ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 3 by nigel, Sat Feb 24 21:38:01 2007 UTC revision 579 by ph10, Wed Nov 24 17:39:25 2010 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
4    Version 8.11 24-Nov-2010
5    ------------------------
7    1.  (*THEN) was not working properly if there were untried alternatives prior
8        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
9        backtracked to try for "b" instead of moving to the next alternative branch
10        at the same level (in this case, to look for "c"). The Perl documentation
11        is clear that when (*THEN) is backtracked onto, it goes to the "next
12        alternative in the innermost enclosing group".
14    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
15        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
16        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
17        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
18        (*THEN).
20    3.  If \s appeared in a character class, it removed the VT character from
21        the class, even if it had been included by some previous item, for example
22        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
23        of \s, but is part of the POSIX "space" class.)
25    4.  A partial match never returns an empty string (because you can always
26        match an empty string at the end of the subject); however the checking for
27        an empty string was starting at the "start of match" point. This has been
28        changed to the "earliest inspected character" point, because the returned
29        data for a partial match starts at this character. This means that, for
30        example, /(?<=abc)def/ gives a partial match for the subject "abc"
31        (previously it gave "no match").
33    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
34        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
35        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
36        has an implication that the given string is incomplete (because a partial
37        match is preferred over a full match). For this reason, these items now
38        give a partial match in this situation. [Aside: previously, the one case
39        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
40        match rather than a full match, which was wrong by the old rules, but is
41        now correct.]
43    6.  There was a bug in the handling of #-introduced comments, recognized when
44        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
45        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
46        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
47        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
48        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
49        places in pcre_compile().
51    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
52        comments when looking ahead for named forward references to subpatterns,
53        the only newline sequence it recognized was NL. It now handles newlines
54        according to the set newline convention.
56    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
57        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
58        cater for a lack of strerror(). These oversights have been fixed.
60    9.  Added --match-limit and --recursion-limit to pcregrep.
62    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
64    11. When the -o option was used, pcregrep was setting a return code of 1, even
65        when matches were found, and --line-buffered was not being honoured.
67    12. Added an optional parentheses number to the -o and --only-matching options
68        of pcregrep.
70    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
71        can match an empty string. The code to do it in pcretest and pcredemo
72        needed fixing:
74        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
75            only one byte after an empty string match just before CRLF (this case
76            just got forgotten; "any" and "anycrlf" were OK).
78        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
79            mode when an empty string match preceded an ASCII character followed by
80            a non-ASCII character. (The code for advancing by one character rather
81            than one byte was nonsense.)
83        (c) The pcredemo.c sample program did not have any code at all to handle
84            the cases when CRLF is a valid newline sequence.
86    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
87        as a starting offset was within the subject string. There is now a new
88        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
89        negative or greater than the length of the string. In order to test this,
90        pcretest is extended to allow the setting of negative starting offsets.
92    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
93        starting offset points to the beginning of a UTF-8 character was
94        unnecessarily clumsy. I tidied it up.
96    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
97        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
99    17. Nobody had reported that the --include_dir option, which was added in
100        release 7.7 should have been called --include-dir (hyphen, not underscore)
101        for compatibility with GNU grep. I have changed it to --include-dir, but
102        left --include_dir as an undocumented synonym, and the same for
103        --exclude-dir, though that is not available in GNU grep, at least as of
104        release 2.5.4.
106    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
107        characters from a string of bytes) have been redefined so as not to use
108        loops, in order to improve performance in some environments. At the same
109        time, I abstracted some of the common code into auxiliary macros to save
110        repetition (this should not affect the compiled code).
112    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
113        compile-time error is now given if \c is not followed by an ASCII
114        character, that is, a byte less than 128. (In EBCDIC mode, the code is
115        different, and any byte value is allowed.)
117    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
118        START_OPTIMIZE option, which is now allowed at compile time - but just
119        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
120        to pcregrep and other applications that have no direct access to PCRE
121        options. The new /Y option in pcretest sets this option when calling
122        pcre_compile().
124    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
125        back references. Groups containing recursive back references were forced to
126        be atomic by that change, but in the case of named groups, the amount of
127        memory required was incorrectly computed, leading to "Failed: internal
128        error: code overflow". This has been fixed.
131    Version 8.10 25-Jun-2010
132    ------------------------
134    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
135        THEN.
137    2.  (*ACCEPT) was not working when inside an atomic group.
139    3.  Inside a character class, \B is treated as a literal by default, but
140        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
141        causes the error). The code is unchanged, but I tidied the documentation.
143    4.  Inside a character class, PCRE always treated \R and \X as literals,
144        whereas Perl faults them if its -w option is set. I have changed PCRE so
145        that it faults them when PCRE_EXTRA is set.
147    5.  Added support for \N, which always matches any character other than
148        newline. (It is the same as "." when PCRE_DOTALL is not set.)
150    6.  When compiling pcregrep with newer versions of gcc which may have
151        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
152        declared with attribute warn_unused_result" were given. Just casting the
153        result to (void) does not stop the warnings; a more elaborate fudge is
154        needed. I've used a macro to implement this.
156    7.  Minor change to pcretest.c to avoid a compiler warning.
158    8.  Added four artifical Unicode properties to help with an option to make
159        \s etc use properties (see next item). The new properties are: Xan
160        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
162    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
163        use Unicode properties. (*UCP) at the start of a pattern can be used to set
164        this option. Modified pcretest to add /W to test this facility. Added
165        REG_UCP to make it available via the POSIX interface.
167    10. Added --line-buffered to pcregrep.
169    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
170        studied, and the match started with a letter with a code point greater than
171        127 whose first byte was different to the first byte of the other case of
172        the letter, the other case of this starting letter was not recognized
173        (#976).
175    12. If a pattern that was studied started with a repeated Unicode property
176        test, for example, \p{Nd}+, there was the theoretical possibility of
177        setting up an incorrect bitmap of starting bytes, but fortunately it could
178        not have actually happened in practice until change 8 above was made (it
179        added property types that matched character-matching opcodes).
181    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
182        possible starting bytes for non-anchored patterns.
184    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
185        \R, and also a number of cases that involve Unicode properties, both
186        explicit and implicit when PCRE_UCP is set.
188    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
189        input, it could crash or give wrong results if characters with values
190        greater than 0xc0 were present in the subject string. (Detail: it assumed
191        UTF-8 input when processing these items.)
193    16. Added a lot of (int) casts to avoid compiler warnings in systems where
194        size_t is 64-bit (#991).
196    17. Added a check for running out of memory when PCRE is compiled with
197        --disable-stack-for-recursion (#990).
199    18. If the last data line in a file for pcretest does not have a newline on
200        the end, a newline was missing in the output.
202    19. The default pcre_chartables.c file recognizes only ASCII characters (values
203        less than 128) in its various bitmaps. However, there is a facility for
204        generating tables according to the current locale when PCRE is compiled. It
205        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
206        characters, are recognized by isspace() and therefore were getting set in
207        these tables, and indeed these tables seem to approximate to ISO 8859. This
208        caused a problem in UTF-8 mode when pcre_study() was used to create a list
209        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
210        which of course cannot start UTF-8 characters. I have changed the code so
211        that only real ASCII characters (less than 128) and the correct starting
212        bytes for UTF-8 encodings are set for characters greater than 127 when in
213        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
214        altogether.)
216    20. Added the /T option to pcretest so as to be able to run tests with non-
217        standard character tables, thus making it possible to include the tests
218        used for 19 above in the standard set of tests.
220    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
221        reference to a subpattern the other side of a comment that contains an
222        opening parenthesis caused either an internal compiling error, or a
223        reference to the wrong subpattern.
226    Version 8.02 19-Mar-2010
227    ------------------------
229    1.  The Unicode data tables have been updated to Unicode 5.2.0.
231    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
232        configured.
234    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
235        original author of that file, following a query about its status.
237    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
238        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
240    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
241        quantifier applied to a forward-referencing subroutine call, could compile
242        incorrect code or give the error "internal error: previously-checked
243        referenced subpattern not found".
245    6.  Both MS Visual Studio and Symbian OS have problems with initializing
246        variables to point to external functions. For these systems, therefore,
247        pcre_malloc etc. are now initialized to local functions that call the
248        relevant global functions.
250    7.  There were two entries missing in the vectors called coptable and poptable
251        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
252        I've fixed the data, and added a kludgy way of testing at compile time that
253        the lengths are correct (equal to the number of opcodes).
255    8.  Following on from 7, I added a similar kludge to check the length of the
256        eint vector in pcreposix.c.
258    9.  Error texts for pcre_compile() are held as one long string to avoid too
259        much relocation at load time. To find a text, the string is searched,
260        counting zeros. There was no check for running off the end of the string,
261        which could happen if a new error number was added without updating the
262        string.
264    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
266    11. \K was not working if it appeared in an atomic group or in a group that
267        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
268        \K is "not well defined" if used in an assertion. PCRE now accepts it if
269        the assertion is positive, but not if it is negative.
271    12. Change 11 fortuitously reduced the size of the stack frame used in the
272        "match()" function of pcre_exec.c by one pointer. Forthcoming
273        implementation of support for (*MARK) will need an extra pointer on the
274        stack; I have reserved it now, so that the stack frame size does not
275        decrease.
277    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
278        item in branch that calls a recursion is a subroutine call - as in the
279        second branch in the above example - was incorrectly given the compile-
280        time error "recursive call could loop indefinitely" because pcre_compile()
281        was not correctly checking the subroutine for matching a non-empty string.
283    14. The checks for overrunning compiling workspace could trigger after an
284        overrun had occurred. This is a "should never occur" error, but it can be
285        triggered by pathological patterns such as hundreds of nested parentheses.
286        The checks now trigger 100 bytes before the end of the workspace.
288    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
291    Version 8.01 19-Jan-2010
292    ------------------------
294    1.  If a pattern contained a conditional subpattern with only one branch (in
295        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
296        computed the wrong minimum data length (which is of course zero for such
297        subpatterns). This could cause incorrect "no match" results.
299    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
300        the pattern is reset in the first branch, pcre_compile() failed with
301        "internal error: code overflow at offset...". This happened only when
302        the reset was to the original external option setting. (An optimization
303        abstracts leading options settings into an external setting, which was the
304        cause of this.)
306    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
307        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
308        assertion pattern did not match (meaning that the assertion was true), it
309        was incorrectly treated as false if the SKIP had been reached during the
310        matching. This also applied to assertions used as conditions.
312    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
313        assertion subpattern, including such a pattern used as a condition,
314        unpredictable results occurred, instead of the error return
317    5.  The C++ GlobalReplace function was not working like Perl for the special
318        situation when an empty string is matched. It now does the fancy magic
319        stuff that is necessary.
321    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
322        removed. (These were left over from very, very early versions of PCRE.)
324    7.  Some cosmetic changes to the code to make life easier when compiling it
325        as part of something else:
327        (a) Change DEBUG to PCRE_DEBUG.
329        (b) In pcre_compile(), rename the member of the "branch_chain" structure
330            called "current" as "current_branch", to prevent a collision with the
331            Linux macro when compiled as a kernel module.
333        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
334            prevent a collision with the Linux macro when compiled as a kernel
335            module.
337    8.  In pcre_compile() there are some checks for integer overflows that used to
338        cast potentially large values to (double). This has been changed to that
339        when building, a check for int64_t is made, and if it is found, it is used
340        instead, thus avoiding the use of floating point arithmetic. (There is no
341        other use of FP in PCRE.) If int64_t is not found, the fallback is to
342        double.
344    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
345        2005 (difference between two addresses compared to an unsigned value).
347    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
348        custom one, because of the following reported problem in Windows:
350          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
351              under Win32.
352          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
353              therefore missing the function definition.
354          - The compiler thus generates a "C" signature for the test function.
355          - The linker fails to find the "C" function.
356          - PCRE fails to configure if asked to do so against libbz2.
358    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
359        messages were output:
361          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
362          rerunning libtoolize, to keep the correct libtool macros in-tree.
363          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
365        I have done both of these things.
367    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
368        most of the time, it *can* run out if it is given a pattern that contains a
369        runaway infinite recursion. I updated the discussion in the pcrestack man
370        page.
372    13. Now that we have gone to the x.xx style of version numbers, the minor
373        version may start with zero. Using 08 or 09 is a bad idea because users
374        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
375        interpreted as invalid octal numbers. I've updated the previous comment in
376        configure.ac, and also added a check that gives an error if 08 or 09 are
377        used.
379    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
380        causing partial matching to fail when the end of the subject matched \W
381        in a UTF-8 pattern where \W was quantified with a minimum of 3.
383    15. There were some discrepancies between the declarations in pcre_internal.h
384        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
385        their definitions. The declarations used "const uschar *" and the
386        definitions used USPTR. Even though USPTR is normally defined as "const
387        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
388        reported that: "This difference in casting confuses some C++ compilers, for
389        example, SunCC recognizes above declarations as different functions and
390        generates broken code for hbpcre." I have changed the declarations to use
391        USPTR.
393    16. GNU libtool is named differently on some systems. The autogen.sh script now
394        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
395        (FreeBSD).
397    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
398        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
399        comment: "Figure out how to create a longlong from a string: strtoll and
400        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
401        instance, but it only takes 2 args instead of 3!"
403    18. A subtle bug concerned with back references has been fixed by a change of
404        specification, with a corresponding code fix. A pattern such as
405        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
406        refers, was giving matches when it shouldn't. For example, xa=xaaa would
407        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
408        same bug. Such groups have to be quantified to be useful, or contained
409        inside another quantified group. (If there's no repetition, the reference
410        can never match.) The problem arises because, having left the group and
411        moved on to the rest of the pattern, a later failure that backtracks into
412        the group uses the captured value from the final iteration of the group
413        rather than the correct earlier one. I have fixed this in PCRE by forcing
414        any group that contains a reference to itself to be an atomic group; that
415        is, there cannot be any backtracking into it once it has completed. This is
416        similar to recursive and subroutine calls.
419    Version 8.00 19-Oct-09
420    ----------------------
422    1.  The table for translating pcre_compile() error codes into POSIX error codes
423        was out-of-date, and there was no check on the pcre_compile() error code
424        being within the table. This could lead to an OK return being given in
425        error.
427    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
428        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
429        in a Windows environment.
431    3.  The pcregrep --count option prints the count for each file even when it is
432        zero, as does GNU grep. However, pcregrep was also printing all files when
433        --files-with-matches was added. Now, when both options are given, it prints
434        counts only for those files that have at least one match. (GNU grep just
435        prints the file name in this circumstance, but including the count seems
436        more useful - otherwise, why use --count?) Also ensured that the
437        combination -clh just lists non-zero counts, with no names.
439    4.  The long form of the pcregrep -F option was incorrectly implemented as
440        --fixed_strings instead of --fixed-strings. This is an incompatible change,
441        but it seems right to fix it, and I didn't think it was worth preserving
442        the old behaviour.
444    5.  The command line items --regex=pattern and --regexp=pattern were not
445        recognized by pcregrep, which required --regex pattern or --regexp pattern
446        (with a space rather than an '='). The man page documented the '=' forms,
447        which are compatible with GNU grep; these now work.
449    6.  No libpcreposix.pc file was created for pkg-config; there was just
450        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
452    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
453        when UCP support is not needed, by modifying the Python script that
454        generates it from Unicode data files. This should not matter if the module
455        is correctly used as a library, but I received one complaint about 50K of
456        unwanted data. My guess is that the person linked everything into his
457        program rather than using a library. Anyway, it does no harm.
