ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 81 by nigel, Sat Feb 24 21:40:59 2007 UTC revision 587 by ph10, Fri Jan 14 19:01:25 2011 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
4    Version 8.12 12-Jan-2011
5    ------------------------
7    1.  Fixed some typos in the markup of the man pages, and wrote a script that
8        checks for such things as part of the documentation building process.
10    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
11        --match-limit and --recursion-limit options (added for 8.11). In
12        particular, this made one of the standard tests crash. (The integer value
13        went into the wrong half of a long int.)
15    3.  If the --colour option was given to pcregrep with -v (invert match), it
16        did strange things, either producing crazy output, or crashing. It should,
17        of course, ignore a request for colour when reporting lines that do not
18        match.
20    4.  Another pcregrep bug caused similar problems if --colour was specified with
21        -M (multiline) and the pattern match finished with a line ending.
23    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
24        matched in multiline mode, the following line was shown as part of the
25        match. This seems wrong, so I have changed it.
27    6.  If pcregrep was compiled under Windows, there was a reference to the
28        function pcregrep_exit() before it was defined. I am assuming this was
29        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
30        reported by a user. I've moved the definition above the reference.
33    Version 8.11 10-Dec-2010
34    ------------------------
36    1.  (*THEN) was not working properly if there were untried alternatives prior
37        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
38        backtracked to try for "b" instead of moving to the next alternative branch
39        at the same level (in this case, to look for "c"). The Perl documentation
40        is clear that when (*THEN) is backtracked onto, it goes to the "next
41        alternative in the innermost enclosing group".
43    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
44        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
45        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
46        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
47        (*THEN).
49    3.  If \s appeared in a character class, it removed the VT character from
50        the class, even if it had been included by some previous item, for example
51        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
52        of \s, but is part of the POSIX "space" class.)
54    4.  A partial match never returns an empty string (because you can always
55        match an empty string at the end of the subject); however the checking for
56        an empty string was starting at the "start of match" point. This has been
57        changed to the "earliest inspected character" point, because the returned
58        data for a partial match starts at this character. This means that, for
59        example, /(?<=abc)def/ gives a partial match for the subject "abc"
60        (previously it gave "no match").
62    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
63        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
64        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
65        has an implication that the given string is incomplete (because a partial
66        match is preferred over a full match). For this reason, these items now
67        give a partial match in this situation. [Aside: previously, the one case
68        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
69        match rather than a full match, which was wrong by the old rules, but is
70        now correct.]
72    6.  There was a bug in the handling of #-introduced comments, recognized when
73        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
74        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
75        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
76        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
77        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
78        places in pcre_compile().
80    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
81        comments when looking ahead for named forward references to subpatterns,
82        the only newline sequence it recognized was NL. It now handles newlines
83        according to the set newline convention.
85    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
86        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
87        cater for a lack of strerror(). These oversights have been fixed.
89    9.  Added --match-limit and --recursion-limit to pcregrep.
91    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
93    11. When the -o option was used, pcregrep was setting a return code of 1, even
94        when matches were found, and --line-buffered was not being honoured.
96    12. Added an optional parentheses number to the -o and --only-matching options
97        of pcregrep.
99    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
100        can match an empty string. The code to do it in pcretest and pcredemo
101        needed fixing:
103        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
104            only one byte after an empty string match just before CRLF (this case
105            just got forgotten; "any" and "anycrlf" were OK).
107        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
108            mode when an empty string match preceded an ASCII character followed by
109            a non-ASCII character. (The code for advancing by one character rather
110            than one byte was nonsense.)
112        (c) The pcredemo.c sample program did not have any code at all to handle
113            the cases when CRLF is a valid newline sequence.
115    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
116        as a starting offset was within the subject string. There is now a new
117        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
118        negative or greater than the length of the string. In order to test this,
119        pcretest is extended to allow the setting of negative starting offsets.
121    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
122        starting offset points to the beginning of a UTF-8 character was
123        unnecessarily clumsy. I tidied it up.
125    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
126        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
128    17. Nobody had reported that the --include_dir option, which was added in
129        release 7.7 should have been called --include-dir (hyphen, not underscore)
130        for compatibility with GNU grep. I have changed it to --include-dir, but
131        left --include_dir as an undocumented synonym, and the same for
132        --exclude-dir, though that is not available in GNU grep, at least as of
133        release 2.5.4.
135    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
136        characters from a string of bytes) have been redefined so as not to use
137        loops, in order to improve performance in some environments. At the same
138        time, I abstracted some of the common code into auxiliary macros to save
139        repetition (this should not affect the compiled code).
141    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
142        compile-time error is now given if \c is not followed by an ASCII
143        character, that is, a byte less than 128. (In EBCDIC mode, the code is
144        different, and any byte value is allowed.)
146    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
147        START_OPTIMIZE option, which is now allowed at compile time - but just
148        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
149        to pcregrep and other applications that have no direct access to PCRE
150        options. The new /Y option in pcretest sets this option when calling
151        pcre_compile().
153    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
154        back references. Groups containing recursive back references were forced to
155        be atomic by that change, but in the case of named groups, the amount of
156        memory required was incorrectly computed, leading to "Failed: internal
157        error: code overflow". This has been fixed.
159    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
160        pcretest.c, to avoid build problems in some Borland environments.
163    Version 8.10 25-Jun-2010
164    ------------------------
166    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
167        THEN.
169    2.  (*ACCEPT) was not working when inside an atomic group.
171    3.  Inside a character class, \B is treated as a literal by default, but
172        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
173        causes the error). The code is unchanged, but I tidied the documentation.
175    4.  Inside a character class, PCRE always treated \R and \X as literals,
176        whereas Perl faults them if its -w option is set. I have changed PCRE so
177        that it faults them when PCRE_EXTRA is set.
179    5.  Added support for \N, which always matches any character other than
180        newline. (It is the same as "." when PCRE_DOTALL is not set.)
182    6.  When compiling pcregrep with newer versions of gcc which may have
183        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
184        declared with attribute warn_unused_result" were given. Just casting the
185        result to (void) does not stop the warnings; a more elaborate fudge is
186        needed. I've used a macro to implement this.
188    7.  Minor change to pcretest.c to avoid a compiler warning.
190    8.  Added four artifical Unicode properties to help with an option to make
191        \s etc use properties (see next item). The new properties are: Xan
192        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
194    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
195        use Unicode properties. (*UCP) at the start of a pattern can be used to set
196        this option. Modified pcretest to add /W to test this facility. Added
197        REG_UCP to make it available via the POSIX interface.
199    10. Added --line-buffered to pcregrep.
201    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
202        studied, and the match started with a letter with a code point greater than
203        127 whose first byte was different to the first byte of the other case of
204        the letter, the other case of this starting letter was not recognized
205        (#976).
207    12. If a pattern that was studied started with a repeated Unicode property
208        test, for example, \p{Nd}+, there was the theoretical possibility of
209        setting up an incorrect bitmap of starting bytes, but fortunately it could
210        not have actually happened in practice until change 8 above was made (it
211        added property types that matched character-matching opcodes).
213    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
214        possible starting bytes for non-anchored patterns.
216    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
217        \R, and also a number of cases that involve Unicode properties, both
218        explicit and implicit when PCRE_UCP is set.
220    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
221        input, it could crash or give wrong results if characters with values
222        greater than 0xc0 were present in the subject string. (Detail: it assumed
223        UTF-8 input when processing these items.)
225    16. Added a lot of (int) casts to avoid compiler warnings in systems where
226        size_t is 64-bit (#991).
228    17. Added a check for running out of memory when PCRE is compiled with
229        --disable-stack-for-recursion (#990).
231    18. If the last data line in a file for pcretest does not have a newline on
232        the end, a newline was missing in the output.
234    19. The default pcre_chartables.c file recognizes only ASCII characters (values
235        less than 128) in its various bitmaps. However, there is a facility for
236        generating tables according to the current locale when PCRE is compiled. It
237        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
238        characters, are recognized by isspace() and therefore were getting set in
239        these tables, and indeed these tables seem to approximate to ISO 8859. This
240        caused a problem in UTF-8 mode when pcre_study() was used to create a list
241        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
242        which of course cannot start UTF-8 characters. I have changed the code so
243        that only real ASCII characters (less than 128) and the correct starting
244        bytes for UTF-8 encodings are set for characters greater than 127 when in
245        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
246        altogether.)
248    20. Added the /T option to pcretest so as to be able to run tests with non-
249        standard character tables, thus making it possible to include the tests
250        used for 19 above in the standard set of tests.
252    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
253        reference to a subpattern the other side of a comment that contains an
254        opening parenthesis caused either an internal compiling error, or a
255        reference to the wrong subpattern.
258    Version 8.02 19-Mar-2010
259    ------------------------
261    1.  The Unicode data tables have been updated to Unicode 5.2.0.
263    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
264        configured.
266    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
267        original author of that file, following a query about its status.
269    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
270        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
272    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
273        quantifier applied to a forward-referencing subroutine call, could compile
274        incorrect code or give the error "internal error: previously-checked
275        referenced subpattern not found".
277    6.  Both MS Visual Studio and Symbian OS have problems with initializing
278        variables to point to external functions. For these systems, therefore,
279        pcre_malloc etc. are now initialized to local functions that call the
280        relevant global functions.
282    7.  There were two entries missing in the vectors called coptable and poptable
283        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
284        I've fixed the data, and added a kludgy way of testing at compile time that
285        the lengths are correct (equal to the number of opcodes).
287    8.  Following on from 7, I added a similar kludge to check the length of the
288        eint vector in pcreposix.c.
290    9.  Error texts for pcre_compile() are held as one long string to avoid too
291        much relocation at load time. To find a text, the string is searched,
292        counting zeros. There was no check for running off the end of the string,
293        which could happen if a new error number was added without updating the
294        string.
296    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
298    11. \K was not working if it appeared in an atomic group or in a group that
299        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
300        \K is "not well defined" if used in an assertion. PCRE now accepts it if
301        the assertion is positive, but not if it is negative.
