixed.
1915    
1916     7. Patches from the folks at Google:
1917    
1918          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1919          real life, but is still worth protecting against".
1920    
1921          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1922          regular expressions".
1923    
1924          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1925          have it.
1926    
1927          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1928          "configure" and the latter not, in order to fix a problem somebody had
1929          with compiling the Arg class on HP-UX.
1930    
1931          (e) Improve the error-handling of the C++ wrapper a little bit.
1932    
1933          (f) New tests for checking recursion limiting.
1934    
1935     8. The pcre_memmove() function, which is used only if the environment does not
1936        have a standard memmove() function (and is therefore rarely compiled),
1937        contained two bugs: (a) use of int instead of size_t, and (b) it was not
1938        returning a result (though PCRE never actually uses the result).
1939    
1940     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1941        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1942        returned instead of calling malloc() with an overflowing number that would
1943        most likely cause subsequent chaos.
1944    
1945    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1946    
1947    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1948        with this option is matched, the nmatch and pmatch options of regexec() are
1949        ignored.
1950    
1951    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1952        provided in case anyone wants to the the POSIX interface with UTF-8
1953        strings.
1954    
1955    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1956        C++ linking (needed for some HP-UX environments).
1957    
1958    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1959        (unused parameter) and in the pcre_printint() function (omitted "default"
1960        switch label when the default is to do nothing).
1961    
1962    15. Added some code to make it possible, when PCRE is compiled as a C++
1963        library, to replace subject pointers for pcre_exec() with a smart pointer
1964        class, thus making it possible to process discontinuous strings.
1965    
1966    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1967        much the same function. They were added by different people who were trying
1968        to make PCRE easy to compile on non-Unix systems. It has been suggested
1969        that PCRE_EXPORT be abolished now that there is more automatic apparatus
1970        for compiling on Windows systems. I have therefore replaced it with
1971        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1972        defaults to "extern" for C or "extern C" for C++, which works fine on
1973        Unix-like systems. It is now possible to override the value of PCRE_DATA_
1974        SCOPE with something explicit in config.h. In addition:
1975    
1976        (a) pcreposix.h still had just "extern" instead of either of these macros;
1977            I have replaced it with PCRE_DATA_SCOPE.
1978    
1979        (b) Functions such as _pcre_xclass(), which are internal to the library,
1980            but external in the C sense, all had PCRE_EXPORT in their definitions.
1981            This is apparently wrong for the Windows case, so I have removed it.
1982            (It makes no difference on Unix-like systems.)
1983    
1984    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1985        of recursive calls to match(). This is different to MATCH_LIMIT because
1986        that limits the total number of calls to match(), not all of which increase
1987        the depth of recursion. Limiting the recursion depth limits the amount of
1988        stack (or heap if NO_RECURSE is set) that is used. The default can be set
1989        when PCRE is compiled, and changed at run time. A patch from Google adds
1990        this functionality to the C++ interface.
1991    
1992    18. Changes to the handling of Unicode character properties:
1993    
1994        (a) Updated the table to Unicode 4.1.0.
1995    
1996        (b) Recognize characters that are not in the table as "Cn" (undefined).
1997    
1998        (c) I revised the way the table is implemented to a much improved format
1999            which includes recognition of ranges. It now supports the ranges that
2000            are defined in UnicodeData.txt, and it also amalgamates other
2001            characters into ranges. This has reduced the number of entries in the
2002            table from around 16,000 to around 3,000, thus reducing its size
2003            considerably. I realized I did not need to use a tree structure after
2004            all - a binary chop search is just as efficient. Having reduced the
2005            number of entries, I extended their size from 6 bytes to 8 bytes to
2006            allow for more data.
2007    
2008        (d) Added support for Unicode script names via properties such as \p{Han}.
2009    
2010    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
2011        matching that character.
2012    
2013    20. When matching a repeated Unicode property with a minimum greater than zero,
2014        (for example \pL{2,}), PCRE could look past the end of the subject if it
2015        reached it while seeking the minimum number of characters. This could
2016        happen only if some of the characters were more than one byte long, because
2017        there is a check for at least the minimum number of bytes.
2018    
2019    21. Refactored the implementation of \p and \P so as to be more general, to
2020        allow for more different types of property in future. This has changed the
2021        compiled form incompatibly. Anybody with saved compiled patterns that use
2022        \p or \P will have to recompile them.
2023    
2024    22. Added "Any" and "L&" to the supported property types.
2025    
2026    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
2027        but give a compile time error if the value is greater than 0xff.
2028    
2029    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
2030        accidentally not being installed or uninstalled.
2031    
2032    25. The pcre.h file was built from pcre.h.in, but the only changes that were
2033        made were to insert the current release number. This seemed silly, because
2034        it made things harder for people building PCRE on systems that don't run
2035        "configure". I have turned pcre.h into a distributed file, no longer built
2036        by "configure", with the version identification directly included. There is
2037        no longer a pcre.h.in file.
2038    
2039        However, this change necessitated a change to the pcre-config script as
2040        well. It is built from pcre-config.in, and one of the substitutions was the
2041        release number. I have updated configure.ac so that ./configure now finds
2042        the release number by grepping pcre.h.
2043    
2044    26. Added the ability to run the tests under valgrind.
2045    
2046    
2047    Version 6.4 05-Sep-05
2048    ---------------------
2049    
2050     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
2051        "--" to be printed when multiple files were scanned, even when none of the
2052        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
2053        consider it to be a bug, and have restored the previous behaviour.
2054    
2055     2. A couple of code tidies to get rid of compiler warnings.
2056    
2057     3. The pcretest program used to cheat by referring to symbols in the library
2058        whose names begin with _pcre_. These are internal symbols that are not
2059        really supposed to be visible externally, and in some environments it is
2060        possible to suppress them. The cheating is now confined to including
2061        certain files from the library's source, which is a bit cleaner.
2062    
2063     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
2064        file's purpose clearer.
2065    
2066     5. Reorganized pcre_ucp_findchar().
2067    
2068    
2069    Version 6.3 15-Aug-05
2070    ---------------------
2071    
2072     1. The file libpcre.pc.in did not have general read permission in the tarball.
2073    
2074     2. There were some problems when building without C++ support:
2075    
2076        (a) If C++ support was not built, "make install" and "make test" still
2077            tried to test it.
2078    
2079        (b) There were problems when the value of CXX was explicitly set. Some
2080            changes have been made to try to fix these, and ...
2081    
2082        (c) --disable-cpp can now be used to explicitly disable C++ support.
2083    
2084        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
2085            backslash in a target when C++ was disabled. This confuses some
2086            versions of "make", apparently. Using an intermediate variable solves
2087            this. (Same for CPP_LOBJ.)
2088    
2089     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
2090        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
2091        necessary on certain architectures.
2092    
2093     4. Added a setting of -export-symbols-regex to the link command to remove
2094        those symbols that are exported in the C sense, but actually are local
2095        within the library, and not documented. Their names all begin with
2096        "_pcre_". This is not a perfect job, because (a) we have to except some
2097        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
2098        available (and never for static libraries). I have made a note to try to
2099        find a way round (a) in the future.
2100    
2101    
2102    Version 6.2 01-Aug-05
2103    ---------------------
2104    
2105     1. There was no test for integer overflow of quantifier values. A construction
2106        such as {1111111111111111} would give undefined results. What is worse, if
2107        a minimum quantifier for a parenthesized subpattern overflowed and became
2108        negative, the calculation of the memory size went wrong. This could have
2109        led to memory overwriting.
2110    
2111     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
2112    
2113     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
2114        operating environments where this matters.
2115    
2116     4. Applied Giuseppe Maxia's patch to add additional features for controlling
2117        PCRE options from within the C++ wrapper.
2118    
2119     5. Named capturing subpatterns were not being correctly counted when a pattern
2120        was compiled. This caused two problems: (a) If there were more than 100
2121        such subpatterns, the calculation of the memory needed for the whole
2122        compiled pattern went wrong, leading to an overflow error. (b) Numerical
2123        back references of the form \12, where the number was greater than 9, were
2124        not recognized as back references, even though there were sufficient
2125        previous subpatterns.