459    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
460        was a minimum greater than 1 for a wide character in a possessive
461        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
462        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
463        character. Chaos in the form of incorrect output or a compiling loop could
464        result.
466    9.  The restrictions on what a pattern can contain when partial matching is
467        requested for pcre_exec() have been removed. All patterns can now be
468        partially matched by this function. In addition, if there are at least two
469        slots in the offset vector, the offset of the earliest inspected character
470        for the match and the offset of the end of the subject are set in them when
471        PCRE_ERROR_PARTIAL is returned.
473    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
474        synonymous with PCRE_PARTIAL, for backwards compatibility, and
475        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
476        and may be more useful for multi-segment matching.
478    11. Partial matching with pcre_exec() is now more intuitive. A partial match
479        used to be given if ever the end of the subject was reached; now it is
480        given only if matching could not proceed because another character was
481        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
482        string "Z", which now yields "no match" instead of "partial match". In the
483        case of pcre_dfa_exec(), "no match" is given if every matching path for the
484        final character ended with (*FAIL).
486    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
487        if the pattern had a "must contain" character that was already found in the
488        earlier partial match, unless partial matching was again requested. For
489        example, with the pattern /dog.(body)?/, the "must contain" character is
490        "g". If the first part-match was for the string "dog", restarting with
491        "sbody" failed. This bug has been fixed.
493    13. The string returned by pcre_dfa_exec() after a partial match has been
494        changed so that it starts at the first inspected character rather than the
495        first character of the match. This makes a difference only if the pattern
496        starts with a lookbehind assertion or \b or \B (\K is not supported by
497        pcre_dfa_exec()). It's an incompatible change, but it makes the two
498        matching functions compatible, and I think it's the right thing to do.
500    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
501        so that the demonstration program is easily available in environments where
502        PCRE has not been installed from source.
504    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
505        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
506        library.
508    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
509        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
510        is not the first non-POSIX option to be added. Clearly some people find
511        these options useful.
513    17. If a caller to the POSIX matching function regexec() passes a non-zero
514        value for nmatch with a NULL value for pmatch, the value of
515        nmatch is forced to zero.
517    18. RunGrepTest did not have a test for the availability of the -u option of
518        the diff command, as RunTest does. It now checks in the same way as
519        RunTest, and also checks for the -b option.
521    19. If an odd number of negated classes containing just a single character
522        interposed, within parentheses, between a forward reference to a named
523        subpattern and the definition of the subpattern, compilation crashed with
524        an internal error, complaining that it could not find the referenced
525        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
526        [The bug was that it was starting one character too far in when skipping
527        over the character class, thus treating the ] as data rather than
528        terminating the class. This meant it could skip too much.]
530    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
531        /g option in pcretest when the pattern contains \K, which makes it possible
532        to have an empty string match not at the start, even when the pattern is
533        anchored. Updated pcretest and pcredemo to use this option.
535    21. If the maximum number of capturing subpatterns in a recursion was greater
536        than the maximum at the outer level, the higher number was returned, but
537        with unset values at the outer level. The correct (outer level) value is
538        now given.
540    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
541        PCRE did not set those parentheses (unlike Perl). I have now found a way to
542        make it do so. The string so far is captured, making this feature
543        compatible with Perl.
545    23. The tests have been re-organized, adding tests 11 and 12, to make it
546        possible to check the Perl 5.10 features against Perl 5.10.
548    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
549        pattern matches a fixed length string. PCRE did not allow this; now it
550        does. Neither allows recursion.
552    25. I finally figured out how to implement a request to provide the minimum
553        length of subject string that was needed in order to match a given pattern.
554        (It was back references and recursion that I had previously got hung up
555        on.) This code has now been added to pcre_study(); it finds a lower bound
556        to the length of subject needed. It is not necessarily the greatest lower
557        bound, but using it to avoid searching strings that are too short does give
558        some useful speed-ups. The value is available to calling programs via
559        pcre_fullinfo().
561    26. While implementing 25, I discovered to my embarrassment that pcretest had
562        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
563        study optimizations had never been tested with that matching function.
564        Oops. What is worse, even when it was passed study data, there was a bug in
565        pcre_dfa_exec() that meant it never actually used it. Double oops. There
566        were also very few tests of studied patterns with pcre_dfa_exec().
568    27. If (?| is used to create subpatterns with duplicate numbers, they are now
569        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
570        on the other side of the coin, they are no longer allowed to have different
571        names, because these cannot be distinguished in PCRE, and this has caused
572        confusion. (This is a difference from Perl.)
574    28. When duplicate subpattern names are present (necessarily with different
575        numbers, as required by 27 above), and a test is made by name in a
576        conditional pattern, either for a subpattern having been matched, or for
577        recursion in such a pattern, all the associated numbered subpatterns are
578        tested, and the overall condition is true if the condition is true for any
579        one of them. This is the way Perl works, and is also more like the way
580        testing by number works.
583    Version 7.9 11-Apr-09
584    ---------------------
586    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
587        (pcretest), all targets were linked against these libraries. This included
588        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
589        libraries. This caused unwanted dependencies to be created. This problem
590        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
591        pcretest is linked with readline.
593    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
594        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
595        moved outside it again, because FALSE and TRUE are already defined in AIX,
596        but BOOL is not.
598    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
599        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
601    4.  The pcregrep documentation said spaces were inserted as well as colons (or
602        hyphens) following file names and line numbers when outputting matching
603        lines. This is not true; no spaces are inserted. I have also clarified the
604        wording for the --colour (or --color) option.
606    5.  In pcregrep, when --colour was used with -o, the list of matching strings
607        was not coloured; this is different to GNU grep, so I have changed it to be
608        the same.
610    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
611        each matching line was coloured. Now it goes on to look for further matches
612        of any of the test patterns, which is the same behaviour as GNU grep.
614    7.  A pattern that could match an empty string could cause pcregrep to loop; it
615        doesn't make sense to accept an empty string match in pcregrep, so I have
616        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
617        seems to be how GNU grep behaves.
619    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
620        start or after a newline", because the conditional assertion was not being
621        correctly handled. The rule now is that both the assertion and what follows
622        in the first alternative must satisfy the test.
624    9.  If auto-callout was enabled in a pattern with a conditional group whose
625        condition was an assertion, PCRE could crash during matching, both with
626        pcre_exec() and pcre_dfa_exec().
628    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
629        used for matching.
631    11. Unicode property support in character classes was not working for
632        characters (bytes) greater than 127 when not in UTF-8 mode.
634    12. Added the -M command line option to pcretest.
636    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
638    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
640    16. Added comments and documentation about mis-use of no_arg in the C++
641        wrapper.
643    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
644        from Martin Jerabek that uses macro names for all relevant character and
645        string constants.
647    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
648        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
649        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
650        these, but not everybody uses configure.
652    19. A conditional group that had only one branch was not being correctly
653        recognized as an item that could match an empty string. This meant that an
654        enclosing group might also not be so recognized, causing infinite looping
655        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
656        with the subject "ab", where knowledge that the repeated group can match
657        nothing is needed in order to break the loop.
659    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
660        exec(), but without supplying a callout function, matching went wrong.
662    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
663        leak if the size of the offset vector was greater than 30. When the vector
664        is smaller, the saved offsets during recursion go onto a local stack
665        vector, but for larger vectors malloc() is used. It was failing to free
666        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
667        error, in fact).
669    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
670        heapframe that is used only when UTF-8 support is enabled. This caused no
671        problem, but was untidy.
673    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
674        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
675        included within another project.
677    24. Steven Van Ingelgem's patches to add more options to the CMake support,
678        slightly modified by me:
680          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
681              not building pcregrep.
683          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
684              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
686    25. Forward references, both numeric and by name, in patterns that made use of
687        duplicate group numbers, could behave incorrectly or give incorrect errors,
688        because when scanning forward to find the reference group, PCRE was not
689        taking into account the duplicate group numbers. A pattern such as
690        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
692    26. Changed a few more instances of "const unsigned char *" to USPTR, making
693        the feature of a custom pointer more persuasive (as requested by a user).
695    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
696        pcretest.c, inside #ifndefs, because it seems they are sometimes already
697        pre-defined.
699    28. Added support for (*UTF8) at the start of a pattern.
701    29. Arrange for flags added by the "release type" setting in CMake to be shown
702        in the configuration summary.
705    Version 7.8 05-Sep-08
706    ---------------------
708    1.  Replaced UCP searching code with optimized version as implemented for Ad
709        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
710        stage table and inline lookup instead of a function, giving speed ups of 2
711        to 5 times on some simple patterns that I tested. Permission was given to
712        distribute the MultiStage2.py script that generates the tables (it's not in
713        the tarball, but is in the Subversion repository).
715    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
716        scripts.
718    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
719        a group with a zero qualifier. The result of the study could be incorrect,
720        or the function might crash, depending on the pattern.
722    4.  Caseless matching was not working for non-ASCII characters in back
723        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
724        It now works when Unicode Property Support is available.
726    5.  In pcretest, an escape such as \x{de} in the data was always generating
727        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
728        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
729        truncation.
731    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
733    7.  Added two (int) casts to pcregrep when printing the difference of two
734        pointers, in case they are 64-bit values.
736    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
737        test 2 if it fails.
739    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
740        and a #define of that name to empty if it is not externally set. This is to
741        allow users of MSVC to set it if necessary.
743    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
744        the convenience functions in the pcre_get.c source file.
746    11. An option change at the start of a pattern that had top-level alternatives
747        could cause overwriting and/or a crash. This command provoked a crash in
748        some environments:
750          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
752        This potential security problem was recorded as CVE-2008-2371.
754    12. For a pattern where the match had to start at the beginning or immediately
755        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
756        pcre_dfa_exec() could read past the end of the passed subject if there was
757        no match. To help with detecting such bugs (e.g. with valgrind), I modified
758        pcretest so that it places the subject at the end of its malloc-ed buffer.
760    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
761        exec() might read past the end of the data buffer in UTF-8 mode.
763    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
764        the data contained the byte 0x85 as part of a UTF-8 character within its
765        first line. This applied both to normal and DFA matching.
767    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
768        /^[^d]*?$/8 failed to match "abc".
770    16. Added a missing copyright notice to pcrecpp_internal.h.
772    17. Make it more clear in the documentation that values returned from
773        pcre_exec() in ovector are byte offsets, not character counts.
775    18. Tidied a few places to stop certain compilers from issuing warnings.
777    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
778        supplied by Stefan Weber. I made a further small update for 7.8 because
779        there is a change of source arrangements: the pcre_searchfuncs.c module is
780        replaced by pcre_ucd.c.
783    Version 7.7 07-May-08
784    ---------------------
786    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
787        a string to a long long, pretend we don't even have a long long." This is
788        done by checking for the strtoq, strtoll, and _strtoi64 functions.
790    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
791        pre-7.6 versions, which defined a global no_arg variable instead of putting
792        it in the RE class. (See also #8 below.)
794    3.  Remove a line of dead code, identified by coverity and reported by Nuno
795        Lopes.
797    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
799        (1) The include/exclude patterns were being applied to the whole pathnames
800            of files, instead of just to the final components.
802        (2) If there was more than one level of directory, the subdirectories were
803            skipped unless they satisfied the include/exclude conditions. This is
804            inconsistent with GNU grep (and could even be seen as contrary to the
805            pcregrep specification - which I improved to make it absolutely clear).
806            The action now is always to scan all levels of directory, and just
807            apply the include/exclude patterns to regular files.
809    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
810        --exclude_dir in the tests to avoid scanning .svn directories.
812    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
813        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
814        doesn't support NULs in patterns.
816    7.  Added some missing "const"s to declarations of static tables in
817        pcre_compile.c and pcre_dfa_exec.c.
819    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
820        caused by fix #2  above. (Subsequently also a second patch to fix the
821        first patch. And a third patch - this was a messy problem.)
823    9.  Applied Craig's patch to remove the use of push_back().
825    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
826        matching function regexec().
828    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
829        which, however, unlike Perl's \g{...}, are subroutine calls, not back
830        references. PCRE supports relative numbers with this syntax (I don't think
831        Oniguruma does).
833    12. Previously, a group with a zero repeat such as (...){0} was completely
834        omitted from the compiled regex. However, this means that if the group
835        was called as a subroutine from elsewhere in the pattern, things went wrong
836        (an internal error was given). Such groups are now left in the compiled
837        pattern, with a new opcode that causes them to be skipped at execution
838        time.
840    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
841        to the way PCRE behaves:
843        (a) A lone ] character is dis-allowed (Perl treats it as data).
845        (b) A back reference to an unmatched subpattern matches an empty string
846            (Perl fails the current match path).
848        (c) A data ] in a character class must be notated as \] because if the
849            first data character in a class is ], it defines an empty class. (In
850            Perl it is not possible to have an empty class.) The empty class []
851            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
852            The negative empty class [^] matches any one character, independently
853            of the DOTALL setting.
855    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
856        non-existent subpattern following a character class starting with ']' and
857        containing () gave an internal compiling error instead of "reference to
858        non-existent subpattern". Fortunately, when the pattern did exist, the
859        compiled code was correct. (When scanning forwards to check for the
860        existencd of the subpattern, it was treating the data ']' as terminating
861        the class, so got the count wrong. When actually compiling, the reference
862        was subsequently set up correctly.)
864    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
865        it was being rejected as not supported by pcre_dfa_exec(), even though
866        other assertions are supported. I have made pcre_dfa_exec() support
867        (*FAIL).
869    16. The implementation of 13c above involved the invention of a new opcode,
870        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
871        cannot be changed at match time, I realized I could make a small
872        improvement to matching performance by compiling OP_ALLANY instead of
873        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
874        on the OP_ANY path.
876    17. Compiling pcretest on Windows with readline support failed without the
877        following two fixes: (1) Make the unistd.h include conditional on
878        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
880    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
881        ncurses library to be included for pcretest when ReadLine support is
882        requested, but also to allow for it to be overridden. This patch came from
883        Daniel Bergström.
885    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
886        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
887        any errors with the current Unicode tables. Thanks to Peter Kankowski for
888        spotting this.
891    Version 7.6 28-Jan-08
892    ---------------------
894    1.  A character class containing a very large number of characters with
895        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
896        overflow.
898    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
899        HAVE_LONG_LONG is not defined.
901    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
902        bring them up to date and include new features. This patch includes:
904        - Fixed PH's badly added libz and libbz2 support.
905        - Fixed a problem with static linking.
906        - Added pcredemo. [But later removed - see 7 below.]
907        - Fixed dftables problem and added an option.
908        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
909            HAVE_LONG_LONG.
910        - Added readline support for pcretest.
911        - Added an listing of the option settings after cmake has run.
913    4.  A user submitted a patch to Makefile that makes it easy to create
914        "pcre.dll" under mingw when using Configure/Make. I added stuff to
915        Makefile.am that cause it to include this special target, without
916        affecting anything else. Note that the same mingw target plus all
917        the other distribution libraries and programs are now supported
918        when configuring with CMake (see 6 below) instead of with
919        Configure/Make.
921    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
922        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
923        exported in the Windows port". It has not yet been confirmed that the patch
924        solves the problem, but it does no harm.
926    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
927        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
928        with CMake, and also correct the comment about stack recursion.