303    12. Change 11 fortuitously reduced the size of the stack frame used in the
304        "match()" function of pcre_exec.c by one pointer. Forthcoming
305        implementation of support for (*MARK) will need an extra pointer on the
306        stack; I have reserved it now, so that the stack frame size does not
307        decrease.
309    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
310        item in branch that calls a recursion is a subroutine call - as in the
311        second branch in the above example - was incorrectly given the compile-
312        time error "recursive call could loop indefinitely" because pcre_compile()
313        was not correctly checking the subroutine for matching a non-empty string.
315    14. The checks for overrunning compiling workspace could trigger after an
316        overrun had occurred. This is a "should never occur" error, but it can be
317        triggered by pathological patterns such as hundreds of nested parentheses.
318        The checks now trigger 100 bytes before the end of the workspace.
320    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
323    Version 8.01 19-Jan-2010
324    ------------------------
326    1.  If a pattern contained a conditional subpattern with only one branch (in
327        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
328        computed the wrong minimum data length (which is of course zero for such
329        subpatterns). This could cause incorrect "no match" results.
331    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
332        the pattern is reset in the first branch, pcre_compile() failed with
333        "internal error: code overflow at offset...". This happened only when
334        the reset was to the original external option setting. (An optimization
335        abstracts leading options settings into an external setting, which was the
336        cause of this.)
338    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
339        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
340        assertion pattern did not match (meaning that the assertion was true), it
341        was incorrectly treated as false if the SKIP had been reached during the
342        matching. This also applied to assertions used as conditions.
344    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
345        assertion subpattern, including such a pattern used as a condition,
346        unpredictable results occurred, instead of the error return
349    5.  The C++ GlobalReplace function was not working like Perl for the special
350        situation when an empty string is matched. It now does the fancy magic
351        stuff that is necessary.
353    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
354        removed. (These were left over from very, very early versions of PCRE.)
356    7.  Some cosmetic changes to the code to make life easier when compiling it
357        as part of something else:
359        (a) Change DEBUG to PCRE_DEBUG.
361        (b) In pcre_compile(), rename the member of the "branch_chain" structure
362            called "current" as "current_branch", to prevent a collision with the
363            Linux macro when compiled as a kernel module.
365        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
366            prevent a collision with the Linux macro when compiled as a kernel
367            module.
369    8.  In pcre_compile() there are some checks for integer overflows that used to
370        cast potentially large values to (double). This has been changed to that
371        when building, a check for int64_t is made, and if it is found, it is used
372        instead, thus avoiding the use of floating point arithmetic. (There is no
373        other use of FP in PCRE.) If int64_t is not found, the fallback is to
374        double.
376    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
377        2005 (difference between two addresses compared to an unsigned value).
379    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
380        custom one, because of the following reported problem in Windows:
382          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
383              under Win32.
384          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
385              therefore missing the function definition.
386          - The compiler thus generates a "C" signature for the test function.
387          - The linker fails to find the "C" function.
388          - PCRE fails to configure if asked to do so against libbz2.
390    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
391        messages were output:
393          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
394          rerunning libtoolize, to keep the correct libtool macros in-tree.
395          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
397        I have done both of these things.
399    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
400        most of the time, it *can* run out if it is given a pattern that contains a
401        runaway infinite recursion. I updated the discussion in the pcrestack man
402        page.
404    13. Now that we have gone to the x.xx style of version numbers, the minor
405        version may start with zero. Using 08 or 09 is a bad idea because users
406        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
407        interpreted as invalid octal numbers. I've updated the previous comment in
408        configure.ac, and also added a check that gives an error if 08 or 09 are
409        used.
411    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
412        causing partial matching to fail when the end of the subject matched \W
413        in a UTF-8 pattern where \W was quantified with a minimum of 3.
415    15. There were some discrepancies between the declarations in pcre_internal.h
416        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
417        their definitions. The declarations used "const uschar *" and the
418        definitions used USPTR. Even though USPTR is normally defined as "const
419        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
420        reported that: "This difference in casting confuses some C++ compilers, for
421        example, SunCC recognizes above declarations as different functions and
422        generates broken code for hbpcre." I have changed the declarations to use
423        USPTR.
425    16. GNU libtool is named differently on some systems. The autogen.sh script now
426        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
427        (FreeBSD).
429    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
430        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
431        comment: "Figure out how to create a longlong from a string: strtoll and
432        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
433        instance, but it only takes 2 args instead of 3!"
435    18. A subtle bug concerned with back references has been fixed by a change of
436        specification, with a corresponding code fix. A pattern such as
437        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
438        refers, was giving matches when it shouldn't. For example, xa=xaaa would
439        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
440        same bug. Such groups have to be quantified to be useful, or contained
441        inside another quantified group. (If there's no repetition, the reference
442        can never match.) The problem arises because, having left the group and
443        moved on to the rest of the pattern, a later failure that backtracks into
444        the group uses the captured value from the final iteration of the group
445        rather than the correct earlier one. I have fixed this in PCRE by forcing
446        any group that contains a reference to itself to be an atomic group; that
447        is, there cannot be any backtracking into it once it has completed. This is
448        similar to recursive and subroutine calls.
451    Version 8.00 19-Oct-09
452    ----------------------
454    1.  The table for translating pcre_compile() error codes into POSIX error codes
455        was out-of-date, and there was no check on the pcre_compile() error code
456        being within the table. This could lead to an OK return being given in
457        error.
459    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
460        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
461        in a Windows environment.
463    3.  The pcregrep --count option prints the count for each file even when it is
464        zero, as does GNU grep. However, pcregrep was also printing all files when
465        --files-with-matches was added. Now, when both options are given, it prints
466        counts only for those files that have at least one match. (GNU grep just
467        prints the file name in this circumstance, but including the count seems
468        more useful - otherwise, why use --count?) Also ensured that the
469        combination -clh just lists non-zero counts, with no names.
471    4.  The long form of the pcregrep -F option was incorrectly implemented as
472        --fixed_strings instead of --fixed-strings. This is an incompatible change,
473        but it seems right to fix it, and I didn't think it was worth preserving
474        the old behaviour.
476    5.  The command line items --regex=pattern and --regexp=pattern were not
477        recognized by pcregrep, which required --regex pattern or --regexp pattern
478        (with a space rather than an '='). The man page documented the '=' forms,
479        which are compatible with GNU grep; these now work.
481    6.  No libpcreposix.pc file was created for pkg-config; there was just
482        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
484    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
485        when UCP support is not needed, by modifying the Python script that
486        generates it from Unicode data files. This should not matter if the module
487        is correctly used as a library, but I received one complaint about 50K of
488        unwanted data. My guess is that the person linked everything into his
489        program rather than using a library. Anyway, it does no harm.
491    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
492        was a minimum greater than 1 for a wide character in a possessive
493        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
494        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
495        character. Chaos in the form of incorrect output or a compiling loop could
496        result.
498    9.  The restrictions on what a pattern can contain when partial matching is
499        requested for pcre_exec() have been removed. All patterns can now be
500        partially matched by this function. In addition, if there are at least two
501        slots in the offset vector, the offset of the earliest inspected character
502        for the match and the offset of the end of the subject are set in them when
503        PCRE_ERROR_PARTIAL is returned.
505    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
506        synonymous with PCRE_PARTIAL, for backwards compatibility, and
507        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
508        and may be more useful for multi-segment matching.
510    11. Partial matching with pcre_exec() is now more intuitive. A partial match
511        used to be given if ever the end of the subject was reached; now it is
512        given only if matching could not proceed because another character was
513        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
514        string "Z", which now yields "no match" instead of "partial match". In the
515        case of pcre_dfa_exec(), "no match" is given if every matching path for the
516        final character ended with (*FAIL).
518    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
519        if the pattern had a "must contain" character that was already found in the
520        earlier partial match, unless partial matching was again requested. For
521        example, with the pattern /dog.(body)?/, the "must contain" character is
522        "g". If the first part-match was for the string "dog", restarting with
523        "sbody" failed. This bug has been fixed.
525    13. The string returned by pcre_dfa_exec() after a partial match has been
526        changed so that it starts at the first inspected character rather than the
527        first character of the match. This makes a difference only if the pattern
528        starts with a lookbehind assertion or \b or \B (\K is not supported by
529        pcre_dfa_exec()). It's an incompatible change, but it makes the two
530        matching functions compatible, and I think it's the right thing to do.
532    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
533        so that the demonstration program is easily available in environments where
534        PCRE has not been installed from source.
536    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
537        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
538        library.
540    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
541        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
542        is not the first non-POSIX option to be added. Clearly some people find
543        these options useful.
545    17. If a caller to the POSIX matching function regexec() passes a non-zero
546        value for nmatch with a NULL value for pmatch, the value of
547        nmatch is forced to zero.
549    18. RunGrepTest did not have a test for the availability of the -u option of
550        the diff command, as RunTest does. It now checks in the same way as
551        RunTest, and also checks for the -b option.
553    19. If an odd number of negated classes containing just a single character
554        interposed, within parentheses, between a forward reference to a named
555        subpattern and the definition of the subpattern, compilation crashed with
556        an internal error, complaining that it could not find the referenced
557        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
558        [The bug was that it was starting one character too far in when skipping
559        over the character class, thus treating the ] as data rather than
560        terminating the class. This meant it could skip too much.]
562    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
563        /g option in pcretest when the pattern contains \K, which makes it possible
564        to have an empty string match not at the start, even when the pattern is
565        anchored. Updated pcretest and pcredemo to use this option.
567    21. If the maximum number of capturing subpatterns in a recursion was greater
568        than the maximum at the outer level, the higher number was returned, but
569        with unset values at the outer level. The correct (outer level) value is
570        now given.
572    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
573        PCRE did not set those parentheses (unlike Perl). I have now found a way to
574        make it do so. The string so far is captured, making this feature
575        compatible with Perl.
577    23. The tests have been re-organized, adding tests 11 and 12, to make it
578        possible to check the Perl 5.10 features against Perl 5.10.
580    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
581        pattern matches a fixed length string. PCRE did not allow this; now it
582        does. Neither allows recursion.