2126    
2127     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
2128        versions of gcc, e.g. 2.95.4.
2129    
2130    
2131    Version 6.1 21-Jun-05
2132    ---------------------
2133    
2134     1. There was one reference to the variable "posix" in pcretest.c that was not
2135        surrounded by "#if !defined NOPOSIX".
2136    
2137     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
2138        the cross-check on the old pcre_info() function, for the benefit of the
2139        cut-down version of PCRE that is currently imported into Exim.
2140    
2141     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
2142        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
2143        patterns that start (?i) but I don't think that matters. The use of (?i) is
2144        just an example; this all applies to the other options as well.
2145    
2146     4. Since libtool seems to echo the compile commands it is issuing, the output
2147        from "make" can be reduced a bit by putting "@" in front of each libtool
2148        compile command.
2149    
2150     5. Patch from the folks at Google for configure.in to be a bit more thorough
2151        in checking for a suitable C++ installation before trying to compile the
2152        C++ stuff. This should fix a reported problem when a compiler was present,
2153        but no suitable headers.
2154    
2155     6. The man pages all had just "PCRE" as their title. I have changed them to
2156        be the relevant file name. I have also arranged that these names are
2157        retained in the file doc/pcre.txt, which is a concatenation in text format
2158        of all the man pages except the little individual ones for each function.
2159    
2160     7. The NON-UNIX-USE file had not been updated for the different set of source
2161        files that come with release 6. I also added a few comments about the C++
2162        wrapper.
2163    
2164    
2165    Version 6.0 07-Jun-05
2166    ---------------------
2167    
2168     1. Some minor internal re-organization to help with my DFA experiments.
2169    
2170     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
2171        didn't matter for the library itself when fully configured, but did matter
2172        when compiling without UCP support, or within Exim, where the ucp files are
2173        not imported.
2174    
2175     3. Refactoring of the library code to split up the various functions into
2176        different source modules. The addition of the new DFA matching code (see
2177        below) to a single monolithic source would have made it really too
2178        unwieldy, quite apart from causing all the code to be include in a
2179        statically linked application, when only some functions are used. This is
2180        relevant even without the DFA addition now that patterns can be compiled in
2181        one application and matched in another.
2182    
2183        The downside of splitting up is that there have to be some external
2184        functions and data tables that are used internally in different modules of
2185        the library but which are not part of the API. These have all had their
2186        names changed to start with "_pcre_" so that they are unlikely to clash
2187        with other external names.
2188    
2189     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
2190        a different (DFA) algorithm. Although it is slower than the original
2191        function, it does have some advantages for certain types of matching
2192        problem.
2193    
2194     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
2195        including restarting after a partial match.
2196    
2197     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
2198        defined when compiling for Windows was sent to me. I have put it into the
2199        code, though I have no means of testing or verifying it.
2200    
2201     7. Added the pcre_refcount() auxiliary function.
2202    
2203     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
2204        match before or at the first newline in the subject string. In pcretest,
2205        the /f option on a pattern can be used to set this.
2206    
2207     9. A repeated \w when used in UTF-8 mode with characters greater than 256
2208        would behave wrongly. This has been present in PCRE since release 4.0.
2209    
2210    10. A number of changes to the pcregrep command:
2211    
2212        (a) Refactored how -x works; insert ^(...)$ instead of setting
2213            PCRE_ANCHORED and checking the length, in preparation for adding
2214            something similar for -w.
2215    
2216        (b) Added the -w (match as a word) option.
2217    
2218        (c) Refactored the way lines are read and buffered so as to have more
2219            than one at a time available.
2220    
2221        (d) Implemented a pcregrep test script.
2222    
2223        (e) Added the -M (multiline match) option. This allows patterns to match
2224            over several lines of the subject. The buffering ensures that at least
2225            8K, or the rest of the document (whichever is the shorter) is available
2226            for matching (and similarly the previous 8K for lookbehind assertions).
2227    
2228        (f) Changed the --help output so that it now says
2229    
2230              -w, --word-regex(p)
2231    
2232            instead of two lines, one with "regex" and the other with "regexp"
2233            because that confused at least one person since the short forms are the
2234            same. (This required a bit of code, as the output is generated
2235            automatically from a table. It wasn't just a text change.)
2236    
2237        (g) -- can be used to terminate pcregrep options if the next thing isn't an
2238            option but starts with a hyphen. Could be a pattern or a path name
2239            starting with a hyphen, for instance.
2240    
2241        (h) "-" can be given as a file name to represent stdin.
2242    
2243        (i) When file names are being printed, "(standard input)" is used for
2244            the standard input, for compatibility with GNU grep. Previously
2245            "<stdin>" was used.
2246    
2247        (j) The option --label=xxx can be used to supply a name to be used for
2248            stdin when file names are being printed. There is no short form.
2249    
2250        (k) Re-factored the options decoding logic because we are going to add
2251            two more options that take data. Such options can now be given in four
2252            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
2253    
2254        (l) Added the -A, -B, and -C options for requesting that lines of context
2255            around matches be printed.
2256    
2257        (m) Added the -L option to print the names of files that do not contain
2258            any matching lines, that is, the complement of -l.
2259    
2260        (n) The return code is 2 if any file cannot be opened, but pcregrep does
2261            continue to scan other files.
2262    
2263        (o) The -s option was incorrectly implemented. For compatibility with other
2264            greps, it now suppresses the error message for a non-existent or non-
2265            accessible file (but not the return code). There is a new option called
2266            -q that suppresses the output of matching lines, which was what -s was
2267            previously doing.
2268    
2269        (p) Added --include and --exclude options to specify files for inclusion
2270            and exclusion when recursing.
2271    
2272    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
2273        Hopefully, it now does.
2274    
2275    12. Missing cast in pcre_study().
2276    
2277    13. Added an "uninstall" target to the makefile.
2278    
2279    14. Replaced "extern" in the function prototypes in Makefile.in with
2280        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
2281        world, but is set differently for Windows.
2282    
2283    15. Added a second compiling function called pcre_compile2(). The only
2284        difference is that it has an extra argument, which is a pointer to an
2285        integer error code. When there is a compile-time failure, this is set
2286        non-zero, in addition to the error test pointer being set to point to an
2287        error message. The new argument may be NULL if no error number is required
2288        (but then you may as well call pcre_compile(), which is now just a
2289        wrapper). This facility is provided because some applications need a
2290        numeric error indication, but it has also enabled me to tidy up the way
2291        compile-time errors are handled in the POSIX wrapper.
2292    
2293    16. Added VPATH=.libs to the makefile; this should help when building with one
2294        prefix path and installing with another. (Or so I'm told by someone who
2295        knows more about this stuff than I do.)
2296    
2297    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
2298        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
2299        match everything, including newlines. This is not POSIX-compatible, but
2300        somebody wanted the feature. From pcretest it can be activated by using
2301        both the P and the s flags.
2302    
2303    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
2304    
2305    19. libpcre.pc was being incorrectly installed as executable.
2306    
2307    20. A couple of places in pcretest check for end-of-line by looking for '\n';
2308        it now also looks for '\r' so that it will work unmodified on Windows.
2309    
2310    21. Added Google's contributed C++ wrapper to the distribution.
2311    
2312    22. Added some untidy missing memory free() calls in pcretest, to keep
2313        Electric Fence happy when testing.
2314    
2315    
2316    
2317    Version 5.0 13-Sep-04
2318    ---------------------
2319    
2320     1. Internal change: literal characters are no longer packed up into items
2321        containing multiple characters in a single byte-string. Each character
2322        is now matched using a separate opcode. However, there may be more than one
2323        byte in the character in UTF-8 mode.
2324    
2325     2. The pcre_callout_block structure has two new fields: pattern_position and
2326        next_item_length. These contain the offset in the pattern to the next match
2327        item, and its length, respectively.
2328    
2329     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
2330        insertion of callouts before each pattern item. Added the /C option to
2331        pcretest to make use of this.
2332    
2333     4. On the advice of a Windows user, the lines
2334    
2335          #if defined(_WIN32) || defined(WIN32)
2336          _setmode( _fileno( stdout ), 0x8000 );
2337          #endif  /* defined(_WIN32) || defined(WIN32) */
2338    
2339        have been added to the source of pcretest. This apparently does useful
2340        magic in relation to line terminators.