930    7.  Remove the automatic building of pcredemo from the ./configure system and
931        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
932        of a program that users should build themselves after PCRE is installed, so
933        building it automatically is not really right. What is more, it gave
934        trouble in some build environments.
936    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
939    Version 7.5 10-Jan-08
940    ---------------------
942    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
943        values in parens when parsing an RE using the C++ wrapper."
945    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
946        Characters greater than 255 were excluded from the class instead of being
947        included.
949    3.  The same bug as (2) above applied to negated POSIX classes such as
950        [:^space:].
952    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
953        defined or documented. It seems to have been a typo for PCRE_STATIC, so
954        I have changed it.
956    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
957        first named subpattern) and a construct such as (?&a) would reference the
958        first named subpattern whose name started with "a" (in other words, the
959        length check was missing). Both these problems are fixed. "Subpattern name
960        expected" is now given for (?&) (a zero-length name), and this patch also
961        makes it give the same error for \k'' (previously it complained that that
962        was a reference to a non-existent subpattern).
964    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
965        this is right because (?- can be followed by option settings as well as by
966        digits. I have, however, made the messages clearer.
968    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
969        than the number used in the conditional) now cause a compile-time error.
970        This is actually not compatible with Perl, which accepts such patterns, but
971        treats the conditional as always being FALSE (as PCRE used to), but it
972        seems to me that giving a diagnostic is better.
974    8.  Change "alphameric" to the more common word "alphanumeric" in comments
975        and messages.
977    9.  Fix two occurrences of "backslash" in comments that should have been
978        "backspace".
980    10. Remove two redundant lines of code that can never be obeyed (their function
981        was moved elsewhere).
983    11. The program that makes PCRE's Unicode character property table had a bug
984        which caused it to generate incorrect table entries for sequences of
985        characters that have the same character type, but are in different scripts.
986        It amalgamated them into a single range, with the script of the first of
987        them. In other words, some characters were in the wrong script. There were
988        thirteen such cases, affecting characters in the following ranges:
990          U+002b0 - U+002c1
991          U+0060c - U+0060d
992          U+0061e - U+00612
993          U+0064b - U+0065e
994          U+0074d - U+0076d
995          U+01800 - U+01805
996          U+01d00 - U+01d77
997          U+01d9b - U+01dbf
998          U+0200b - U+0200f
999          U+030fc - U+030fe
1000          U+03260 - U+0327f
1001          U+0fb46 - U+0fbb1
1002          U+10450 - U+1049d
1004    12. The -o option (show only the matching part of a line) for pcregrep was not
1005        compatible with GNU grep in that, if there was more than one match in a
1006        line, it showed only the first of them. It now behaves in the same way as
1007        GNU grep.
1009    13. If the -o and -v options were combined for pcregrep, it printed a blank
1010        line for every non-matching line. GNU grep prints nothing, and pcregrep now
1011        does the same. The return code can be used to tell if there were any
1012        non-matching lines.
1014    14. Added --file-offsets and --line-offsets to pcregrep.
1016    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
1017        infinitely looping recursion. The bug was that positive lookaheads were not
1018        being skipped when checking for a possible empty match (negative lookaheads
1019        and both kinds of lookbehind were skipped).
1021    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
1022        inclusion of <windows.h> to before rather than after the definition of
1023        INVALID_FILE_ATTRIBUTES (patch from David Byron).
1025    17. Specifying a possessive quantifier with a specific limit for a Unicode
1026        character property caused pcre_compile() to compile bad code, which led at
1027        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
1028        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
1029        caused the error; without that there was no problem.
1031    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
1033    19. Added --enable-pcretest-libreadline.
1035    20. In pcrecpp.cc, the variable 'count' was incremented twice in
1036        RE::GlobalReplace(). As a result, the number of replacements returned was
1037        double what it should be. I removed one of the increments, but Craig sent a
1038        later patch that removed the other one (the right fix) and added unit tests
1039        that check the return values (which was not done before).
1041    21. Several CMake things:
1043        (1) Arranged that, when cmake is used on Unix, the libraries end up with
1044            the names libpcre and libpcreposix, not just pcre and pcreposix.
1046        (2) The above change means that pcretest and pcregrep are now correctly
1047            linked with the newly-built libraries, not previously installed ones.
1051    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
1052        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
1053        UTF-8 newline character). The key issue is that the pattern starts .*;
1054        this means that the match must be either at the beginning, or after a
1055        newline. The bug was in the code for advancing after a failed match and
1056        checking that the new position followed a newline. It was not taking
1057        account of UTF-8 characters correctly.
1059    23. PCRE was behaving differently from Perl in the way it recognized POSIX
1060        character classes. PCRE was not treating the sequence [:...:] as a
1061        character class unless the ... were all letters. Perl, however, seems to
1062        allow any characters between [: and :], though of course it rejects as
1063        unknown any "names" that contain non-letters, because all the known class
1064        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
1065        for example, whereas PCRE did not - it did not recognize a POSIX character
1066        class. This seemed a bit dangerous, so the code has been changed to be
1067        closer to Perl. The behaviour is not identical to Perl, because PCRE will
1068        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
1069        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
1070        Perl does, and where it didn't before.
1072    24. Rewrite so as to remove the single use of %n from pcregrep because in some
1073        Windows environments %n is disabled by default.
1076    Version 7.4 21-Sep-07
1077    ---------------------
1079    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
1080        means that a class such as [\s] counted as "explicit reference to CR or
1081        LF". That isn't really right - the whole point of the change was to try to
1082        help when there was an actual mention of one of the two characters. So now
1083        the change happens only if \r or \n (or a literal CR or LF) character is
1084        encountered.
1086    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
1087        of both had grown to the point where there were only 3 bits left.
1088        Fortunately, there was spare space in the data structure, and so I have
1089        moved the internal flags into a new 16-bit field to free up more option
1090        bits.
1092    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
1093        but did not set the internal JCHANGED flag - either of these is enough to
1094        control the way the "get" function works - but the PCRE_INFO_JCHANGED
1095        facility is supposed to tell if (?J) was ever used, so now (?J) at the
1096        start sets both bits.
1098    4.  Added options (at build time, compile time, exec time) to change \R from
1099        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
1101    5.  doc/pcresyntax.html was missing from the distribution.
1103    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
1104        compatibility, even though it is no longer used.
1106    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
1107        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
1108        windows.h file is present (where different names are used). [This was
1109        reversed later after testing - see 16 below.]
1111    8.  Changed all #include <config.h> to #include "config.h". There were also
1112        some further <pcre.h> cases that I changed to "pcre.h".
1114    9.  When pcregrep was used with the --colour option, it missed the line ending
1115        sequence off the lines that it output.
1117    10. It was pointed out to me that arrays of string pointers cause lots of
1118        relocations when a shared library is dynamically loaded. A technique of
1119        using a single long string with a table of offsets can drastically reduce
1120        these. I have refactored PCRE in four places to do this. The result is
1121        dramatic:
1123          Originally:                          290
1124          After changing UCP table:            187
1125          After changing error message table:   43
1126          After changing table of "verbs"       36
1127          After changing table of Posix names   22
1129        Thanks to the folks working on Gregex for glib for this insight.
1131    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
1132        unicode-properties was also set.
1134    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
1136    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
1137        checked only for CRLF.
1139    14. Added casts to pcretest.c to avoid compiler warnings.
1141    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
1143    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
1144        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
1145        entirely. This removes changes made in 7 above.
1147    17. The CMake files have been updated, and there is now more information about
1148        building with CMake in the NON-UNIX-USE document.
1151    Version 7.3 28-Aug-07
1152    ---------------------
1154     1. In the rejigging of the build system that eventually resulted in 7.1, the
1155        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
1156        brackets there is not right, since it causes compilers to look for an
1157        installed pcre.h, not the version that is in the source that is being
1158        compiled (which of course may be different). I have changed it back to:
1160          #include "pcre.h"
1162        I have a vague recollection that the change was concerned with compiling in
1163        different directories, but in the new build system, that is taken care of
1164        by the VPATH setting the Makefile.
1166     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
1167        when the subject happened to end in the byte 0x85 (e.g. if the last
1168        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
1169        characters but of course it shouldn't be taken as a newline when it is part
1170        of another character. The bug was that, for an unlimited repeat of . in
1171        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
1172        characters when looking for a newline.
1174     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
1176     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
1177        in debug output.
1179     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
1180        long printing in the pcrecpp unittest when running under MinGW.
1182     6. ESC_K was left out of the EBCDIC table.
1184     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
1185        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
1186        limit also applies to "virtual nesting" when a pattern is recursive, and in
1187        this case 1000 isn't so big. I have been able to remove this limit at the
1188        expense of backing off one optimization in certain circumstances. Normally,
1189        when pcre_exec() would call its internal match() function recursively and
1190        immediately return the result unconditionally, it uses a "tail recursion"
1191        feature to save stack. However, when a subpattern that can match an empty
1192        string has an unlimited repetition quantifier, it no longer makes this
1193        optimization. That gives it a stack frame in which to save the data for
1194        checking that an empty string has been matched. Previously this was taken
1195        from the 1000-entry workspace that had been reserved. So now there is no
1196        explicit limit, but more stack is used.
1198     8. Applied Daniel's patches to solve problems with the import/export magic
1199        syntax that is required for Windows, and which was going wrong for the
1200        pcreposix and pcrecpp parts of the library. These were overlooked when this
1201        problem was solved for the main library.
1203     9. There were some crude static tests to avoid integer overflow when computing
1204        the size of patterns that contain repeated groups with explicit upper
1205        limits. As the maximum quantifier is 65535, the maximum group length was
1206        set at 30,000 so that the product of these two numbers did not overflow a
1207        32-bit integer. However, it turns out that people want to use groups that
1208        are longer than 30,000 bytes (though not repeat them that many times).
1209        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1210        made it possible to implement the integer overflow checks in a much more
1211        dynamic way, which I have now done. The artificial limitation on group
1212        length has been removed - we now have only the limit on the total length of
1213        the compiled pattern, which depends on the LINK_SIZE setting.
1215    10. Fixed a bug in the documentation for get/copy named substring when
1216        duplicate names are permitted. If none of the named substrings are set, the
1217        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1218        empty string.
1220    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1221        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1222        because the ] is interpreted as the first data character and the
1223        terminating ] is not found. PCRE has been made compatible with Perl in this
1224        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1225        cause memory overwriting.
1227    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1228        string has been matched (to stop an infinite loop). It was not recognizing
1229        a conditional subpattern that could match an empty string if that
1230        subpattern was within another subpattern. For example, it looped when
1231        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
1232        condition was not nested. This bug has been fixed.
1234    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1235        past the start of the subject in the presence of bytes with the top bit
1236        set, for example "\x8aBCD".
1238    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1239        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1241    14. Optimized (?!) to (*FAIL).
1243    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1244        This restricts code points to be within the range 0 to 0x10FFFF, excluding
1245        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1246        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1247        does: it's just the validity check that is more restrictive.
1249    16. Inserted checks for integer overflows during escape sequence (backslash)
1250        processing, and also fixed erroneous offset values for syntax errors during
1251        backslash processing.
1253    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1254        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1256    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1257        caused an overrun.
1259    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1260        something other than just ASCII characters) inside a group that had an
1261        unlimited repeat caused a loop at compile time (while checking to see
1262        whether the group could match an empty string).
1264    20. Debugging a pattern containing \p or \P could cause a crash. For example,
1265        [\P{Any}] did so. (Error in the code for printing property names.)
1267    21. An orphan \E inside a character class could cause a crash.
1269    22. A repeated capturing bracket such as (A)? could cause a wild memory
1270        reference during compilation.
1272    23. There are several functions in pcre_compile() that scan along a compiled
1273        expression for various reasons (e.g. to see if it's fixed length for look
1274        behind). There were bugs in these functions when a repeated \p or \P was
1275        present in the pattern. These operators have additional parameters compared
1276        with \d, etc, and these were not being taken into account when moving along
1277        the compiled data. Specifically:
1279        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1280            length.
1282        (b) An item such as \pL+ within a repeated group could cause crashes or
1283            loops.
1285        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1286            "reference to non-existent subpattern" error.
1288        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1290    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1291        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1293    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1295    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1296        character were causing crashes (broken optimization).
1298    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1299        \p or \P) caused a compile-time loop.
1301    28. More problems have arisen in unanchored patterns when CRLF is a valid line
1302        break. For example, the unstudied pattern [\r\n]A does not match the string
1303        "\r\nA" because change 7.0/46 below moves the current point on by two
1304        characters after failing to match at the start. However, the pattern \nA
1305        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
1306        the same is true. There doesn't seem any very clean way out of this, but
1307        what I have chosen to do makes the common cases work: PCRE now takes note
1308        of whether there can be an explicit match for \r or \n anywhere in the
1309        pattern, and if so, 7.0/46 no longer applies. As part of this change,
1310        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1311        pattern has explicit CR or LF references.
1313    29. Added (*CR) etc for changing newline setting at start of pattern.
1316    Version 7.2 19-Jun-07
1317    ---------------------
1319     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
1320        which is apparently normally available under Windows.
1322     2. Re-jig the pcregrep tests with different newline settings in an attempt
1323        to make them independent of the local environment's newline setting.
1325     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
1327     4. Some of the "internals" tests were previously cut out when the link size
1328        was not 2, because the output contained actual offsets. The recent new
1329        "Z" feature of pcretest means that these can be cut out, making the tests
1330        usable with all link sizes.
1332     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
1333        stack recursion. This gives a massive performance boost under BSD, but just
1334        a small improvement under Linux. However, it saves one field in the frame
1335        in all cases.
1337     6. Added more features from the forthcoming Perl 5.10:
1339        (a) (?-n) (where n is a string of digits) is a relative subroutine or
1340            recursion call. It refers to the nth most recently opened parentheses.
1342        (b) (?+n) is also a relative subroutine call; it refers to the nth next
1343            to be opened parentheses.
1345        (c) Conditions that refer to capturing parentheses can be specified
1346            relatively, for example, (?(-2)... or (?(+3)...
1348        (d) \K resets the start of the current match so that everything before
1349            is not part of it.
1351        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1353        (f) \g{name} is another synonym - part of Perl 5.10's unification of
1354            reference syntax.
1356        (g) (?| introduces a group in which the numbering of parentheses in each
1357            alternative starts with the same number.
1359        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
1361     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
1364     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
1365        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
1366        for detecting groups that can match an empty string.
1368     9. A pattern with a very large number of alternatives (more than several
1369        hundred) was running out of internal workspace during the pre-compile
1370        phase, where pcre_compile() figures out how much memory will be needed. A
1371        bit of new cunning has reduced the workspace needed for groups with
1372        alternatives. The 1000-alternative test pattern now uses 12 bytes of
1373        workspace instead of running out of the 4096 that are available.
1375    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
1377    11. Applied patch from Google to remove an optimization that didn't quite work.
1378        The report of the bug said:
1380          pcrecpp::RE("a*").FullMatch("aaa") matches, while
1381          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
1382          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
1384    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
1385        it matched the wrong number of bytes.
1388    Version 7.1 24-Apr-07
1389    ---------------------
1391     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
1392        that is more "standard", making use of automake and other Autotools. There
1393        is some re-arrangement of the files and adjustment of comments consequent
1394        on this.
1396     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
1397        for recursive directory scanning broke on some systems because the files
1398        are not scanned in any specific order and on different systems the order
1399        was different. A call to "sort" has been inserted into RunGrepTest for the
1400        approprate test as a short-term fix. In the longer term there may be an
1401        alternative.