584    25. I finally figured out how to implement a request to provide the minimum
585        length of subject string that was needed in order to match a given pattern.
586        (It was back references and recursion that I had previously got hung up
587        on.) This code has now been added to pcre_study(); it finds a lower bound
588        to the length of subject needed. It is not necessarily the greatest lower
589        bound, but using it to avoid searching strings that are too short does give
590        some useful speed-ups. The value is available to calling programs via
591        pcre_fullinfo().
593    26. While implementing 25, I discovered to my embarrassment that pcretest had
594        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
595        study optimizations had never been tested with that matching function.
596        Oops. What is worse, even when it was passed study data, there was a bug in
597        pcre_dfa_exec() that meant it never actually used it. Double oops. There
598        were also very few tests of studied patterns with pcre_dfa_exec().
600    27. If (?| is used to create subpatterns with duplicate numbers, they are now
601        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
602        on the other side of the coin, they are no longer allowed to have different
603        names, because these cannot be distinguished in PCRE, and this has caused
604        confusion. (This is a difference from Perl.)
606    28. When duplicate subpattern names are present (necessarily with different
607        numbers, as required by 27 above), and a test is made by name in a
608        conditional pattern, either for a subpattern having been matched, or for
609        recursion in such a pattern, all the associated numbered subpatterns are
610        tested, and the overall condition is true if the condition is true for any
611        one of them. This is the way Perl works, and is also more like the way
612        testing by number works.
615    Version 7.9 11-Apr-09
616    ---------------------
618    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
619        (pcretest), all targets were linked against these libraries. This included
620        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
621        libraries. This caused unwanted dependencies to be created. This problem
622        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
623        pcretest is linked with readline.
625    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
626        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
627        moved outside it again, because FALSE and TRUE are already defined in AIX,
628        but BOOL is not.
630    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
631        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
633    4.  The pcregrep documentation said spaces were inserted as well as colons (or
634        hyphens) following file names and line numbers when outputting matching
635        lines. This is not true; no spaces are inserted. I have also clarified the
636        wording for the --colour (or --color) option.
638    5.  In pcregrep, when --colour was used with -o, the list of matching strings
639        was not coloured; this is different to GNU grep, so I have changed it to be
640        the same.
642    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
643        each matching line was coloured. Now it goes on to look for further matches
644        of any of the test patterns, which is the same behaviour as GNU grep.
646    7.  A pattern that could match an empty string could cause pcregrep to loop; it
647        doesn't make sense to accept an empty string match in pcregrep, so I have
648        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
649        seems to be how GNU grep behaves.
651    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
652        start or after a newline", because the conditional assertion was not being
653        correctly handled. The rule now is that both the assertion and what follows
654        in the first alternative must satisfy the test.
656    9.  If auto-callout was enabled in a pattern with a conditional group whose
657        condition was an assertion, PCRE could crash during matching, both with
658        pcre_exec() and pcre_dfa_exec().
660    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
661        used for matching.
663    11. Unicode property support in character classes was not working for
664        characters (bytes) greater than 127 when not in UTF-8 mode.
666    12. Added the -M command line option to pcretest.
668    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
670    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
672    16. Added comments and documentation about mis-use of no_arg in the C++
673        wrapper.
675    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
676        from Martin Jerabek that uses macro names for all relevant character and
677        string constants.
679    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
680        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
681        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
682        these, but not everybody uses configure.
684    19. A conditional group that had only one branch was not being correctly
685        recognized as an item that could match an empty string. This meant that an
686        enclosing group might also not be so recognized, causing infinite looping
687        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
688        with the subject "ab", where knowledge that the repeated group can match
689        nothing is needed in order to break the loop.
691    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
692        exec(), but without supplying a callout function, matching went wrong.
694    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
695        leak if the size of the offset vector was greater than 30. When the vector
696        is smaller, the saved offsets during recursion go onto a local stack
697        vector, but for larger vectors malloc() is used. It was failing to free
698        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
699        error, in fact).
701    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
702        heapframe that is used only when UTF-8 support is enabled. This caused no
703        problem, but was untidy.
705    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
706        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
707        included within another project.
709    24. Steven Van Ingelgem's patches to add more options to the CMake support,
710        slightly modified by me:
712          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
713              not building pcregrep.
715          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
716              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
718    25. Forward references, both numeric and by name, in patterns that made use of
719        duplicate group numbers, could behave incorrectly or give incorrect errors,
720        because when scanning forward to find the reference group, PCRE was not
721        taking into account the duplicate group numbers. A pattern such as
722        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
724    26. Changed a few more instances of "const unsigned char *" to USPTR, making
725        the feature of a custom pointer more persuasive (as requested by a user).
727    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
728        pcretest.c, inside #ifndefs, because it seems they are sometimes already
729        pre-defined.
731    28. Added support for (*UTF8) at the start of a pattern.
733    29. Arrange for flags added by the "release type" setting in CMake to be shown
734        in the configuration summary.
737    Version 7.8 05-Sep-08
738    ---------------------
740    1.  Replaced UCP searching code with optimized version as implemented for Ad
741        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
742        stage table and inline lookup instead of a function, giving speed ups of 2
743        to 5 times on some simple patterns that I tested. Permission was given to
744        distribute the MultiStage2.py script that generates the tables (it's not in
745        the tarball, but is in the Subversion repository).
747    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
748        scripts.
750    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
751        a group with a zero qualifier. The result of the study could be incorrect,
752        or the function might crash, depending on the pattern.
754    4.  Caseless matching was not working for non-ASCII characters in back
755        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
756        It now works when Unicode Property Support is available.
758    5.  In pcretest, an escape such as \x{de} in the data was always generating
759        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
760        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
761        truncation.
763    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
765    7.  Added two (int) casts to pcregrep when printing the difference of two
766        pointers, in case they are 64-bit values.
768    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
769        test 2 if it fails.
771    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
772        and a #define of that name to empty if it is not externally set. This is to
773        allow users of MSVC to set it if necessary.
775    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
776        the convenience functions in the pcre_get.c source file.
778    11. An option change at the start of a pattern that had top-level alternatives
779        could cause overwriting and/or a crash. This command provoked a crash in
780        some environments:
782          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
784        This potential security problem was recorded as CVE-2008-2371.
786    12. For a pattern where the match had to start at the beginning or immediately
787        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
788        pcre_dfa_exec() could read past the end of the passed subject if there was
789        no match. To help with detecting such bugs (e.g. with valgrind), I modified
790        pcretest so that it places the subject at the end of its malloc-ed buffer.
792    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
793        exec() might read past the end of the data buffer in UTF-8 mode.
795    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
796        the data contained the byte 0x85 as part of a UTF-8 character within its
797        first line. This applied both to normal and DFA matching.
799    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
800        /^[^d]*?$/8 failed to match "abc".
802    16. Added a missing copyright notice to pcrecpp_internal.h.
804    17. Make it more clear in the documentation that values returned from
805        pcre_exec() in ovector are byte offsets, not character counts.
807    18. Tidied a few places to stop certain compilers from issuing warnings.
809    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
810        supplied by Stefan Weber. I made a further small update for 7.8 because
811        there is a change of source arrangements: the pcre_searchfuncs.c module is
812        replaced by pcre_ucd.c.
815    Version 7.7 07-May-08
816    ---------------------
818    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
819        a string to a long long, pretend we don't even have a long long." This is
820        done by checking for the strtoq, strtoll, and _strtoi64 functions.
822    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
823        pre-7.6 versions, which defined a global no_arg variable instead of putting
824        it in the RE class. (See also #8 below.)
826    3.  Remove a line of dead code, identified by coverity and reported by Nuno
827        Lopes.
829    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
831        (1) The include/exclude patterns were being applied to the whole pathnames
832            of files, instead of just to the final components.
834        (2) If there was more than one level of directory, the subdirectories were
835            skipped unless they satisfied the include/exclude conditions. This is
836            inconsistent with GNU grep (and could even be seen as contrary to the
837            pcregrep specification - which I improved to make it absolutely clear).
838            The action now is always to scan all levels of directory, and just
839            apply the include/exclude patterns to regular files.
841    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
842        --exclude_dir in the tests to avoid scanning .svn directories.
844    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
845        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
846        doesn't support NULs in patterns.
848    7.  Added some missing "const"s to declarations of static tables in
849        pcre_compile.c and pcre_dfa_exec.c.
851    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
852        caused by fix #2  above. (Subsequently also a second patch to fix the
853        first patch. And a third patch - this was a messy problem.)
855    9.  Applied Craig's patch to remove the use of push_back().
857    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
858        matching function regexec().
860    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
861        which, however, unlike Perl's \g{...}, are subroutine calls, not back
862        references. PCRE supports relative numbers with this syntax (I don't think
863        Oniguruma does).
865    12. Previously, a group with a zero repeat such as (...){0} was completely
866        omitted from the compiled regex. However, this means that if the group
867        was called as a subroutine from elsewhere in the pattern, things went wrong
868        (an internal error was given). Such groups are now left in the compiled
869        pattern, with a new opcode that causes them to be skipped at execution
870        time.
872    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
873        to the way PCRE behaves:
875        (a) A lone ] character is dis-allowed (Perl treats it as data).
877        (b) A back reference to an unmatched subpattern matches an empty string
878            (Perl fails the current match path).
880        (c) A data ] in a character class must be notated as \] because if the
881            first data character in a class is ], it defines an empty class. (In
882            Perl it is not possible to have an empty class.) The empty class []
883            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
884            The negative empty class [^] matches any one character, independently
885            of the DOTALL setting.
887    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
888        non-existent subpattern following a character class starting with ']' and
889        containing () gave an internal compiling error instead of "reference to
890        non-existent subpattern". Fortunately, when the pattern did exist, the
891        compiled code was correct. (When scanning forwards to check for the
892        existencd of the subpattern, it was treating the data ']' as terminating
893        the class, so got the count wrong. When actually compiling, the reference
894        was subsequently set up correctly.)
896    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
897        it was being rejected as not supported by pcre_dfa_exec(), even though
898        other assertions are supported. I have made pcre_dfa_exec() support
899        (*FAIL).
