ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 35 by nigel, Sat Feb 24 21:39:05 2007 UTC revision 609 by ph10, Wed Jun 15 18:09:23 2011 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
4    Version 8.13 30-Apr-2011
5    ------------------------
7    1.  The Unicode data tables have been updated to Unicode 6.0.0.
9    2.  Two minor typos in pcre_internal.h have been fixed.
11    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
12        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
13        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
15    4.  There were a number of related bugs in the code for matching backrefences
16        caselessly in UTF-8 mode when codes for the characters concerned were
17        different numbers of bytes. For example, U+023A and U+2C65 are an upper
18        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
19        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
20        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
21        2-byte code at the end of the subject (it thought there wasn't enough data
22        left).
24    5.  Comprehensive information about what went wrong is now returned by
25        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
26        as the output vector has at least 2 elements. The offset of the start of
27        the failing character and a reason code are placed in the vector.
29    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
30        now returned is for the first byte of the failing character, instead of the
31        last byte inspected. This is an incompatible change, but I hope it is small
32        enough not to be a problem. It makes the returned offset consistent with
33        pcre_exec() and pcre_dfa_exec().
35    7.  pcretest now gives a text phrase as well as the error number when
36        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
37        failure, the offset and reason code are output.
39    8.  When \R was used with a maximizing quantifier it failed to skip backwards
40        over a \r\n pair if the subsequent match failed. Instead, it just skipped
41        back over a single character (\n). This seems wrong (because it treated the
42        two characters as a single entity when going forwards), conflicts with the
43        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
44        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
45        has been changed.
47    9.  Some internal refactoring has changed the processing so that the handling
48        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
49        time (the PCRE_DOTALL option was changed this way some time ago: version
50        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
51        which was always a bit of a fudge. It also means that there is one less
52        argument for the match() function, which reduces its stack requirements
53        slightly. This change also fixes an incompatibility with Perl: the pattern
54        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
56    10. More internal refactoring has drastically reduced the number of recursive
57        calls to match() for possessively repeated groups such as (abc)++ when
58        using pcre_exec().
60    11. While implementing 10, a number of bugs in the handling of groups were
61        discovered and fixed:
63        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
64        (a|)*(?1) gave a compile-time internal error.
65        ((a|)+)+  did not notice that the outer group could match an empty string.
66        (^a|^)+   was not marked as anchored.
67        (.*a|.*)+ was not marked as matching at start or after a newline.
69    12. Yet more internal refactoring has removed another argument from the match()
70        function. Special calls to this function are now indicated by setting a
71        value in a variable in the "match data" data block.
73    13. Be more explicit in pcre_study() instead of relying on "default" for
74        opcodes that mean there is no starting character; this means that when new
75        ones are added and accidentally left out of pcre_study(), testing should
76        pick them up.
78    14. The -s option of pcretest has been documented for ages as being an old
79        synonym of -m (show memory usage). I have changed it to mean "force study
80        for every regex", that is, assume /S for every regex. This is similar to -i
81        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
82        using it. It makes it easier to run collection of tests with study enabled,
83        and thereby test pcre_study() more easily.
85    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
86        restoration of the capturing data to the outer values was not happening
87        correctly.
89    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
90        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
91        pattern had matched an empty string, and so incorrectly returned a no
92        match.
94    17. There was optimizing code for the last branch of non-capturing parentheses,
95        and also for the obeyed branch of a conditional subexpression, which used
96        tail recursion to cut down on stack usage. Unfortunately, not that there is
97        the possibility of (*THEN) occurring in these branches, tail recursion is
98        no longer possible because the return has to be checked for (*THEN). These
99        two optimizations have therefore been removed.
102    Version 8.12 15-Jan-2011
103    ------------------------
105    1.  Fixed some typos in the markup of the man pages, and wrote a script that
106        checks for such things as part of the documentation building process.
108    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
109        --match-limit and --recursion-limit options (added for 8.11). In
110        particular, this made one of the standard tests fail. (The integer value
111        went into the wrong half of a long int.)
113    3.  If the --colour option was given to pcregrep with -v (invert match), it
114        did strange things, either producing crazy output, or crashing. It should,
115        of course, ignore a request for colour when reporting lines that do not
116        match.
118    4.  Another pcregrep bug caused similar problems if --colour was specified with
119        -M (multiline) and the pattern match finished with a line ending.
121    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
122        matched in multiline mode, the following line was shown as part of the
123        match. This seems wrong, so I have changed it.
125    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
126        the check for further matches in the same line (so they could be coloured)
127        to overrun the end of the current line. If another match was found, it was
128        incorrectly shown (and then shown again when found in the next line).
130    7.  If pcregrep was compiled under Windows, there was a reference to the
131        function pcregrep_exit() before it was defined. I am assuming this was
132        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
133        reported by a user. I've moved the definition above the reference.
136    Version 8.11 10-Dec-2010
137    ------------------------
139    1.  (*THEN) was not working properly if there were untried alternatives prior
140        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
141        backtracked to try for "b" instead of moving to the next alternative branch
142        at the same level (in this case, to look for "c"). The Perl documentation
143        is clear that when (*THEN) is backtracked onto, it goes to the "next
144        alternative in the innermost enclosing group".
146    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
147        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
148        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
149        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
150        (*THEN).
152    3.  If \s appeared in a character class, it removed the VT character from
153        the class, even if it had been included by some previous item, for example
154        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
155        of \s, but is part of the POSIX "space" class.)
157    4.  A partial match never returns an empty string (because you can always
158        match an empty string at the end of the subject); however the checking for
159        an empty string was starting at the "start of match" point. This has been
160        changed to the "earliest inspected character" point, because the returned
161        data for a partial match starts at this character. This means that, for
162        example, /(?<=abc)def/ gives a partial match for the subject "abc"
163        (previously it gave "no match").
165    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
166        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
167        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
168        has an implication that the given string is incomplete (because a partial
169        match is preferred over a full match). For this reason, these items now
170        give a partial match in this situation. [Aside: previously, the one case
171        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
172        match rather than a full match, which was wrong by the old rules, but is
173        now correct.]
175    6.  There was a bug in the handling of #-introduced comments, recognized when
176        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
177        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
178        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
179        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
180        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
181        places in pcre_compile().
183    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
184        comments when looking ahead for named forward references to subpatterns,
185        the only newline sequence it recognized was NL. It now handles newlines
186        according to the set newline convention.
188    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
189        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
190        cater for a lack of strerror(). These oversights have been fixed.
192    9.  Added --match-limit and --recursion-limit to pcregrep.
194    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
196    11. When the -o option was used, pcregrep was setting a return code of 1, even
197        when matches were found, and --line-buffered was not being honoured.
199    12. Added an optional parentheses number to the -o and --only-matching options
200        of pcregrep.
202    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
203        can match an empty string. The code to do it in pcretest and pcredemo
204        needed fixing:
206        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
207            only one byte after an empty string match just before CRLF (this case
208            just got forgotten; "any" and "anycrlf" were OK).
210        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
211            mode when an empty string match preceded an ASCII character followed by
212            a non-ASCII character. (The code for advancing by one character rather
213            than one byte was nonsense.)
215        (c) The pcredemo.c sample program did not have any code at all to handle
216            the cases when CRLF is a valid newline sequence.
218    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
219        as a starting offset was within the subject string. There is now a new
220        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
221        negative or greater than the length of the string. In order to test this,
222        pcretest is extended to allow the setting of negative starting offsets.
224    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
225        starting offset points to the beginning of a UTF-8 character was
226        unnecessarily clumsy. I tidied it up.
228    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
229        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
231    17. Nobody had reported that the --include_dir option, which was added in
232        release 7.7 should have been called --include-dir (hyphen, not underscore)
233        for compatibility with GNU grep. I have changed it to --include-dir, but
234        left --include_dir as an undocumented synonym, and the same for
235        --exclude-dir, though that is not available in GNU grep, at least as of
236        release 2.5.4.
238    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
239        characters from a string of bytes) have been redefined so as not to use
240        loops, in order to improve performance in some environments. At the same
241        time, I abstracted some of the common code into auxiliary macros to save
242        repetition (this should not affect the compiled code).
244    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
245        compile-time error is now given if \c is not followed by an ASCII
246        character, that is, a byte less than 128. (In EBCDIC mode, the code is
247        different, and any byte value is allowed.)
249    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
250        START_OPTIMIZE option, which is now allowed at compile time - but just
251        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
252        to pcregrep and other applications that have no direct access to PCRE
253        options. The new /Y option in pcretest sets this option when calling
254        pcre_compile().
256    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
257        back references. Groups containing recursive back references were forced to
258        be atomic by that change, but in the case of named groups, the amount of
259        memory required was incorrectly computed, leading to "Failed: internal
260        error: code overflow". This has been fixed.
262    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
263        pcretest.c, to avoid build problems in some Borland environments.
266    Version 8.10 25-Jun-2010
267    ------------------------
269    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
270        THEN.
272    2.  (*ACCEPT) was not working when inside an atomic group.
274    3.  Inside a character class, \B is treated as a literal by default, but
275        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
276        causes the error). The code is unchanged, but I tidied the documentation.
278    4.  Inside a character class, PCRE always treated \R and \X as literals,
279        whereas Perl faults them if its -w option is set. I have changed PCRE so
280        that it faults them when PCRE_EXTRA is set.
282    5.  Added support for \N, which always matches any character other than
283        newline. (It is the same as "." when PCRE_DOTALL is not set.)
285    6.  When compiling pcregrep with newer versions of gcc which may have
286        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
287        declared with attribute warn_unused_result" were given. Just casting the
288        result to (void) does not stop the warnings; a more elaborate fudge is
289        needed. I've used a macro to implement this.
291    7.  Minor change to pcretest.c to avoid a compiler warning.
293    8.  Added four artifical Unicode properties to help with an option to make
294        \s etc use properties (see next item). The new properties are: Xan
295        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
297    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
298        use Unicode properties. (*UCP) at the start of a pattern can be used to set
299        this option. Modified pcretest to add /W to test this facility. Added
300        REG_UCP to make it available via the POSIX interface.
302    10. Added --line-buffered to pcregrep.
304    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
305        studied, and the match started with a letter with a code point greater than
306        127 whose first byte was different to the first byte of the other case of
307        the letter, the other case of this starting letter was not recognized
308        (#976).
310    12. If a pattern that was studied started with a repeated Unicode property
311        test, for example, \p{Nd}+, there was the theoretical possibility of
312        setting up an incorrect bitmap of starting bytes, but fortunately it could
313        not have actually happened in practice until change 8 above was made (it
314        added property types that matched character-matching opcodes).
316    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
317        possible starting bytes for non-anchored patterns.
319    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
320        \R, and also a number of cases that involve Unicode properties, both
321        explicit and implicit when PCRE_UCP is set.
323    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
324        input, it could crash or give wrong results if characters with values
325        greater than 0xc0 were present in the subject string. (Detail: it assumed
326        UTF-8 input when processing these items.)
328    16. Added a lot of (int) casts to avoid compiler warnings in systems where
329        size_t is 64-bit (#991).
331    17. Added a check for running out of memory when PCRE is compiled with
332        --disable-stack-for-recursion (#990).
334    18. If the last data line in a file for pcretest does not have a newline on
335        the end, a newline was missing in the output.
337    19. The default pcre_chartables.c file recognizes only ASCII characters (values
338        less than 128) in its various bitmaps. However, there is a facility for
339        generating tables according to the current locale when PCRE is compiled. It
340        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
341        characters, are recognized by isspace() and therefore were getting set in
342        these tables, and indeed these tables seem to approximate to ISO 8859. This
343        caused a problem in UTF-8 mode when pcre_study() was used to create a list
344        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
345        which of course cannot start UTF-8 characters. I have changed the code so
346        that only real ASCII characters (less than 128) and the correct starting
347        bytes for UTF-8 encodings are set for characters greater than 127 when in
348        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
349        altogether.)
351    20. Added the /T option to pcretest so as to be able to run tests with non-
352        standard character tables, thus making it possible to include the tests
353        used for 19 above in the standard set of tests.
355    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
356        reference to a subpattern the other side of a comment that contains an
357        opening parenthesis caused either an internal compiling error, or a
358        reference to the wrong subpattern.
361    Version 8.02 19-Mar-2010
362    ------------------------
364    1.  The Unicode data tables have been updated to Unicode 5.2.0.
366    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
367        configured.
369    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
370        original author of that file, following a query about its status.
372    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
373        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
375    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
376        quantifier applied to a forward-referencing subroutine call, could compile
377        incorrect code or give the error "internal error: previously-checked
378        referenced subpattern not found".