1403     3. I had an email from Eric Raymond about problems translating some of PCRE's
1404        man pages to HTML (despite the fact that I distribute HTML pages, some
1405        people do their own conversions for various reasons). The problems
1406        concerned the use of low-level troff macros .br and .in. I have therefore
1407        removed all such uses from the man pages (some were redundant, some could
1408        be replaced by .nf/.fi pairs). The 132html script that I use to generate
1409        HTML has been updated to handle .nf/.fi and to complain if it encounters
1410        .br or .in.
1412     4. Updated comments in configure.ac that get placed in config.h.in and also
1413        arranged for config.h to be included in the distribution, with the name
1414        config.h.generic, for the benefit of those who have to compile without
1415        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
1417     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
1418        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
1419        makevp.bat for the current PCRE, using the additional files
1420        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
1422     6. A Windows user reported a minor discrepancy with test 2, which turned out
1423        to be caused by a trailing space on an input line that had got lost in his
1424        copy. The trailing space was an accident, so I've just removed it.
1426     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
1427        that is needed.
1429     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
1430        as "const" (a) because they are and (b) because it helps the PHP
1431        maintainers who have recently made a script to detect big data structures
1432        in the php code that should be moved to the .rodata section. I remembered
1433        to update Builducptable as well, so it won't revert if ucptable.h is ever
1434        re-created.
1436     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
1437        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
1438        order to be able to cut out the UTF-8 tables in the latter when UTF-8
1439        support is not required. This saves 1.5-2K of code, which is important in
1440        some applications.
1442        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
1443        so as not to refer to the tables, even though these functions will never be
1444        called when UTF-8 support is disabled. Otherwise there are problems with a
1445        shared library.
1447    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
1449        (a) It was defining its arguments as char * instead of void *.
1451        (b) It was assuming that all moves were upwards in memory; this was true
1452            a long time ago when I wrote it, but is no longer the case.
1454        The emulated memove() is provided for those environments that have neither
1455        memmove() nor bcopy(). I didn't think anyone used it these days, but that
1456        is clearly not the case, as these two bugs were recently reported.
1458    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
1459        and Detrail to create the HTML documentation, the .txt form of the man
1460        pages, and it removes trailing spaces from listed files. It also creates
1461        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
1462        case, it wraps all the #defines with #ifndefs. This script should be run
1463        before "make dist".
1465    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1466        with Unicode property support.
1468        (a) For a maximizing quantifier, if the two different cases of the
1469            character were of different lengths in their UTF-8 codings (there are
1470            some cases like this - I found 11), and the matching function had to
1471            back up over a mixture of the two cases, it incorrectly assumed they
1472            were both the same length.
1474        (b) When PCRE was configured to use the heap rather than the stack for
1475            recursion during matching, it was not correctly preserving the data for
1476            the other case of a UTF-8 character when checking ahead for a match
1477            while processing a minimizing repeat. If the check also involved
1478            matching a wide character, but failed, corruption could cause an
1479            erroneous result when trying to check for a repeat of the original
1480            character.
1482    13. Some tidying changes to the testing mechanism:
1484        (a) The RunTest script now detects the internal link size and whether there
1485            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1486            values substituted by "configure". (The RunGrepTest script already did
1487            this for UTF-8.) The configure.ac script no longer substitutes the
1488            relevant variables.
1490        (b) The debugging options /B and /D in pcretest show the compiled bytecode
1491            with length and offset values. This means that the output is different
1492            for different internal link sizes. Test 2 is skipped for link sizes
1493            other than 2 because of this, bypassing the problem. Unfortunately,
1494            there was also a test in test 3 (the locale tests) that used /B and
1495            failed for link sizes other than 2. Rather than cut the whole test out,
1496            I have added a new /Z option to pcretest that replaces the length and
1497            offset values with spaces. This is now used to make test 3 independent
1498            of link size. (Test 2 will be tidied up later.)
1500    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1501        segmentation fault instead of returning the appropriate error message.
1503    15. In multiline mode when the newline sequence was set to "any", the pattern
1504        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1505        This doesn't seem right; it now treats the CRLF combination as the line
1506        ending, and so does not match in that case. It's only a pattern such as ^$
1507        that would hit this one: something like ^ABC$ would have failed after \r
1508        and then tried again after \r\n.
1510    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1511        in an attempt to make files that differ only in their line terminators
1512        compare equal. This works on Linux.
1514    17. Under certain error circumstances pcregrep might try to free random memory
1515        as it exited. This is now fixed, thanks to valgrind.
1517    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1518        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1519        was because its rules for how to advance for /g after matching an empty
1520        string at the end of a line did not allow for this case. They now check for
1521        it specially.
1523    20. pcretest is supposed to handle patterns and data of any length, by
1524        extending its buffers when necessary. It was getting this wrong when the
1525        buffer for a data line had to be extended.
1527    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1528        CRLF as a newline sequence.
1530    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1531        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1532        I have nevertheless tidied it up.
1534    23. Added some casts to kill warnings from HP-UX ia64 compiler.
1536    24. Added a man page for pcre-config.
1539    Version 7.0 19-Dec-06
1540    ---------------------
1542     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1543        moving to gcc 4.1.1.
1545     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1546        sys/time.h, which is documented as needed for this function. It doesn't
1547        seem to matter on Linux, but it showed up on some releases of OS X.
1549     3. It seems that there are systems where bytes whose values are greater than
1550        127 match isprint() in the "C" locale. The "C" locale should be the
1551        default when a C program starts up. In most systems, only ASCII printing
1552        characters match isprint(). This difference caused the output from pcretest
1553        to vary, making some of the tests fail. I have changed pcretest so that:
1555        (a) When it is outputting text in the compiled version of a pattern, bytes
1556            other than 32-126 are always shown as hex escapes.
1558        (b) When it is outputting text that is a matched part of a subject string,
1559            it does the same, unless a different locale has been set for the match
1560            (using the /L modifier). In this case, it uses isprint() to decide.
1562     4. Fixed a major bug that caused incorrect computation of the amount of memory
1563        required for a compiled pattern when options that changed within the
1564        pattern affected the logic of the preliminary scan that determines the
1565        length. The relevant options are -x, and -i in UTF-8 mode. The result was
1566        that the computed length was too small. The symptoms of this bug were
1567        either the PCRE error "internal error: code overflow" from pcre_compile(),
1568        or a glibc crash with a message such as "pcretest: free(): invalid next
1569        size (fast)". Examples of patterns that provoked this bug (shown in
1570        pcretest format) are:
1572          /(?-x: )/x
1573          /(?x)(?-x: \s*#\s*)/
1574          /((?i)[\x{c0}])/8
1575          /(?i:[\x{c0}])/8
1577        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1578        is now done differently.
1580     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1581        wrapper classes; (b) implement a new function in the C++ scanner that is
1582        more efficient than the old way of doing things because it avoids levels of
1583        recursion in the regex matching; (c) add a paragraph to the documentation
1584        for the FullMatch() function.
1586     6. The escape sequence \n was being treated as whatever was defined as
1587        "newline". Not only was this contrary to the documentation, which states
1588        that \n is character 10 (hex 0A), but it also went horribly wrong when
1589        "newline" was defined as CRLF. This has been fixed.
1591     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1592        was being set to -1 for the "end of line" case (supposedly a value that no
1593        character can have). Though this value is never used (the check for end of
1594        line is "zero bytes in current character"), it caused compiler complaints.
1595        I've changed it to 0xffffffff.
1597     8. In pcre_version.c, the version string was being built by a sequence of
1598        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1599        string (as it is for production releases) called a macro with an empty
1600        argument. The C standard says the result of this is undefined. The gcc
1601        compiler treats it as an empty string (which was what was wanted) but it is
1602        reported that Visual C gives an error. The source has been hacked around to
1603        avoid this problem.
1605     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1606        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1607        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1608        of them did).
1610    10. Originally, pcretest opened its input and output without "b"; then I was
1611        told that "b" was needed in some environments, so it was added for release
1612        5.0 to both the input and output. (It makes no difference on Unix-like
1613        systems.) Later I was told that it is wrong for the input on Windows. I've
1614        now abstracted the modes into two macros, to make it easier to fiddle with
1615        them, and removed "b" from the input mode under Windows.
1617    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1619    12. Added -help and --help to pcretest as an official way of being reminded
1620        of the options.
1622    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1623        and pcrecpp.cc because they annoy compilers at high warning levels.
1625    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1627    15. Fixed an occurrence of == in configure.ac that should have been = (shell
1628        scripts are not C programs :-) and which was not noticed because it works
1629        on Linux.
1631    16. pcretest is supposed to handle any length of pattern and data line (as one
1632        line or as a continued sequence of lines) by extending its input buffer if
1633        necessary. This feature was broken for very long pattern lines, leading to
1634        a string of junk being passed to pcre_compile() if the pattern was longer
1635        than about 50K.
1637    17. I have done a major re-factoring of the way pcre_compile() computes the
1638        amount of memory needed for a compiled pattern. Previously, there was code
1639        that made a preliminary scan of the pattern in order to do this. That was
1640        OK when PCRE was new, but as the facilities have expanded, it has become
1641        harder and harder to keep it in step with the real compile phase, and there
1642        have been a number of bugs (see for example, 4 above). I have now found a
1643        cunning way of running the real compile function in a "fake" mode that
1644        enables it to compute how much memory it would need, while actually only
1645        ever using a few hundred bytes of working memory and without too many
1646        tests of the mode. This should make future maintenance and development
1647        easier. A side effect of this work is that the limit of 200 on the nesting
1648        depth of parentheses has been removed (though this was never a serious
1649        limitation, I suspect). However, there is a downside: pcre_compile() now
1650        runs more slowly than before (30% or more, depending on the pattern). I
1651        hope this isn't a big issue. There is no effect on runtime performance.
1653    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1654        newline (only possible for the last line of a file) and it was a
1655        pattern that set a locale (followed by /Lsomething), pcretest crashed.
1657    19. Added additional timing features to pcretest. (1) The -tm option now times
1658        matching only, not compiling. (2) Both -t and -tm can be followed, as a
1659        separate command line item, by a number that specifies the number of
1660        repeats to use when timing. The default is 50000; this gives better
1661        precision, but takes uncomfortably long for very large patterns.
1663    20. Extended pcre_study() to be more clever in cases where a branch of a
1664        subpattern has no definite first character. For example, (a*|b*)[cd] would
1665        previously give no result from pcre_study(). Now it recognizes that the
1666        first character must be a, b, c, or d.
1668    21. There was an incorrect error "recursive call could loop indefinitely" if
1669        a subpattern (or the entire pattern) that was being tested for matching an
1670        empty string contained only one non-empty item after a nested subpattern.
1671        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1672        incorrectly, because the \d was being skipped in the check.
1674    22. The pcretest program now has a new pattern option /B and a command line
1675        option -b, which is equivalent to adding /B to every pattern. This causes
1676        it to show the compiled bytecode, without the additional information that
1677        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1678        is the same as /B/I).
1680    23. A new optimization is now able automatically to treat some sequences such
1681        as a*b as a*+b. More specifically, if something simple (such as a character
1682        or a simple class like \d) has an unlimited quantifier, and is followed by
1683        something that cannot possibly match the quantified thing, the quantifier
1684        is automatically "possessified".
1686    24. A recursive reference to a subpattern whose number was greater than 39
1687        went wrong under certain circumstances in UTF-8 mode. This bug could also
1688        have affected the operation of pcre_study().
1690    25. Realized that a little bit of performance could be had by replacing
1691        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1693    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1695    27. Possessive quantifiers such as a++ were previously implemented by turning
1696        them into atomic groups such as ($>a+). Now they have their own opcodes,
1697        which improves performance. This includes the automatically created ones
1698        from 23 above.
1700    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1701        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1702        the first matched character to be a colon. This applied both to named and
1703        numbered groups.
1705    29. The ucpinternal.h header file was missing its idempotency #ifdef.
1707    30. I was sent a "project" file called libpcre.a.dev which I understand makes
1708        building PCRE on Windows easier, so I have included it in the distribution.
1710    31. There is now a check in pcretest against a ridiculously large number being
1711        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1712        loop, the loop is abandoned.
1714    32. Forward references to subpatterns in conditions such as (?(2)...) where
1715        subpattern 2 is defined later cause pcre_compile() to search forwards in
1716        the pattern for the relevant set of parentheses. This search went wrong
1717        when there were unescaped parentheses in a character class, parentheses
1718        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1720    33. "Subroutine" calls and backreferences were previously restricted to
1721        referencing subpatterns earlier in the regex. This restriction has now
1722        been removed.
1724    34. Added a number of extra features that are going to be in Perl 5.10. On the
1725        whole, these are just syntactic alternatives for features that PCRE had
1726        previously implemented using the Python syntax or my own invention. The
1727        other formats are all retained for compatibility.
1729        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1730            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1731            also .NET compatible.
1733        (b) A recursion or subroutine call to a named group can now be defined as
1734            (?&name) as well as (?P>name).
1736        (c) A backreference to a named group can now be defined as \k<name> or
1737            \k'name' as well as (?P=name). The new forms, as well as being in Perl
1738            5.10, are also .NET compatible.
1740        (d) A conditional reference to a named group can now use the syntax
1741            (?(<name>) or (?('name') as well as (?(name).
1743        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1744            groups (named and numbered) that are never evaluated inline, but can be
1745            called as "subroutines" from elsewhere. In effect, the DEFINE condition
1746            is always false. There may be only one alternative in such a group.
1748        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1749            as the simple (?(R). The condition is true only if the most recent
1750            recursion is that of the given number or name. It does not search out
1751            through the entire recursion stack.
1753        (g) The escape \gN or \g{N} has been added, where N is a positive or
1754            negative number, specifying an absolute or relative reference.
1756    35. Tidied to get rid of some further signed/unsigned compiler warnings and
1757        some "unreachable code" warnings.
1759    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1760        things, this adds five new scripts.
1762    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1763        There were also incompatibilities regarding the handling of \Q..\E inside
1764        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1765        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1767    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1768        matches an empty string, and forcibly breaks the loop. There were bugs in
1769        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
1770        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
1771        separate and independent bugs (that affected different cases) have been
1772        fixed.
1774    39. Refactored the code to abolish the use of different opcodes for small
1775        capturing bracket numbers. This is a tidy that I avoided doing when I
1776        removed the limit on the number of capturing brackets for 3.5 back in 2001.
1777        The new approach is not only tidier, it makes it possible to reduce the
1778        memory needed to fix the previous bug (38).
1780    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1781        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1782        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1783        mode.
1785    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1786        report.
1788    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1789        copy construction and assignment in the C++ wrapper.
1791    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1792        couple of bugs that could have given wrong results in the "--newline=crlf"
1793        case.
1795    44. Added a number of casts and did some reorganization of signed/unsigned int
1796        variables following suggestions from Dair Grant. Also renamed the variable
1797        "this" as "item" because it is a C++ keyword.
1799    45. Arranged for dftables to add
1801          #include "pcre_internal.h"
1803        to pcre_chartables.c because without it, gcc 4.x may remove the array
1804        definition from the final binary if PCRE is built into a static library and
1805        dead code stripping is activated.
1807    46. For an unanchored pattern, if a match attempt fails at the start of a
1808        newline sequence, and the newline setting is CRLF or ANY, and the next two
1809        characters are CRLF, advance by two characters instead of one.