2341    
2342     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
2343        for the benefit of those environments where the "b" makes a difference.
2344    
2345     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
2346        to know about it. I have put a hack into configure.in that adds in code
2347        to set GCC=yes if CC=icc. This seems to end up at a point in the
2348        generated configure script that is early enough to affect the setting of
2349        compiler options, which is what is needed, but I have no means of testing
2350        whether it really works. (The user who reported this had patched the
2351        generated configure script, which of course I cannot do.)
2352    
2353        LATER: After change 22 below (new libtool files), the configure script
2354        seems to know about icc (and also ecc). Therefore, I have commented out
2355        this hack in configure.in.
2356    
2357     7. Added support for pkg-config (2 patches were sent in).
2358    
2359     8. Negated POSIX character classes that used a combination of internal tables
2360        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2361        [[:^ascii]]. Typically, they would match almost any characters. The other
2362        POSIX classes were not broken in this way.
2363    
2364     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2365        to find the match, as PCRE was deluded into thinking that the match had to
2366        start at the start point or following a newline. The same bug applied to
2367        patterns with negative forward assertions or any backward assertions
2368        preceding ".*" at the start, unless the pattern required a fixed first
2369        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2370    
2371    10. In UTF-8 mode, when moving forwards in the subject after a failed match
2372        starting at the last subject character, bytes beyond the end of the subject
2373        string were read.
2374    
2375    11. Renamed the variable "class" as "classbits" to make life easier for C++
2376        users. (Previously there was a macro definition, but it apparently wasn't
2377        enough.)
2378    
2379    12. Added the new field "tables" to the extra data so that tables can be passed
2380        in at exec time, or the internal tables can be re-selected. This allows
2381        a compiled regex to be saved and re-used at a later time by a different
2382        program that might have everything at different addresses.
2383    
2384    13. Modified the pcre-config script so that, when run on Solaris, it shows a
2385        -R library as well as a -L library.
2386    
2387    14. The debugging options of pcretest (-d on the command line or D on a
2388        pattern) showed incorrect output for anything following an extended class
2389        that contained multibyte characters and which was followed by a quantifier.
2390    
2391    15. Added optional support for general category Unicode character properties
2392        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2393        support. It adds about 90K to the size of the library. The meanings of the
2394        inbuilt class escapes such as \d and \s have NOT been changed.
2395    
2396    16. Updated pcredemo.c to include calls to free() to release the memory for the
2397        compiled pattern.
2398    
2399    17. The generated file chartables.c was being created in the source directory
2400        instead of in the building directory. This caused the build to fail if the
2401        source directory was different from the building directory, and was
2402        read-only.
2403    
2404    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2405        file. No doubt somebody will tell me if they don't make sense... Also added
2406        Dan Mooney's comments about building on OpenVMS.
2407    
2408    19. Added support for partial matching via the PCRE_PARTIAL option for
2409        pcre_exec() and the \P data escape in pcretest.
2410    
2411    20. Extended pcretest with 3 new pattern features:
2412    
2413        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
2414              write the compiled pattern to the file whose name is "rest-of-line".
2415              This is a straight binary dump of the data, with the saved pointer to
2416              the character tables forced to be NULL. The study data, if any, is
2417              written too. After writing, pcretest reads a new pattern.
2418    
2419        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2420              compiled pattern from the given file. There must not be any
2421              occurrences of "<" in the file name (pretty unlikely); if there are,
2422              pcretest will instead treat the initial "<" as a pattern delimiter.
2423              After reading in the pattern, pcretest goes on to read data lines as
2424              usual.
2425    
2426        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2427              and 16-bit fields in a compiled pattern, to simulate a pattern that
2428              was compiled on a host of opposite endianness.
2429    
2430    21. The pcre-exec() function can now cope with patterns that were compiled on
2431        hosts of opposite endianness, with this restriction:
2432    
2433          As for any compiled expression that is saved and used later, the tables
2434          pointer field cannot be preserved; the extra_data field in the arguments
2435          to pcre_exec() should be used to pass in a tables address if a value
2436          other than the default internal tables were used at compile time.
2437    
2438    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2439        now diagnosed as an error. Previously, most of the time, a negative number
2440        would have been treated as zero, but if in addition "ovector" was passed as
2441        NULL, a crash could occur.
2442    
2443    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2444        new versions from the libtool 1.5 distribution (the last one is a copy of
2445        a file called libtool.m4). This seems to have fixed the need to patch
2446        "configure" to support Darwin 1.3 (which I used to do). However, I still
2447        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2448        workstation).
2449    
2450    24. Changed the PCRE licence to be the more standard "BSD" licence.
2451    
2452    
2453    Version 4.5 01-Dec-03
2454    ---------------------
2455    
2456     1. There has been some re-arrangement of the code for the match() function so
2457        that it can be compiled in a version that does not call itself recursively.
2458        Instead, it keeps those local variables that need separate instances for
2459        each "recursion" in a frame on the heap, and gets/frees frames whenever it
2460        needs to "recurse". Keeping track of where control must go is done by means
2461        of setjmp/longjmp. The whole thing is implemented by a set of macros that
2462        hide most of the details from the main code, and operates only if
2463        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2464        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2465        operating.
2466    
2467        To make it easier for callers to provide specially tailored get/free
2468        functions for this usage, two new functions, pcre_stack_malloc, and
2469        pcre_stack_free, are used. They are always called in strict stacking order,
2470        and the size of block requested is always the same.
2471    
2472        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2473        PCRE has been compiled to use the stack or the heap for recursion. The
2474        -C option of pcretest uses this to show which version is compiled.
2475    
2476        A new data escape \S, is added to pcretest; it causes the amounts of store
2477        obtained and freed by both kinds of malloc/free at match time to be added
2478        to the output.
2479    
2480     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2481        what's available on my current Linux desktop machine.
2482    
2483     3. When matching a UTF-8 string, the test for a valid string at the start has
2484        been extended. If start_offset is not zero, PCRE now checks that it points
2485        to a byte that is the start of a UTF-8 character. If not, it returns
2486        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2487        this is necessary because there may be backward assertions in the pattern.
2488        When matching the same subject several times, it may save resources to use
2489        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2490    
2491     4. The code for checking the validity of UTF-8 strings has been tightened so
2492        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2493        containing "overlong sequences".
2494    
2495     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2496        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2497        should have been "&", but it just so happened that all the cases this let
2498        through by mistake were picked up later in the function.
2499    
2500     6. I had used a variable called "isblank" - this is a C99 function, causing
2501        some compilers to warn. To avoid this, I renamed it (as "blankclass").
2502    
2503     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2504        prompting; (b) run "./pcretest /dev/null" at the start of the test script
2505        so the version is shown; (c) stop "make test" echoing "./RunTest".
2506    
2507     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2508    
2509     9. The prototype for memmove() for systems that don't have it was using
2510        size_t, but the inclusion of the header that defines size_t was later. I've
2511        moved the #includes for the C headers earlier to avoid this.
2512    
2513    10. Added some adjustments to the code to make it easier to compiler on certain
2514        special systems:
2515    
2516          (a) Some "const" qualifiers were missing.
2517          (b) Added the macro EXPORT before all exported functions; by default this
2518              is defined to be empty.
2519          (c) Changed the dftables auxiliary program (that builds chartables.c) so
2520              that it reads its output file name as an argument instead of writing
2521              to the standard output and assuming this can be redirected.
2522    
2523    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2524        class containing characters with values greater than 255, PCRE compilation
2525        went into a loop.
2526    
2527    12. A recursive reference to a subpattern that was within another subpattern
2528        that had a minimum quantifier of zero caused PCRE to crash. For example,
2529        (x(y(?2))z)? provoked this bug with a subject that got as far as the
2530        recursion. If the recursively-called subpattern itself had a zero repeat,
2531        that was OK.
2532    
2533    13. In pcretest, the buffer for reading a data line was set at 30K, but the
2534        buffer into which it was copied (for escape processing) was still set at
2535        1024, so long lines caused crashes.