380    6.  Both MS Visual Studio and Symbian OS have problems with initializing
381        variables to point to external functions. For these systems, therefore,
382        pcre_malloc etc. are now initialized to local functions that call the
383        relevant global functions.
385    7.  There were two entries missing in the vectors called coptable and poptable
386        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
387        I've fixed the data, and added a kludgy way of testing at compile time that
388        the lengths are correct (equal to the number of opcodes).
390    8.  Following on from 7, I added a similar kludge to check the length of the
391        eint vector in pcreposix.c.
393    9.  Error texts for pcre_compile() are held as one long string to avoid too
394        much relocation at load time. To find a text, the string is searched,
395        counting zeros. There was no check for running off the end of the string,
396        which could happen if a new error number was added without updating the
397        string.
399    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
401    11. \K was not working if it appeared in an atomic group or in a group that
402        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
403        \K is "not well defined" if used in an assertion. PCRE now accepts it if
404        the assertion is positive, but not if it is negative.
406    12. Change 11 fortuitously reduced the size of the stack frame used in the
407        "match()" function of pcre_exec.c by one pointer. Forthcoming
408        implementation of support for (*MARK) will need an extra pointer on the
409        stack; I have reserved it now, so that the stack frame size does not
410        decrease.
412    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
413        item in branch that calls a recursion is a subroutine call - as in the
414        second branch in the above example - was incorrectly given the compile-
415        time error "recursive call could loop indefinitely" because pcre_compile()
416        was not correctly checking the subroutine for matching a non-empty string.
418    14. The checks for overrunning compiling workspace could trigger after an
419        overrun had occurred. This is a "should never occur" error, but it can be
420        triggered by pathological patterns such as hundreds of nested parentheses.
421        The checks now trigger 100 bytes before the end of the workspace.
423    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
426    Version 8.01 19-Jan-2010
427    ------------------------
429    1.  If a pattern contained a conditional subpattern with only one branch (in
430        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
431        computed the wrong minimum data length (which is of course zero for such
432        subpatterns). This could cause incorrect "no match" results.
434    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
435        the pattern is reset in the first branch, pcre_compile() failed with
436        "internal error: code overflow at offset...". This happened only when
437        the reset was to the original external option setting. (An optimization
438        abstracts leading options settings into an external setting, which was the
439        cause of this.)
441    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
442        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
443        assertion pattern did not match (meaning that the assertion was true), it
444        was incorrectly treated as false if the SKIP had been reached during the
445        matching. This also applied to assertions used as conditions.
447    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
448        assertion subpattern, including such a pattern used as a condition,
449        unpredictable results occurred, instead of the error return
452    5.  The C++ GlobalReplace function was not working like Perl for the special
453        situation when an empty string is matched. It now does the fancy magic
454        stuff that is necessary.
456    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
457        removed. (These were left over from very, very early versions of PCRE.)
459    7.  Some cosmetic changes to the code to make life easier when compiling it
460        as part of something else:
462        (a) Change DEBUG to PCRE_DEBUG.
464        (b) In pcre_compile(), rename the member of the "branch_chain" structure
465            called "current" as "current_branch", to prevent a collision with the
466            Linux macro when compiled as a kernel module.
468        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
469            prevent a collision with the Linux macro when compiled as a kernel
470            module.
472    8.  In pcre_compile() there are some checks for integer overflows that used to
473        cast potentially large values to (double). This has been changed to that
474        when building, a check for int64_t is made, and if it is found, it is used
475        instead, thus avoiding the use of floating point arithmetic. (There is no
476        other use of FP in PCRE.) If int64_t is not found, the fallback is to
477        double.
479    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
480        2005 (difference between two addresses compared to an unsigned value).
482    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
483        custom one, because of the following reported problem in Windows:
485          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
486              under Win32.
487          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
488              therefore missing the function definition.
489          - The compiler thus generates a "C" signature for the test function.
490          - The linker fails to find the "C" function.
491          - PCRE fails to configure if asked to do so against libbz2.
493    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
494        messages were output:
496          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
497          rerunning libtoolize, to keep the correct libtool macros in-tree.
498          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
500        I have done both of these things.
502    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
503        most of the time, it *can* run out if it is given a pattern that contains a
504        runaway infinite recursion. I updated the discussion in the pcrestack man
505        page.
507    13. Now that we have gone to the x.xx style of version numbers, the minor
508        version may start with zero. Using 08 or 09 is a bad idea because users
509        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
510        interpreted as invalid octal numbers. I've updated the previous comment in
511        configure.ac, and also added a check that gives an error if 08 or 09 are
512        used.
514    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
515        causing partial matching to fail when the end of the subject matched \W
516        in a UTF-8 pattern where \W was quantified with a minimum of 3.
518    15. There were some discrepancies between the declarations in pcre_internal.h
519        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
520        their definitions. The declarations used "const uschar *" and the
521        definitions used USPTR. Even though USPTR is normally defined as "const
522        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
523        reported that: "This difference in casting confuses some C++ compilers, for
524        example, SunCC recognizes above declarations as different functions and
525        generates broken code for hbpcre." I have changed the declarations to use
526        USPTR.
528    16. GNU libtool is named differently on some systems. The autogen.sh script now
529        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
530        (FreeBSD).
532    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
533        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
534        comment: "Figure out how to create a longlong from a string: strtoll and
535        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
536        instance, but it only takes 2 args instead of 3!"
538    18. A subtle bug concerned with back references has been fixed by a change of
539        specification, with a corresponding code fix. A pattern such as
540        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
541        refers, was giving matches when it shouldn't. For example, xa=xaaa would
542        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
543        same bug. Such groups have to be quantified to be useful, or contained
544        inside another quantified group. (If there's no repetition, the reference
545        can never match.) The problem arises because, having left the group and
546        moved on to the rest of the pattern, a later failure that backtracks into
547        the group uses the captured value from the final iteration of the group
548        rather than the correct earlier one. I have fixed this in PCRE by forcing
549        any group that contains a reference to itself to be an atomic group; that
550        is, there cannot be any backtracking into it once it has completed. This is
551        similar to recursive and subroutine calls.
554    Version 8.00 19-Oct-09
555    ----------------------
557    1.  The table for translating pcre_compile() error codes into POSIX error codes
558        was out-of-date, and there was no check on the pcre_compile() error code
559        being within the table. This could lead to an OK return being given in
560        error.
562    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
563        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
564        in a Windows environment.
566    3.  The pcregrep --count option prints the count for each file even when it is
567        zero, as does GNU grep. However, pcregrep was also printing all files when
568        --files-with-matches was added. Now, when both options are given, it prints
569        counts only for those files that have at least one match. (GNU grep just
570        prints the file name in this circumstance, but including the count seems
571        more useful - otherwise, why use --count?) Also ensured that the
572        combination -clh just lists non-zero counts, with no names.
574    4.  The long form of the pcregrep -F option was incorrectly implemented as
575        --fixed_strings instead of --fixed-strings. This is an incompatible change,
576        but it seems right to fix it, and I didn't think it was worth preserving
577        the old behaviour.
579    5.  The command line items --regex=pattern and --regexp=pattern were not
580        recognized by pcregrep, which required --regex pattern or --regexp pattern
581        (with a space rather than an '='). The man page documented the '=' forms,
582        which are compatible with GNU grep; these now work.
584    6.  No libpcreposix.pc file was created for pkg-config; there was just
585        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
587    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
588        when UCP support is not needed, by modifying the Python script that
589        generates it from Unicode data files. This should not matter if the module
590        is correctly used as a library, but I received one complaint about 50K of
591        unwanted data. My guess is that the person linked everything into his
592        program rather than using a library. Anyway, it does no harm.
594    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
595        was a minimum greater than 1 for a wide character in a possessive
596        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
597        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
598        character. Chaos in the form of incorrect output or a compiling loop could
599        result.
601    9.  The restrictions on what a pattern can contain when partial matching is
602        requested for pcre_exec() have been removed. All patterns can now be
603        partially matched by this function. In addition, if there are at least two
604        slots in the offset vector, the offset of the earliest inspected character
605        for the match and the offset of the end of the subject are set in them when
606        PCRE_ERROR_PARTIAL is returned.
608    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
609        synonymous with PCRE_PARTIAL, for backwards compatibility, and
610        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
611        and may be more useful for multi-segment matching.
613    11. Partial matching with pcre_exec() is now more intuitive. A partial match
614        used to be given if ever the end of the subject was reached; now it is
615        given only if matching could not proceed because another character was
616        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
617        string "Z", which now yields "no match" instead of "partial match". In the
618        case of pcre_dfa_exec(), "no match" is given if every matching path for the
619        final character ended with (*FAIL).
621    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
622        if the pattern had a "must contain" character that was already found in the
623        earlier partial match, unless partial matching was again requested. For
624        example, with the pattern /dog.(body)?/, the "must contain" character is
625        "g". If the first part-match was for the string "dog", restarting with
626        "sbody" failed. This bug has been fixed.
628    13. The string returned by pcre_dfa_exec() after a partial match has been
629        changed so that it starts at the first inspected character rather than the
630        first character of the match. This makes a difference only if the pattern
631        starts with a lookbehind assertion or \b or \B (\K is not supported by
632        pcre_dfa_exec()). It's an incompatible change, but it makes the two
633        matching functions compatible, and I think it's the right thing to do.
635    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
636        so that the demonstration program is easily available in environments where
637        PCRE has not been installed from source.
639    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
640        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
641        library.
643    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
644        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
645        is not the first non-POSIX option to be added. Clearly some people find
646        these options useful.
648    17. If a caller to the POSIX matching function regexec() passes a non-zero
649        value for nmatch with a NULL value for pmatch, the value of
650        nmatch is forced to zero.
652    18. RunGrepTest did not have a test for the availability of the -u option of
653        the diff command, as RunTest does. It now checks in the same way as
654        RunTest, and also checks for the -b option.
656    19. If an odd number of negated classes containing just a single character
657        interposed, within parentheses, between a forward reference to a named
658        subpattern and the definition of the subpattern, compilation crashed with
659        an internal error, complaining that it could not find the referenced
660        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
661        [The bug was that it was starting one character too far in when skipping
662        over the character class, thus treating the ] as data rather than
663        terminating the class. This meant it could skip too much.]
665    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
666        /g option in pcretest when the pattern contains \K, which makes it possible
667        to have an empty string match not at the start, even when the pattern is
668        anchored. Updated pcretest and pcredemo to use this option.
670    21. If the maximum number of capturing subpatterns in a recursion was greater
671        than the maximum at the outer level, the higher number was returned, but
672        with unset values at the outer level. The correct (outer level) value is
673        now given.
675    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
676        PCRE did not set those parentheses (unlike Perl). I have now found a way to
677        make it do so. The string so far is captured, making this feature
678        compatible with Perl.
680    23. The tests have been re-organized, adding tests 11 and 12, to make it
681        possible to check the Perl 5.10 features against Perl 5.10.
683    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
684        pattern matches a fixed length string. PCRE did not allow this; now it
685        does. Neither allows recursion.
687    25. I finally figured out how to implement a request to provide the minimum
688        length of subject string that was needed in order to match a given pattern.
689        (It was back references and recursion that I had previously got hung up
690        on.) This code has now been added to pcre_study(); it finds a lower bound
691        to the length of subject needed. It is not necessarily the greatest lower
692        bound, but using it to avoid searching strings that are too short does give
693        some useful speed-ups. The value is available to calling programs via
694        pcre_fullinfo().
696    26. While implementing 25, I discovered to my embarrassment that pcretest had
697        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
698        study optimizations had never been tested with that matching function.
699        Oops. What is worse, even when it was passed study data, there was a bug in
700        pcre_dfa_exec() that meant it never actually used it. Double oops. There
701        were also very few tests of studied patterns with pcre_dfa_exec().
703    27. If (?| is used to create subpatterns with duplicate numbers, they are now
704        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
705        on the other side of the coin, they are no longer allowed to have different
706        names, because these cannot be distinguished in PCRE, and this has caused
707        confusion. (This is a difference from Perl.)
709    28. When duplicate subpattern names are present (necessarily with different
710        numbers, as required by 27 above), and a test is made by name in a
711        conditional pattern, either for a subpattern having been matched, or for
712        recursion in such a pattern, all the associated numbered subpatterns are
713        tested, and the overall condition is true if the condition is true for any
714        one of them. This is the way Perl works, and is also more like the way
715        testing by number works.