1812    Version 6.7 04-Jul-06
1813    ---------------------
1815     1. In order to handle tests when input lines are enormously long, pcretest has
1816        been re-factored so that it automatically extends its buffers when
1817        necessary. The code is crude, but this _is_ just a test program. The
1818        default size has been increased from 32K to 50K.
1820     2. The code in pcre_study() was using the value of the re argument before
1821        testing it for NULL. (Of course, in any sensible call of the function, it
1822        won't be NULL.)
1824     3. The memmove() emulation function in pcre_internal.h, which is used on
1825        systems that lack both memmove() and bcopy() - that is, hardly ever -
1826        was missing a "static" storage class specifier.
1828     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1829        containing an extended class (one that cannot be represented by a bitmap
1830        because it contains high-valued characters or Unicode property items, e.g.
1831        [\pZ]). Almost always one would set UTF-8 mode when processing such a
1832        pattern, but PCRE should not loop if you do not (it no longer does).
1833        [Detail: two cases were found: (a) a repeated subpattern containing an
1834        extended class; (b) a recursive reference to a subpattern that followed a
1835        previous extended class. It wasn't skipping over the extended class
1836        correctly when UTF-8 mode was not set.]
1838     5. A negated single-character class was not being recognized as fixed-length
1839        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1840        compile error "lookbehind assertion is not fixed length".
1842     6. The RunPerlTest auxiliary script was showing an unexpected difference
1843        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1844        write a Perl script that can interpret lines of an input file either as
1845        byte characters or as UTF-8, which is what "perltest" was being required to
1846        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1847        can't do is switch easily at run time between having the "use utf8;" pragma
1848        or not. In the end, I fudged it by using the RunPerlTest script to insert
1849        "use utf8;" explicitly for the UTF-8 tests.
1851     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1852        the end of the subject string, contrary to the documentation and to what
1853        Perl does. This was true of both matching functions. Now it matches only at
1854        the start of the subject and immediately after *internal* newlines.
1856     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1857        a pointer to an int instead of a pointer to an unsigned long int. This
1858        caused problems on 64-bit systems.
1860     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1861        instance of the 'standard' template library not being so standard".
1863    10. There was no check on the number of named subpatterns nor the maximum
1864        length of a subpattern name. The product of these values is used to compute
1865        the size of the memory block for a compiled pattern. By supplying a very
1866        long subpattern name and a large number of named subpatterns, the size
1867        computation could be caused to overflow. This is now prevented by limiting
1868        the length of names to 32 characters, and the number of named subpatterns
1869        to 10,000.
1871    11. Subpatterns that are repeated with specific counts have to be replicated in
1872        the compiled pattern. The size of memory for this was computed from the
1873        length of the subpattern and the repeat count. The latter is limited to
1874        65535, but there was no limit on the former, meaning that integer overflow
1875        could in principle occur. The compiled length of a repeated subpattern is
1876        now limited to 30,000 bytes in order to prevent this.
1878    12. Added the optional facility to have named substrings with the same name.
1880    13. Added the ability to use a named substring as a condition, using the
1881        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1882        are numbers (not recommended). Forward references are permitted.
1884    14. Added forward references in named backreferences (if you see what I mean).
1886    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1887        pattern could run off the end of the subject. For example, the pattern
1888        "(?s)(.{1,5})"8 did this with the subject "ab".
1890    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1891        PCRE_CASELESS was set when matching characters that were quantified with ?
1892        or *.
1894    17. A character class other than a single negated character that had a minimum
1895        but no maximum quantifier - for example [ab]{6,} - was not handled
1896        correctly by pce_dfa_exec(). It would match only one character.
1898    18. A valid (though odd) pattern that looked like a POSIX character
1899        class but used an invalid character after [ (for example [[,abc,]]) caused
1900        pcre_compile() to give the error "Failed: internal error: code overflow" or
1901        in some cases to crash with a glibc free() error. This could even happen if
1902        the pattern terminated after [[ but there just happened to be a sequence of
1903        letters, a binary zero, and a closing ] in the memory that followed.
1905    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1906        over the years. Originally (before any Unicode support), just the bottom 8
1907        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1908        output from "man perlunicode" includes this:
1910          The regular expression compiler produces polymorphic opcodes.  That
1911          is, the pattern adapts to the data and automatically switches to
1912          the Unicode character scheme when presented with Unicode data--or
1913          instead uses a traditional byte scheme when presented with byte
1914          data.
1916        Sadly, a wide octal escape does not cause a switch, and in a string with
1917        no other multibyte characters, these octal escapes are treated as before.
1918        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
1919        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1920        Unicode string.
1922        I have not perpetrated such confusion in PCRE. Up till now, it took just
1923        the bottom 8 bits, as in old Perl. I have now made octal escapes with
1924        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1925        translate to the appropriate multibyte character.
1927    29. Applied some refactoring to reduce the number of warnings from Microsoft
1928        and Borland compilers. This has included removing the fudge introduced
1929        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1930        a warning about an unused variable.
1932    21. PCRE has not included VT (character 0x0b) in the set of whitespace
1933        characters since release 4.0, because Perl (from release 5.004) does not.
1934        [Or at least, is documented not to: some releases seem to be in conflict
1935        with the documentation.] However, when a pattern was studied with
1936        pcre_study() and all its branches started with \s, PCRE still included VT
1937        as a possible starting character. Of course, this did no harm; it just
1938        caused an unnecessary match attempt.
1940    22. Removed a now-redundant internal flag bit that recorded the fact that case
1941        dependency changed within the pattern. This was once needed for "required
1942        byte" processing, but is no longer used. This recovers a now-scarce options
1943        bit. Also moved the least significant internal flag bit to the most-
1944        significant bit of the word, which was not previously used (hangover from
1945        the days when it was an int rather than a uint) to free up another bit for
1946        the future.
1948    23. Added support for CRLF line endings as well as CR and LF. As well as the
1949        default being selectable at build time, it can now be changed at runtime
1950        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1951        specify that it is scanning data with non-default line endings.
1953    24. Changed the definition of CXXLINK to make it agree with the definition of
1954        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1956    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1957        recursions. This makes a big different to stack usage for some patterns.
1959    26. If a subpattern containing a named recursion or subroutine reference such
1960        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1961        the space required for the compiled pattern went wrong and gave too small a
1962        value. Depending on the environment, this could lead to "Failed: internal
1963        error: code overflow at offset 49" or "glibc detected double free or
1964        corruption" errors.
1966    27. Applied patches from Google (a) to support the new newline modes and (b) to
1967        advance over multibyte UTF-8 characters in GlobalReplace.
1969    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1970        difference for some implementation of PCRE in some Windows version.
1972    29. Added some extra testing facilities to pcretest:
1974        \q<number>   in a data line sets the "match limit" value
1975        \Q<number>   in a data line sets the "match recursion limt" value
1976        -S <number>  sets the stack size, where <number> is in megabytes
1978        The -S option isn't available for Windows.
1981    Version 6.6 06-Feb-06
1982    ---------------------
1984     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1985        in pcreposix.h. I have copied the definition from pcre.h.
1987     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1988        because pcre.h is no longer a built file.
1990     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1991        not normally included in the compiled code.
1994    Version 6.5 01-Feb-06
1995    ---------------------
1997     1. When using the partial match feature with pcre_dfa_exec(), it was not
1998        anchoring the second and subsequent partial matches at the new starting
1999        point. This could lead to incorrect results. For example, with the pattern
2000        /1234/, partially matching against "123" and then "a4" gave a match.
2002     2. Changes to pcregrep:
2004        (a) All non-match returns from pcre_exec() were being treated as failures
2005            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
2006            error message is output. Some extra information is given for the
2007            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
2008            probably the only errors that are likely to be caused by users (by
2009            specifying a regex that has nested indefinite repeats, for instance).
2010            If there are more than 20 of these errors, pcregrep is abandoned.
2012        (b) A binary zero was treated as data while matching, but terminated the
2013            output line if it was written out. This has been fixed: binary zeroes
2014            are now no different to any other data bytes.
2016        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
2017            used to set a locale for matching. The --locale=xxxx long option has
2018            been added (no short equivalent) to specify a locale explicitly on the
2019            pcregrep command, overriding the environment variables.
2021        (d) When -B was used with -n, some line numbers in the output were one less
2022            than they should have been.
2024        (e) Added the -o (--only-matching) option.
2026        (f) If -A or -C was used with -c (count only), some lines of context were
2027            accidentally printed for the final match.
2029        (g) Added the -H (--with-filename) option.
2031        (h) The combination of options -rh failed to suppress file names for files
2032            that were found from directory arguments.
2034        (i) Added the -D (--devices) and -d (--directories) options.
2036        (j) Added the -F (--fixed-strings) option.
2038        (k) Allow "-" to be used as a file name for -f as well as for a data file.
2040        (l) Added the --colo(u)r option.
2042        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
2043            is not present by default.
2045     3. A nasty bug was discovered in the handling of recursive patterns, that is,
2046        items such as (?R) or (?1), when the recursion could match a number of
2047        alternatives. If it matched one of the alternatives, but subsequently,
2048        outside the recursion, there was a failure, the code tried to back up into
2049        the recursion. However, because of the way PCRE is implemented, this is not
2050        possible, and the result was an incorrect result from the match.
2052        In order to prevent this happening, the specification of recursion has
2053        been changed so that all such subpatterns are automatically treated as
2054        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
2056     4. I had overlooked the fact that, in some locales, there are characters for
2057        which isalpha() is true but neither isupper() nor islower() are true. In
2058        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
2059        and ordfeminine) are like this. This affected the treatment of \w and \W
2060        when they appeared in character classes, but not when they appeared outside
2061        a character class. The bit map for "word" characters is now created
2062        separately from the results of isalnum() instead of just taking it from the
2063        upper, lower, and digit maps. (Plus the underscore character, of course.)
2065     5. The above bug also affected the handling of POSIX character classes such as
2066        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
2067        permanent tables. Instead, the bit maps for such a class were previously
2068        created as the appropriate unions of the upper, lower, and digit bitmaps.
2069        Now they are created by subtraction from the [[:word:]] class, which has
2070        its own bitmap.
2072     6. The [[:blank:]] character class matches horizontal, but not vertical space.
2073        It is created by subtracting the vertical space characters (\x09, \x0a,
2074        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
2075        subtraction was done in the overall bitmap for a character class, meaning
2076        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
2077        be recognized. This bug has been fixed.
2079     7. Patches from the folks at Google:
2081          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
2082          real life, but is still worth protecting against".
2084          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
2085          regular expressions".
2087          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
2088          have it.
2090          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
2091          "configure" and the latter not, in order to fix a problem somebody had
2092          with compiling the Arg class on HP-UX.
2094          (e) Improve the error-handling of the C++ wrapper a little bit.
2096          (f) New tests for checking recursion limiting.
2098     8. The pcre_memmove() function, which is used only if the environment does not
2099        have a standard memmove() function (and is therefore rarely compiled),
2100        contained two bugs: (a) use of int instead of size_t, and (b) it was not
2101        returning a result (though PCRE never actually uses the result).
2103     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
2104        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
2105        returned instead of calling malloc() with an overflowing number that would
2106        most likely cause subsequent chaos.
2108    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
2110    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
2111        with this option is matched, the nmatch and pmatch options of regexec() are
2112        ignored.
2114    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
2115        provided in case anyone wants to the the POSIX interface with UTF-8
2116        strings.
2118    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
2119        C++ linking (needed for some HP-UX environments).
2121    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
2122        (unused parameter) and in the pcre_printint() function (omitted "default"
2123        switch label when the default is to do nothing).
2125    15. Added some code to make it possible, when PCRE is compiled as a C++
2126        library, to replace subject pointers for pcre_exec() with a smart pointer
2127        class, thus making it possible to process discontinuous strings.
2129    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
2130        much the same function. They were added by different people who were trying
2131        to make PCRE easy to compile on non-Unix systems. It has been suggested
2132        that PCRE_EXPORT be abolished now that there is more automatic apparatus
2133        for compiling on Windows systems. I have therefore replaced it with
2134        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
2135        defaults to "extern" for C or "extern C" for C++, which works fine on
2136        Unix-like systems. It is now possible to override the value of PCRE_DATA_
2137        SCOPE with something explicit in config.h. In addition:
2139        (a) pcreposix.h still had just "extern" instead of either of these macros;
2140            I have replaced it with PCRE_DATA_SCOPE.
2142        (b) Functions such as _pcre_xclass(), which are internal to the library,
2143            but external in the C sense, all had PCRE_EXPORT in their definitions.
2144            This is apparently wrong for the Windows case, so I have removed it.
2145            (It makes no difference on Unix-like systems.)
2147    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
2148        of recursive calls to match(). This is different to MATCH_LIMIT because
2149        that limits the total number of calls to match(), not all of which increase
2150        the depth of recursion. Limiting the recursion depth limits the amount of
2151        stack (or heap if NO_RECURSE is set) that is used. The default can be set
2152        when PCRE is compiled, and changed at run time. A patch from Google adds
2153        this functionality to the C++ interface.
2155    18. Changes to the handling of Unicode character properties:
2157        (a) Updated the table to Unicode 4.1.0.
2159        (b) Recognize characters that are not in the table as "Cn" (undefined).
2161        (c) I revised the way the table is implemented to a much improved format
2162            which includes recognition of ranges. It now supports the ranges that
2163            are defined in UnicodeData.txt, and it also amalgamates other
2164            characters into ranges. This has reduced the number of entries in the
2165            table from around 16,000 to around 3,000, thus reducing its size
2166            considerably. I realized I did not need to use a tree structure after
2167            all - a binary chop search is just as efficient. Having reduced the
2168            number of entries, I extended their size from 6 bytes to 8 bytes to
2169            allow for more data.
2171        (d) Added support for Unicode script names via properties such as \p{Han}.
2173    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
2174        matching that character.
2176    20. When matching a repeated Unicode property with a minimum greater than zero,
2177        (for example \pL{2,}), PCRE could look past the end of the subject if it
2178        reached it while seeking the minimum number of characters. This could
2179        happen only if some of the characters were more than one byte long, because
2180        there is a check for at least the minimum number of bytes.
2182    21. Refactored the implementation of \p and \P so as to be more general, to
2183        allow for more different types of property in future. This has changed the
2184        compiled form incompatibly. Anybody with saved compiled patterns that use
2185        \p or \P will have to recompile them.
2187    22. Added "Any" and "L&" to the supported property types.
2189    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
2190        but give a compile time error if the value is greater than 0xff.
2192    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
2193        accidentally not being installed or uninstalled.
2195    25. The pcre.h file was built from pcre.h.in, but the only changes that were
2196        made were to insert the current release number. This seemed silly, because
2197        it made things harder for people building PCRE on systems that don't run
2198        "configure". I have turned pcre.h into a distributed file, no longer built
2199        by "configure", with the version identification directly included. There is
2200        no longer a pcre.h.in file.
2202        However, this change necessitated a change to the pcre-config script as
2203        well. It is built from pcre-config.in, and one of the substitutions was the
2204        release number. I have updated configure.ac so that ./configure now finds
2205        the release number by grepping pcre.h.
2207    26. Added the ability to run the tests under valgrind.
2210    Version 6.4 05-Sep-05
2211    ---------------------
2213     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
2214        "--" to be printed when multiple files were scanned, even when none of the
2215        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
2216        consider it to be a bug, and have restored the previous behaviour.