2536    
2537    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2538        "internal error: code overflow...". This applied to any character class
2539        that was followed by a possessive quantifier.
2540    
2541    15. Modified the Makefile to add libpcre.la as a prerequisite for
2542        libpcreposix.la because I was told this is needed for a parallel build to
2543        work.
2544    
2545    16. If a pattern that contained .* following optional items at the start was
2546        studied, the wrong optimizing data was generated, leading to matching
2547        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2548        matching string must start with a or b or c. The correct conclusion for
2549        this pattern is that a match can start with any character.
2550    
2551    
2552    Version 4.4 13-Aug-03
2553    ---------------------
2554    
2555     1. In UTF-8 mode, a character class containing characters with values between
2556        127 and 255 was not handled correctly if the compiled pattern was studied.
2557        In fixing this, I have also improved the studying algorithm for such
2558        classes (slightly).
2559    
2560     2. Three internal functions had redundant arguments passed to them. Removal
2561        might give a very teeny performance improvement.
2562    
2563     3. Documentation bug: the value of the capture_top field in a callout is *one
2564        more than* the number of the hightest numbered captured substring.
2565    
2566     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2567        in incorrectly linking with a previously installed version. They now link
2568        explicitly with libpcre.la.
2569    
2570     5. configure.in no longer needs to recognize Cygwin specially.
2571    
2572     6. A problem in pcre.in for Windows platforms is fixed.
2573    
2574     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2575        pcretest, it used to include the size of the study block as part of its
2576        output. Unfortunately, the structure contains a field that has a different
2577        size on different hardware architectures. This meant that the tests that
2578        showed this size failed. As the block is currently always of a fixed size,
2579        this information isn't actually particularly useful in pcretest output, so
2580        I have just removed it.
2581    
2582     8. Three pre-processor statements accidentally did not start in column 1.
2583        Sadly, there are *still* compilers around that complain, even though
2584        standard C has not required this for well over a decade. Sigh.
2585    
2586     9. In pcretest, the code for checking callouts passed small integers in the
2587        callout_data field, which is a void * field. However, some picky compilers
2588        complained about the casts involved for this on 64-bit systems. Now
2589        pcretest passes the address of the small integer instead, which should get
2590        rid of the warnings.
2591    
2592    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2593        both compile and run time, and gives an error if an invalid UTF-8 sequence
2594        is found. There is a option for disabling this check in cases where the
2595        string is known to be correct and/or the maximum performance is wanted.
2596    
2597    11. In response to a bug report, I changed one line in Makefile.in from
2598    
2599            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2600        to
2601            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2602    
2603        to look similar to other lines, but I have no way of telling whether this
2604        is the right thing to do, as I do not use Windows. No doubt I'll get told
2605        if it's wrong...
2606    
2607    
2608    Version 4.3 21-May-03
2609    ---------------------
2610    
2611    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2612       Makefile.
2613    
2614    2. Some refactoring to improve the quality of the code:
2615    
2616       (i)   The utf8_table... variables are now declared "const".
2617    
2618       (ii)  The code for \cx, which used the "case flipping" table to upper case
2619             lower case letters, now just substracts 32. This is ASCII-specific,
2620             but the whole concept of \cx is ASCII-specific, so it seems
2621             reasonable.
2622    
2623       (iii) PCRE was using its character types table to recognize decimal and
2624             hexadecimal digits in the pattern. This is silly, because it handles
2625             only 0-9, a-f, and A-F, but the character types table is locale-
2626             specific, which means strange things might happen. A private
2627             table is now used for this - though it costs 256 bytes, a table is
2628             much faster than multiple explicit tests. Of course, the standard
2629             character types table is still used for matching digits in subject
2630             strings against \d.
2631    
2632       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2633             ending in _t are). So I've renamed it as ESC_tee.
2634    
2635    3. The first argument for regexec() in the POSIX wrapper should have been
2636       defined as "const".
2637    
2638    4. Changed pcretest to use malloc() for its buffers so that they can be
2639       Electric Fenced for debugging.
2640    
2641    5. There were several places in the code where, in UTF-8 mode, PCRE would try
2642       to read one or more bytes before the start of the subject string. Often this
2643       had no effect on PCRE's behaviour, but in some circumstances it could
2644       provoke a segmentation fault.
2645    
2646    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2647       to try to read one or more bytes before the start of the subject string.
2648    
2649    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
2650       UTF-8 support could misbehave in various ways if the subject string
2651       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
2652       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
2653       back over UTF-8 characters.)
2654    
2655    
2656    Version 4.2 14-Apr-03
2657    ---------------------
2658    
2659    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
2660    
2661    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
2662         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
2663         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
2664         [WIN_PREFIX]: new variable, "cyg" for Cygwin
2665         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
2666           and BUILD_EXEEXT
2667         Note: automatic setting of the BUILD variables is not yet working
2668         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
2669           compile-time but not at link-time
2670         [LINK]: use for linking executables only
2671         make different versions for Windows and non-Windows
2672         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
2673           libraries
2674         [LINK_FOR_BUILD]: new variable
2675         [OBJEXT]: use throughout
2676         [EXEEXT]: use throughout
2677         <winshared>: new target
2678         <wininstall>: new target
2679         <dftables.o>: use native compiler
2680         <dftables>: use native linker
2681         <install>: handle Windows platform correctly
2682         <clean>: ditto
2683         <check>: ditto
2684         copy DLL to top builddir before testing
2685    
2686       As part of these changes, -no-undefined was removed again. This was reported
2687       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
2688       in any case.
2689    
2690    3. Some tidies to get rid of compiler warnings:
2691    
2692       . In the match_data structure, match_limit was an unsigned long int, whereas
2693         match_call_count was an int. I've made them both unsigned long ints.
2694    
2695       . In pcretest the fact that a const uschar * doesn't automatically cast to
2696         a void * provoked a warning.
2697    
2698       . Turning on some more compiler warnings threw up some "shadow" variables
2699         and a few more missing casts.
2700    
2701    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
2702       option, a class that contained a single character with a value between 128
2703       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
2704    
2705    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
2706       option, a class that contained several characters, but with at least one
2707       whose value was between 128 and 255 caused PCRE to crash.
2708    
2709    
2710    Version 4.1 12-Mar-03
2711    ---------------------
2712    
2713    1. Compiling with gcc -pedantic found a couple of places where casts were
2714    needed, and a string in dftables.c that was longer than standard compilers are
2715    required to support.
2716    
2717    2. Compiling with Sun's compiler found a few more places where the code could
2718    be tidied up in order to avoid warnings.
2719    
2720    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
2721    first of these names is deprecated in the latest Autoconf in favour of the name
2722    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
2723    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2724    analogy I have changed it to CFLAGS_FOR_BUILD.
2725    
2726    4. Added -no-undefined to the linking command in the Makefile, because this is
2727    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2728    linking step for the pcreposix library.
2729    
2730    5. PCRE was failing to diagnose the case of two named groups with the same
2731    name.
2732    
2733    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2734    literal character that is needed in the subject for a match, and scans along to
2735    ensure that it is present before embarking on the full matching process. This
2736    saves time in cases of nested unlimited repeats that are never going to match.
2737    Problem: the scan can take a lot of time if the subject is very long (e.g.
2738    megabytes), thus penalizing straightforward matches. It is now done only if the
2739    amount of subject to be scanned is less than 1000 bytes.
2740    
2741    7. A lesser problem with the same optimization is that it was recording the
2742    first character of an anchored pattern as "needed", thus provoking a search
2743    right along the subject, even when the first match of the pattern was going to
2744    fail. The "needed" character is now not set for anchored patterns, unless it
2745    follows something in the pattern that is of non-fixed length. Thus, it still
2746    fulfils its original purpose of finding quick non-matches in cases of nested
2747    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
2748    
2749    
2750    Version 4.0 17-Feb-03
2751    ---------------------
2752    
2753    1. If a comment in an extended regex that started immediately after a meta-item
2754    extended to the end of string, PCRE compiled incorrect data. This could lead to
2755    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
2756    
2757    2. Moved to autoconf 2.53 and libtool 1.4.2.
2758    
2759    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
2760    the special perltest8 script is no longer needed - all the tests can be run
2761    from a single perltest script.