901    16. The implementation of 13c above involved the invention of a new opcode,
902        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
903        cannot be changed at match time, I realized I could make a small
904        improvement to matching performance by compiling OP_ALLANY instead of
905        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
906        on the OP_ANY path.
908    17. Compiling pcretest on Windows with readline support failed without the
909        following two fixes: (1) Make the unistd.h include conditional on
910        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
912    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
913        ncurses library to be included for pcretest when ReadLine support is
914        requested, but also to allow for it to be overridden. This patch came from
915        Daniel Bergström.
917    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
918        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
919        any errors with the current Unicode tables. Thanks to Peter Kankowski for
920        spotting this.
923    Version 7.6 28-Jan-08
924    ---------------------
926    1.  A character class containing a very large number of characters with
927        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
928        overflow.
930    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
931        HAVE_LONG_LONG is not defined.
933    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
934        bring them up to date and include new features. This patch includes:
936        - Fixed PH's badly added libz and libbz2 support.
937        - Fixed a problem with static linking.
938        - Added pcredemo. [But later removed - see 7 below.]
939        - Fixed dftables problem and added an option.
940        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
941            HAVE_LONG_LONG.
942        - Added readline support for pcretest.
943        - Added an listing of the option settings after cmake has run.
945    4.  A user submitted a patch to Makefile that makes it easy to create
946        "pcre.dll" under mingw when using Configure/Make. I added stuff to
947        Makefile.am that cause it to include this special target, without
948        affecting anything else. Note that the same mingw target plus all
949        the other distribution libraries and programs are now supported
950        when configuring with CMake (see 6 below) instead of with
951        Configure/Make.
953    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
954        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
955        exported in the Windows port". It has not yet been confirmed that the patch
956        solves the problem, but it does no harm.
958    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
959        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
960        with CMake, and also correct the comment about stack recursion.
962    7.  Remove the automatic building of pcredemo from the ./configure system and
963        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
964        of a program that users should build themselves after PCRE is installed, so
965        building it automatically is not really right. What is more, it gave
966        trouble in some build environments.
968    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
971    Version 7.5 10-Jan-08
972    ---------------------
974    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
975        values in parens when parsing an RE using the C++ wrapper."
977    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
978        Characters greater than 255 were excluded from the class instead of being
979        included.
981    3.  The same bug as (2) above applied to negated POSIX classes such as
982        [:^space:].
984    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
985        defined or documented. It seems to have been a typo for PCRE_STATIC, so
986        I have changed it.
988    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
989        first named subpattern) and a construct such as (?&a) would reference the
990        first named subpattern whose name started with "a" (in other words, the
991        length check was missing). Both these problems are fixed. "Subpattern name
992        expected" is now given for (?&) (a zero-length name), and this patch also
993        makes it give the same error for \k'' (previously it complained that that
994        was a reference to a non-existent subpattern).
996    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
997        this is right because (?- can be followed by option settings as well as by
998        digits. I have, however, made the messages clearer.
1000    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
1001        than the number used in the conditional) now cause a compile-time error.
1002        This is actually not compatible with Perl, which accepts such patterns, but
1003        treats the conditional as always being FALSE (as PCRE used to), but it
1004        seems to me that giving a diagnostic is better.
1006    8.  Change "alphameric" to the more common word "alphanumeric" in comments
1007        and messages.
1009    9.  Fix two occurrences of "backslash" in comments that should have been
1010        "backspace".
1012    10. Remove two redundant lines of code that can never be obeyed (their function
1013        was moved elsewhere).
1015    11. The program that makes PCRE's Unicode character property table had a bug
1016        which caused it to generate incorrect table entries for sequences of
1017        characters that have the same character type, but are in different scripts.
1018        It amalgamated them into a single range, with the script of the first of
1019        them. In other words, some characters were in the wrong script. There were
1020        thirteen such cases, affecting characters in the following ranges:
1022          U+002b0 - U+002c1
1023          U+0060c - U+0060d
1024          U+0061e - U+00612
1025          U+0064b - U+0065e
1026          U+0074d - U+0076d
1027          U+01800 - U+01805
1028          U+01d00 - U+01d77
1029          U+01d9b - U+01dbf
1030          U+0200b - U+0200f
1031          U+030fc - U+030fe
1032          U+03260 - U+0327f
1033          U+0fb46 - U+0fbb1
1034          U+10450 - U+1049d
1036    12. The -o option (show only the matching part of a line) for pcregrep was not
1037        compatible with GNU grep in that, if there was more than one match in a
1038        line, it showed only the first of them. It now behaves in the same way as
1039        GNU grep.
1041    13. If the -o and -v options were combined for pcregrep, it printed a blank
1042        line for every non-matching line. GNU grep prints nothing, and pcregrep now
1043        does the same. The return code can be used to tell if there were any
1044        non-matching lines.
1046    14. Added --file-offsets and --line-offsets to pcregrep.
1048    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
1049        infinitely looping recursion. The bug was that positive lookaheads were not
1050        being skipped when checking for a possible empty match (negative lookaheads
1051        and both kinds of lookbehind were skipped).
1053    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
1054        inclusion of <windows.h> to before rather than after the definition of
1055        INVALID_FILE_ATTRIBUTES (patch from David Byron).
1057    17. Specifying a possessive quantifier with a specific limit for a Unicode
1058        character property caused pcre_compile() to compile bad code, which led at
1059        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
1060        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
1061        caused the error; without that there was no problem.
1063    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
1065    19. Added --enable-pcretest-libreadline.
1067    20. In pcrecpp.cc, the variable 'count' was incremented twice in
1068        RE::GlobalReplace(). As a result, the number of replacements returned was
1069        double what it should be. I removed one of the increments, but Craig sent a
1070        later patch that removed the other one (the right fix) and added unit tests
1071        that check the return values (which was not done before).
1073    21. Several CMake things:
1075        (1) Arranged that, when cmake is used on Unix, the libraries end up with
1076            the names libpcre and libpcreposix, not just pcre and pcreposix.
1078        (2) The above change means that pcretest and pcregrep are now correctly
1079            linked with the newly-built libraries, not previously installed ones.
1083    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
1084        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
1085        UTF-8 newline character). The key issue is that the pattern starts .*;
1086        this means that the match must be either at the beginning, or after a
1087        newline. The bug was in the code for advancing after a failed match and
1088        checking that the new position followed a newline. It was not taking
1089        account of UTF-8 characters correctly.
1091    23. PCRE was behaving differently from Perl in the way it recognized POSIX
1092        character classes. PCRE was not treating the sequence [:...:] as a
1093        character class unless the ... were all letters. Perl, however, seems to
1094        allow any characters between [: and :], though of course it rejects as
1095        unknown any "names" that contain non-letters, because all the known class
1096        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
1097        for example, whereas PCRE did not - it did not recognize a POSIX character
1098        class. This seemed a bit dangerous, so the code has been changed to be
1099        closer to Perl. The behaviour is not identical to Perl, because PCRE will
1100        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
1101        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
1102        Perl does, and where it didn't before.
1104    24. Rewrite so as to remove the single use of %n from pcregrep because in some
1105        Windows environments %n is disabled by default.
1108    Version 7.4 21-Sep-07
1109    ---------------------
1111    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
1112        means that a class such as [\s] counted as "explicit reference to CR or
1113        LF". That isn't really right - the whole point of the change was to try to
1114        help when there was an actual mention of one of the two characters. So now
1115        the change happens only if \r or \n (or a literal CR or LF) character is
1116        encountered.
1118    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
1119        of both had grown to the point where there were only 3 bits left.
1120        Fortunately, there was spare space in the data structure, and so I have
1121        moved the internal flags into a new 16-bit field to free up more option
1122        bits.
1124    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
1125        but did not set the internal JCHANGED flag - either of these is enough to
1126        control the way the "get" function works - but the PCRE_INFO_JCHANGED
1127        facility is supposed to tell if (?J) was ever used, so now (?J) at the
1128        start sets both bits.
1130    4.  Added options (at build time, compile time, exec time) to change \R from
1131        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
1133    5.  doc/pcresyntax.html was missing from the distribution.
1135    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
1136        compatibility, even though it is no longer used.
1138    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
1139        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
1140        windows.h file is present (where different names are used). [This was
1141        reversed later after testing - see 16 below.]
1143    8.  Changed all #include <config.h> to #include "config.h". There were also
1144        some further <pcre.h> cases that I changed to "pcre.h".
1146    9.  When pcregrep was used with the --colour option, it missed the line ending
1147        sequence off the lines that it output.
1149    10. It was pointed out to me that arrays of string pointers cause lots of
1150        relocations when a shared library is dynamically loaded. A technique of
1151        using a single long string with a table of offsets can drastically reduce
1152        these. I have refactored PCRE in four places to do this. The result is
1153        dramatic:
1155          Originally:                          290
1156          After changing UCP table:            187
1157          After changing error message table:   43
1158          After changing table of "verbs"       36
1159          After changing table of Posix names   22
1161        Thanks to the folks working on Gregex for glib for this insight.
1163    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
1164        unicode-properties was also set.
1166    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
1168    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
1169        checked only for CRLF.
1171    14. Added casts to pcretest.c to avoid compiler warnings.
1173    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
1175    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
1176        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
1177        entirely. This removes changes made in 7 above.
1179    17. The CMake files have been updated, and there is now more information about
1180        building with CMake in the NON-UNIX-USE document.
1183    Version 7.3 28-Aug-07
1184    ---------------------
1186     1. In the rejigging of the build system that eventually resulted in 7.1, the
1187        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
1188        brackets there is not right, since it causes compilers to look for an
1189        installed pcre.h, not the version that is in the source that is being
1190        compiled (which of course may be different). I have changed it back to:
1192          #include "pcre.h"
1194        I have a vague recollection that the change was concerned with compiling in
1195        different directories, but in the new build system, that is taken care of
1196        by the VPATH setting the Makefile.
1198     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
1199        when the subject happened to end in the byte 0x85 (e.g. if the last
1200        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
1201        characters but of course it shouldn't be taken as a newline when it is part
1202        of another character. The bug was that, for an unlimited repeat of . in
1203        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
1204        characters when looking for a newline.