2218     2. A couple of code tidies to get rid of compiler warnings.
2220     3. The pcretest program used to cheat by referring to symbols in the library
2221        whose names begin with _pcre_. These are internal symbols that are not
2222        really supposed to be visible externally, and in some environments it is
2223        possible to suppress them. The cheating is now confined to including
2224        certain files from the library's source, which is a bit cleaner.
2226     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
2227        file's purpose clearer.
2229     5. Reorganized pcre_ucp_findchar().
2232    Version 6.3 15-Aug-05
2233    ---------------------
2235     1. The file libpcre.pc.in did not have general read permission in the tarball.
2237     2. There were some problems when building without C++ support:
2239        (a) If C++ support was not built, "make install" and "make test" still
2240            tried to test it.
2242        (b) There were problems when the value of CXX was explicitly set. Some
2243            changes have been made to try to fix these, and ...
2245        (c) --disable-cpp can now be used to explicitly disable C++ support.
2247        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
2248            backslash in a target when C++ was disabled. This confuses some
2249            versions of "make", apparently. Using an intermediate variable solves
2250            this. (Same for CPP_LOBJ.)
2252     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
2253        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
2254        necessary on certain architectures.
2256     4. Added a setting of -export-symbols-regex to the link command to remove
2257        those symbols that are exported in the C sense, but actually are local
2258        within the library, and not documented. Their names all begin with
2259        "_pcre_". This is not a perfect job, because (a) we have to except some
2260        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
2261        available (and never for static libraries). I have made a note to try to
2262        find a way round (a) in the future.
2265    Version 6.2 01-Aug-05
2266    ---------------------
2268     1. There was no test for integer overflow of quantifier values. A construction
2269        such as {1111111111111111} would give undefined results. What is worse, if
2270        a minimum quantifier for a parenthesized subpattern overflowed and became
2271        negative, the calculation of the memory size went wrong. This could have
2272        led to memory overwriting.
2274     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
2276     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
2277        operating environments where this matters.
2279     4. Applied Giuseppe Maxia's patch to add additional features for controlling
2280        PCRE options from within the C++ wrapper.
2282     5. Named capturing subpatterns were not being correctly counted when a pattern
2283        was compiled. This caused two problems: (a) If there were more than 100
2284        such subpatterns, the calculation of the memory needed for the whole
2285        compiled pattern went wrong, leading to an overflow error. (b) Numerical
2286        back references of the form \12, where the number was greater than 9, were
2287        not recognized as back references, even though there were sufficient
2288        previous subpatterns.
2290     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
2291        versions of gcc, e.g. 2.95.4.
2294    Version 6.1 21-Jun-05
2295    ---------------------
2297     1. There was one reference to the variable "posix" in pcretest.c that was not
2298        surrounded by "#if !defined NOPOSIX".
2300     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
2301        the cross-check on the old pcre_info() function, for the benefit of the
2302        cut-down version of PCRE that is currently imported into Exim.
2304     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
2305        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
2306        patterns that start (?i) but I don't think that matters. The use of (?i) is
2307        just an example; this all applies to the other options as well.
2309     4. Since libtool seems to echo the compile commands it is issuing, the output
2310        from "make" can be reduced a bit by putting "@" in front of each libtool
2311        compile command.
2313     5. Patch from the folks at Google for configure.in to be a bit more thorough
2314        in checking for a suitable C++ installation before trying to compile the
2315        C++ stuff. This should fix a reported problem when a compiler was present,
2316        but no suitable headers.
2318     6. The man pages all had just "PCRE" as their title. I have changed them to
2319        be the relevant file name. I have also arranged that these names are
2320        retained in the file doc/pcre.txt, which is a concatenation in text format
2321        of all the man pages except the little individual ones for each function.
2323     7. The NON-UNIX-USE file had not been updated for the different set of source
2324        files that come with release 6. I also added a few comments about the C++
2325        wrapper.
2328    Version 6.0 07-Jun-05
2329    ---------------------
2331     1. Some minor internal re-organization to help with my DFA experiments.
2333     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
2334        didn't matter for the library itself when fully configured, but did matter
2335        when compiling without UCP support, or within Exim, where the ucp files are
2336        not imported.
2338     3. Refactoring of the library code to split up the various functions into
2339        different source modules. The addition of the new DFA matching code (see
2340        below) to a single monolithic source would have made it really too
2341        unwieldy, quite apart from causing all the code to be include in a
2342        statically linked application, when only some functions are used. This is
2343        relevant even without the DFA addition now that patterns can be compiled in
2344        one application and matched in another.
2346        The downside of splitting up is that there have to be some external
2347        functions and data tables that are used internally in different modules of
2348        the library but which are not part of the API. These have all had their
2349        names changed to start with "_pcre_" so that they are unlikely to clash
2350        with other external names.
2352     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
2353        a different (DFA) algorithm. Although it is slower than the original
2354        function, it does have some advantages for certain types of matching
2355        problem.
2357     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
2358        including restarting after a partial match.
2360     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
2361        defined when compiling for Windows was sent to me. I have put it into the
2362        code, though I have no means of testing or verifying it.
2364     7. Added the pcre_refcount() auxiliary function.
2366     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
2367        match before or at the first newline in the subject string. In pcretest,
2368        the /f option on a pattern can be used to set this.
2370     9. A repeated \w when used in UTF-8 mode with characters greater than 256
2371        would behave wrongly. This has been present in PCRE since release 4.0.
2373    10. A number of changes to the pcregrep command:
2375        (a) Refactored how -x works; insert ^(...)$ instead of setting
2376            PCRE_ANCHORED and checking the length, in preparation for adding
2377            something similar for -w.
2379        (b) Added the -w (match as a word) option.
2381        (c) Refactored the way lines are read and buffered so as to have more
2382            than one at a time available.
2384        (d) Implemented a pcregrep test script.
2386        (e) Added the -M (multiline match) option. This allows patterns to match
2387            over several lines of the subject. The buffering ensures that at least
2388            8K, or the rest of the document (whichever is the shorter) is available
2389            for matching (and similarly the previous 8K for lookbehind assertions).
2391        (f) Changed the --help output so that it now says
2393              -w, --word-regex(p)
2395            instead of two lines, one with "regex" and the other with "regexp"
2396            because that confused at least one person since the short forms are the
2397            same. (This required a bit of code, as the output is generated
2398            automatically from a table. It wasn't just a text change.)
2400        (g) -- can be used to terminate pcregrep options if the next thing isn't an
2401            option but starts with a hyphen. Could be a pattern or a path name
2402            starting with a hyphen, for instance.
2404        (h) "-" can be given as a file name to represent stdin.
2406        (i) When file names are being printed, "(standard input)" is used for
2407            the standard input, for compatibility with GNU grep. Previously
2408            "<stdin>" was used.
2410        (j) The option --label=xxx can be used to supply a name to be used for
2411            stdin when file names are being printed. There is no short form.
2413        (k) Re-factored the options decoding logic because we are going to add
2414            two more options that take data. Such options can now be given in four
2415            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
2417        (l) Added the -A, -B, and -C options for requesting that lines of context
2418            around matches be printed.
2420        (m) Added the -L option to print the names of files that do not contain
2421            any matching lines, that is, the complement of -l.
2423        (n) The return code is 2 if any file cannot be opened, but pcregrep does
2424            continue to scan other files.
2426        (o) The -s option was incorrectly implemented. For compatibility with other
2427            greps, it now suppresses the error message for a non-existent or non-
2428            accessible file (but not the return code). There is a new option called
2429            -q that suppresses the output of matching lines, which was what -s was
2430            previously doing.
2432        (p) Added --include and --exclude options to specify files for inclusion
2433            and exclusion when recursing.
2435    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
2436        Hopefully, it now does.
2438    12. Missing cast in pcre_study().
2440    13. Added an "uninstall" target to the makefile.
2442    14. Replaced "extern" in the function prototypes in Makefile.in with
2443        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
2444        world, but is set differently for Windows.
2446    15. Added a second compiling function called pcre_compile2(). The only
2447        difference is that it has an extra argument, which is a pointer to an
2448        integer error code. When there is a compile-time failure, this is set
2449        non-zero, in addition to the error test pointer being set to point to an
2450        error message. The new argument may be NULL if no error number is required
2451        (but then you may as well call pcre_compile(), which is now just a
2452        wrapper). This facility is provided because some applications need a
2453        numeric error indication, but it has also enabled me to tidy up the way
2454        compile-time errors are handled in the POSIX wrapper.
2456    16. Added VPATH=.libs to the makefile; this should help when building with one
2457        prefix path and installing with another. (Or so I'm told by someone who
2458        knows more about this stuff than I do.)
2460    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
2461        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
2462        match everything, including newlines. This is not POSIX-compatible, but
2463        somebody wanted the feature. From pcretest it can be activated by using
2464        both the P and the s flags.
2466    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
2468    19. libpcre.pc was being incorrectly installed as executable.
2470    20. A couple of places in pcretest check for end-of-line by looking for '\n';
2471        it now also looks for '\r' so that it will work unmodified on Windows.
2473    21. Added Google's contributed C++ wrapper to the distribution.
2475    22. Added some untidy missing memory free() calls in pcretest, to keep
2476        Electric Fence happy when testing.
2480    Version 5.0 13-Sep-04
2481    ---------------------
2483     1. Internal change: literal characters are no longer packed up into items
2484        containing multiple characters in a single byte-string. Each character
2485        is now matched using a separate opcode. However, there may be more than one
2486        byte in the character in UTF-8 mode.
2488     2. The pcre_callout_block structure has two new fields: pattern_position and
2489        next_item_length. These contain the offset in the pattern to the next match
2490        item, and its length, respectively.
2492     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
2493        insertion of callouts before each pattern item. Added the /C option to
2494        pcretest to make use of this.
2496     4. On the advice of a Windows user, the lines
2498          #if defined(_WIN32) || defined(WIN32)
2499          _setmode( _fileno( stdout ), 0x8000 );
2500          #endif  /* defined(_WIN32) || defined(WIN32) */
2502        have been added to the source of pcretest. This apparently does useful
2503        magic in relation to line terminators.
2505     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
2506        for the benefit of those environments where the "b" makes a difference.
2508     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
2509        to know about it. I have put a hack into configure.in that adds in code
2510        to set GCC=yes if CC=icc. This seems to end up at a point in the
2511        generated configure script that is early enough to affect the setting of
2512        compiler options, which is what is needed, but I have no means of testing
2513        whether it really works. (The user who reported this had patched the
2514        generated configure script, which of course I cannot do.)
2516        LATER: After change 22 below (new libtool files), the configure script
2517        seems to know about icc (and also ecc). Therefore, I have commented out
2518        this hack in configure.in.
2520     7. Added support for pkg-config (2 patches were sent in).
2522     8. Negated POSIX character classes that used a combination of internal tables
2523        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2524        [[:^ascii]]. Typically, they would match almost any characters. The other
2525        POSIX classes were not broken in this way.
2527     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2528        to find the match, as PCRE was deluded into thinking that the match had to
2529        start at the start point or following a newline. The same bug applied to
2530        patterns with negative forward assertions or any backward assertions
2531        preceding ".*" at the start, unless the pattern required a fixed first
2532        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2534    10. In UTF-8 mode, when moving forwards in the subject after a failed match
2535        starting at the last subject character, bytes beyond the end of the subject
2536        string were read.
2538    11. Renamed the variable "class" as "classbits" to make life easier for C++
2539        users. (Previously there was a macro definition, but it apparently wasn't
2540        enough.)
2542    12. Added the new field "tables" to the extra data so that tables can be passed
2543        in at exec time, or the internal tables can be re-selected. This allows
2544        a compiled regex to be saved and re-used at a later time by a different
2545        program that might have everything at different addresses.
2547    13. Modified the pcre-config script so that, when run on Solaris, it shows a
2548        -R library as well as a -L library.
2550    14. The debugging options of pcretest (-d on the command line or D on a
2551        pattern) showed incorrect output for anything following an extended class
2552        that contained multibyte characters and which was followed by a quantifier.
2554    15. Added optional support for general category Unicode character properties
2555        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2556        support. It adds about 90K to the size of the library. The meanings of the
2557        inbuilt class escapes such as \d and \s have NOT been changed.
2559    16. Updated pcredemo.c to include calls to free() to release the memory for the
2560        compiled pattern.
2562    17. The generated file chartables.c was being created in the source directory
2563        instead of in the building directory. This caused the build to fail if the
2564        source directory was different from the building directory, and was
2565        read-only.
2567    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2568        file. No doubt somebody will tell me if they don't make sense... Also added
2569        Dan Mooney's comments about building on OpenVMS.
2571    19. Added support for partial matching via the PCRE_PARTIAL option for
2572        pcre_exec() and the \P data escape in pcretest.
2574    20. Extended pcretest with 3 new pattern features:
2576        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
2577              write the compiled pattern to the file whose name is "rest-of-line".
2578              This is a straight binary dump of the data, with the saved pointer to
2579              the character tables forced to be NULL. The study data, if any, is
2580              written too. After writing, pcretest reads a new pattern.
2582        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2583              compiled pattern from the given file. There must not be any
2584              occurrences of "<" in the file name (pretty unlikely); if there are,
2585              pcretest will instead treat the initial "<" as a pattern delimiter.
2586              After reading in the pattern, pcretest goes on to read data lines as
2587              usual.
2589        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2590              and 16-bit fields in a compiled pattern, to simulate a pattern that
2591              was compiled on a host of opposite endianness.
2593    21. The pcre-exec() function can now cope with patterns that were compiled on
2594        hosts of opposite endianness, with this restriction:
2596          As for any compiled expression that is saved and used later, the tables
2597          pointer field cannot be preserved; the extra_data field in the arguments
2598          to pcre_exec() should be used to pass in a tables address if a value
2599          other than the default internal tables were used at compile time.
2601    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2602        now diagnosed as an error. Previously, most of the time, a negative number
2603        would have been treated as zero, but if in addition "ovector" was passed as
2604        NULL, a crash could occur.
2606    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2607        new versions from the libtool 1.5 distribution (the last one is a copy of
2608        a file called libtool.m4). This seems to have fixed the need to patch
2609        "configure" to support Darwin 1.3 (which I used to do). However, I still
2610        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2611        workstation).
2613    24. Changed the PCRE licence to be the more standard "BSD" licence.
2616    Version 4.5 01-Dec-03
2617    ---------------------
2619     1. There has been some re-arrangement of the code for the match() function so
2620        that it can be compiled in a version that does not call itself recursively.
2621        Instead, it keeps those local variables that need separate instances for
2622        each "recursion" in a frame on the heap, and gets/frees frames whenever it
2623        needs to "recurse". Keeping track of where control must go is done by means
2624        of setjmp/longjmp. The whole thing is implemented by a set of macros that
2625        hide most of the details from the main code, and operates only if
2626        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2627        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2628        operating.
2630        To make it easier for callers to provide specially tailored get/free
2631        functions for this usage, two new functions, pcre_stack_malloc, and
2632        pcre_stack_free, are used. They are always called in strict stacking order,
2633        and the size of block requested is always the same.
2635        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2636        PCRE has been compiled to use the stack or the heap for recursion. The
2637        -C option of pcretest uses this to show which version is compiled.