2762    
2763    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
2764    by \s. It has now been removed in PCRE. This means it isn't recognized as
2765    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
2766    class [:space:] *does* include VT, thereby creating a mess.
2767    
2768    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
2769    space and tab.
2770    
2771    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
2772    its new features into the main test script, reducing the number of scripts.
2773    
2774    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
2775    were backward compatible, and made the (?i) apply to the whole pattern, as if
2776    /i were given. Now it behaves more logically, and applies the option setting
2777    only to what follows. PCRE has been changed to follow suit. However, if it
2778    finds options settings right at the start of the pattern, it extracts them into
2779    the global options, as before. Thus, they show up in the info data.
2780    
2781    8. Added support for the \Q...\E escape sequence. Characters in between are
2782    treated as literals. This is slightly different from Perl in that $ and @ are
2783    also handled as literals inside the quotes. In Perl, they will cause variable
2784    interpolation. Note the following examples:
2785    
2786        Pattern            PCRE matches      Perl matches
2787    
2788        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
2789        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
2790        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
2791    
2792    For compatibility with Perl, \Q...\E sequences are recognized inside character
2793    classes as well as outside them.
2794    
2795    9. Re-organized 3 code statements in pcretest to avoid "overflow in
2796    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2797    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2798    signed/unsigned warnings.
2799    
2800    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2801    option for pcretest, so I've replaced it by a simple function that does just
2802    that job.
2803    
2804    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2805    "pcregrep -".
2806    
2807    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2808    Java package. This provides some syntactic sugar for simple cases of what my
2809    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2810    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2811    item, you can use this simplified notation. Note that only makes sense with
2812    greedy quantifiers. Consequently, the use of the possessive quantifier forces
2813    greediness, whatever the setting of the PCRE_UNGREEDY option.
2814    
2815    13. A change of greediness default within a pattern was not taking effect at
2816    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2817    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2818    was abstracted outside.
2819    
2820    14. PCRE now supports the \G assertion. It is true when the current matching
2821    position is at the start point of the match. This differs from \A when the
2822    starting offset is non-zero. Used with the /g option of pcretest (or similar
2823    code), it works in the same way as it does for Perl's /g option. If all
2824    alternatives of a regex begin with \G, the expression is anchored to the start
2825    match position, and the "anchored" flag is set in the compiled expression.
2826    
2827    15. Some bugs concerning the handling of certain option changes within patterns
2828    have been fixed. These applied to options other than (?ims). For example,
2829    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2830    been the other way round. Some of this was related to change 7 above.
2831    
2832    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2833    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2834    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2835    POSIX classes only within a class (e.g. /[[:alpha:]]/).
2836    
2837    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2838    mode. Unlike ".", it always matches newline, whatever the setting of
2839    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2840    assertions. Perl allows it, but it doesn't (in general) work because it can't
2841    calculate the length of the lookbehind. At least, that's the case for Perl
2842    5.8.0 - I've been told they are going to document that it doesn't work in
2843    future.
2844    
2845    18. Added an error diagnosis for escapes that PCRE does not support: these are
2846    \L, \l, \N, \P, \p, \U, \u, and \X.
2847    
2848    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2849    reading past the end of the pattern in cases such as /[abcd/.
2850    
2851    20. PCRE was getting more memory than necessary for patterns with classes that
2852    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2853    
2854    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2855    compiling PCRE for use with Virtual Pascal.
2856    
2857    22. Small fix to the Makefile to make it work properly if the build is done
2858    outside the source tree.
2859    
2860    23. Added a new extension: a condition to go with recursion. If a conditional
2861    subpattern starts with (?(R) the "true" branch is used if recursion has
2862    happened, whereas the "false" branch is used only at the top level.
2863    
2864    24. When there was a very long string of literal characters (over 255 bytes
2865    without UTF support, over 250 bytes with UTF support), the computation of how
2866    much memory was required could be incorrect, leading to segfaults or other
2867    strange effects.
2868    
2869    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2870    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2871    there was a subsequent back reference to those brackets. This meant that, for
2872    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2873    possible to check for precisely this case. All we can do is abandon the
2874    optimization if .* occurs inside capturing brackets when there are any back
2875    references whatsoever. (See below for a better fix that came later.)
2876    
2877    26. The handling of the optimization for finding the first character of a
2878    non-anchored pattern, and for finding a character that is required later in the
2879    match were failing in some cases. This didn't break the matching; it just
2880    failed to optimize when it could. The way this is done has been re-implemented.
2881    
2882    27. Fixed typo in error message for invalid (?R item (it said "(?p").
2883    
2884    28. Added a new feature that provides some of the functionality that Perl
2885    provides with (?{...}). The facility is termed a "callout". The way it is done
2886    in PCRE is for the caller to provide an optional function, by setting
2887    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2888    global variable. By default it is unset, which disables all calling out. To get
2889    the function called, the regex must include (?C) at appropriate points. This
2890    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2891    This provides a means of identifying different callout points. When PCRE
2892    reaches such a point in the regex, if pcre_callout has been set, the external
2893    function is called. It is provided with data in a structure called
2894    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2895    matching continues; if it returns a non-zero value, the match at the current
2896    point fails. However, backtracking will occur if possible. [This was changed
2897    later and other features added - see item 49 below.]
2898    
2899    29. pcretest is upgraded to test the callout functionality. It provides a
2900    callout function that displays information. By default, it shows the start of
2901    the match and the current position in the text. There are some new data escapes
2902    to vary what happens:
2903    
2904        \C+         in addition, show current contents of captured substrings
2905        \C-         do not supply a callout function
2906        \C!n        return 1 when callout number n is reached
2907        \C!n!m      return 1 when callout number n is reached for the mth time
2908    
2909    30. If pcregrep was called with the -l option and just a single file name, it
2910    output "<stdin>" if a match was found, instead of the file name.
2911    
2912    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2913    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2914    pcre_exec(). This saves a malloc/free per call. The default value of
2915    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2916    when configuring.
2917    
2918    32. The default maximum size of a compiled pattern is 64K. There have been a
2919    few cases of people hitting this limit. The code now uses macros to handle the
2920    storing of links as offsets within the compiled pattern. It defaults to 2-byte
2921    links, but this can be changed to 3 or 4 bytes by --with-link-size when
2922    configuring. Tests 2 and 5 work only with 2-byte links because they output
2923    debugging information about compiled patterns.
2924    
2925    33. Internal code re-arrangements:
2926    
2927    (a) Moved the debugging function for printing out a compiled regex into
2928        its own source file (printint.c) and used #include to pull it into
2929        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2930        separate copies.
2931    
2932    (b) Defined the list of op-code names for debugging as a macro in
2933        internal.h so that it is next to the definition of the opcodes.
2934    
2935    (c) Defined a table of op-code lengths for simpler skipping along compiled
2936        code. This is again a macro in internal.h so that it is next to the
2937        definition of the opcodes.
2938    
2939    34. Added support for recursive calls to individual subpatterns, along the
2940    lines of Robin Houston's patch (but implemented somewhat differently).
2941    
2942    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2943    allow it to read and process whole directories in Win32. This code was
2944    contributed by Lionel Fourquaux; it has not been tested by me.
2945    
2946    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2947    used to name a group. Names consist of alphanumerics and underscores, and must
2948    be unique. Back references use the syntax (?P=name) and recursive calls use
2949    (?P>name) which is a PCRE extension to the Python extension. Groups still have
2950    numbers. The function pcre_fullinfo() can be used after compilation to extract
2951    a name/number map. There are three relevant calls:
2952    
2953      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
2954      PCRE_INFO_NAMECOUNT            yields the number of entries
2955      PCRE_INFO_NAMETABLE            yields a pointer to the map.
2956    
2957    The map is a vector of fixed-size entries. The size of each entry depends on
2958    the length of the longest name used. The first two bytes of each entry are the
2959    group number, most significant byte first. There follows the corresponding
2960    name, zero terminated. The names are in alphabetical order.
2961    
2962    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2963    case instead of 255. Making it the same both with and without UTF-8 support
2964    means that the same test output works with both.
2965    
2966    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2967    calling malloc() with a zero argument.