718    Version 7.9 11-Apr-09
719    ---------------------
721    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
722        (pcretest), all targets were linked against these libraries. This included
723        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
724        libraries. This caused unwanted dependencies to be created. This problem
725        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
726        pcretest is linked with readline.
728    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
729        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
730        moved outside it again, because FALSE and TRUE are already defined in AIX,
731        but BOOL is not.
733    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
734        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
736    4.  The pcregrep documentation said spaces were inserted as well as colons (or
737        hyphens) following file names and line numbers when outputting matching
738        lines. This is not true; no spaces are inserted. I have also clarified the
739        wording for the --colour (or --color) option.
741    5.  In pcregrep, when --colour was used with -o, the list of matching strings
742        was not coloured; this is different to GNU grep, so I have changed it to be
743        the same.
745    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
746        each matching line was coloured. Now it goes on to look for further matches
747        of any of the test patterns, which is the same behaviour as GNU grep.
749    7.  A pattern that could match an empty string could cause pcregrep to loop; it
750        doesn't make sense to accept an empty string match in pcregrep, so I have
751        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
752        seems to be how GNU grep behaves.
754    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
755        start or after a newline", because the conditional assertion was not being
756        correctly handled. The rule now is that both the assertion and what follows
757        in the first alternative must satisfy the test.
759    9.  If auto-callout was enabled in a pattern with a conditional group whose
760        condition was an assertion, PCRE could crash during matching, both with
761        pcre_exec() and pcre_dfa_exec().
763    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
764        used for matching.
766    11. Unicode property support in character classes was not working for
767        characters (bytes) greater than 127 when not in UTF-8 mode.
769    12. Added the -M command line option to pcretest.
771    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
773    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
775    16. Added comments and documentation about mis-use of no_arg in the C++
776        wrapper.
778    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
779        from Martin Jerabek that uses macro names for all relevant character and
780        string constants.
782    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
783        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
784        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
785        these, but not everybody uses configure.
787    19. A conditional group that had only one branch was not being correctly
788        recognized as an item that could match an empty string. This meant that an
789        enclosing group might also not be so recognized, causing infinite looping
790        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
791        with the subject "ab", where knowledge that the repeated group can match
792        nothing is needed in order to break the loop.
794    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
795        exec(), but without supplying a callout function, matching went wrong.
797    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
798        leak if the size of the offset vector was greater than 30. When the vector
799        is smaller, the saved offsets during recursion go onto a local stack
800        vector, but for larger vectors malloc() is used. It was failing to free
801        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
802        error, in fact).
804    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
805        heapframe that is used only when UTF-8 support is enabled. This caused no
806        problem, but was untidy.
808    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
809        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
810        included within another project.
812    24. Steven Van Ingelgem's patches to add more options to the CMake support,
813        slightly modified by me:
815          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
816              not building pcregrep.
818          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
819              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
821    25. Forward references, both numeric and by name, in patterns that made use of
822        duplicate group numbers, could behave incorrectly or give incorrect errors,
823        because when scanning forward to find the reference group, PCRE was not
824        taking into account the duplicate group numbers. A pattern such as
825        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
827    26. Changed a few more instances of "const unsigned char *" to USPTR, making
828        the feature of a custom pointer more persuasive (as requested by a user).
830    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
831        pcretest.c, inside #ifndefs, because it seems they are sometimes already
832        pre-defined.
834    28. Added support for (*UTF8) at the start of a pattern.
836    29. Arrange for flags added by the "release type" setting in CMake to be shown
837        in the configuration summary.
840    Version 7.8 05-Sep-08
841    ---------------------
843    1.  Replaced UCP searching code with optimized version as implemented for Ad
844        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
845        stage table and inline lookup instead of a function, giving speed ups of 2
846        to 5 times on some simple patterns that I tested. Permission was given to
847        distribute the MultiStage2.py script that generates the tables (it's not in
848        the tarball, but is in the Subversion repository).
850    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
851        scripts.
853    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
854        a group with a zero qualifier. The result of the study could be incorrect,
855        or the function might crash, depending on the pattern.
857    4.  Caseless matching was not working for non-ASCII characters in back
858        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
859        It now works when Unicode Property Support is available.
861    5.  In pcretest, an escape such as \x{de} in the data was always generating
862        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
863        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
864        truncation.
866    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
868    7.  Added two (int) casts to pcregrep when printing the difference of two
869        pointers, in case they are 64-bit values.
871    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
872        test 2 if it fails.
874    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
875        and a #define of that name to empty if it is not externally set. This is to
876        allow users of MSVC to set it if necessary.
878    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
879        the convenience functions in the pcre_get.c source file.
881    11. An option change at the start of a pattern that had top-level alternatives
882        could cause overwriting and/or a crash. This command provoked a crash in
883        some environments:
885          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
887        This potential security problem was recorded as CVE-2008-2371.
889    12. For a pattern where the match had to start at the beginning or immediately
890        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
891        pcre_dfa_exec() could read past the end of the passed subject if there was
892        no match. To help with detecting such bugs (e.g. with valgrind), I modified
893        pcretest so that it places the subject at the end of its malloc-ed buffer.
895    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
896        exec() might read past the end of the data buffer in UTF-8 mode.
898    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
899        the data contained the byte 0x85 as part of a UTF-8 character within its
900        first line. This applied both to normal and DFA matching.
902    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
903        /^[^d]*?$/8 failed to match "abc".
905    16. Added a missing copyright notice to pcrecpp_internal.h.
907    17. Make it more clear in the documentation that values returned from
908        pcre_exec() in ovector are byte offsets, not character counts.
910    18. Tidied a few places to stop certain compilers from issuing warnings.
912    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
913        supplied by Stefan Weber. I made a further small update for 7.8 because
914        there is a change of source arrangements: the pcre_searchfuncs.c module is
915        replaced by pcre_ucd.c.
918    Version 7.7 07-May-08
919    ---------------------
921    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
922        a string to a long long, pretend we don't even have a long long." This is
923        done by checking for the strtoq, strtoll, and _strtoi64 functions.
925    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
926        pre-7.6 versions, which defined a global no_arg variable instead of putting
927        it in the RE class. (See also #8 below.)
929    3.  Remove a line of dead code, identified by coverity and reported by Nuno
930        Lopes.
932    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
934        (1) The include/exclude patterns were being applied to the whole pathnames
935            of files, instead of just to the final components.
937        (2) If there was more than one level of directory, the subdirectories were
938            skipped unless they satisfied the include/exclude conditions. This is
939            inconsistent with GNU grep (and could even be seen as contrary to the
940            pcregrep specification - which I improved to make it absolutely clear).
941            The action now is always to scan all levels of directory, and just
942            apply the include/exclude patterns to regular files.
944    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
945        --exclude_dir in the tests to avoid scanning .svn directories.
947    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
948        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
949        doesn't support NULs in patterns.
951    7.  Added some missing "const"s to declarations of static tables in
952        pcre_compile.c and pcre_dfa_exec.c.
954    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
955        caused by fix #2  above. (Subsequently also a second patch to fix the
956        first patch. And a third patch - this was a messy problem.)
958    9.  Applied Craig's patch to remove the use of push_back().
960    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
961        matching function regexec().
963    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
964        which, however, unlike Perl's \g{...}, are subroutine calls, not back
965        references. PCRE supports relative numbers with this syntax (I don't think
966        Oniguruma does).
968    12. Previously, a group with a zero repeat such as (...){0} was completely
969        omitted from the compiled regex. However, this means that if the group
970        was called as a subroutine from elsewhere in the pattern, things went wrong
971        (an internal error was given). Such groups are now left in the compiled
972        pattern, with a new opcode that causes them to be skipped at execution
973        time.
975    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
976        to the way PCRE behaves:
978        (a) A lone ] character is dis-allowed (Perl treats it as data).
980        (b) A back reference to an unmatched subpattern matches an empty string
981            (Perl fails the current match path).
983        (c) A data ] in a character class must be notated as \] because if the
984            first data character in a class is ], it defines an empty class. (In
985            Perl it is not possible to have an empty class.) The empty class []
986            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
987            The negative empty class [^] matches any one character, independently
988            of the DOTALL setting.
990    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
991        non-existent subpattern following a character class starting with ']' and
992        containing () gave an internal compiling error instead of "reference to
993        non-existent subpattern". Fortunately, when the pattern did exist, the
994        compiled code was correct. (When scanning forwards to check for the
995        existencd of the subpattern, it was treating the data ']' as terminating
996        the class, so got the count wrong. When actually compiling, the reference
997        was subsequently set up correctly.)
999    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
1000        it was being rejected as not supported by pcre_dfa_exec(), even though
1001        other assertions are supported. I have made pcre_dfa_exec() support
1002        (*FAIL).
1004    16. The implementation of 13c above involved the invention of a new opcode,
1005        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
1006        cannot be changed at match time, I realized I could make a small
1007        improvement to matching performance by compiling OP_ALLANY instead of
1008        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
1009        on the OP_ANY path.
1011    17. Compiling pcretest on Windows with readline support failed without the
1012        following two fixes: (1) Make the unistd.h include conditional on
1013        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
1015    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
1016        ncurses library to be included for pcretest when ReadLine support is
1017        requested, but also to allow for it to be overridden. This patch came from
1018        Daniel Bergström.
1020    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
1021        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
1022        any errors with the current Unicode tables. Thanks to Peter Kankowski for
1023        spotting this.
1026    Version 7.6 28-Jan-08
1027    ---------------------
1029    1.  A character class containing a very large number of characters with
1030        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
1031        overflow.
1033    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
1034        HAVE_LONG_LONG is not defined.
1036    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
1037        bring them up to date and include new features. This patch includes:
1039        - Fixed PH's badly added libz and libbz2 support.
1040        - Fixed a problem with static linking.
1041        - Added pcredemo. [But later removed - see 7 below.]
1042        - Fixed dftables problem and added an option.
1043        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
1044            HAVE_LONG_LONG.
1045        - Added readline support for pcretest.
1046        - Added an listing of the option settings after cmake has run.
1048    4.  A user submitted a patch to Makefile that makes it easy to create
1049        "pcre.dll" under mingw when using Configure/Make. I added stuff to
1050        Makefile.am that cause it to include this special target, without
1051        affecting anything else. Note that the same mingw target plus all
1052        the other distribution libraries and programs are now supported
1053        when configuring with CMake (see 6 below) instead of with
1054        Configure/Make.
1056    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
1057        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
1058        exported in the Windows port". It has not yet been confirmed that the patch
1059        solves the problem, but it does no harm.
1061    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
1062        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
1063        with CMake, and also correct the comment about stack recursion.
1065    7.  Remove the automatic building of pcredemo from the ./configure system and
1066        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
1067        of a program that users should build themselves after PCRE is installed, so
1068        building it automatically is not really right. What is more, it gave
1069        trouble in some build environments.
1071    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
1074    Version 7.5 10-Jan-08
1075    ---------------------
1077    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
1078        values in parens when parsing an RE using the C++ wrapper."
1080    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
1081        Characters greater than 255 were excluded from the class instead of being
1082        included.
1084    3.  The same bug as (2) above applied to negated POSIX classes such as
1085        [:^space:].
1087    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
1088        defined or documented. It seems to have been a typo for PCRE_STATIC, so
1089        I have changed it.
1091    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
1092        first named subpattern) and a construct such as (?&a) would reference the
1093        first named subpattern whose name started with "a" (in other words, the
1094        length check was missing). Both these problems are fixed. "Subpattern name
1095        expected" is now given for (?&) (a zero-length name), and this patch also
1096        makes it give the same error for \k'' (previously it complained that that
1097        was a reference to a non-existent subpattern).
1099    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
1100        this is right because (?- can be followed by option settings as well as by
1101        digits. I have, however, made the messages clearer.
1103    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
1104        than the number used in the conditional) now cause a compile-time error.
1105        This is actually not compatible with Perl, which accepts such patterns, but
1106        treats the conditional as always being FALSE (as PCRE used to), but it
1107        seems to me that giving a diagnostic is better.
1109    8.  Change "alphameric" to the more common word "alphanumeric" in comments
1110        and messages.
1112    9.  Fix two occurrences of "backslash" in comments that should have been
1113        "backspace".
1115    10. Remove two redundant lines of code that can never be obeyed (their function
1116        was moved elsewhere).
1118    11. The program that makes PCRE's Unicode character property table had a bug
1119        which caused it to generate incorrect table entries for sequences of
1120        characters that have the same character type, but are in different scripts.