1206     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
1208     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
1209        in debug output.
1211     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
1212        long printing in the pcrecpp unittest when running under MinGW.
1214     6. ESC_K was left out of the EBCDIC table.
1216     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
1217        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
1218        limit also applies to "virtual nesting" when a pattern is recursive, and in
1219        this case 1000 isn't so big. I have been able to remove this limit at the
1220        expense of backing off one optimization in certain circumstances. Normally,
1221        when pcre_exec() would call its internal match() function recursively and
1222        immediately return the result unconditionally, it uses a "tail recursion"
1223        feature to save stack. However, when a subpattern that can match an empty
1224        string has an unlimited repetition quantifier, it no longer makes this
1225        optimization. That gives it a stack frame in which to save the data for
1226        checking that an empty string has been matched. Previously this was taken
1227        from the 1000-entry workspace that had been reserved. So now there is no
1228        explicit limit, but more stack is used.
1230     8. Applied Daniel's patches to solve problems with the import/export magic
1231        syntax that is required for Windows, and which was going wrong for the
1232        pcreposix and pcrecpp parts of the library. These were overlooked when this
1233        problem was solved for the main library.
1235     9. There were some crude static tests to avoid integer overflow when computing
1236        the size of patterns that contain repeated groups with explicit upper
1237        limits. As the maximum quantifier is 65535, the maximum group length was
1238        set at 30,000 so that the product of these two numbers did not overflow a
1239        32-bit integer. However, it turns out that people want to use groups that
1240        are longer than 30,000 bytes (though not repeat them that many times).
1241        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1242        made it possible to implement the integer overflow checks in a much more
1243        dynamic way, which I have now done. The artificial limitation on group
1244        length has been removed - we now have only the limit on the total length of
1245        the compiled pattern, which depends on the LINK_SIZE setting.
1247    10. Fixed a bug in the documentation for get/copy named substring when
1248        duplicate names are permitted. If none of the named substrings are set, the
1249        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1250        empty string.
1252    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1253        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1254        because the ] is interpreted as the first data character and the
1255        terminating ] is not found. PCRE has been made compatible with Perl in this
1256        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1257        cause memory overwriting.
1259    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1260        string has been matched (to stop an infinite loop). It was not recognizing
1261        a conditional subpattern that could match an empty string if that
1262        subpattern was within another subpattern. For example, it looped when
1263        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
1264        condition was not nested. This bug has been fixed.
1266    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1267        past the start of the subject in the presence of bytes with the top bit
1268        set, for example "\x8aBCD".
1270    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1271        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1273    14. Optimized (?!) to (*FAIL).
1275    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1276        This restricts code points to be within the range 0 to 0x10FFFF, excluding
1277        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1278        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1279        does: it's just the validity check that is more restrictive.
1281    16. Inserted checks for integer overflows during escape sequence (backslash)
1282        processing, and also fixed erroneous offset values for syntax errors during
1283        backslash processing.
1285    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1286        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1288    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1289        caused an overrun.
1291    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1292        something other than just ASCII characters) inside a group that had an
1293        unlimited repeat caused a loop at compile time (while checking to see
1294        whether the group could match an empty string).
1296    20. Debugging a pattern containing \p or \P could cause a crash. For example,
1297        [\P{Any}] did so. (Error in the code for printing property names.)
1299    21. An orphan \E inside a character class could cause a crash.
1301    22. A repeated capturing bracket such as (A)? could cause a wild memory
1302        reference during compilation.
1304    23. There are several functions in pcre_compile() that scan along a compiled
1305        expression for various reasons (e.g. to see if it's fixed length for look
1306        behind). There were bugs in these functions when a repeated \p or \P was
1307        present in the pattern. These operators have additional parameters compared
1308        with \d, etc, and these were not being taken into account when moving along
1309        the compiled data. Specifically:
1311        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1312            length.
1314        (b) An item such as \pL+ within a repeated group could cause crashes or
1315            loops.
1317        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1318            "reference to non-existent subpattern" error.
1320        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1322    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1323        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1325    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1327    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1328        character were causing crashes (broken optimization).
1330    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1331        \p or \P) caused a compile-time loop.
1333    28. More problems have arisen in unanchored patterns when CRLF is a valid line
1334        break. For example, the unstudied pattern [\r\n]A does not match the string
1335        "\r\nA" because change 7.0/46 below moves the current point on by two
1336        characters after failing to match at the start. However, the pattern \nA
1337        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
1338        the same is true. There doesn't seem any very clean way out of this, but
1339        what I have chosen to do makes the common cases work: PCRE now takes note
1340        of whether there can be an explicit match for \r or \n anywhere in the
1341        pattern, and if so, 7.0/46 no longer applies. As part of this change,
1342        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1343        pattern has explicit CR or LF references.
1345    29. Added (*CR) etc for changing newline setting at start of pattern.
1348    Version 7.2 19-Jun-07
1349    ---------------------
1351     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
1352        which is apparently normally available under Windows.
1354     2. Re-jig the pcregrep tests with different newline settings in an attempt
1355        to make them independent of the local environment's newline setting.
1357     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
1359     4. Some of the "internals" tests were previously cut out when the link size
1360        was not 2, because the output contained actual offsets. The recent new
1361        "Z" feature of pcretest means that these can be cut out, making the tests
1362        usable with all link sizes.
1364     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
1365        stack recursion. This gives a massive performance boost under BSD, but just
1366        a small improvement under Linux. However, it saves one field in the frame
1367        in all cases.
1369     6. Added more features from the forthcoming Perl 5.10:
1371        (a) (?-n) (where n is a string of digits) is a relative subroutine or
1372            recursion call. It refers to the nth most recently opened parentheses.
1374        (b) (?+n) is also a relative subroutine call; it refers to the nth next
1375            to be opened parentheses.
1377        (c) Conditions that refer to capturing parentheses can be specified
1378            relatively, for example, (?(-2)... or (?(+3)...
1380        (d) \K resets the start of the current match so that everything before
1381            is not part of it.
1383        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1385        (f) \g{name} is another synonym - part of Perl 5.10's unification of
1386            reference syntax.
1388        (g) (?| introduces a group in which the numbering of parentheses in each
1389            alternative starts with the same number.
1391        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
1393     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
1396     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
1397        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
1398        for detecting groups that can match an empty string.
1400     9. A pattern with a very large number of alternatives (more than several
1401        hundred) was running out of internal workspace during the pre-compile
1402        phase, where pcre_compile() figures out how much memory will be needed. A
1403        bit of new cunning has reduced the workspace needed for groups with
1404        alternatives. The 1000-alternative test pattern now uses 12 bytes of
1405        workspace instead of running out of the 4096 that are available.
1407    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
1409    11. Applied patch from Google to remove an optimization that didn't quite work.
1410        The report of the bug said:
1412          pcrecpp::RE("a*").FullMatch("aaa") matches, while
1413          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
1414          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
1416    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
1417        it matched the wrong number of bytes.
1420    Version 7.1 24-Apr-07
1421    ---------------------
1423     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
1424        that is more "standard", making use of automake and other Autotools. There
1425        is some re-arrangement of the files and adjustment of comments consequent
1426        on this.
1428     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
1429        for recursive directory scanning broke on some systems because the files
1430        are not scanned in any specific order and on different systems the order
1431        was different. A call to "sort" has been inserted into RunGrepTest for the
1432        approprate test as a short-term fix. In the longer term there may be an
1433        alternative.
1435     3. I had an email from Eric Raymond about problems translating some of PCRE's
1436        man pages to HTML (despite the fact that I distribute HTML pages, some
1437        people do their own conversions for various reasons). The problems
1438        concerned the use of low-level troff macros .br and .in. I have therefore
1439        removed all such uses from the man pages (some were redundant, some could
1440        be replaced by .nf/.fi pairs). The 132html script that I use to generate
1441        HTML has been updated to handle .nf/.fi and to complain if it encounters
1442        .br or .in.
1444     4. Updated comments in configure.ac that get placed in config.h.in and also
1445        arranged for config.h to be included in the distribution, with the name
1446        config.h.generic, for the benefit of those who have to compile without
1447        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
1449     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
1450        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
1451        makevp.bat for the current PCRE, using the additional files
1452        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
1454     6. A Windows user reported a minor discrepancy with test 2, which turned out
1455        to be caused by a trailing space on an input line that had got lost in his
1456        copy. The trailing space was an accident, so I've just removed it.
1458     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
1459        that is needed.
1461     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
1462        as "const" (a) because they are and (b) because it helps the PHP
1463        maintainers who have recently made a script to detect big data structures
1464        in the php code that should be moved to the .rodata section. I remembered
1465        to update Builducptable as well, so it won't revert if ucptable.h is ever
1466        re-created.
1468     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
1469        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
1470        order to be able to cut out the UTF-8 tables in the latter when UTF-8
1471        support is not required. This saves 1.5-2K of code, which is important in
1472        some applications.
1474        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
1475        so as not to refer to the tables, even though these functions will never be
1476        called when UTF-8 support is disabled. Otherwise there are problems with a
1477        shared library.
1479    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
1481        (a) It was defining its arguments as char * instead of void *.
1483        (b) It was assuming that all moves were upwards in memory; this was true
1484            a long time ago when I wrote it, but is no longer the case.
1486        The emulated memove() is provided for those environments that have neither
1487        memmove() nor bcopy(). I didn't think anyone used it these days, but that
1488        is clearly not the case, as these two bugs were recently reported.
1490    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
1491        and Detrail to create the HTML documentation, the .txt form of the man
1492        pages, and it removes trailing spaces from listed files. It also creates
1493        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
1494        case, it wraps all the #defines with #ifndefs. This script should be run
1495        before "make dist".
1497    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1498        with Unicode property support.
1500        (a) For a maximizing quantifier, if the two different cases of the
1501            character were of different lengths in their UTF-8 codings (there are
1502            some cases like this - I found 11), and the matching function had to
1503            back up over a mixture of the two cases, it incorrectly assumed they
1504            were both the same length.