2968    
2969    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2970    optimization. I've improved things by keeping a bitmap of backreferences with
2971    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2972    fact referenced, the optimization can be applied. It is unlikely that a
2973    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2974    the match to follow \n) will appear inside brackets with a number greater than
2975    31, but if it does, any back reference > 31 suppresses the optimization.
2976    
2977    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2978    of disabling numbered capturing parentheses. Any opening parenthesis that is
2979    not followed by ? behaves as if it were followed by ?: but named parentheses
2980    can still be used for capturing (and they will acquire numbers in the usual
2981    way).
2982    
2983    41. Redesigned the return codes from the match() function into yes/no/error so
2984    that errors can be passed back from deep inside the nested calls. A malloc
2985    failure while inside a recursive subpattern call now causes the
2986    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2987    
2988    42. It is now possible to set a limit on the number of times the match()
2989    function is called in a call to pcre_exec(). This facility makes it possible to
2990    limit the amount of recursion and backtracking, though not in a directly
2991    obvious way, because the match() function is used in a number of different
2992    circumstances. The count starts from zero for each position in the subject
2993    string (for non-anchored patterns). The default limit is, for compatibility, a
2994    large number, namely 10 000 000. You can change this in two ways:
2995    
2996    (a) When configuring PCRE before making, you can use --with-match-limit=n
2997        to set a default value for the compiled library.
2998    
2999    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
3000        a different value is set. See 45 below.
3001    
3002    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
3003    
3004    43. Added a new function pcre_config(int, void *) to enable run-time extraction
3005    of things that can be changed at compile time. The first argument specifies
3006    what is wanted and the second points to where the information is to be placed.
3007    The current list of available information is:
3008    
3009      PCRE_CONFIG_UTF8
3010    
3011    The output is an integer that is set to one if UTF-8 support is available;
3012    otherwise it is set to zero.
3013    
3014      PCRE_CONFIG_NEWLINE
3015    
3016    The output is an integer that it set to the value of the code that is used for
3017    newline. It is either LF (10) or CR (13).
3018    
3019      PCRE_CONFIG_LINK_SIZE
3020    
3021    The output is an integer that contains the number of bytes used for internal
3022    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
3023    
3024      PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
3025    
3026    The output is an integer that contains the threshold above which the POSIX
3027    interface uses malloc() for output vectors. See item 31 above.
3028    
3029      PCRE_CONFIG_MATCH_LIMIT
3030    
3031    The output is an unsigned integer that contains the default limit of the number
3032    of match() calls in a pcre_exec() execution. See 42 above.
3033    
3034    44. pcretest has been upgraded by the addition of the -C option. This causes it
3035    to extract all the available output from the new pcre_config() function, and to
3036    output it. The program then exits immediately.
3037    
3038    45. A need has arisen to pass over additional data with calls to pcre_exec() in
3039    order to support additional features. One way would have been to define
3040    pcre_exec2() (for example) with extra arguments, but this would not have been
3041    extensible, and would also have required all calls to the original function to
3042    be mapped to the new one. Instead, I have chosen to extend the mechanism that
3043    is used for passing in "extra" data from pcre_study().
3044    
3045    The pcre_extra structure is now exposed and defined in pcre.h. It currently
3046    contains the following fields:
3047    
3048      flags         a bitmap indicating which of the following fields are set
3049      study_data    opaque data from pcre_study()
3050      match_limit   a way of specifying a limit on match() calls for a specific
3051                      call to pcre_exec()
3052      callout_data  data for callouts (see 49 below)
3053    
3054    The flag bits are also defined in pcre.h, and are
3055    
3056      PCRE_EXTRA_STUDY_DATA
3057      PCRE_EXTRA_MATCH_LIMIT
3058      PCRE_EXTRA_CALLOUT_DATA
3059    
3060    The pcre_study() function now returns one of these new pcre_extra blocks, with
3061    the actual study data pointed to by the study_data field, and the
3062    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
3063    before. That is, this change is entirely upwards-compatible and requires no
3064    change to existing code.
3065    
3066    If you want to pass in additional data to pcre_exec(), you can either place it
3067    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
3068    block.
3069    
3070    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
3071    data string contains the escape sequence \M, pcretest calls pcre_exec() several
3072    times with different match limits, until it finds the minimum value needed for
3073    pcre_exec() to complete. The value is then output. This can be instructive; for
3074    most simple matches the number is quite small, but for pathological cases it
3075    gets very large very quickly.
3076    
3077    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
3078    returns the size of the data block pointed to by the study_data field in a
3079    pcre_extra block, that is, the value that was passed as the argument to
3080    pcre_malloc() when PCRE was getting memory in which to place the information
3081    created by pcre_study(). The fourth argument should point to a size_t variable.
3082    pcretest has been extended so that this information is shown after a successful
3083    pcre_study() call when information about the compiled regex is being displayed.
3084    
3085    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
3086    because what follows is always an absolute path. (Later: it turns out that this
3087    is more than cosmetic for MinGW, because it doesn't like empty path
3088    components.)
3089    
3090    49. Some changes have been made to the callout feature (see 28 above):
3091    
3092    (i)  A callout function now has three choices for what it returns:
3093    
3094           0  =>  success, carry on matching
3095         > 0  =>  failure at this point, but backtrack if possible
3096         < 0  =>  serious error, return this value from pcre_exec()
3097    
3098         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
3099         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
3100         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
3101         use by callout functions. It will never be used by PCRE itself.
3102    
3103    (ii) The pcre_extra structure (see 45 above) has a void * field called
3104         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
3105         pcre_callout_block structure has a field of the same name. The contents of
3106         the field passed in the pcre_extra structure are passed to the callout
3107         function in the corresponding field in the callout block. This makes it
3108         easier to use the same callout-containing regex from multiple threads. For
3109         testing, the pcretest program has a new data escape
3110    
3111           \C*n        pass the number n (may be negative) as callout_data
3112    
3113         If the callout function in pcretest receives a non-zero value as
3114         callout_data, it returns that value.
3115    
3116    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
3117    there were some redundant $(CFLAGS) in commands that are now specified as
3118    $(LINK), which already includes $(CFLAGS).
3119    
3120    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
3121    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
3122    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
3123    one-byte characters throughout. Note that case-insensitive matching applies
3124    only to characters whose values are less than 256. PCRE doesn't support the
3125    notion of cases for higher-valued characters.
3126    
3127    (i)   A character class whose characters are all within 0-255 is handled as
3128          a bit map, and the map is inverted for negative classes. Previously, a
3129          character > 255 always failed to match such a class; however it should
3130          match if the class was a negative one (e.g. [^ab]). This has been fixed.
3131    
3132    (ii)  A negated character class with a single character < 255 is coded as
3133          "not this character" (OP_NOT). This wasn't working properly when the test
3134          character was multibyte, either singly or repeated.
3135    
3136    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
3137          mode, for example: \x{100}{2,3}.
3138    
3139    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
3140          singly or repeated) now correctly test multibyte characters. However,
3141          PCRE doesn't recognize any characters with values greater than 255 as
3142          digits, spaces, or word characters. Such characters always match \D, \S,
3143          and \W, and never match \d, \s, or \w.
3144    
3145    (v)   Classes may now contain characters and character ranges with values
3146          greater than 255. For example: [ab\x{100}-\x{400}].
3147    
3148    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
3149          PCRE in UTF-8 mode.
3150    
3151    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
3152    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
3153    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
3154    value.)
3155    
3156    53. The single man page has become too large. I have therefore split it up into
3157    a number of separate man pages. These also give rise to individual HTML pages;
3158    these are now put in a separate directory, and there is an index.html page that
3159    lists them all. Some hyperlinking between the pages has been installed.
3160    
3161    54. Added convenience functions for handling named capturing parentheses.
3162    
3163    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
3164    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
3165    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
3166    are faulted.
3167    
3168    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
3169    calling configure. These values are used when compiling the dftables.c program
3170    which is run to generate the source of the default character tables. They
3171    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
3172    you will need to set these values.
3173    
3174    57. Updated the building process for Windows DLL, as provided by Fred Cox.
3175    
3176    
3177    Version 3.9 02-Jan-02
3178    ---------------------
3179    
3180    1. A bit of extraneous text had somehow crept into the pcregrep documentation.