1121        It amalgamated them into a single range, with the script of the first of
1122        them. In other words, some characters were in the wrong script. There were
1123        thirteen such cases, affecting characters in the following ranges:
1125          U+002b0 - U+002c1
1126          U+0060c - U+0060d
1127          U+0061e - U+00612
1128          U+0064b - U+0065e
1129          U+0074d - U+0076d
1130          U+01800 - U+01805
1131          U+01d00 - U+01d77
1132          U+01d9b - U+01dbf
1133          U+0200b - U+0200f
1134          U+030fc - U+030fe
1135          U+03260 - U+0327f
1136          U+0fb46 - U+0fbb1
1137          U+10450 - U+1049d
1139    12. The -o option (show only the matching part of a line) for pcregrep was not
1140        compatible with GNU grep in that, if there was more than one match in a
1141        line, it showed only the first of them. It now behaves in the same way as
1142        GNU grep.
1144    13. If the -o and -v options were combined for pcregrep, it printed a blank
1145        line for every non-matching line. GNU grep prints nothing, and pcregrep now
1146        does the same. The return code can be used to tell if there were any
1147        non-matching lines.
1149    14. Added --file-offsets and --line-offsets to pcregrep.
1151    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
1152        infinitely looping recursion. The bug was that positive lookaheads were not
1153        being skipped when checking for a possible empty match (negative lookaheads
1154        and both kinds of lookbehind were skipped).
1156    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
1157        inclusion of <windows.h> to before rather than after the definition of
1158        INVALID_FILE_ATTRIBUTES (patch from David Byron).
1160    17. Specifying a possessive quantifier with a specific limit for a Unicode
1161        character property caused pcre_compile() to compile bad code, which led at
1162        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
1163        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
1164        caused the error; without that there was no problem.
1166    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
1168    19. Added --enable-pcretest-libreadline.
1170    20. In pcrecpp.cc, the variable 'count' was incremented twice in
1171        RE::GlobalReplace(). As a result, the number of replacements returned was
1172        double what it should be. I removed one of the increments, but Craig sent a
1173        later patch that removed the other one (the right fix) and added unit tests
1174        that check the return values (which was not done before).
1176    21. Several CMake things:
1178        (1) Arranged that, when cmake is used on Unix, the libraries end up with
1179            the names libpcre and libpcreposix, not just pcre and pcreposix.
1181        (2) The above change means that pcretest and pcregrep are now correctly
1182            linked with the newly-built libraries, not previously installed ones.
1186    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
1187        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
1188        UTF-8 newline character). The key issue is that the pattern starts .*;
1189        this means that the match must be either at the beginning, or after a
1190        newline. The bug was in the code for advancing after a failed match and
1191        checking that the new position followed a newline. It was not taking
1192        account of UTF-8 characters correctly.
1194    23. PCRE was behaving differently from Perl in the way it recognized POSIX
1195        character classes. PCRE was not treating the sequence [:...:] as a
1196        character class unless the ... were all letters. Perl, however, seems to
1197        allow any characters between [: and :], though of course it rejects as
1198        unknown any "names" that contain non-letters, because all the known class
1199        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
1200        for example, whereas PCRE did not - it did not recognize a POSIX character
1201        class. This seemed a bit dangerous, so the code has been changed to be
1202        closer to Perl. The behaviour is not identical to Perl, because PCRE will
1203        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
1204        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
1205        Perl does, and where it didn't before.
1207    24. Rewrite so as to remove the single use of %n from pcregrep because in some
1208        Windows environments %n is disabled by default.
1211    Version 7.4 21-Sep-07
1212    ---------------------
1214    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
1215        means that a class such as [\s] counted as "explicit reference to CR or
1216        LF". That isn't really right - the whole point of the change was to try to
1217        help when there was an actual mention of one of the two characters. So now
1218        the change happens only if \r or \n (or a literal CR or LF) character is
1219        encountered.
1221    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
1222        of both had grown to the point where there were only 3 bits left.
1223        Fortunately, there was spare space in the data structure, and so I have
1224        moved the internal flags into a new 16-bit field to free up more option
1225        bits.
1227    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
1228        but did not set the internal JCHANGED flag - either of these is enough to
1229        control the way the "get" function works - but the PCRE_INFO_JCHANGED
1230        facility is supposed to tell if (?J) was ever used, so now (?J) at the
1231        start sets both bits.
1233    4.  Added options (at build time, compile time, exec time) to change \R from
1234        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
1236    5.  doc/pcresyntax.html was missing from the distribution.
1238    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
1239        compatibility, even though it is no longer used.
1241    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
1242        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
1243        windows.h file is present (where different names are used). [This was
1244        reversed later after testing - see 16 below.]
1246    8.  Changed all #include <config.h> to #include "config.h". There were also
1247        some further <pcre.h> cases that I changed to "pcre.h".
1249    9.  When pcregrep was used with the --colour option, it missed the line ending
1250        sequence off the lines that it output.
1252    10. It was pointed out to me that arrays of string pointers cause lots of
1253        relocations when a shared library is dynamically loaded. A technique of
1254        using a single long string with a table of offsets can drastically reduce
1255        these. I have refactored PCRE in four places to do this. The result is
1256        dramatic:
1258          Originally:                          290
1259          After changing UCP table:            187
1260          After changing error message table:   43
1261          After changing table of "verbs"       36
1262          After changing table of Posix names   22
1264        Thanks to the folks working on Gregex for glib for this insight.
1266    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
1267        unicode-properties was also set.
1269    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
1271    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
1272        checked only for CRLF.
1274    14. Added casts to pcretest.c to avoid compiler warnings.
1276    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
1278    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
1279        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
1280        entirely. This removes changes made in 7 above.
1282    17. The CMake files have been updated, and there is now more information about
1283        building with CMake in the NON-UNIX-USE document.
1286    Version 7.3 28-Aug-07
1287    ---------------------
1289     1. In the rejigging of the build system that eventually resulted in 7.1, the
1290        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
1291        brackets there is not right, since it causes compilers to look for an
1292        installed pcre.h, not the version that is in the source that is being
1293        compiled (which of course may be different). I have changed it back to:
1295          #include "pcre.h"
1297        I have a vague recollection that the change was concerned with compiling in
1298        different directories, but in the new build system, that is taken care of
1299        by the VPATH setting the Makefile.
1301     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
1302        when the subject happened to end in the byte 0x85 (e.g. if the last
1303        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
1304        characters but of course it shouldn't be taken as a newline when it is part
1305        of another character. The bug was that, for an unlimited repeat of . in
1306        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
1307        characters when looking for a newline.
1309     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
1311     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
1312        in debug output.
1314     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
1315        long printing in the pcrecpp unittest when running under MinGW.
1317     6. ESC_K was left out of the EBCDIC table.
1319     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
1320        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
1321        limit also applies to "virtual nesting" when a pattern is recursive, and in
1322        this case 1000 isn't so big. I have been able to remove this limit at the
1323        expense of backing off one optimization in certain circumstances. Normally,
1324        when pcre_exec() would call its internal match() function recursively and
1325        immediately return the result unconditionally, it uses a "tail recursion"
1326        feature to save stack. However, when a subpattern that can match an empty
1327        string has an unlimited repetition quantifier, it no longer makes this
1328        optimization. That gives it a stack frame in which to save the data for
1329        checking that an empty string has been matched. Previously this was taken
1330        from the 1000-entry workspace that had been reserved. So now there is no
1331        explicit limit, but more stack is used.
1333     8. Applied Daniel's patches to solve problems with the import/export magic
1334        syntax that is required for Windows, and which was going wrong for the
1335        pcreposix and pcrecpp parts of the library. These were overlooked when this
1336        problem was solved for the main library.
1338     9. There were some crude static tests to avoid integer overflow when computing
1339        the size of patterns that contain repeated groups with explicit upper
1340        limits. As the maximum quantifier is 65535, the maximum group length was
1341        set at 30,000 so that the product of these two numbers did not overflow a
1342        32-bit integer. However, it turns out that people want to use groups that
1343        are longer than 30,000 bytes (though not repeat them that many times).
1344        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1345        made it possible to implement the integer overflow checks in a much more
1346        dynamic way, which I have now done. The artificial limitation on group
1347        length has been removed - we now have only the limit on the total length of
1348        the compiled pattern, which depends on the LINK_SIZE setting.
1350    10. Fixed a bug in the documentation for get/copy named substring when
1351        duplicate names are permitted. If none of the named substrings are set, the
1352        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1353        empty string.
1355    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1356        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1357        because the ] is interpreted as the first data character and the
1358        terminating ] is not found. PCRE has been made compatible with Perl in this
1359        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1360        cause memory overwriting.
1362    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1363        string has been matched (to stop an infinite loop). It was not recognizing
1364        a conditional subpattern that could match an empty string if that
1365        subpattern was within another subpattern. For example, it looped when
1366        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
1367        condition was not nested. This bug has been fixed.
1369    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1370        past the start of the subject in the presence of bytes with the top bit
1371        set, for example "\x8aBCD".
1373    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1374        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1376    14. Optimized (?!) to (*FAIL).
1378    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1379        This restricts code points to be within the range 0 to 0x10FFFF, excluding
1380        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1381        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1382        does: it's just the validity check that is more restrictive.
1384    16. Inserted checks for integer overflows during escape sequence (backslash)
1385        processing, and also fixed erroneous offset values for syntax errors during
1386        backslash processing.
1388    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1389        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1391    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1392        caused an overrun.
1394    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1395        something other than just ASCII characters) inside a group that had an
1396        unlimited repeat caused a loop at compile time (while checking to see
1397        whether the group could match an empty string).
1399    20. Debugging a pattern containing \p or \P could cause a crash. For example,
1400        [\P{Any}] did so. (Error in the code for printing property names.)
1402    21. An orphan \E inside a character class could cause a crash.
1404    22. A repeated capturing bracket such as (A)? could cause a wild memory
1405        reference during compilation.
1407    23. There are several functions in pcre_compile() that scan along a compiled
1408        expression for various reasons (e.g. to see if it's fixed length for look
1409        behind). There were bugs in these functions when a repeated \p or \P was
1410        present in the pattern. These operators have additional parameters compared
1411        with \d, etc, and these were not being taken into account when moving along
1412        the compiled data. Specifically:
1414        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1415            length.
1417        (b) An item such as \pL+ within a repeated group could cause crashes or
1418            loops.
1420        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1421            "reference to non-existent subpattern" error.
1423        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1425    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1426        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1428    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1430    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1431        character were causing crashes (broken optimization).
1433    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1434        \p or \P) caused a compile-time loop.
1436    28. More problems have arisen in unanchored patterns when CRLF is a valid line
1437        break. For example, the unstudied pattern [\r\n]A does not match the string
1438        "\r\nA" because change 7.0/46 below moves the current point on by two
1439        characters after failing to match at the start. However, the pattern \nA
1440        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
1441        the same is true. There doesn't seem any very clean way out of this, but
1442        what I have chosen to do makes the common cases work: PCRE now takes note
1443        of whether there can be an explicit match for \r or \n anywhere in the
1444        pattern, and if so, 7.0/46 no longer applies. As part of this change,
1445        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1446        pattern has explicit CR or LF references.
1448    29. Added (*CR) etc for changing newline setting at start of pattern.
1451    Version 7.2 19-Jun-07
1452    ---------------------
1454     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
1455        which is apparently normally available under Windows.
1457     2. Re-jig the pcregrep tests with different newline settings in an attempt
1458        to make them independent of the local environment's newline setting.
1460     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
1462     4. Some of the "internals" tests were previously cut out when the link size
1463        was not 2, because the output contained actual offsets. The recent new
1464        "Z" feature of pcretest means that these can be cut out, making the tests
1465        usable with all link sizes.
1467     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
1468        stack recursion. This gives a massive performance boost under BSD, but just
1469        a small improvement under Linux. However, it saves one field in the frame
1470        in all cases.
1472     6. Added more features from the forthcoming Perl 5.10:
1474        (a) (?-n) (where n is a string of digits) is a relative subroutine or
1475            recursion call. It refers to the nth most recently opened parentheses.
1477        (b) (?+n) is also a relative subroutine call; it refers to the nth next
1478            to be opened parentheses.
1480        (c) Conditions that refer to capturing parentheses can be specified
1481            relatively, for example, (?(-2)... or (?(+3)...
1483        (d) \K resets the start of the current match so that everything before
1484            is not part of it.
1486        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1488        (f) \g{name} is another synonym - part of Perl 5.10's unification of
1489            reference syntax.
1491        (g) (?| introduces a group in which the numbering of parentheses in each
1492            alternative starts with the same number.
1494        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
1496     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
1499     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
1500        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
1501        for detecting groups that can match an empty string.