1506        (b) When PCRE was configured to use the heap rather than the stack for
1507            recursion during matching, it was not correctly preserving the data for
1508            the other case of a UTF-8 character when checking ahead for a match
1509            while processing a minimizing repeat. If the check also involved
1510            matching a wide character, but failed, corruption could cause an
1511            erroneous result when trying to check for a repeat of the original
1512            character.
1514    13. Some tidying changes to the testing mechanism:
1516        (a) The RunTest script now detects the internal link size and whether there
1517            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1518            values substituted by "configure". (The RunGrepTest script already did
1519            this for UTF-8.) The configure.ac script no longer substitutes the
1520            relevant variables.
1522        (b) The debugging options /B and /D in pcretest show the compiled bytecode
1523            with length and offset values. This means that the output is different
1524            for different internal link sizes. Test 2 is skipped for link sizes
1525            other than 2 because of this, bypassing the problem. Unfortunately,
1526            there was also a test in test 3 (the locale tests) that used /B and
1527            failed for link sizes other than 2. Rather than cut the whole test out,
1528            I have added a new /Z option to pcretest that replaces the length and
1529            offset values with spaces. This is now used to make test 3 independent
1530            of link size. (Test 2 will be tidied up later.)
1532    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1533        segmentation fault instead of returning the appropriate error message.
1535    15. In multiline mode when the newline sequence was set to "any", the pattern
1536        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1537        This doesn't seem right; it now treats the CRLF combination as the line
1538        ending, and so does not match in that case. It's only a pattern such as ^$
1539        that would hit this one: something like ^ABC$ would have failed after \r
1540        and then tried again after \r\n.
1542    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1543        in an attempt to make files that differ only in their line terminators
1544        compare equal. This works on Linux.
1546    17. Under certain error circumstances pcregrep might try to free random memory
1547        as it exited. This is now fixed, thanks to valgrind.
1549    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1550        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1551        was because its rules for how to advance for /g after matching an empty
1552        string at the end of a line did not allow for this case. They now check for
1553        it specially.
1555    20. pcretest is supposed to handle patterns and data of any length, by
1556        extending its buffers when necessary. It was getting this wrong when the
1557        buffer for a data line had to be extended.
1559    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1560        CRLF as a newline sequence.
1562    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1563        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1564        I have nevertheless tidied it up.
1566    23. Added some casts to kill warnings from HP-UX ia64 compiler.
1568    24. Added a man page for pcre-config.
1571    Version 7.0 19-Dec-06
1572    ---------------------
1574     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1575        moving to gcc 4.1.1.
1577     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1578        sys/time.h, which is documented as needed for this function. It doesn't
1579        seem to matter on Linux, but it showed up on some releases of OS X.
1581     3. It seems that there are systems where bytes whose values are greater than
1582        127 match isprint() in the "C" locale. The "C" locale should be the
1583        default when a C program starts up. In most systems, only ASCII printing
1584        characters match isprint(). This difference caused the output from pcretest
1585        to vary, making some of the tests fail. I have changed pcretest so that:
1587        (a) When it is outputting text in the compiled version of a pattern, bytes
1588            other than 32-126 are always shown as hex escapes.
1590        (b) When it is outputting text that is a matched part of a subject string,
1591            it does the same, unless a different locale has been set for the match
1592            (using the /L modifier). In this case, it uses isprint() to decide.
1594     4. Fixed a major bug that caused incorrect computation of the amount of memory
1595        required for a compiled pattern when options that changed within the
1596        pattern affected the logic of the preliminary scan that determines the
1597        length. The relevant options are -x, and -i in UTF-8 mode. The result was
1598        that the computed length was too small. The symptoms of this bug were
1599        either the PCRE error "internal error: code overflow" from pcre_compile(),
1600        or a glibc crash with a message such as "pcretest: free(): invalid next
1601        size (fast)". Examples of patterns that provoked this bug (shown in
1602        pcretest format) are:
1604          /(?-x: )/x
1605          /(?x)(?-x: \s*#\s*)/
1606          /((?i)[\x{c0}])/8
1607          /(?i:[\x{c0}])/8
1609        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1610        is now done differently.
1612     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1613        wrapper classes; (b) implement a new function in the C++ scanner that is
1614        more efficient than the old way of doing things because it avoids levels of
1615        recursion in the regex matching; (c) add a paragraph to the documentation
1616        for the FullMatch() function.
1618     6. The escape sequence \n was being treated as whatever was defined as
1619        "newline". Not only was this contrary to the documentation, which states
1620        that \n is character 10 (hex 0A), but it also went horribly wrong when
1621        "newline" was defined as CRLF. This has been fixed.
1623     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1624        was being set to -1 for the "end of line" case (supposedly a value that no
1625        character can have). Though this value is never used (the check for end of
1626        line is "zero bytes in current character"), it caused compiler complaints.
1627        I've changed it to 0xffffffff.
1629     8. In pcre_version.c, the version string was being built by a sequence of
1630        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1631        string (as it is for production releases) called a macro with an empty
1632        argument. The C standard says the result of this is undefined. The gcc
1633        compiler treats it as an empty string (which was what was wanted) but it is
1634        reported that Visual C gives an error. The source has been hacked around to
1635        avoid this problem.
1637     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1638        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1639        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1640        of them did).
1642    10. Originally, pcretest opened its input and output without "b"; then I was
1643        told that "b" was needed in some environments, so it was added for release
1644        5.0 to both the input and output. (It makes no difference on Unix-like
1645        systems.) Later I was told that it is wrong for the input on Windows. I've
1646        now abstracted the modes into two macros, to make it easier to fiddle with
1647        them, and removed "b" from the input mode under Windows.
1649    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1651    12. Added -help and --help to pcretest as an official way of being reminded
1652        of the options.
1654    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1655        and pcrecpp.cc because they annoy compilers at high warning levels.
1657    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1659    15. Fixed an occurrence of == in configure.ac that should have been = (shell
1660        scripts are not C programs :-) and which was not noticed because it works
1661        on Linux.
1663    16. pcretest is supposed to handle any length of pattern and data line (as one
1664        line or as a continued sequence of lines) by extending its input buffer if
1665        necessary. This feature was broken for very long pattern lines, leading to
1666        a string of junk being passed to pcre_compile() if the pattern was longer
1667        than about 50K.
1669    17. I have done a major re-factoring of the way pcre_compile() computes the
1670        amount of memory needed for a compiled pattern. Previously, there was code
1671        that made a preliminary scan of the pattern in order to do this. That was
1672        OK when PCRE was new, but as the facilities have expanded, it has become
1673        harder and harder to keep it in step with the real compile phase, and there
1674        have been a number of bugs (see for example, 4 above). I have now found a
1675        cunning way of running the real compile function in a "fake" mode that
1676        enables it to compute how much memory it would need, while actually only
1677        ever using a few hundred bytes of working memory and without too many
1678        tests of the mode. This should make future maintenance and development
1679        easier. A side effect of this work is that the limit of 200 on the nesting
1680        depth of parentheses has been removed (though this was never a serious
1681        limitation, I suspect). However, there is a downside: pcre_compile() now
1682        runs more slowly than before (30% or more, depending on the pattern). I
1683        hope this isn't a big issue. There is no effect on runtime performance.
1685    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1686        newline (only possible for the last line of a file) and it was a
1687        pattern that set a locale (followed by /Lsomething), pcretest crashed.
1689    19. Added additional timing features to pcretest. (1) The -tm option now times
1690        matching only, not compiling. (2) Both -t and -tm can be followed, as a
1691        separate command line item, by a number that specifies the number of
1692        repeats to use when timing. The default is 50000; this gives better
1693        precision, but takes uncomfortably long for very large patterns.
1695    20. Extended pcre_study() to be more clever in cases where a branch of a
1696        subpattern has no definite first character. For example, (a*|b*)[cd] would
1697        previously give no result from pcre_study(). Now it recognizes that the
1698        first character must be a, b, c, or d.
1700    21. There was an incorrect error "recursive call could loop indefinitely" if
1701        a subpattern (or the entire pattern) that was being tested for matching an
1702        empty string contained only one non-empty item after a nested subpattern.
1703        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1704        incorrectly, because the \d was being skipped in the check.
1706    22. The pcretest program now has a new pattern option /B and a command line
1707        option -b, which is equivalent to adding /B to every pattern. This causes
1708        it to show the compiled bytecode, without the additional information that
1709        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1710        is the same as /B/I).
1712    23. A new optimization is now able automatically to treat some sequences such
1713        as a*b as a*+b. More specifically, if something simple (such as a character
1714        or a simple class like \d) has an unlimited quantifier, and is followed by
1715        something that cannot possibly match the quantified thing, the quantifier
1716        is automatically "possessified".
1718    24. A recursive reference to a subpattern whose number was greater than 39
1719        went wrong under certain circumstances in UTF-8 mode. This bug could also
1720        have affected the operation of pcre_study().
1722    25. Realized that a little bit of performance could be had by replacing
1723        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1725    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1727    27. Possessive quantifiers such as a++ were previously implemented by turning
1728        them into atomic groups such as ($>a+). Now they have their own opcodes,
1729        which improves performance. This includes the automatically created ones
1730        from 23 above.
1732    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1733        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1734        the first matched character to be a colon. This applied both to named and
1735        numbered groups.
1737    29. The ucpinternal.h header file was missing its idempotency #ifdef.
1739    30. I was sent a "project" file called libpcre.a.dev which I understand makes
1740        building PCRE on Windows easier, so I have included it in the distribution.
1742    31. There is now a check in pcretest against a ridiculously large number being
1743        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1744        loop, the loop is abandoned.
1746    32. Forward references to subpatterns in conditions such as (?(2)...) where
1747        subpattern 2 is defined later cause pcre_compile() to search forwards in
1748        the pattern for the relevant set of parentheses. This search went wrong
1749        when there were unescaped parentheses in a character class, parentheses
1750        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1752    33. "Subroutine" calls and backreferences were previously restricted to
1753        referencing subpatterns earlier in the regex. This restriction has now
1754        been removed.