3181    
3182    2. If --disable-static was given, the building process failed when trying to
3183    build pcretest and pcregrep. (For some reason it was using libtool to compile
3184    them, which is not right, as they aren't part of the library.)
3185    
3186    
3187    Version 3.8 18-Dec-01
3188    ---------------------
3189    
3190    1. The experimental UTF-8 code was completely screwed up. It was packing the
3191    bytes in the wrong order. How dumb can you get?
3192    
3193    
3194    Version 3.7 29-Oct-01
3195    ---------------------
3196    
3197    1. In updating pcretest to check change 1 of version 3.6, I screwed up.
3198    This caused pcretest, when used on the test data, to segfault. Unfortunately,
3199    this didn't happen under Solaris 8, where I normally test things.
3200    
3201    2. The Makefile had to be changed to make it work on BSD systems, where 'make'
3202    doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
3203    isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
3204    this fix an hour or so after the initial 3.7 release.)
3205    
3206    
3207    Version 3.6 23-Oct-01
3208    ---------------------
3209    
3210    1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
3211    offsets passed as NULL with zero offset count.
3212    
3213    2. The config.guess and config.sub files had not been updated when I moved to
3214    the latest autoconf.
3215    
3216    
3217    Version 3.5 15-Aug-01
3218    ---------------------
3219    
3220    1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
3221    had been forgotten.
3222    
3223    2. By using declared but undefined structures, we can avoid using "void"
3224    definitions in pcre.h while keeping the internal definitions of the structures
3225    private.
3226    
3227    3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
3228    user point of view, this means that both static and shared libraries are built
3229    by default, but this can be individually controlled. More of the work of
3230    handling this static/shared cases is now inside libtool instead of PCRE's make
3231    file.
3232    
3233    4. The pcretest utility is now installed along with pcregrep because it is
3234    useful for users (to test regexs) and by doing this, it automatically gets
3235    relinked by libtool. The documentation has been turned into a man page, so
3236    there are now .1, .txt, and .html versions in /doc.
3237    
3238    5. Upgrades to pcregrep:
3239       (i)   Added long-form option names like gnu grep.
3240       (ii)  Added --help to list all options with an explanatory phrase.
3241       (iii) Added -r, --recursive to recurse into sub-directories.
3242       (iv)  Added -f, --file to read patterns from a file.
3243    
3244    6. pcre_exec() was referring to its "code" argument before testing that
3245    argument for NULL (and giving an error if it was NULL).
3246    
3247    7. Upgraded Makefile.in to allow for compiling in a different directory from
3248    the source directory.
3249    
3250    8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
3251    options bits, the pointer it was passed was to an int instead of to an unsigned
3252    long int. This mattered only on 64-bit systems.
3253    
3254    9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
3255    generated) instead of pcre.in, which it its source. Also made the same change
3256    in several of the .c files.
3257    
3258    10. A new release of gcc defines printf() as a macro, which broke pcretest
3259    because it had an ifdef in the middle of a string argument for printf(). Fixed
3260    by using separate calls to printf().
3261    
3262    11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
3263    script, to force use of CR or LF instead of \n in the source. On non-Unix
3264    systems, the value can be set in config.h.
3265    
3266    12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
3267    absolute limit. Changed the text of the error message to make this clear, and
3268    likewise updated the man page.
3269    
3270    13. The limit of 99 on the number of capturing subpatterns has been removed.
3271    The new limit is 65535, which I hope will not be a "real" limit.
3272    
3273    
3274    Version 3.4 22-Aug-00
3275    ---------------------
3276    
3277    1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
3278    
3279    2. Diagnose condition (?(0) as an error instead of crashing on matching.
3280    
3281    
3282    Version 3.3 01-Aug-00
3283    ---------------------
3284    
3285    1. If an octal character was given, but the value was greater than \377, it
3286    was not getting masked to the least significant bits, as documented. This could
3287    lead to crashes in some systems.
3288    
3289    2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
3290    the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
3291    
3292    3. Added the functions pcre_free_substring() and pcre_free_substring_list().
3293    These just pass their arguments on to (pcre_free)(), but they are provided
3294    because some uses of PCRE bind it to non-C systems that can call its functions,
3295    but cannot call free() or pcre_free() directly.
3296    
3297    4. Add "make test" as a synonym for "make check". Corrected some comments in
3298    the Makefile.
3299    
3300    5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
3301    Makefile.
3302    
3303    6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
3304    command called pgrep for grepping around the active processes.
3305    
3306    7. Added the beginnings of support for UTF-8 character strings.
3307    
3308    8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
3309    RANLIB to ./ltconfig so that they are used by libtool. I think these are all
3310    the relevant ones. (AR is not passed because ./ltconfig does its own figuring
3311    out for the ar command.)
3312    
3313    
3314    Version 3.2 12-May-00
3315    ---------------------
3316    
3317    This is purely a bug fixing release.
3318    
3319    1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
3320    of ZA. This was just one example of several cases that could provoke this bug,
3321    which was introduced by change 9 of version 2.00. The code for breaking
3322    infinite loops after an iteration that matches an empty string was't working
3323    correctly.
3324    
3325    2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
3326    when matched against abbab (for example). After matching an empty string, it
3327    wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
3328    caused it to match further down the string than it should.
3329    
3330    3. The code contained an inclusion of sys/types.h. It isn't clear why this
3331    was there because it doesn't seem to be needed, and it causes trouble on some
3332    systems, as it is not a Standard C header. It has been removed.
3333    
3334    4. Made 4 silly changes to the source to avoid stupid compiler warnings that
3335    were reported on the Macintosh. The changes were from
3336    
3337      while ((c = *(++ptr)) != 0 && c != '\n');
3338    to
3339      while ((c = *(++ptr)) != 0 && c != '\n') ;
3340    
3341    Totally extraordinary, but if that's what it takes...
3342    
3343    5. PCRE is being used in one environment where neither memmove() nor bcopy() is
3344    available. Added HAVE_BCOPY and an autoconf test for it; if neither
3345    HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
3346    assumes the way PCRE uses memmove() (always moving upwards).
3347    
3348    6. PCRE is being used in one environment where strchr() is not available. There
3349    was only one use in pcre.c, and writing it out to avoid strchr() probably gives
3350    faster code anyway.
3351    
3352    
3353    Version 3.1 09-Feb-00
3354    ---------------------
3355    
3356    The only change in this release is the fixing of some bugs in Makefile.in for
3357    the "install" target:
3358    
3359    (1) It was failing to install pcreposix.h.
3360    
3361    (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
3362    
3363    
3364    Version 3.0 01-Feb-00
3365    ---------------------
3366    
3367    1. Add support for the /+ modifier to perltest (to output $` like it does in
3368    pcretest).
3369    
3370    2. Add support for the /g modifier to perltest.
3371    
3372    3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
3373    matches null strings.
3374    
3375    4. Fix perltest so that it doesn't do unwanted things when fed an empty
3376    pattern. Perl treats empty patterns specially - it reuses the most recent
3377    pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
3378    effect.
3379    
3380    5. The POSIX interface was broken in that it was just handing over the POSIX
3381    captured string vector to pcre_exec(), but (since release 2.00) PCRE has
3382    required a bigger vector, with some working space on the end. This means that
3383    the POSIX wrapper now has to get and free some memory, and copy the results.
3384    
3385    6. Added some simple autoconf support, placing the test data and the
3386    documentation in separate directories, re-organizing some of the
3387    information files, and making it build pcre-config (a GNU standard). Also added
3388    libtool support for building PCRE as a shared library, which is now the
3389    default.
3390    
3391    7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
3392    09 are not valid octal constants. Single digits will be used for minor values
3393    less than 10.
3394    
3395    8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
3396    existing programs that set these in the POSIX interface can use PCRE without
3397    modification.
3398    
3399    9. Added a new function, pcre_fullinfo() with an extensible interface. It can
3400    return all that pcre_info() returns, plus additional data. The pcre_info()
3401    function is retained for compatibility, but is considered to be obsolete.