1503     9. A pattern with a very large number of alternatives (more than several
1504        hundred) was running out of internal workspace during the pre-compile
1505        phase, where pcre_compile() figures out how much memory will be needed. A
1506        bit of new cunning has reduced the workspace needed for groups with
1507        alternatives. The 1000-alternative test pattern now uses 12 bytes of
1508        workspace instead of running out of the 4096 that are available.
1510    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
1512    11. Applied patch from Google to remove an optimization that didn't quite work.
1513        The report of the bug said:
1515          pcrecpp::RE("a*").FullMatch("aaa") matches, while
1516          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
1517          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
1519    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
1520        it matched the wrong number of bytes.
1523    Version 7.1 24-Apr-07
1524    ---------------------
1526     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
1527        that is more "standard", making use of automake and other Autotools. There
1528        is some re-arrangement of the files and adjustment of comments consequent
1529        on this.
1531     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
1532        for recursive directory scanning broke on some systems because the files
1533        are not scanned in any specific order and on different systems the order
1534        was different. A call to "sort" has been inserted into RunGrepTest for the
1535        approprate test as a short-term fix. In the longer term there may be an
1536        alternative.
1538     3. I had an email from Eric Raymond about problems translating some of PCRE's
1539        man pages to HTML (despite the fact that I distribute HTML pages, some
1540        people do their own conversions for various reasons). The problems
1541        concerned the use of low-level troff macros .br and .in. I have therefore
1542        removed all such uses from the man pages (some were redundant, some could
1543        be replaced by .nf/.fi pairs). The 132html script that I use to generate
1544        HTML has been updated to handle .nf/.fi and to complain if it encounters
1545        .br or .in.
1547     4. Updated comments in configure.ac that get placed in config.h.in and also
1548        arranged for config.h to be included in the distribution, with the name
1549        config.h.generic, for the benefit of those who have to compile without
1550        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
1552     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
1553        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
1554        makevp.bat for the current PCRE, using the additional files
1555        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
1557     6. A Windows user reported a minor discrepancy with test 2, which turned out
1558        to be caused by a trailing space on an input line that had got lost in his
1559        copy. The trailing space was an accident, so I've just removed it.
1561     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
1562        that is needed.
1564     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
1565        as "const" (a) because they are and (b) because it helps the PHP
1566        maintainers who have recently made a script to detect big data structures
1567        in the php code that should be moved to the .rodata section. I remembered
1568        to update Builducptable as well, so it won't revert if ucptable.h is ever
1569        re-created.
1571     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
1572        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
1573        order to be able to cut out the UTF-8 tables in the latter when UTF-8
1574        support is not required. This saves 1.5-2K of code, which is important in
1575        some applications.
1577        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
1578        so as not to refer to the tables, even though these functions will never be
1579        called when UTF-8 support is disabled. Otherwise there are problems with a
1580        shared library.
1582    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
1584        (a) It was defining its arguments as char * instead of void *.
1586        (b) It was assuming that all moves were upwards in memory; this was true
1587            a long time ago when I wrote it, but is no longer the case.
1589        The emulated memove() is provided for those environments that have neither
1590        memmove() nor bcopy(). I didn't think anyone used it these days, but that
1591        is clearly not the case, as these two bugs were recently reported.
1593    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
1594        and Detrail to create the HTML documentation, the .txt form of the man
1595        pages, and it removes trailing spaces from listed files. It also creates
1596        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
1597        case, it wraps all the #defines with #ifndefs. This script should be run
1598        before "make dist".
1600    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1601        with Unicode property support.
1603        (a) For a maximizing quantifier, if the two different cases of the
1604            character were of different lengths in their UTF-8 codings (there are
1605            some cases like this - I found 11), and the matching function had to
1606            back up over a mixture of the two cases, it incorrectly assumed they
1607            were both the same length.
1609        (b) When PCRE was configured to use the heap rather than the stack for
1610            recursion during matching, it was not correctly preserving the data for
1611            the other case of a UTF-8 character when checking ahead for a match
1612            while processing a minimizing repeat. If the check also involved
1613            matching a wide character, but failed, corruption could cause an
1614            erroneous result when trying to check for a repeat of the original
1615            character.
1617    13. Some tidying changes to the testing mechanism:
1619        (a) The RunTest script now detects the internal link size and whether there
1620            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1621            values substituted by "configure". (The RunGrepTest script already did
1622            this for UTF-8.) The configure.ac script no longer substitutes the
1623            relevant variables.
1625        (b) The debugging options /B and /D in pcretest show the compiled bytecode
1626            with length and offset values. This means that the output is different
1627            for different internal link sizes. Test 2 is skipped for link sizes
1628            other than 2 because of this, bypassing the problem. Unfortunately,
1629            there was also a test in test 3 (the locale tests) that used /B and
1630            failed for link sizes other than 2. Rather than cut the whole test out,
1631            I have added a new /Z option to pcretest that replaces the length and
1632            offset values with spaces. This is now used to make test 3 independent
1633            of link size. (Test 2 will be tidied up later.)
1635    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1636        segmentation fault instead of returning the appropriate error message.
1638    15. In multiline mode when the newline sequence was set to "any", the pattern
1639        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1640        This doesn't seem right; it now treats the CRLF combination as the line
1641        ending, and so does not match in that case. It's only a pattern such as ^$
1642        that would hit this one: something like ^ABC$ would have failed after \r
1643        and then tried again after \r\n.
1645    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1646        in an attempt to make files that differ only in their line terminators
1647        compare equal. This works on Linux.
1649    17. Under certain error circumstances pcregrep might try to free random memory
1650        as it exited. This is now fixed, thanks to valgrind.
1652    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1653        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1654        was because its rules for how to advance for /g after matching an empty
1655        string at the end of a line did not allow for this case. They now check for
1656        it specially.
1658    20. pcretest is supposed to handle patterns and data of any length, by
1659        extending its buffers when necessary. It was getting this wrong when the
1660        buffer for a data line had to be extended.
1662    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1663        CRLF as a newline sequence.
1665    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1666        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1667        I have nevertheless tidied it up.
1669    23. Added some casts to kill warnings from HP-UX ia64 compiler.
1671    24. Added a man page for pcre-config.
1674    Version 7.0 19-Dec-06
1675    ---------------------
1677     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1678        moving to gcc 4.1.1.
1680     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1681        sys/time.h, which is documented as needed for this function. It doesn't
1682        seem to matter on Linux, but it showed up on some releases of OS X.
1684     3. It seems that there are systems where bytes whose values are greater than
1685        127 match isprint() in the "C" locale. The "C" locale should be the
1686        default when a C program starts up. In most systems, only ASCII printing
1687        characters match isprint(). This difference caused the output from pcretest
1688        to vary, making some of the tests fail. I have changed pcretest so that:
1690        (a) When it is outputting text in the compiled version of a pattern, bytes
1691            other than 32-126 are always shown as hex escapes.
1693        (b) When it is outputting text that is a matched part of a subject string,
1694            it does the same, unless a different locale has been set for the match
1695            (using the /L modifier). In this case, it uses isprint() to decide.
1697     4. Fixed a major bug that caused incorrect computation of the amount of memory
1698        required for a compiled pattern when options that changed within the
1699        pattern affected the logic of the preliminary scan that determines the
1700        length. The relevant options are -x, and -i in UTF-8 mode. The result was
1701        that the computed length was too small. The symptoms of this bug were
1702        either the PCRE error "internal error: code overflow" from pcre_compile(),
1703        or a glibc crash with a message such as "pcretest: free(): invalid next
1704        size (fast)". Examples of patterns that provoked this bug (shown in
1705        pcretest format) are:
1707          /(?-x: )/x
1708          /(?x)(?-x: \s*#\s*)/
1709          /((?i)[\x{c0}])/8
1710          /(?i:[\x{c0}])/8
1712        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1713        is now done differently.
1715     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1716        wrapper classes; (b) implement a new function in the C++ scanner that is
1717        more efficient than the old way of doing things because it avoids levels of
1718        recursion in the regex matching; (c) add a paragraph to the documentation
1719        for the FullMatch() function.
1721     6. The escape sequence \n was being treated as whatever was defined as
1722        "newline". Not only was this contrary to the documentation, which states
1723        that \n is character 10 (hex 0A), but it also went horribly wrong when
1724        "newline" was defined as CRLF. This has been fixed.
1726     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1727        was being set to -1 for the "end of line" case (supposedly a value that no
1728        character can have). Though this value is never used (the check for end of
1729        line is "zero bytes in current character"), it caused compiler complaints.
1730        I've changed it to 0xffffffff.
1732     8. In pcre_version.c, the version string was being built by a sequence of
1733        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1734        string (as it is for production releases) called a macro with an empty
1735        argument. The C standard says the result of this is undefined. The gcc
1736        compiler treats it as an empty string (which was what was wanted) but it is
1737        reported that Visual C gives an error. The source has been hacked around to
1738        avoid this problem.
1740     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1741        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1742        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1743        of them did).
1745    10. Originally, pcretest opened its input and output without "b"; then I was
1746        told that "b" was needed in some environments, so it was added for release
1747        5.0 to both the input and output. (It makes no difference on Unix-like
1748        systems.) Later I was told that it is wrong for the input on Windows. I've
1749        now abstracted the modes into two macros, to make it easier to fiddle with
1750        them, and removed "b" from the input mode under Windows.
1752    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1754    12. Added -help and --help to pcretest as an official way of being reminded
1755        of the options.
1757    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1758        and pcrecpp.cc because they annoy compilers at high warning levels.
1760    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1762    15. Fixed an occurrence of == in configure.ac that should have been = (shell
1763        scripts are not C programs :-) and which was not noticed because it works
1764        on Linux.
1766    16. pcretest is supposed to handle any length of pattern and data line (as one
1767        line or as a continued sequence of lines) by extending its input buffer if
1768        necessary. This feature was broken for very long pattern lines, leading to
1769        a string of junk being passed to pcre_compile() if the pattern was longer
1770        than about 50K.
1772    17. I have done a major re-factoring of the way pcre_compile() computes the
1773        amount of memory needed for a compiled pattern. Previously, there was code
1774        that made a preliminary scan of the pattern in order to do this. That was
1775        OK when PCRE was new, but as the facilities have expanded, it has become
1776        harder and harder to keep it in step with the real compile phase, and there
1777        have been a number of bugs (see for example, 4 above). I have now found a
1778        cunning way of running the real compile function in a "fake" mode that
1779        enables it to compute how much memory it would need, while actually only
1780        ever using a few hundred bytes of working memory and without too many
1781        tests of the mode. This should make future maintenance and development
1782        easier. A side effect of this work is that the limit of 200 on the nesting
1783        depth of parentheses has been removed (though this was never a serious
1784        limitation, I suspect). However, there is a downside: pcre_compile() now
1785        runs more slowly than before (30% or more, depending on the pattern). I
1786        hope this isn't a big issue. There is no effect on runtime performance.
1788    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1789        newline (only possible for the last line of a file) and it was a
1790        pattern that set a locale (followed by /Lsomething), pcretest crashed.
1792    19. Added additional timing features to pcretest. (1) The -tm option now times
1793        matching only, not compiling. (2) Both -t and -tm can be followed, as a
1794        separate command line item, by a number that specifies the number of
1795        repeats to use when timing. The default is 50000; this gives better
1796        precision, but takes uncomfortably long for very large patterns.
1798    20. Extended pcre_study() to be more clever in cases where a branch of a
1799        subpattern has no definite first character. For example, (a*|b*)[cd] would
1800        previously give no result from pcre_study(). Now it recognizes that the
1801        first character must be a, b, c, or d.
1803    21. There was an incorrect error "recursive call could loop indefinitely" if
1804        a subpattern (or the entire pattern) that was being tested for matching an
1805        empty string contained only one non-empty item after a nested subpattern.
1806        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1807        incorrectly, because the \d was being skipped in the check.
1809    22. The pcretest program now has a new pattern option /B and a command line
1810        option -b, which is equivalent to adding /B to every pattern. This causes
1811        it to show the compiled bytecode, without the additional information that
1812        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1813        is the same as /B/I).
1815    23. A new optimization is now able automatically to treat some sequences such
1816        as a*b as a*+b. More specifically, if something simple (such as a character
1817        or a simple class like \d) has an unlimited quantifier, and is followed by
1818        something that cannot possibly match the quantified thing, the quantifier
1819        is automatically "possessified".
1821    24. A recursive reference to a subpattern whose number was greater than 39
1822        went wrong under certain circumstances in UTF-8 mode. This bug could also
1823        have affected the operation of pcre_study().