2639        A new data escape \S, is added to pcretest; it causes the amounts of store
2640        obtained and freed by both kinds of malloc/free at match time to be added
2641        to the output.
2643     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2644        what's available on my current Linux desktop machine.
2646     3. When matching a UTF-8 string, the test for a valid string at the start has
2647        been extended. If start_offset is not zero, PCRE now checks that it points
2648        to a byte that is the start of a UTF-8 character. If not, it returns
2649        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2650        this is necessary because there may be backward assertions in the pattern.
2651        When matching the same subject several times, it may save resources to use
2652        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2654     4. The code for checking the validity of UTF-8 strings has been tightened so
2655        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2656        containing "overlong sequences".
2658     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2659        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2660        should have been "&", but it just so happened that all the cases this let
2661        through by mistake were picked up later in the function.
2663     6. I had used a variable called "isblank" - this is a C99 function, causing
2664        some compilers to warn. To avoid this, I renamed it (as "blankclass").
2666     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2667        prompting; (b) run "./pcretest /dev/null" at the start of the test script
2668        so the version is shown; (c) stop "make test" echoing "./RunTest".
2670     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2672     9. The prototype for memmove() for systems that don't have it was using
2673        size_t, but the inclusion of the header that defines size_t was later. I've
2674        moved the #includes for the C headers earlier to avoid this.
2676    10. Added some adjustments to the code to make it easier to compiler on certain
2677        special systems:
2679          (a) Some "const" qualifiers were missing.
2680          (b) Added the macro EXPORT before all exported functions; by default this
2681              is defined to be empty.
2682          (c) Changed the dftables auxiliary program (that builds chartables.c) so
2683              that it reads its output file name as an argument instead of writing
2684              to the standard output and assuming this can be redirected.
2686    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2687        class containing characters with values greater than 255, PCRE compilation
2688        went into a loop.
2690    12. A recursive reference to a subpattern that was within another subpattern
2691        that had a minimum quantifier of zero caused PCRE to crash. For example,
2692        (x(y(?2))z)? provoked this bug with a subject that got as far as the
2693        recursion. If the recursively-called subpattern itself had a zero repeat,
2694        that was OK.
2696    13. In pcretest, the buffer for reading a data line was set at 30K, but the
2697        buffer into which it was copied (for escape processing) was still set at
2698        1024, so long lines caused crashes.
2700    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2701        "internal error: code overflow...". This applied to any character class
2702        that was followed by a possessive quantifier.
2704    15. Modified the Makefile to add libpcre.la as a prerequisite for
2705        libpcreposix.la because I was told this is needed for a parallel build to
2706        work.
2708    16. If a pattern that contained .* following optional items at the start was
2709        studied, the wrong optimizing data was generated, leading to matching
2710        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2711        matching string must start with a or b or c. The correct conclusion for
2712        this pattern is that a match can start with any character.
2715    Version 4.4 13-Aug-03
2716    ---------------------
2718     1. In UTF-8 mode, a character class containing characters with values between
2719        127 and 255 was not handled correctly if the compiled pattern was studied.
2720        In fixing this, I have also improved the studying algorithm for such
2721        classes (slightly).
2723     2. Three internal functions had redundant arguments passed to them. Removal
2724        might give a very teeny performance improvement.
2726     3. Documentation bug: the value of the capture_top field in a callout is *one
2727        more than* the number of the hightest numbered captured substring.
2729     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2730        in incorrectly linking with a previously installed version. They now link
2731        explicitly with libpcre.la.
2733     5. configure.in no longer needs to recognize Cygwin specially.
2735     6. A problem in pcre.in for Windows platforms is fixed.
2737     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2738        pcretest, it used to include the size of the study block as part of its
2739        output. Unfortunately, the structure contains a field that has a different
2740        size on different hardware architectures. This meant that the tests that
2741        showed this size failed. As the block is currently always of a fixed size,
2742        this information isn't actually particularly useful in pcretest output, so
2743        I have just removed it.
2745     8. Three pre-processor statements accidentally did not start in column 1.
2746        Sadly, there are *still* compilers around that complain, even though
2747        standard C has not required this for well over a decade. Sigh.
2749     9. In pcretest, the code for checking callouts passed small integers in the
2750        callout_data field, which is a void * field. However, some picky compilers
2751        complained about the casts involved for this on 64-bit systems. Now
2752        pcretest passes the address of the small integer instead, which should get
2753        rid of the warnings.
2755    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2756        both compile and run time, and gives an error if an invalid UTF-8 sequence
2757        is found. There is a option for disabling this check in cases where the
2758        string is known to be correct and/or the maximum performance is wanted.
2760    11. In response to a bug report, I changed one line in Makefile.in from
2762            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2763        to
2764            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2766        to look similar to other lines, but I have no way of telling whether this
2767        is the right thing to do, as I do not use Windows. No doubt I'll get told
2768        if it's wrong...
2771    Version 4.3 21-May-03
2772    ---------------------
2774    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2775       Makefile.
2777    2. Some refactoring to improve the quality of the code:
2779       (i)   The utf8_table... variables are now declared "const".
2781       (ii)  The code for \cx, which used the "case flipping" table to upper case
2782             lower case letters, now just substracts 32. This is ASCII-specific,
2783             but the whole concept of \cx is ASCII-specific, so it seems
2784             reasonable.
2786       (iii) PCRE was using its character types table to recognize decimal and
2787             hexadecimal digits in the pattern. This is silly, because it handles
2788             only 0-9, a-f, and A-F, but the character types table is locale-
2789             specific, which means strange things might happen. A private
2790             table is now used for this - though it costs 256 bytes, a table is
2791             much faster than multiple explicit tests. Of course, the standard
2792             character types table is still used for matching digits in subject
2793             strings against \d.
2795       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2796             ending in _t are). So I've renamed it as ESC_tee.
2798    3. The first argument for regexec() in the POSIX wrapper should have been
2799       defined as "const".
2801    4. Changed pcretest to use malloc() for its buffers so that they can be
2802       Electric Fenced for debugging.
2804    5. There were several places in the code where, in UTF-8 mode, PCRE would try
2805       to read one or more bytes before the start of the subject string. Often this
2806       had no effect on PCRE's behaviour, but in some circumstances it could
2807       provoke a segmentation fault.
2809    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2810       to try to read one or more bytes before the start of the subject string.
2812    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
2813       UTF-8 support could misbehave in various ways if the subject string
2814       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
2815       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
2816       back over UTF-8 characters.)
2819    Version 4.2 14-Apr-03
2820    ---------------------
2822    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
2824    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
2825         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
2826         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
2827         [WIN_PREFIX]: new variable, "cyg" for Cygwin
2828         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
2829           and BUILD_EXEEXT
2830         Note: automatic setting of the BUILD variables is not yet working
2831         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
2832           compile-time but not at link-time
2833         [LINK]: use for linking executables only
2834         make different versions for Windows and non-Windows
2835         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
2836           libraries
2837         [LINK_FOR_BUILD]: new variable
2838         [OBJEXT]: use throughout
2839         [EXEEXT]: use throughout
2840         <winshared>: new target
2841         <wininstall>: new target
2842         <dftables.o>: use native compiler
2843         <dftables>: use native linker
2844         <install>: handle Windows platform correctly
2845         <clean>: ditto
2846         <check>: ditto
2847         copy DLL to top builddir before testing
2849       As part of these changes, -no-undefined was removed again. This was reported
2850       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
2851       in any case.
2853    3. Some tidies to get rid of compiler warnings:
2855       . In the match_data structure, match_limit was an unsigned long int, whereas
2856         match_call_count was an int. I've made them both unsigned long ints.
2858       . In pcretest the fact that a const uschar * doesn't automatically cast to
2859         a void * provoked a warning.
2861       . Turning on some more compiler warnings threw up some "shadow" variables
2862         and a few more missing casts.
2864    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
2865       option, a class that contained a single character with a value between 128
2866       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
2868    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
2869       option, a class that contained several characters, but with at least one
2870       whose value was between 128 and 255 caused PCRE to crash.
2873    Version 4.1 12-Mar-03
2874    ---------------------
2876    1. Compiling with gcc -pedantic found a couple of places where casts were
2877    needed, and a string in dftables.c that was longer than standard compilers are
2878    required to support.
2880    2. Compiling with Sun's compiler found a few more places where the code could
2881    be tidied up in order to avoid warnings.
2883    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
2884    first of these names is deprecated in the latest Autoconf in favour of the name
2885    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
2886    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2887    analogy I have changed it to CFLAGS_FOR_BUILD.
2889    4. Added -no-undefined to the linking command in the Makefile, because this is
2890    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2891    linking step for the pcreposix library.
2893    5. PCRE was failing to diagnose the case of two named groups with the same
2894    name.
2896    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2897    literal character that is needed in the subject for a match, and scans along to
2898    ensure that it is present before embarking on the full matching process. This
2899    saves time in cases of nested unlimited repeats that are never going to match.
2900    Problem: the scan can take a lot of time if the subject is very long (e.g.
2901    megabytes), thus penalizing straightforward matches. It is now done only if the
2902    amount of subject to be scanned is less than 1000 bytes.
2904    7. A lesser problem with the same optimization is that it was recording the
2905    first character of an anchored pattern as "needed", thus provoking a search
2906    right along the subject, even when the first match of the pattern was going to
2907    fail. The "needed" character is now not set for anchored patterns, unless it
2908    follows something in the pattern that is of non-fixed length. Thus, it still
2909    fulfils its original purpose of finding quick non-matches in cases of nested
2910    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
2913    Version 4.0 17-Feb-03
2914    ---------------------
2916    1. If a comment in an extended regex that started immediately after a meta-item
2917    extended to the end of string, PCRE compiled incorrect data. This could lead to
2918    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
2920    2. Moved to autoconf 2.53 and libtool 1.4.2.
2922    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
2923    the special perltest8 script is no longer needed - all the tests can be run
2924    from a single perltest script.
2926    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
2927    by \s. It has now been removed in PCRE. This means it isn't recognized as
2928    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
2929    class [:space:] *does* include VT, thereby creating a mess.
2931    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
2932    space and tab.
2934    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
2935    its new features into the main test script, reducing the number of scripts.
2937    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
2938    were backward compatible, and made the (?i) apply to the whole pattern, as if
2939    /i were given. Now it behaves more logically, and applies the option setting
2940    only to what follows. PCRE has been changed to follow suit. However, if it
2941    finds options settings right at the start of the pattern, it extracts them into
2942    the global options, as before. Thus, they show up in the info data.
2944    8. Added support for the \Q...\E escape sequence. Characters in between are
2945    treated as literals. This is slightly different from Perl in that $ and @ are
2946    also handled as literals inside the quotes. In Perl, they will cause variable
2947    interpolation. Note the following examples:
2949        Pattern            PCRE matches      Perl matches
2951        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
2952        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
2953        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
2955    For compatibility with Perl, \Q...\E sequences are recognized inside character
2956    classes as well as outside them.
2958    9. Re-organized 3 code statements in pcretest to avoid "overflow in
2959    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2960    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2961    signed/unsigned warnings.
2963    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2964    option for pcretest, so I've replaced it by a simple function that does just
2965    that job.
2967    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2968    "pcregrep -".
2970    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2971    Java package. This provides some syntactic sugar for simple cases of what my
2972    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2973    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2974    item, you can use this simplified notation. Note that only makes sense with
2975    greedy quantifiers. Consequently, the use of the possessive quantifier forces
2976    greediness, whatever the setting of the PCRE_UNGREEDY option.
2978    13. A change of greediness default within a pattern was not taking effect at
2979    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2980    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2981    was abstracted outside.
2983    14. PCRE now supports the \G assertion. It is true when the current matching
2984    position is at the start point of the match. This differs from \A when the
2985    starting offset is non-zero. Used with the /g option of pcretest (or similar
2986    code), it works in the same way as it does for Perl's /g option. If all
2987    alternatives of a regex begin with \G, the expression is anchored to the start
2988    match position, and the "anchored" flag is set in the compiled expression.
2990    15. Some bugs concerning the handling of certain option changes within patterns
2991    have been fixed. These applied to options other than (?ims). For example,
2992    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2993    been the other way round. Some of this was related to change 7 above.
2995    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2996    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2997    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2998    POSIX classes only within a class (e.g. /[[:alpha:]]/).
3000    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
3001    mode. Unlike ".", it always matches newline, whatever the setting of
3002    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
3003    assertions. Perl allows it, but it doesn't (in general) work because it can't
3004    calculate the length of the lookbehind. At least, that's the case for Perl
3005    5.8.0 - I've been told they are going to document that it doesn't work in
3006    future.
3008    18. Added an error diagnosis for escapes that PCRE does not support: these are
3009    \L, \l, \N, \P, \p, \U, \u, and \X.
3011    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
3012    reading past the end of the pattern in cases such as /[abcd/.
3014    20. PCRE was getting more memory than necessary for patterns with classes that
3015    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
3017    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
3018    compiling PCRE for use with Virtual Pascal.
3020    22. Small fix to the Makefile to make it work properly if the build is done
3021    outside the source tree.
3023    23. Added a new extension: a condition to go with recursion. If a conditional
3024    subpattern starts with (?(R) the "true" branch is used if recursion has
3025    happened, whereas the "false" branch is used only at the top level.
3027    24. When there was a very long string of literal characters (over 255 bytes
3028    without UTF support, over 250 bytes with UTF support), the computation of how
3029    much memory was required could be incorrect, leading to segfaults or other
3030    strange effects.
3032    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
3033    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
3034    there was a subsequent back reference to those brackets. This meant that, for
3035    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
3036    possible to check for precisely this case. All we can do is abandon the
3037    optimization if .* occurs inside capturing brackets when there are any back
3038    references whatsoever. (See below for a better fix that came later.)
3040    26. The handling of the optimization for finding the first character of a
3041    non-anchored pattern, and for finding a character that is required later in the
3042    match were failing in some cases. This didn't break the matching; it just
3043    failed to optimize when it could. The way this is done has been re-implemented.
3045    27. Fixed typo in error message for invalid (?R item (it said "(?p").
3047    28. Added a new feature that provides some of the functionality that Perl
3048    provides with (?{...}). The facility is termed a "callout". The way it is done
3049    in PCRE is for the caller to provide an optional function, by setting
3050    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
3051    global variable. By default it is unset, which disables all calling out. To get
3052    the function called, the regex must include (?C) at appropriate points. This
3053    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
3054    This provides a means of identifying different callout points. When PCRE
3055    reaches such a point in the regex, if pcre_callout has been set, the external
3056    function is called. It is provided with data in a structure called
3057    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
3058    matching continues; if it returns a non-zero value, the match at the current
3059    point fails. However, backtracking will occur if possible. [This was changed
3060    later and other features added - see item 49 below.]
3062    29. pcretest is upgraded to test the callout functionality. It provides a
3063    callout function that displays information. By default, it shows the start of
3064    the match and the current position in the text. There are some new data escapes
3065    to vary what happens:
3067        \C+         in addition, show current contents of captured substrings
3068        \C-         do not supply a callout function
3069        \C!n        return 1 when callout number n is reached
3070        \C!n!m      return 1 when callout number n is reached for the mth time
3072    30. If pcregrep was called with the -l option and just a single file name, it
3073    output "<stdin>" if a match was found, instead of the file name.
3075    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
3076    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
3077    pcre_exec(). This saves a malloc/free per call. The default value of
3078    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
3079    when configuring.