1756    34. Added a number of extra features that are going to be in Perl 5.10. On the
1757        whole, these are just syntactic alternatives for features that PCRE had
1758        previously implemented using the Python syntax or my own invention. The
1759        other formats are all retained for compatibility.
1761        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1762            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1763            also .NET compatible.
1765        (b) A recursion or subroutine call to a named group can now be defined as
1766            (?&name) as well as (?P>name).
1768        (c) A backreference to a named group can now be defined as \k<name> or
1769            \k'name' as well as (?P=name). The new forms, as well as being in Perl
1770            5.10, are also .NET compatible.
1772        (d) A conditional reference to a named group can now use the syntax
1773            (?(<name>) or (?('name') as well as (?(name).
1775        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1776            groups (named and numbered) that are never evaluated inline, but can be
1777            called as "subroutines" from elsewhere. In effect, the DEFINE condition
1778            is always false. There may be only one alternative in such a group.
1780        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1781            as the simple (?(R). The condition is true only if the most recent
1782            recursion is that of the given number or name. It does not search out
1783            through the entire recursion stack.
1785        (g) The escape \gN or \g{N} has been added, where N is a positive or
1786            negative number, specifying an absolute or relative reference.
1788    35. Tidied to get rid of some further signed/unsigned compiler warnings and
1789        some "unreachable code" warnings.
1791    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1792        things, this adds five new scripts.
1794    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1795        There were also incompatibilities regarding the handling of \Q..\E inside
1796        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1797        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1799    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1800        matches an empty string, and forcibly breaks the loop. There were bugs in
1801        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
1802        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
1803        separate and independent bugs (that affected different cases) have been
1804        fixed.
1806    39. Refactored the code to abolish the use of different opcodes for small
1807        capturing bracket numbers. This is a tidy that I avoided doing when I
1808        removed the limit on the number of capturing brackets for 3.5 back in 2001.
1809        The new approach is not only tidier, it makes it possible to reduce the
1810        memory needed to fix the previous bug (38).
1812    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1813        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1814        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1815        mode.
1817    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1818        report.
1820    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1821        copy construction and assignment in the C++ wrapper.
1823    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1824        couple of bugs that could have given wrong results in the "--newline=crlf"
1825        case.
1827    44. Added a number of casts and did some reorganization of signed/unsigned int
1828        variables following suggestions from Dair Grant. Also renamed the variable
1829        "this" as "item" because it is a C++ keyword.
1831    45. Arranged for dftables to add
1833          #include "pcre_internal.h"
1835        to pcre_chartables.c because without it, gcc 4.x may remove the array
1836        definition from the final binary if PCRE is built into a static library and
1837        dead code stripping is activated.
1839    46. For an unanchored pattern, if a match attempt fails at the start of a
1840        newline sequence, and the newline setting is CRLF or ANY, and the next two
1841        characters are CRLF, advance by two characters instead of one.
1844    Version 6.7 04-Jul-06
1845    ---------------------
1847     1. In order to handle tests when input lines are enormously long, pcretest has
1848        been re-factored so that it automatically extends its buffers when
1849        necessary. The code is crude, but this _is_ just a test program. The
1850        default size has been increased from 32K to 50K.
1852     2. The code in pcre_study() was using the value of the re argument before
1853        testing it for NULL. (Of course, in any sensible call of the function, it
1854        won't be NULL.)
1856     3. The memmove() emulation function in pcre_internal.h, which is used on
1857        systems that lack both memmove() and bcopy() - that is, hardly ever -
1858        was missing a "static" storage class specifier.
1860     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1861        containing an extended class (one that cannot be represented by a bitmap
1862        because it contains high-valued characters or Unicode property items, e.g.
1863        [\pZ]). Almost always one would set UTF-8 mode when processing such a
1864        pattern, but PCRE should not loop if you do not (it no longer does).
1865        [Detail: two cases were found: (a) a repeated subpattern containing an
1866        extended class; (b) a recursive reference to a subpattern that followed a
1867        previous extended class. It wasn't skipping over the extended class
1868        correctly when UTF-8 mode was not set.]
1870     5. A negated single-character class was not being recognized as fixed-length
1871        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1872        compile error "lookbehind assertion is not fixed length".
1874     6. The RunPerlTest auxiliary script was showing an unexpected difference
1875        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1876        write a Perl script that can interpret lines of an input file either as
1877        byte characters or as UTF-8, which is what "perltest" was being required to
1878        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1879        can't do is switch easily at run time between having the "use utf8;" pragma
1880        or not. In the end, I fudged it by using the RunPerlTest script to insert
1881        "use utf8;" explicitly for the UTF-8 tests.
1883     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1884        the end of the subject string, contrary to the documentation and to what
1885        Perl does. This was true of both matching functions. Now it matches only at
1886        the start of the subject and immediately after *internal* newlines.
1888     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1889        a pointer to an int instead of a pointer to an unsigned long int. This
1890        caused problems on 64-bit systems.
1892     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1893        instance of the 'standard' template library not being so standard".
1895    10. There was no check on the number of named subpatterns nor the maximum
1896        length of a subpattern name. The product of these values is used to compute
1897        the size of the memory block for a compiled pattern. By supplying a very
1898        long subpattern name and a large number of named subpatterns, the size
1899        computation could be caused to overflow. This is now prevented by limiting
1900        the length of names to 32 characters, and the number of named subpatterns
1901        to 10,000.
1903    11. Subpatterns that are repeated with specific counts have to be replicated in
1904        the compiled pattern. The size of memory for this was computed from the
1905        length of the subpattern and the repeat count. The latter is limited to
1906        65535, but there was no limit on the former, meaning that integer overflow
1907        could in principle occur. The compiled length of a repeated subpattern is
1908        now limited to 30,000 bytes in order to prevent this.
1910    12. Added the optional facility to have named substrings with the same name.
1912    13. Added the ability to use a named substring as a condition, using the
1913        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1914        are numbers (not recommended). Forward references are permitted.
1916    14. Added forward references in named backreferences (if you see what I mean).
1918    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1919        pattern could run off the end of the subject. For example, the pattern
1920        "(?s)(.{1,5})"8 did this with the subject "ab".
1922    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1923        PCRE_CASELESS was set when matching characters that were quantified with ?
1924        or *.
1926    17. A character class other than a single negated character that had a minimum
1927        but no maximum quantifier - for example [ab]{6,} - was not handled
1928        correctly by pce_dfa_exec(). It would match only one character.
1930    18. A valid (though odd) pattern that looked like a POSIX character
1931        class but used an invalid character after [ (for example [[,abc,]]) caused
1932        pcre_compile() to give the error "Failed: internal error: code overflow" or
1933        in some cases to crash with a glibc free() error. This could even happen if
1934        the pattern terminated after [[ but there just happened to be a sequence of
1935        letters, a binary zero, and a closing ] in the memory that followed.
1937    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1938        over the years. Originally (before any Unicode support), just the bottom 8
1939        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1940        output from "man perlunicode" includes this:
1942          The regular expression compiler produces polymorphic opcodes.  That
1943          is, the pattern adapts to the data and automatically switches to
1944          the Unicode character scheme when presented with Unicode data--or
1945          instead uses a traditional byte scheme when presented with byte
1946          data.
1948        Sadly, a wide octal escape does not cause a switch, and in a string with
1949        no other multibyte characters, these octal escapes are treated as before.
1950        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
1951        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1952        Unicode string.
1954        I have not perpetrated such confusion in PCRE. Up till now, it took just
1955        the bottom 8 bits, as in old Perl. I have now made octal escapes with
1956        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1957        translate to the appropriate multibyte character.
1959    29. Applied some refactoring to reduce the number of warnings from Microsoft
1960        and Borland compilers. This has included removing the fudge introduced
1961        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1962        a warning about an unused variable.
1964    21. PCRE has not included VT (character 0x0b) in the set of whitespace
1965        characters since release 4.0, because Perl (from release 5.004) does not.
1966        [Or at least, is documented not to: some releases seem to be in conflict
1967        with the documentation.] However, when a pattern was studied with
1968        pcre_study() and all its branches started with \s, PCRE still included VT
1969        as a possible starting character. Of course, this did no harm; it just
1970        caused an unnecessary match attempt.
1972    22. Removed a now-redundant internal flag bit that recorded the fact that case
1973        dependency changed within the pattern. This was once needed for "required
1974        byte" processing, but is no longer used. This recovers a now-scarce options
1975        bit. Also moved the least significant internal flag bit to the most-
1976        significant bit of the word, which was not previously used (hangover from
1977        the days when it was an int rather than a uint) to free up another bit for
1978        the future.
1980    23. Added support for CRLF line endings as well as CR and LF. As well as the
1981        default being selectable at build time, it can now be changed at runtime
1982        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1983        specify that it is scanning data with non-default line endings.
1985    24. Changed the definition of CXXLINK to make it agree with the definition of
1986        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1988    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1989        recursions. This makes a big different to stack usage for some patterns.
1991    26. If a subpattern containing a named recursion or subroutine reference such
1992        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1993        the space required for the compiled pattern went wrong and gave too small a
1994        value. Depending on the environment, this could lead to "Failed: internal
1995        error: code overflow at offset 49" or "glibc detected double free or
1996        corruption" errors.
1998    27. Applied patches from Google (a) to support the new newline modes and (b) to
1999        advance over multibyte UTF-8 characters in GlobalReplace.
2001    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
2002        difference for some implementation of PCRE in some Windows version.
2004    29. Added some extra testing facilities to pcretest:
2006        \q<number>   in a data line sets the "match limit" value
2007        \Q<number>   in a data line sets the "match recursion limt" value
2008        -S <number>  sets the stack size, where <number> is in megabytes
2010        The -S option isn't available for Windows.
2013    Version 6.6 06-Feb-06
2014    ---------------------
2016     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
2017        in pcreposix.h. I have copied the definition from pcre.h.
2019     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
2020        because pcre.h is no longer a built file.
2022     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
2023        not normally included in the compiled code.