1825    25. Realized that a little bit of performance could be had by replacing
1826        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1828    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1830    27. Possessive quantifiers such as a++ were previously implemented by turning
1831        them into atomic groups such as ($>a+). Now they have their own opcodes,
1832        which improves performance. This includes the automatically created ones
1833        from 23 above.
1835    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1836        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1837        the first matched character to be a colon. This applied both to named and
1838        numbered groups.
1840    29. The ucpinternal.h header file was missing its idempotency #ifdef.
1842    30. I was sent a "project" file called libpcre.a.dev which I understand makes
1843        building PCRE on Windows easier, so I have included it in the distribution.
1845    31. There is now a check in pcretest against a ridiculously large number being
1846        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1847        loop, the loop is abandoned.
1849    32. Forward references to subpatterns in conditions such as (?(2)...) where
1850        subpattern 2 is defined later cause pcre_compile() to search forwards in
1851        the pattern for the relevant set of parentheses. This search went wrong
1852        when there were unescaped parentheses in a character class, parentheses
1853        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1855    33. "Subroutine" calls and backreferences were previously restricted to
1856        referencing subpatterns earlier in the regex. This restriction has now
1857        been removed.
1859    34. Added a number of extra features that are going to be in Perl 5.10. On the
1860        whole, these are just syntactic alternatives for features that PCRE had
1861        previously implemented using the Python syntax or my own invention. The
1862        other formats are all retained for compatibility.
1864        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1865            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1866            also .NET compatible.
1868        (b) A recursion or subroutine call to a named group can now be defined as
1869            (?&name) as well as (?P>name).
1871        (c) A backreference to a named group can now be defined as \k<name> or
1872            \k'name' as well as (?P=name). The new forms, as well as being in Perl
1873            5.10, are also .NET compatible.
1875        (d) A conditional reference to a named group can now use the syntax
1876            (?(<name>) or (?('name') as well as (?(name).
1878        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1879            groups (named and numbered) that are never evaluated inline, but can be
1880            called as "subroutines" from elsewhere. In effect, the DEFINE condition
1881            is always false. There may be only one alternative in such a group.
1883        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1884            as the simple (?(R). The condition is true only if the most recent
1885            recursion is that of the given number or name. It does not search out
1886            through the entire recursion stack.
1888        (g) The escape \gN or \g{N} has been added, where N is a positive or
1889            negative number, specifying an absolute or relative reference.
1891    35. Tidied to get rid of some further signed/unsigned compiler warnings and
1892        some "unreachable code" warnings.
1894    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1895        things, this adds five new scripts.
1897    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1898        There were also incompatibilities regarding the handling of \Q..\E inside
1899        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1900        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1902    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1903        matches an empty string, and forcibly breaks the loop. There were bugs in
1904        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
1905        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
1906        separate and independent bugs (that affected different cases) have been
1907        fixed.
1909    39. Refactored the code to abolish the use of different opcodes for small
1910        capturing bracket numbers. This is a tidy that I avoided doing when I
1911        removed the limit on the number of capturing brackets for 3.5 back in 2001.
1912        The new approach is not only tidier, it makes it possible to reduce the
1913        memory needed to fix the previous bug (38).
1915    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1916        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1917        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1918        mode.
1920    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1921        report.
1923    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1924        copy construction and assignment in the C++ wrapper.
1926    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1927        couple of bugs that could have given wrong results in the "--newline=crlf"
1928        case.
1930    44. Added a number of casts and did some reorganization of signed/unsigned int
1931        variables following suggestions from Dair Grant. Also renamed the variable
1932        "this" as "item" because it is a C++ keyword.
1934    45. Arranged for dftables to add
1936          #include "pcre_internal.h"
1938        to pcre_chartables.c because without it, gcc 4.x may remove the array
1939        definition from the final binary if PCRE is built into a static library and
1940        dead code stripping is activated.
1942    46. For an unanchored pattern, if a match attempt fails at the start of a
1943        newline sequence, and the newline setting is CRLF or ANY, and the next two
1944        characters are CRLF, advance by two characters instead of one.
1947    Version 6.7 04-Jul-06
1948    ---------------------
1950     1. In order to handle tests when input lines are enormously long, pcretest has
1951        been re-factored so that it automatically extends its buffers when
1952        necessary. The code is crude, but this _is_ just a test program. The
1953        default size has been increased from 32K to 50K.
1955     2. The code in pcre_study() was using the value of the re argument before
1956        testing it for NULL. (Of course, in any sensible call of the function, it
1957        won't be NULL.)
1959     3. The memmove() emulation function in pcre_internal.h, which is used on
1960        systems that lack both memmove() and bcopy() - that is, hardly ever -
1961        was missing a "static" storage class specifier.
1963     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1964        containing an extended class (one that cannot be represented by a bitmap
1965        because it contains high-valued characters or Unicode property items, e.g.
1966        [\pZ]). Almost always one would set UTF-8 mode when processing such a
1967        pattern, but PCRE should not loop if you do not (it no longer does).
1968        [Detail: two cases were found: (a) a repeated subpattern containing an
1969        extended class; (b) a recursive reference to a subpattern that followed a
1970        previous extended class. It wasn't skipping over the extended class
1971        correctly when UTF-8 mode was not set.]
1973     5. A negated single-character class was not being recognized as fixed-length
1974        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1975        compile error "lookbehind assertion is not fixed length".
1977     6. The RunPerlTest auxiliary script was showing an unexpected difference
1978        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1979        write a Perl script that can interpret lines of an input file either as
1980        byte characters or as UTF-8, which is what "perltest" was being required to
1981        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1982        can't do is switch easily at run time between having the "use utf8;" pragma
1983        or not. In the end, I fudged it by using the RunPerlTest script to insert
1984        "use utf8;" explicitly for the UTF-8 tests.
1986     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1987        the end of the subject string, contrary to the documentation and to what
1988        Perl does. This was true of both matching functions. Now it matches only at
1989        the start of the subject and immediately after *internal* newlines.
1991     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1992        a pointer to an int instead of a pointer to an unsigned long int. This
1993        caused problems on 64-bit systems.
1995     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1996        instance of the 'standard' template library not being so standard".
1998    10. There was no check on the number of named subpatterns nor the maximum
1999        length of a subpattern name. The product of these values is used to compute
2000        the size of the memory block for a compiled pattern. By supplying a very
2001        long subpattern name and a large number of named subpatterns, the size
2002        computation could be caused to overflow. This is now prevented by limiting
2003        the length of names to 32 characters, and the number of named subpatterns
2004        to 10,000.
2006    11. Subpatterns that are repeated with specific counts have to be replicated in
2007        the compiled pattern. The size of memory for this was computed from the
2008        length of the subpattern and the repeat count. The latter is limited to
2009        65535, but there was no limit on the former, meaning that integer overflow
2010        could in principle occur. The compiled length of a repeated subpattern is
2011        now limited to 30,000 bytes in order to prevent this.
2013    12. Added the optional facility to have named substrings with the same name.
2015    13. Added the ability to use a named substring as a condition, using the
2016        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
2017        are numbers (not recommended). Forward references are permitted.
2019    14. Added forward references in named backreferences (if you see what I mean).
2021    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
2022        pattern could run off the end of the subject. For example, the pattern
2023        "(?s)(.{1,5})"8 did this with the subject "ab".
2025    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
2026        PCRE_CASELESS was set when matching characters that were quantified with ?
2027        or *.
2029    17. A character class other than a single negated character that had a minimum
2030        but no maximum quantifier - for example [ab]{6,} - was not handled
2031        correctly by pce_dfa_exec(). It would match only one character.
2033    18. A valid (though odd) pattern that looked like a POSIX character
2034        class but used an invalid character after [ (for example [[,abc,]]) caused
2035        pcre_compile() to give the error "Failed: internal error: code overflow" or
2036        in some cases to crash with a glibc free() error. This could even happen if
2037        the pattern terminated after [[ but there just happened to be a sequence of
2038        letters, a binary zero, and a closing ] in the memory that followed.
2040    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
2041        over the years. Originally (before any Unicode support), just the bottom 8
2042        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
2043        output from "man perlunicode" includes this:
2045          The regular expression compiler produces polymorphic opcodes.  That
2046          is, the pattern adapts to the data and automatically switches to
2047          the Unicode character scheme when presented with Unicode data--or
2048          instead uses a traditional byte scheme when presented with byte
2049          data.
2051        Sadly, a wide octal escape does not cause a switch, and in a string with
2052        no other multibyte characters, these octal escapes are treated as before.
2053        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
2054        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
2055        Unicode string.
2057        I have not perpetrated such confusion in PCRE. Up till now, it took just
2058        the bottom 8 bits, as in old Perl. I have now made octal escapes with
2059        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
2060        translate to the appropriate multibyte character.
2062    29. Applied some refactoring to reduce the number of warnings from Microsoft
2063        and Borland compilers. This has included removing the fudge introduced
2064        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
2065        a warning about an unused variable.
2067    21. PCRE has not included VT (character 0x0b) in the set of whitespace
2068        characters since release 4.0, because Perl (from release 5.004) does not.
2069        [Or at least, is documented not to: some releases seem to be in conflict
2070        with the documentation.] However, when a pattern was studied with
2071        pcre_study() and all its branches started with \s, PCRE still included VT
2072        as a possible starting character. Of course, this did no harm; it just
2073        caused an unnecessary match attempt.
2075    22. Removed a now-redundant internal flag bit that recorded the fact that case
2076        dependency changed within the pattern. This was once needed for "required
2077        byte" processing, but is no longer used. This recovers a now-scarce options
2078        bit. Also moved the least significant internal flag bit to the most-
2079        significant bit of the word, which was not previously used (hangover from
2080        the days when it was an int rather than a uint) to free up another bit for
2081        the future.
2083    23. Added support for CRLF line endings as well as CR and LF. As well as the
2084        default being selectable at build time, it can now be changed at runtime
2085        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
2086        specify that it is scanning data with non-default line endings.
2088    24. Changed the definition of CXXLINK to make it agree with the definition of
2089        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
2091    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
2092        recursions. This makes a big different to stack usage for some patterns.
2094    26. If a subpattern containing a named recursion or subroutine reference such
2095        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
2096        the space required for the compiled pattern went wrong and gave too small a
2097        value. Depending on the environment, this could lead to "Failed: internal
2098        error: code overflow at offset 49" or "glibc detected double free or
2099        corruption" errors.
2101    27. Applied patches from Google (a) to support the new newline modes and (b) to
2102        advance over multibyte UTF-8 characters in GlobalReplace.
2104    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
2105        difference for some implementation of PCRE in some Windows version.
2107    29. Added some extra testing facilities to pcretest:
2109        \q<number>   in a data line sets the "match limit" value
2110        \Q<number>   in a data line sets the "match recursion limt" value
2111        -S <number>  sets the stack size, where <number> is in megabytes
2113        The -S option isn't available for Windows.
2116    Version 6.6 06-Feb-06
2117    ---------------------
2119     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
2120        in pcreposix.h. I have copied the definition from pcre.h.
2122     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
2123        because pcre.h is no longer a built file.
2125     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
2126        not normally included in the compiled code.
2129    Version 6.5 01-Feb-06
2130    ---------------------
2132     1. When using the partial match feature with pcre_dfa_exec(), it was not
2133        anchoring the second and subsequent partial matches at the new starting
2134        point. This could lead to incorrect results. For example, with the pattern
2135        /1234/, partially matching against "123" and then "a4" gave a match.
2137     2. Changes to pcregrep:
2139        (a) All non-match returns from pcre_exec() were being treated as failures
2140            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
2141            error message is output. Some extra information is given for the
2142            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
2143            probably the only errors that are likely to be caused by users (by
2144            specifying a regex that has nested indefinite repeats, for instance).
2145            If there are more than 20 of these errors, pcregrep is abandoned.
2147        (b) A binary zero was treated as data while matching, but terminated the
2148            output line if it was written out. This has been fixed: binary zeroes
2149            are now no different to any other data bytes.
2151        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
2152            used to set a locale for matching. The --locale=xxxx long option has
2153            been added (no short equivalent) to specify a locale explicitly on the
2154            pcregrep command, overriding the environment variables.
2156        (d) When -B was used with -n, some line numbers in the output were one less
2157            than they should have been.
2159        (e) Added the -o (--only-matching) option.
2161        (f) If -A or -C was used with -c (count only), some lines of context were
2162            accidentally printed for the final match.
2164        (g) Added the -H (--with-filename) option.
2166        (h) The combination of options -rh failed to suppress file names for files
2167            that were found from directory arguments.