3081    32. The default maximum size of a compiled pattern is 64K. There have been a
3082    few cases of people hitting this limit. The code now uses macros to handle the
3083    storing of links as offsets within the compiled pattern. It defaults to 2-byte
3084    links, but this can be changed to 3 or 4 bytes by --with-link-size when
3085    configuring. Tests 2 and 5 work only with 2-byte links because they output
3086    debugging information about compiled patterns.
3088    33. Internal code re-arrangements:
3090    (a) Moved the debugging function for printing out a compiled regex into
3091        its own source file (printint.c) and used #include to pull it into
3092        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
3093        separate copies.
3095    (b) Defined the list of op-code names for debugging as a macro in
3096        internal.h so that it is next to the definition of the opcodes.
3098    (c) Defined a table of op-code lengths for simpler skipping along compiled
3099        code. This is again a macro in internal.h so that it is next to the
3100        definition of the opcodes.
3102    34. Added support for recursive calls to individual subpatterns, along the
3103    lines of Robin Houston's patch (but implemented somewhat differently).
3105    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
3106    allow it to read and process whole directories in Win32. This code was
3107    contributed by Lionel Fourquaux; it has not been tested by me.
3109    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
3110    used to name a group. Names consist of alphanumerics and underscores, and must
3111    be unique. Back references use the syntax (?P=name) and recursive calls use
3112    (?P>name) which is a PCRE extension to the Python extension. Groups still have
3113    numbers. The function pcre_fullinfo() can be used after compilation to extract
3114    a name/number map. There are three relevant calls:
3116      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
3117      PCRE_INFO_NAMECOUNT            yields the number of entries
3118      PCRE_INFO_NAMETABLE            yields a pointer to the map.
3120    The map is a vector of fixed-size entries. The size of each entry depends on
3121    the length of the longest name used. The first two bytes of each entry are the
3122    group number, most significant byte first. There follows the corresponding
3123    name, zero terminated. The names are in alphabetical order.
3125    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
3126    case instead of 255. Making it the same both with and without UTF-8 support
3127    means that the same test output works with both.
3129    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
3130    calling malloc() with a zero argument.
3132    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
3133    optimization. I've improved things by keeping a bitmap of backreferences with
3134    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
3135    fact referenced, the optimization can be applied. It is unlikely that a
3136    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
3137    the match to follow \n) will appear inside brackets with a number greater than
3138    31, but if it does, any back reference > 31 suppresses the optimization.
3140    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
3141    of disabling numbered capturing parentheses. Any opening parenthesis that is
3142    not followed by ? behaves as if it were followed by ?: but named parentheses
3143    can still be used for capturing (and they will acquire numbers in the usual
3144    way).
3146    41. Redesigned the return codes from the match() function into yes/no/error so
3147    that errors can be passed back from deep inside the nested calls. A malloc
3148    failure while inside a recursive subpattern call now causes the
3149    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
3151    42. It is now possible to set a limit on the number of times the match()
3152    function is called in a call to pcre_exec(). This facility makes it possible to
3153    limit the amount of recursion and backtracking, though not in a directly
3154    obvious way, because the match() function is used in a number of different
3155    circumstances. The count starts from zero for each position in the subject
3156    string (for non-anchored patterns). The default limit is, for compatibility, a
3157    large number, namely 10 000 000. You can change this in two ways:
3159    (a) When configuring PCRE before making, you can use --with-match-limit=n
3160        to set a default value for the compiled library.
3162    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
3163        a different value is set. See 45 below.
3165    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
3167    43. Added a new function pcre_config(int, void *) to enable run-time extraction
3168    of things that can be changed at compile time. The first argument specifies
3169    what is wanted and the second points to where the information is to be placed.
3170    The current list of available information is:
3172      PCRE_CONFIG_UTF8
3174    The output is an integer that is set to one if UTF-8 support is available;
3175    otherwise it is set to zero.
3179    The output is an integer that it set to the value of the code that is used for
3180    newline. It is either LF (10) or CR (13).
3184    The output is an integer that contains the number of bytes used for internal
3185    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
3189    The output is an integer that contains the threshold above which the POSIX
3190    interface uses malloc() for output vectors. See item 31 above.
3194    The output is an unsigned integer that contains the default limit of the number
3195    of match() calls in a pcre_exec() execution. See 42 above.
3197    44. pcretest has been upgraded by the addition of the -C option. This causes it
3198    to extract all the available output from the new pcre_config() function, and to
3199    output it. The program then exits immediately.
3201    45. A need has arisen to pass over additional data with calls to pcre_exec() in
3202    order to support additional features. One way would have been to define
3203    pcre_exec2() (for example) with extra arguments, but this would not have been
3204    extensible, and would also have required all calls to the original function to
3205    be mapped to the new one. Instead, I have chosen to extend the mechanism that
3206    is used for passing in "extra" data from pcre_study().
3208    The pcre_extra structure is now exposed and defined in pcre.h. It currently
3209    contains the following fields:
3211      flags         a bitmap indicating which of the following fields are set
3212      study_data    opaque data from pcre_study()
3213      match_limit   a way of specifying a limit on match() calls for a specific
3214                      call to pcre_exec()
3215      callout_data  data for callouts (see 49 below)
3217    The flag bits are also defined in pcre.h, and are
3223    The pcre_study() function now returns one of these new pcre_extra blocks, with
3224    the actual study data pointed to by the study_data field, and the
3225    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
3226    before. That is, this change is entirely upwards-compatible and requires no
3227    change to existing code.
3229    If you want to pass in additional data to pcre_exec(), you can either place it
3230    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
3231    block.
3233    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
3234    data string contains the escape sequence \M, pcretest calls pcre_exec() several
3235    times with different match limits, until it finds the minimum value needed for
3236    pcre_exec() to complete. The value is then output. This can be instructive; for
3237    most simple matches the number is quite small, but for pathological cases it
3238    gets very large very quickly.
3240    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
3241    returns the size of the data block pointed to by the study_data field in a
3242    pcre_extra block, that is, the value that was passed as the argument to
3243    pcre_malloc() when PCRE was getting memory in which to place the information
3244    created by pcre_study(). The fourth argument should point to a size_t variable.
3245    pcretest has been extended so that this information is shown after a successful
3246    pcre_study() call when information about the compiled regex is being displayed.
3248    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
3249    because what follows is always an absolute path. (Later: it turns out that this
3250    is more than cosmetic for MinGW, because it doesn't like empty path
3251    components.)
3253    49. Some changes have been made to the callout feature (see 28 above):
3255    (i)  A callout function now has three choices for what it returns:
3257           0  =>  success, carry on matching
3258         > 0  =>  failure at this point, but backtrack if possible
3259         < 0  =>  serious error, return this value from pcre_exec()
3261         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
3262         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
3263         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
3264         use by callout functions. It will never be used by PCRE itself.
3266    (ii) The pcre_extra structure (see 45 above) has a void * field called
3267         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
3268         pcre_callout_block structure has a field of the same name. The contents of
3269         the field passed in the pcre_extra structure are passed to the callout
3270         function in the corresponding field in the callout block. This makes it
3271         easier to use the same callout-containing regex from multiple threads. For
3272         testing, the pcretest program has a new data escape
3274           \C*n        pass the number n (may be negative) as callout_data
3276         If the callout function in pcretest receives a non-zero value as
3277         callout_data, it returns that value.
3279    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
3280    there were some redundant $(CFLAGS) in commands that are now specified as
3281    $(LINK), which already includes $(CFLAGS).
3283    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
3284    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
3285    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
3286    one-byte characters throughout. Note that case-insensitive matching applies
3287    only to characters whose values are less than 256. PCRE doesn't support the
3288    notion of cases for higher-valued characters.
3290    (i)   A character class whose characters are all within 0-255 is handled as
3291          a bit map, and the map is inverted for negative classes. Previously, a
3292          character > 255 always failed to match such a class; however it should
3293          match if the class was a negative one (e.g. [^ab]). This has been fixed.
3295    (ii)  A negated character class with a single character < 255 is coded as
3296          "not this character" (OP_NOT). This wasn't working properly when the test
3297          character was multibyte, either singly or repeated.
3299    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
3300          mode, for example: \x{100}{2,3}.
3302    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
3303          singly or repeated) now correctly test multibyte characters. However,
3304          PCRE doesn't recognize any characters with values greater than 255 as
3305          digits, spaces, or word characters. Such characters always match \D, \S,
3306          and \W, and never match \d, \s, or \w.
3308    (v)   Classes may now contain characters and character ranges with values
3309          greater than 255. For example: [ab\x{100}-\x{400}].
3311    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
3312          PCRE in UTF-8 mode.
3314    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
3315    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
3316    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
3317    value.)
3319    53. The single man page has become too large. I have therefore split it up into
3320    a number of separate man pages. These also give rise to individual HTML pages;
3321    these are now put in a separate directory, and there is an index.html page that
3322    lists them all. Some hyperlinking between the pages has been installed.
3324    54. Added convenience functions for handling named capturing parentheses.
3326    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
3327    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
3328    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
3329    are faulted.
3331    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
3332    calling configure. These values are used when compiling the dftables.c program
3333    which is run to generate the source of the default character tables. They
3334    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
3335    you will need to set these values.
3337    57. Updated the building process for Windows DLL, as provided by Fred Cox.
3340    Version 3.9 02-Jan-02
3341    ---------------------
3343    1. A bit of extraneous text had somehow crept into the pcregrep documentation.
3345    2. If --disable-static was given, the building process failed when trying to
3346    build pcretest and pcregrep. (For some reason it was using libtool to compile
3347    them, which is not right, as they aren't part of the library.)
3350    Version 3.8 18-Dec-01
3351    ---------------------
3353    1. The experimental UTF-8 code was completely screwed up. It was packing the
3354    bytes in the wrong order. How dumb can you get?
3357    Version 3.7 29-Oct-01
3358    ---------------------
3360    1. In updating pcretest to check change 1 of version 3.6, I screwed up.
3361    This caused pcretest, when used on the test data, to segfault. Unfortunately,
3362    this didn't happen under Solaris 8, where I normally test things.
3364    2. The Makefile had to be changed to make it work on BSD systems, where 'make'
3365    doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
3366    isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
3367    this fix an hour or so after the initial 3.7 release.)
3370    Version 3.6 23-Oct-01
3371    ---------------------
3373    1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
3374    offsets passed as NULL with zero offset count.
3376    2. The config.guess and config.sub files had not been updated when I moved to
3377    the latest autoconf.
3380    Version 3.5 15-Aug-01
3381    ---------------------
3383    1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
3384    had been forgotten.
3386    2. By using declared but undefined structures, we can avoid using "void"
3387    definitions in pcre.h while keeping the internal definitions of the structures
3388    private.
3390    3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
3391    user point of view, this means that both static and shared libraries are built
3392    by default, but this can be individually controlled. More of the work of
3393    handling this static/shared cases is now inside libtool instead of PCRE's make
3394    file.
3396    4. The pcretest utility is now installed along with pcregrep because it is
3397    useful for users (to test regexs) and by doing this, it automatically gets
3398    relinked by libtool. The documentation has been turned into a man page, so
3399    there are now .1, .txt, and .html versions in /doc.
3401    5. Upgrades to pcregrep:
3402       (i)   Added long-form option names like gnu grep.
3403       (ii)  Added --help to list all options with an explanatory phrase.
3404       (iii) Added -r, --recursive to recurse into sub-directories.
3405       (iv)  Added -f, --file to read patterns from a file.
3407    6. pcre_exec() was referring to its "code" argument before testing that
3408    argument for NULL (and giving an error if it was NULL).
3410    7. Upgraded Makefile.in to allow for compiling in a different directory from
3411    the source directory.
3413    8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
3414    options bits, the pointer it was passed was to an int instead of to an unsigned
3415    long int. This mattered only on 64-bit systems.
3417    9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
3418    generated) instead of pcre.in, which it its source. Also made the same change
3419    in several of the .c files.
3421    10. A new release of gcc defines printf() as a macro, which broke pcretest
3422    because it had an ifdef in the middle of a string argument for printf(). Fixed
3423    by using separate calls to printf().
3425    11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
3426    script, to force use of CR or LF instead of \n in the source. On non-Unix
3427    systems, the value can be set in config.h.
3429    12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
3430    absolute limit. Changed the text of the error message to make this clear, and
3431    likewise updated the man page.
3433    13. The limit of 99 on the number of capturing subpatterns has been removed.
3434    The new limit is 65535, which I hope will not be a "real" limit.
3437    Version 3.4 22-Aug-00
3438    ---------------------
3440    1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
3442    2. Diagnose condition (?(0) as an error instead of crashing on matching.
3445    Version 3.3 01-Aug-00
3446    ---------------------
3448    1. If an octal character was given, but the value was greater than \377, it
3449    was not getting masked to the least significant bits, as documented. This could
3450    lead to crashes in some systems.
3452    2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
3453    the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
3455    3. Added the functions pcre_free_substring() and pcre_free_substring_list().
3456    These just pass their arguments on to (pcre_free)(), but they are provided
3457    because some uses of PCRE bind it to non-C systems that can call its functions,
3458    but cannot call free() or pcre_free() directly.
3460    4. Add "make test" as a synonym for "make check". Corrected some comments in
3461    the Makefile.
3463    5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
3464    Makefile.
3466    6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
3467    command called pgrep for grepping around the active processes.
3469    7. Added the beginnings of support for UTF-8 character strings.
3471    8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
3472    RANLIB to ./ltconfig so that they are used by libtool. I think these are all
3473    the relevant ones. (AR is not passed because ./ltconfig does its own figuring
3474    out for the ar command.)
3477    Version 3.2 12-May-00
3478    ---------------------
3480    This is purely a bug fixing release.
3482    1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
3483    of ZA. This was just one example of several cases that could provoke this bug,
3484    which was introduced by change 9 of version 2.00. The code for breaking
3485    infinite loops after an iteration that matches an empty string was't working
3486    correctly.
3488    2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
3489    when matched against abbab (for example). After matching an empty string, it
3490    wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
3491    caused it to match further down the string than it should.
3493    3. The code contained an inclusion of sys/types.h. It isn't clear why this
3494    was there because it doesn't seem to be needed, and it causes trouble on some
3495    systems, as it is not a Standard C header. It has been removed.
3497    4. Made 4 silly changes to the source to avoid stupid compiler warnings that
3498    were reported on the Macintosh. The changes were from
3500      while ((c = *(++ptr)) != 0 && c != '\n');
3501    to
3502      while ((c = *(++ptr)) != 0 && c != '\n') ;
3504    Totally extraordinary, but if that's what it takes...
3506    5. PCRE is being used in one environment where neither memmove() nor bcopy() is
3507    available. Added HAVE_BCOPY and an autoconf test for it; if neither
3508    HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
3509    assumes the way PCRE uses memmove() (always moving upwards).
3511    6. PCRE is being used in one environment where strchr() is not available. There
3512    was only one use in pcre.c, and writing it out to avoid strchr() probably gives
3513    faster code anyway.
3516    Version 3.1 09-Feb-00
3517    ---------------------
3519    The only change in this release is the fixing of some bugs in Makefile.in for
3520    the "install" target:
3522    (1) It was failing to install pcreposix.h.
3524    (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
3527    Version 3.0 01-Feb-00
3528    ---------------------
3530    1. Add support for the /+ modifier to perltest (to output $` like it does in
3531    pcretest).
3533    2. Add support for the /g modifier to perltest.