2026    Version 6.5 01-Feb-06
2027    ---------------------
2029     1. When using the partial match feature with pcre_dfa_exec(), it was not
2030        anchoring the second and subsequent partial matches at the new starting
2031        point. This could lead to incorrect results. For example, with the pattern
2032        /1234/, partially matching against "123" and then "a4" gave a match.
2034     2. Changes to pcregrep:
2036        (a) All non-match returns from pcre_exec() were being treated as failures
2037            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
2038            error message is output. Some extra information is given for the
2039            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
2040            probably the only errors that are likely to be caused by users (by
2041            specifying a regex that has nested indefinite repeats, for instance).
2042            If there are more than 20 of these errors, pcregrep is abandoned.
2044        (b) A binary zero was treated as data while matching, but terminated the
2045            output line if it was written out. This has been fixed: binary zeroes
2046            are now no different to any other data bytes.
2048        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
2049            used to set a locale for matching. The --locale=xxxx long option has
2050            been added (no short equivalent) to specify a locale explicitly on the
2051            pcregrep command, overriding the environment variables.
2053        (d) When -B was used with -n, some line numbers in the output were one less
2054            than they should have been.
2056        (e) Added the -o (--only-matching) option.
2058        (f) If -A or -C was used with -c (count only), some lines of context were
2059            accidentally printed for the final match.
2061        (g) Added the -H (--with-filename) option.
2063        (h) The combination of options -rh failed to suppress file names for files
2064            that were found from directory arguments.
2066        (i) Added the -D (--devices) and -d (--directories) options.
2068        (j) Added the -F (--fixed-strings) option.
2070        (k) Allow "-" to be used as a file name for -f as well as for a data file.
2072        (l) Added the --colo(u)r option.
2074        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
2075            is not present by default.
2077     3. A nasty bug was discovered in the handling of recursive patterns, that is,
2078        items such as (?R) or (?1), when the recursion could match a number of
2079        alternatives. If it matched one of the alternatives, but subsequently,
2080        outside the recursion, there was a failure, the code tried to back up into
2081        the recursion. However, because of the way PCRE is implemented, this is not
2082        possible, and the result was an incorrect result from the match.
2084        In order to prevent this happening, the specification of recursion has
2085        been changed so that all such subpatterns are automatically treated as
2086        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
2088     4. I had overlooked the fact that, in some locales, there are characters for
2089        which isalpha() is true but neither isupper() nor islower() are true. In
2090        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
2091        and ordfeminine) are like this. This affected the treatment of \w and \W
2092        when they appeared in character classes, but not when they appeared outside
2093        a character class. The bit map for "word" characters is now created
2094        separately from the results of isalnum() instead of just taking it from the
2095        upper, lower, and digit maps. (Plus the underscore character, of course.)
2097     5. The above bug also affected the handling of POSIX character classes such as
2098        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
2099        permanent tables. Instead, the bit maps for such a class were previously
2100        created as the appropriate unions of the upper, lower, and digit bitmaps.
2101        Now they are created by subtraction from the [[:word:]] class, which has
2102        its own bitmap.
2104     6. The [[:blank:]] character class matches horizontal, but not vertical space.
2105        It is created by subtracting the vertical space characters (\x09, \x0a,
2106        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
2107        subtraction was done in the overall bitmap for a character class, meaning
2108        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
2109        be recognized. This bug has been fixed.
2111     7. Patches from the folks at Google:
2113          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
2114          real life, but is still worth protecting against".
2116          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
2117          regular expressions".
2119          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
2120          have it.
2122          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
2123          "configure" and the latter not, in order to fix a problem somebody had
2124          with compiling the Arg class on HP-UX.
2126          (e) Improve the error-handling of the C++ wrapper a little bit.
2128          (f) New tests for checking recursion limiting.
2130     8. The pcre_memmove() function, which is used only if the environment does not
2131        have a standard memmove() function (and is therefore rarely compiled),
2132        contained two bugs: (a) use of int instead of size_t, and (b) it was not
2133        returning a result (though PCRE never actually uses the result).
2135     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
2136        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
2137        returned instead of calling malloc() with an overflowing number that would
2138        most likely cause subsequent chaos.
2140    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
2142    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
2143        with this option is matched, the nmatch and pmatch options of regexec() are
2144        ignored.
2146    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
2147        provided in case anyone wants to the the POSIX interface with UTF-8
2148        strings.
2150    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
2151        C++ linking (needed for some HP-UX environments).
2153    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
2154        (unused parameter) and in the pcre_printint() function (omitted "default"
2155        switch label when the default is to do nothing).
2157    15. Added some code to make it possible, when PCRE is compiled as a C++
2158        library, to replace subject pointers for pcre_exec() with a smart pointer
2159        class, thus making it possible to process discontinuous strings.
2161    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
2162        much the same function. They were added by different people who were trying
2163        to make PCRE easy to compile on non-Unix systems. It has been suggested
2164        that PCRE_EXPORT be abolished now that there is more automatic apparatus
2165        for compiling on Windows systems. I have therefore replaced it with
2166        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
2167        defaults to "extern" for C or "extern C" for C++, which works fine on
2168        Unix-like systems. It is now possible to override the value of PCRE_DATA_
2169        SCOPE with something explicit in config.h. In addition:
2171        (a) pcreposix.h still had just "extern" instead of either of these macros;
2172            I have replaced it with PCRE_DATA_SCOPE.
2174        (b) Functions such as _pcre_xclass(), which are internal to the library,
2175            but external in the C sense, all had PCRE_EXPORT in their definitions.
2176            This is apparently wrong for the Windows case, so I have removed it.
2177            (It makes no difference on Unix-like systems.)
2179    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
2180        of recursive calls to match(). This is different to MATCH_LIMIT because
2181        that limits the total number of calls to match(), not all of which increase
2182        the depth of recursion. Limiting the recursion depth limits the amount of
2183        stack (or heap if NO_RECURSE is set) that is used. The default can be set
2184        when PCRE is compiled, and changed at run time. A patch from Google adds
2185        this functionality to the C++ interface.
2187    18. Changes to the handling of Unicode character properties:
2189        (a) Updated the table to Unicode 4.1.0.
2191        (b) Recognize characters that are not in the table as "Cn" (undefined).
2193        (c) I revised the way the table is implemented to a much improved format
2194            which includes recognition of ranges. It now supports the ranges that
2195            are defined in UnicodeData.txt, and it also amalgamates other
2196            characters into ranges. This has reduced the number of entries in the
2197            table from around 16,000 to around 3,000, thus reducing its size
2198            considerably. I realized I did not need to use a tree structure after
2199            all - a binary chop search is just as efficient. Having reduced the
2200            number of entries, I extended their size from 6 bytes to 8 bytes to
2201            allow for more data.
2203        (d) Added support for Unicode script names via properties such as \p{Han}.
2205    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
2206        matching that character.
2208    20. When matching a repeated Unicode property with a minimum greater than zero,
2209        (for example \pL{2,}), PCRE could look past the end of the subject if it
2210        reached it while seeking the minimum number of characters. This could
2211        happen only if some of the characters were more than one byte long, because
2212        there is a check for at least the minimum number of bytes.
2214    21. Refactored the implementation of \p and \P so as to be more general, to
2215        allow for more different types of property in future. This has changed the
2216        compiled form incompatibly. Anybody with saved compiled patterns that use
2217        \p or \P will have to recompile them.
2219    22. Added "Any" and "L&" to the supported property types.
2221    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
2222        but give a compile time error if the value is greater than 0xff.
2224    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
2225        accidentally not being installed or uninstalled.
2227    25. The pcre.h file was built from pcre.h.in, but the only changes that were
2228        made were to insert the current release number. This seemed silly, because
2229        it made things harder for people building PCRE on systems that don't run
2230        "configure". I have turned pcre.h into a distributed file, no longer built
2231        by "configure", with the version identification directly included. There is
2232        no longer a pcre.h.in file.
2234        However, this change necessitated a change to the pcre-config script as
2235        well. It is built from pcre-config.in, and one of the substitutions was the
2236        release number. I have updated configure.ac so that ./configure now finds
2237        the release number by grepping pcre.h.
2239    26. Added the ability to run the tests under valgrind.
2242    Version 6.4 05-Sep-05
2243    ---------------------
2245     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
2246        "--" to be printed when multiple files were scanned, even when none of the
2247        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
2248        consider it to be a bug, and have restored the previous behaviour.
2250     2. A couple of code tidies to get rid of compiler warnings.
2252     3. The pcretest program used to cheat by referring to symbols in the library
2253        whose names begin with _pcre_. These are internal symbols that are not
2254        really supposed to be visible externally, and in some environments it is
2255        possible to suppress them. The cheating is now confined to including
2256        certain files from the library's source, which is a bit cleaner.
2258     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
2259        file's purpose clearer.
2261     5. Reorganized pcre_ucp_findchar().
2264    Version 6.3 15-Aug-05
2265    ---------------------
2267     1. The file libpcre.pc.in did not have general read permission in the tarball.
2269     2. There were some problems when building without C++ support:
2271        (a) If C++ support was not built, "make install" and "make test" still
2272            tried to test it.
2274        (b) There were problems when the value of CXX was explicitly set. Some
2275            changes have been made to try to fix these, and ...
2277        (c) --disable-cpp can now be used to explicitly disable C++ support.
2279        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
2280            backslash in a target when C++ was disabled. This confuses some
2281            versions of "make", apparently. Using an intermediate variable solves
2282            this. (Same for CPP_LOBJ.)
2284     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
2285        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
2286        necessary on certain architectures.
2288     4. Added a setting of -export-symbols-regex to the link command to remove
2289        those symbols that are exported in the C sense, but actually are local
2290        within the library, and not documented. Their names all begin with
2291        "_pcre_". This is not a perfect job, because (a) we have to except some
2292        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
2293        available (and never for static libraries). I have made a note to try to
2294        find a way round (a) in the future.
2297  Version 6.2 01-Aug-05  Version 6.2 01-Aug-05
2298  ---------------------  ---------------------

Removed from v.81  
changed lines
  Added in v.587

  ViewVC Help
Powered by ViewVC 1.1.5