2169        (i) Added the -D (--devices) and -d (--directories) options.
2171        (j) Added the -F (--fixed-strings) option.
2173        (k) Allow "-" to be used as a file name for -f as well as for a data file.
2175        (l) Added the --colo(u)r option.
2177        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
2178            is not present by default.
2180     3. A nasty bug was discovered in the handling of recursive patterns, that is,
2181        items such as (?R) or (?1), when the recursion could match a number of
2182        alternatives. If it matched one of the alternatives, but subsequently,
2183        outside the recursion, there was a failure, the code tried to back up into
2184        the recursion. However, because of the way PCRE is implemented, this is not
2185        possible, and the result was an incorrect result from the match.
2187        In order to prevent this happening, the specification of recursion has
2188        been changed so that all such subpatterns are automatically treated as
2189        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
2191     4. I had overlooked the fact that, in some locales, there are characters for
2192        which isalpha() is true but neither isupper() nor islower() are true. In
2193        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
2194        and ordfeminine) are like this. This affected the treatment of \w and \W
2195        when they appeared in character classes, but not when they appeared outside
2196        a character class. The bit map for "word" characters is now created
2197        separately from the results of isalnum() instead of just taking it from the
2198        upper, lower, and digit maps. (Plus the underscore character, of course.)
2200     5. The above bug also affected the handling of POSIX character classes such as
2201        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
2202        permanent tables. Instead, the bit maps for such a class were previously
2203        created as the appropriate unions of the upper, lower, and digit bitmaps.
2204        Now they are created by subtraction from the [[:word:]] class, which has
2205        its own bitmap.
2207     6. The [[:blank:]] character class matches horizontal, but not vertical space.
2208        It is created by subtracting the vertical space characters (\x09, \x0a,
2209        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
2210        subtraction was done in the overall bitmap for a character class, meaning
2211        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
2212        be recognized. This bug has been fixed.
2214     7. Patches from the folks at Google:
2216          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
2217          real life, but is still worth protecting against".
2219          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
2220          regular expressions".
2222          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
2223          have it.
2225          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
2226          "configure" and the latter not, in order to fix a problem somebody had
2227          with compiling the Arg class on HP-UX.
2229          (e) Improve the error-handling of the C++ wrapper a little bit.
2231          (f) New tests for checking recursion limiting.
2233     8. The pcre_memmove() function, which is used only if the environment does not
2234        have a standard memmove() function (and is therefore rarely compiled),
2235        contained two bugs: (a) use of int instead of size_t, and (b) it was not
2236        returning a result (though PCRE never actually uses the result).
2238     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
2239        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
2240        returned instead of calling malloc() with an overflowing number that would
2241        most likely cause subsequent chaos.
2243    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
2245    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
2246        with this option is matched, the nmatch and pmatch options of regexec() are
2247        ignored.
2249    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
2250        provided in case anyone wants to the the POSIX interface with UTF-8
2251        strings.
2253    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
2254        C++ linking (needed for some HP-UX environments).
2256    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
2257        (unused parameter) and in the pcre_printint() function (omitted "default"
2258        switch label when the default is to do nothing).
2260    15. Added some code to make it possible, when PCRE is compiled as a C++
2261        library, to replace subject pointers for pcre_exec() with a smart pointer
2262        class, thus making it possible to process discontinuous strings.
2264    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
2265        much the same function. They were added by different people who were trying
2266        to make PCRE easy to compile on non-Unix systems. It has been suggested
2267        that PCRE_EXPORT be abolished now that there is more automatic apparatus
2268        for compiling on Windows systems. I have therefore replaced it with
2269        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
2270        defaults to "extern" for C or "extern C" for C++, which works fine on
2271        Unix-like systems. It is now possible to override the value of PCRE_DATA_
2272        SCOPE with something explicit in config.h. In addition:
2274        (a) pcreposix.h still had just "extern" instead of either of these macros;
2275            I have replaced it with PCRE_DATA_SCOPE.
2277        (b) Functions such as _pcre_xclass(), which are internal to the library,
2278            but external in the C sense, all had PCRE_EXPORT in their definitions.
2279            This is apparently wrong for the Windows case, so I have removed it.
2280            (It makes no difference on Unix-like systems.)
2282    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
2283        of recursive calls to match(). This is different to MATCH_LIMIT because
2284        that limits the total number of calls to match(), not all of which increase
2285        the depth of recursion. Limiting the recursion depth limits the amount of
2286        stack (or heap if NO_RECURSE is set) that is used. The default can be set
2287        when PCRE is compiled, and changed at run time. A patch from Google adds
2288        this functionality to the C++ interface.
2290    18. Changes to the handling of Unicode character properties:
2292        (a) Updated the table to Unicode 4.1.0.
2294        (b) Recognize characters that are not in the table as "Cn" (undefined).
2296        (c) I revised the way the table is implemented to a much improved format
2297            which includes recognition of ranges. It now supports the ranges that
2298            are defined in UnicodeData.txt, and it also amalgamates other
2299            characters into ranges. This has reduced the number of entries in the
2300            table from around 16,000 to around 3,000, thus reducing its size
2301            considerably. I realized I did not need to use a tree structure after
2302            all - a binary chop search is just as efficient. Having reduced the
2303            number of entries, I extended their size from 6 bytes to 8 bytes to
2304            allow for more data.
2306        (d) Added support for Unicode script names via properties such as \p{Han}.
2308    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
2309        matching that character.
2311    20. When matching a repeated Unicode property with a minimum greater than zero,
2312        (for example \pL{2,}), PCRE could look past the end of the subject if it
2313        reached it while seeking the minimum number of characters. This could
2314        happen only if some of the characters were more than one byte long, because
2315        there is a check for at least the minimum number of bytes.
2317    21. Refactored the implementation of \p and \P so as to be more general, to
2318        allow for more different types of property in future. This has changed the
2319        compiled form incompatibly. Anybody with saved compiled patterns that use
2320        \p or \P will have to recompile them.
2322    22. Added "Any" and "L&" to the supported property types.
2324    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
2325        but give a compile time error if the value is greater than 0xff.
2327    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
2328        accidentally not being installed or uninstalled.
2330    25. The pcre.h file was built from pcre.h.in, but the only changes that were
2331        made were to insert the current release number. This seemed silly, because
2332        it made things harder for people building PCRE on systems that don't run
2333        "configure". I have turned pcre.h into a distributed file, no longer built
2334        by "configure", with the version identification directly included. There is
2335        no longer a pcre.h.in file.
2337        However, this change necessitated a change to the pcre-config script as
2338        well. It is built from pcre-config.in, and one of the substitutions was the
2339        release number. I have updated configure.ac so that ./configure now finds
2340        the release number by grepping pcre.h.
2342    26. Added the ability to run the tests under valgrind.
2345    Version 6.4 05-Sep-05
2346    ---------------------
2348     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
2349        "--" to be printed when multiple files were scanned, even when none of the
2350        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
2351        consider it to be a bug, and have restored the previous behaviour.
2353     2. A couple of code tidies to get rid of compiler warnings.
2355     3. The pcretest program used to cheat by referring to symbols in the library
2356        whose names begin with _pcre_. These are internal symbols that are not
2357        really supposed to be visible externally, and in some environments it is
2358        possible to suppress them. The cheating is now confined to including
2359        certain files from the library's source, which is a bit cleaner.
2361     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
2362        file's purpose clearer.
2364     5. Reorganized pcre_ucp_findchar().
2367    Version 6.3 15-Aug-05
2368    ---------------------
2370     1. The file libpcre.pc.in did not have general read permission in the tarball.
2372     2. There were some problems when building without C++ support:
2374        (a) If C++ support was not built, "make install" and "make test" still
2375            tried to test it.
2377        (b) There were problems when the value of CXX was explicitly set. Some
2378            changes have been made to try to fix these, and ...
2380        (c) --disable-cpp can now be used to explicitly disable C++ support.
2382        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
2383            backslash in a target when C++ was disabled. This confuses some
2384            versions of "make", apparently. Using an intermediate variable solves
2385            this. (Same for CPP_LOBJ.)
2387     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
2388        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
2389        necessary on certain architectures.
2391     4. Added a setting of -export-symbols-regex to the link command to remove
2392        those symbols that are exported in the C sense, but actually are local
2393        within the library, and not documented. Their names all begin with
2394        "_pcre_". This is not a perfect job, because (a) we have to except some
2395        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
2396        available (and never for static libraries). I have made a note to try to
2397        find a way round (a) in the future.
2400    Version 6.2 01-Aug-05
2401    ---------------------
2403     1. There was no test for integer overflow of quantifier values. A construction
2404        such as {1111111111111111} would give undefined results. What is worse, if
2405        a minimum quantifier for a parenthesized subpattern overflowed and became
2406        negative, the calculation of the memory size went wrong. This could have
2407        led to memory overwriting.
2409     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
2411     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
2412        operating environments where this matters.
2414     4. Applied Giuseppe Maxia's patch to add additional features for controlling
2415        PCRE options from within the C++ wrapper.
2417     5. Named capturing subpatterns were not being correctly counted when a pattern
2418        was compiled. This caused two problems: (a) If there were more than 100
2419        such subpatterns, the calculation of the memory needed for the whole
2420        compiled pattern went wrong, leading to an overflow error. (b) Numerical
2421        back references of the form \12, where the number was greater than 9, were
2422        not recognized as back references, even though there were sufficient
2423        previous subpatterns.
2425     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
2426        versions of gcc, e.g. 2.95.4.
2429    Version 6.1 21-Jun-05
2430    ---------------------
2432     1. There was one reference to the variable "posix" in pcretest.c that was not
2433        surrounded by "#if !defined NOPOSIX".
2435     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
2436        the cross-check on the old pcre_info() function, for the benefit of the
2437        cut-down version of PCRE that is currently imported into Exim.
2439     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
2440        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
2441        patterns that start (?i) but I don't think that matters. The use of (?i) is
2442        just an example; this all applies to the other options as well.
2444     4. Since libtool seems to echo the compile commands it is issuing, the output
2445        from "make" can be reduced a bit by putting "@" in front of each libtool
2446        compile command.
2448     5. Patch from the folks at Google for configure.in to be a bit more thorough
2449        in checking for a suitable C++ installation before trying to compile the
2450        C++ stuff. This should fix a reported problem when a compiler was present,
2451        but no suitable headers.
2453     6. The man pages all had just "PCRE" as their title. I have changed them to
2454        be the relevant file name. I have also arranged that these names are
2455        retained in the file doc/pcre.txt, which is a concatenation in text format
2456        of all the man pages except the little individual ones for each function.
2458     7. The NON-UNIX-USE file had not been updated for the different set of source
2459        files that come with release 6. I also added a few comments about the C++
2460        wrapper.
2463    Version 6.0 07-Jun-05
2464    ---------------------
2466     1. Some minor internal re-organization to help with my DFA experiments.
2468     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
2469        didn't matter for the library itself when fully configured, but did matter
2470        when compiling without UCP support, or within Exim, where the ucp files are
2471        not imported.
2473     3. Refactoring of the library code to split up the various functions into
2474        different source modules. The addition of the new DFA matching code (see
2475        below) to a single monolithic source would have made it really too
2476        unwieldy, quite apart from causing all the code to be include in a
2477        statically linked application, when only some functions are used. This is
2478        relevant even without the DFA addition now that patterns can be compiled in
2479        one application and matched in another.
2481        The downside of splitting up is that there have to be some external
2482        functions and data tables that are used internally in different modules of
2483        the library but which are not part of the API. These have all had their
2484        names changed to start with "_pcre_" so that they are unlikely to clash
2485        with other external names.
2487     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
2488        a different (DFA) algorithm. Although it is slower than the original
2489        function, it does have some advantages for certain types of matching
2490        problem.
2492     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
2493        including restarting after a partial match.
2495     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
2496        defined when compiling for Windows was sent to me. I have put it into the
2497        code, though I have no means of testing or verifying it.
2499     7. Added the pcre_refcount() auxiliary function.
2501     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
2502        match before or at the first newline in the subject string. In pcretest,
2503        the /f option on a pattern can be used to set this.
2505     9. A repeated \w when used in UTF-8 mode with characters greater than 256
2506        would behave wrongly. This has been present in PCRE since release 4.0.
2508    10. A number of changes to the pcregrep command:
2510        (a) Refactored how -x works; insert ^(...)$ instead of setting
2511            PCRE_ANCHORED and checking the length, in preparation for adding
2512            something similar for -w.
2514        (b) Added the -w (match as a word) option.