3402    
3403    10. Added experimental recursion feature (?R) to handle one common case that
3404    Perl 5.6 will be able to do with (?p{...}).
3405    
3406    11. Added support for POSIX character classes like [:alpha:], which Perl is
3407    adopting.
3408    
3409    
3410    Version 2.08 31-Aug-99
3411    ----------------------
3412    
3413    1. When startoffset was not zero and the pattern began with ".*", PCRE was not
3414    trying to match at the startoffset position, but instead was moving forward to
3415    the next newline as if a previous match had failed.
3416    
3417    2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
3418    and could get into a loop if a null string was matched other than at the start
3419    of the subject.
3420    
3421    3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
3422    be distinguished at compile time, and for completeness also added PCRE_DATE.
3423    
3424    5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
3425    in GnuWin32 environments.
3426    
3427    
3428    Version 2.07 29-Jul-99
3429    ----------------------
3430    
3431    1. The documentation is now supplied in plain text form and HTML as well as in
3432    the form of man page sources.
3433    
3434    2. C++ compilers don't like assigning (void *) values to other pointer types.
3435    In particular this affects malloc(). Although there is no problem in Standard
3436    C, I've put in casts to keep C++ compilers happy.
3437    
3438    3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
3439    should be (const char *).
3440    
3441    4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
3442    be useful for non-Unix systems who don't want to bother with the POSIX stuff.
3443    However, I haven't made this a standard facility. The documentation doesn't
3444    mention it, and the Makefile doesn't support it.
3445    
3446    5. The Makefile now contains an "install" target, with editable destinations at
3447    the top of the file. The pcretest program is not installed.
3448    
3449    6. pgrep -V now gives the PCRE version number and date.
3450    
3451    7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
3452    causing the entire string to be ignored, instead of just the last character.
3453    
3454    8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
3455    non-matching string, it can take a very, very long time, even for strings of
3456    quite modest length, because of the nested recursion. PCRE now does better in
3457    some of these cases. It does this by remembering the last required literal
3458    character in the pattern, and pre-searching the subject to ensure it is present
3459    before running the real match. In other words, it applies a heuristic to detect
3460    some types of certain failure quickly, and in the above example, if presented
3461    with a string that has no trailing " it gives "no match" very quickly.
3462    
3463    9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
3464    other alternatives are tried instead.
3465    
3466    
3467    Version 2.06 09-Jun-99
3468    ----------------------
3469    
3470    1. Change pcretest's output for amount of store used to show just the code
3471    space, because the remainder (the data block) varies in size between 32-bit and
3472    64-bit systems.
3473    
3474    2. Added an extra argument to pcre_exec() to supply an offset in the subject to
3475    start matching at. This allows lookbehinds to work when searching for multiple
3476    occurrences in a string.
3477    
3478    3. Added additional options to pcretest for testing multiple occurrences:
3479    
3480       /+   outputs the rest of the string that follows a match
3481       /g   loops for multiple occurrences, using the new startoffset argument
3482       /G   loops for multiple occurrences by passing an incremented pointer
3483    
3484    4. PCRE wasn't doing the "first character" optimization for patterns starting
3485    with \b or \B, though it was doing it for other lookbehind assertions. That is,
3486    it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
3487    the letter 'x'. On long subject strings, this gives a significant speed-up.
3488    
3489    
3490    Version 2.05 21-Apr-99
3491    ----------------------
3492    
3493    1. Changed the type of magic_number from int to long int so that it works
3494    properly on 16-bit systems.
3495    
3496    2. Fixed a bug which caused patterns starting with .* not to work correctly
3497    when the subject string contained newline characters. PCRE was assuming
3498    anchoring for such patterns in all cases, which is not correct because .* will
3499    not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
3500    DOTALL is set at top level; otherwise it knows that patterns starting with .*
3501    must be retried after every newline in the subject.
3502    
3503    
3504    Version 2.04 18-Feb-99
3505    ----------------------
3506    
3507    1. For parenthesized subpatterns with repeats whose minimum was zero, the
3508    computation of the store needed to hold the pattern was incorrect (too large).
3509    If such patterns were nested a few deep, this could multiply and become a real
3510    problem.
3511    
3512    2. Added /M option to pcretest to show the memory requirement of a specific
3513    pattern. Made -m a synonym of -s (which does this globally) for compatibility.
3514    
3515    3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
3516    compiled in such a way that the backtracking after subsequent failure was
3517    pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
3518    ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
3519    
3520    
3521    Version 2.03 02-Feb-99
3522    ----------------------
3523    
3524    1. Fixed typo and small mistake in man page.
3525    
3526    2. Added 4th condition (GPL supersedes if conflict) and created separate
3527    LICENCE file containing the conditions.
3528    
3529    3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
3530    Perl, that is the internal \ allows the delimiter to be included in the
3531    pattern. Locked out the use of \ as a delimiter. If \ immediately follows
3532    the final delimiter, add \ to the end of the pattern (to test the error).
3533    
3534    4. Added the convenience functions for extracting substrings after a successful
3535    match. Updated pcretest to make it able to test these functions.
3536    
3537    
3538    Version 2.02 14-Jan-99
3539    ----------------------
3540    
3541    1. Initialized the working variables associated with each extraction so that
3542    their saving and restoring doesn't refer to uninitialized store.
3543    
3544    2. Put dummy code into study.c in order to trick the optimizer of the IBM C
3545    compiler for OS/2 into generating correct code. Apparently IBM isn't going to
3546    fix the problem.
3547    
3548    3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
3549    calls, and wasn't printing the correct value for compiling calls. Increased the
3550    default value of LOOPREPEAT, and the number of significant figures in the
3551    times.
3552    
3553    4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
3554    
3555    5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
3556    a building problem on Windows NT with a FAT file system.
3557    
3558    
3559    Version 2.01 21-Oct-98
3560    ----------------------
3561    
3562    1. Changed the API for pcre_compile() to allow for the provision of a pointer
3563    to character tables built by pcre_maketables() in the current locale. If NULL
3564    is passed, the default tables are used.
3565    
3566    
3567    Version 2.00 24-Sep-98
3568    ----------------------
3569    
3570    1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
3571    it any more.
3572    
3573    2. Allow quantification of (?>) groups, and make it work correctly.
3574    
3575    3. The first character computation wasn't working for (?>) groups.
3576    
3577    4. Correct the implementation of \Z (it is permitted to match on the \n at the
3578    end of the subject) and add 5.005's \z, which really does match only at the
3579    very end of the subject.
3580    
3581    5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
3582    
3583    6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
3584    DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
3585    localized options. All options to pcre_study() were also removed.
3586    
3587    7. Add other new features from 5.005:
3588    
3589       $(?<=           positive lookbehind
3590       $(?<!           negative lookbehind
3591       (?imsx-imsx)    added the unsetting capability
3592                       such a setting is global if at outer level; local otherwise
3593       (?imsx-imsx:)   non-capturing groups with option setting
3594       (?(cond)re|re)  conditional pattern matching
3595    
3596       A backreference to itself in a repeated group matches the previous
3597       captured string.
3598    
3599    8. General tidying up of studying (both automatic and via "study")
3600    consequential on the addition of new assertions.
3601    
3602    9. As in 5.005, unlimited repeated groups that could match an empty substring
3603    are no longer faulted at compile time. Instead, the loop is forcibly broken at
3604    runtime if any iteration does actually match an empty substring.
3605    
3606    10. Include the RunTest script in the distribution.
3607    
3608    11. Added tests from the Perl 5.005_02 distribution. This showed up a few
3609    discrepancies, some of which were old and were also with respect to 5.004. They
3610    have now been fixed.
3611    
3612    
3613    Version 1.09 28-Apr-98
3614    ----------------------
3615    
3616    1. A negated single character class followed by a quantifier with a minimum
3617    value of one (e.g.  [^x]{1,6}  ) was not compiled correctly. This could lead to
3618    program crashes, or just wrong answers. This did not apply to negated classes
3619    containing more than one character, or to minima other than one.
3620    
3621    
3622  Version 1.08 27-Mar-98  Version 1.08 27-Mar-98
3623  ----------------------  ----------------------

Legend:
Removed from v.19  
changed lines
  Added in v.524

  ViewVC Help
Powered by ViewVC 1.1.5