2516        (c) Refactored the way lines are read and buffered so as to have more
2517            than one at a time available.
2519        (d) Implemented a pcregrep test script.
2521        (e) Added the -M (multiline match) option. This allows patterns to match
2522            over several lines of the subject. The buffering ensures that at least
2523            8K, or the rest of the document (whichever is the shorter) is available
2524            for matching (and similarly the previous 8K for lookbehind assertions).
2526        (f) Changed the --help output so that it now says
2528              -w, --word-regex(p)
2530            instead of two lines, one with "regex" and the other with "regexp"
2531            because that confused at least one person since the short forms are the
2532            same. (This required a bit of code, as the output is generated
2533            automatically from a table. It wasn't just a text change.)
2535        (g) -- can be used to terminate pcregrep options if the next thing isn't an
2536            option but starts with a hyphen. Could be a pattern or a path name
2537            starting with a hyphen, for instance.
2539        (h) "-" can be given as a file name to represent stdin.
2541        (i) When file names are being printed, "(standard input)" is used for
2542            the standard input, for compatibility with GNU grep. Previously
2543            "<stdin>" was used.
2545        (j) The option --label=xxx can be used to supply a name to be used for
2546            stdin when file names are being printed. There is no short form.
2548        (k) Re-factored the options decoding logic because we are going to add
2549            two more options that take data. Such options can now be given in four
2550            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
2552        (l) Added the -A, -B, and -C options for requesting that lines of context
2553            around matches be printed.
2555        (m) Added the -L option to print the names of files that do not contain
2556            any matching lines, that is, the complement of -l.
2558        (n) The return code is 2 if any file cannot be opened, but pcregrep does
2559            continue to scan other files.
2561        (o) The -s option was incorrectly implemented. For compatibility with other
2562            greps, it now suppresses the error message for a non-existent or non-
2563            accessible file (but not the return code). There is a new option called
2564            -q that suppresses the output of matching lines, which was what -s was
2565            previously doing.
2567        (p) Added --include and --exclude options to specify files for inclusion
2568            and exclusion when recursing.
2570    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
2571        Hopefully, it now does.
2573    12. Missing cast in pcre_study().
2575    13. Added an "uninstall" target to the makefile.
2577    14. Replaced "extern" in the function prototypes in Makefile.in with
2578        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
2579        world, but is set differently for Windows.
2581    15. Added a second compiling function called pcre_compile2(). The only
2582        difference is that it has an extra argument, which is a pointer to an
2583        integer error code. When there is a compile-time failure, this is set
2584        non-zero, in addition to the error test pointer being set to point to an
2585        error message. The new argument may be NULL if no error number is required
2586        (but then you may as well call pcre_compile(), which is now just a
2587        wrapper). This facility is provided because some applications need a
2588        numeric error indication, but it has also enabled me to tidy up the way
2589        compile-time errors are handled in the POSIX wrapper.
2591    16. Added VPATH=.libs to the makefile; this should help when building with one
2592        prefix path and installing with another. (Or so I'm told by someone who
2593        knows more about this stuff than I do.)
2595    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
2596        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
2597        match everything, including newlines. This is not POSIX-compatible, but
2598        somebody wanted the feature. From pcretest it can be activated by using
2599        both the P and the s flags.
2601    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
2603    19. libpcre.pc was being incorrectly installed as executable.
2605    20. A couple of places in pcretest check for end-of-line by looking for '\n';
2606        it now also looks for '\r' so that it will work unmodified on Windows.
2608    21. Added Google's contributed C++ wrapper to the distribution.
2610    22. Added some untidy missing memory free() calls in pcretest, to keep
2611        Electric Fence happy when testing.
2615    Version 5.0 13-Sep-04
2616    ---------------------
2618     1. Internal change: literal characters are no longer packed up into items
2619        containing multiple characters in a single byte-string. Each character
2620        is now matched using a separate opcode. However, there may be more than one
2621        byte in the character in UTF-8 mode.
2623     2. The pcre_callout_block structure has two new fields: pattern_position and
2624        next_item_length. These contain the offset in the pattern to the next match
2625        item, and its length, respectively.
2627     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
2628        insertion of callouts before each pattern item. Added the /C option to
2629        pcretest to make use of this.
2631     4. On the advice of a Windows user, the lines
2633          #if defined(_WIN32) || defined(WIN32)
2634          _setmode( _fileno( stdout ), 0x8000 );
2635          #endif  /* defined(_WIN32) || defined(WIN32) */
2637        have been added to the source of pcretest. This apparently does useful
2638        magic in relation to line terminators.
2640     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
2641        for the benefit of those environments where the "b" makes a difference.
2643     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
2644        to know about it. I have put a hack into configure.in that adds in code
2645        to set GCC=yes if CC=icc. This seems to end up at a point in the
2646        generated configure script that is early enough to affect the setting of
2647        compiler options, which is what is needed, but I have no means of testing
2648        whether it really works. (The user who reported this had patched the
2649        generated configure script, which of course I cannot do.)
2651        LATER: After change 22 below (new libtool files), the configure script
2652        seems to know about icc (and also ecc). Therefore, I have commented out
2653        this hack in configure.in.
2655     7. Added support for pkg-config (2 patches were sent in).
2657     8. Negated POSIX character classes that used a combination of internal tables
2658        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2659        [[:^ascii]]. Typically, they would match almost any characters. The other
2660        POSIX classes were not broken in this way.
2662     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2663        to find the match, as PCRE was deluded into thinking that the match had to
2664        start at the start point or following a newline. The same bug applied to
2665        patterns with negative forward assertions or any backward assertions
2666        preceding ".*" at the start, unless the pattern required a fixed first
2667        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2669    10. In UTF-8 mode, when moving forwards in the subject after a failed match
2670        starting at the last subject character, bytes beyond the end of the subject
2671        string were read.
2673    11. Renamed the variable "class" as "classbits" to make life easier for C++
2674        users. (Previously there was a macro definition, but it apparently wasn't
2675        enough.)
2677    12. Added the new field "tables" to the extra data so that tables can be passed
2678        in at exec time, or the internal tables can be re-selected. This allows
2679        a compiled regex to be saved and re-used at a later time by a different
2680        program that might have everything at different addresses.
2682    13. Modified the pcre-config script so that, when run on Solaris, it shows a
2683        -R library as well as a -L library.
2685    14. The debugging options of pcretest (-d on the command line or D on a
2686        pattern) showed incorrect output for anything following an extended class
2687        that contained multibyte characters and which was followed by a quantifier.
2689    15. Added optional support for general category Unicode character properties
2690        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2691        support. It adds about 90K to the size of the library. The meanings of the
2692        inbuilt class escapes such as \d and \s have NOT been changed.
2694    16. Updated pcredemo.c to include calls to free() to release the memory for the
2695        compiled pattern.
2697    17. The generated file chartables.c was being created in the source directory
2698        instead of in the building directory. This caused the build to fail if the
2699        source directory was different from the building directory, and was
2700        read-only.
2702    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2703        file. No doubt somebody will tell me if they don't make sense... Also added
2704        Dan Mooney's comments about building on OpenVMS.
2706    19. Added support for partial matching via the PCRE_PARTIAL option for
2707        pcre_exec() and the \P data escape in pcretest.
2709    20. Extended pcretest with 3 new pattern features:
2711        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
2712              write the compiled pattern to the file whose name is "rest-of-line".
2713              This is a straight binary dump of the data, with the saved pointer to
2714              the character tables forced to be NULL. The study data, if any, is
2715              written too. After writing, pcretest reads a new pattern.
2717        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2718              compiled pattern from the given file. There must not be any
2719              occurrences of "<" in the file name (pretty unlikely); if there are,
2720              pcretest will instead treat the initial "<" as a pattern delimiter.
2721              After reading in the pattern, pcretest goes on to read data lines as
2722              usual.
2724        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2725              and 16-bit fields in a compiled pattern, to simulate a pattern that
2726              was compiled on a host of opposite endianness.
2728    21. The pcre-exec() function can now cope with patterns that were compiled on
2729        hosts of opposite endianness, with this restriction:
2731          As for any compiled expression that is saved and used later, the tables
2732          pointer field cannot be preserved; the extra_data field in the arguments
2733          to pcre_exec() should be used to pass in a tables address if a value
2734          other than the default internal tables were used at compile time.
2736    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2737        now diagnosed as an error. Previously, most of the time, a negative number
2738        would have been treated as zero, but if in addition "ovector" was passed as
2739        NULL, a crash could occur.
2741    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2742        new versions from the libtool 1.5 distribution (the last one is a copy of
2743        a file called libtool.m4). This seems to have fixed the need to patch
2744        "configure" to support Darwin 1.3 (which I used to do). However, I still
2745        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2746        workstation).
2748    24. Changed the PCRE licence to be the more standard "BSD" licence.
2751    Version 4.5 01-Dec-03
2752    ---------------------
2754     1. There has been some re-arrangement of the code for the match() function so
2755        that it can be compiled in a version that does not call itself recursively.
2756        Instead, it keeps those local variables that need separate instances for
2757        each "recursion" in a frame on the heap, and gets/frees frames whenever it
2758        needs to "recurse". Keeping track of where control must go is done by means
2759        of setjmp/longjmp. The whole thing is implemented by a set of macros that
2760        hide most of the details from the main code, and operates only if
2761        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2762        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2763        operating.
2765        To make it easier for callers to provide specially tailored get/free
2766        functions for this usage, two new functions, pcre_stack_malloc, and
2767        pcre_stack_free, are used. They are always called in strict stacking order,
2768        and the size of block requested is always the same.
2770        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2771        PCRE has been compiled to use the stack or the heap for recursion. The
2772        -C option of pcretest uses this to show which version is compiled.
2774        A new data escape \S, is added to pcretest; it causes the amounts of store
2775        obtained and freed by both kinds of malloc/free at match time to be added
2776        to the output.
2778     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2779        what's available on my current Linux desktop machine.
2781     3. When matching a UTF-8 string, the test for a valid string at the start has
2782        been extended. If start_offset is not zero, PCRE now checks that it points
2783        to a byte that is the start of a UTF-8 character. If not, it returns
2784        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2785        this is necessary because there may be backward assertions in the pattern.
2786        When matching the same subject several times, it may save resources to use
2787        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2789     4. The code for checking the validity of UTF-8 strings has been tightened so
2790        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2791        containing "overlong sequences".
2793     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2794        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2795        should have been "&", but it just so happened that all the cases this let
2796        through by mistake were picked up later in the function.
2798     6. I had used a variable called "isblank" - this is a C99 function, causing
2799        some compilers to warn. To avoid this, I renamed it (as "blankclass").
2801     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2802        prompting; (b) run "./pcretest /dev/null" at the start of the test script
2803        so the version is shown; (c) stop "make test" echoing "./RunTest".
2805     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2807     9. The prototype for memmove() for systems that don't have it was using
2808        size_t, but the inclusion of the header that defines size_t was later. I've
2809        moved the #includes for the C headers earlier to avoid this.
2811    10. Added some adjustments to the code to make it easier to compiler on certain
2812        special systems:
2814          (a) Some "const" qualifiers were missing.
2815          (b) Added the macro EXPORT before all exported functions; by default this
2816              is defined to be empty.
2817          (c) Changed the dftables auxiliary program (that builds chartables.c) so
2818              that it reads its output file name as an argument instead of writing
2819              to the standard output and assuming this can be redirected.
2821    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2822        class containing characters with values greater than 255, PCRE compilation
2823        went into a loop.
2825    12. A recursive reference to a subpattern that was within another subpattern
2826        that had a minimum quantifier of zero caused PCRE to crash. For example,
2827        (x(y(?2))z)? provoked this bug with a subject that got as far as the
2828        recursion. If the recursively-called subpattern itself had a zero repeat,
2829        that was OK.
2831    13. In pcretest, the buffer for reading a data line was set at 30K, but the
2832        buffer into which it was copied (for escape processing) was still set at
2833        1024, so long lines caused crashes.
2835    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2836        "internal error: code overflow...". This applied to any character class
2837        that was followed by a possessive quantifier.
2839    15. Modified the Makefile to add libpcre.la as a prerequisite for
2840        libpcreposix.la because I was told this is needed for a parallel build to
2841        work.
2843    16. If a pattern that contained .* following optional items at the start was
2844        studied, the wrong optimizing data was generated, leading to matching
2845        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2846        matching string must start with a or b or c. The correct conclusion for
2847        this pattern is that a match can start with any character.
2850    Version 4.4 13-Aug-03
2851    ---------------------
2853     1. In UTF-8 mode, a character class containing characters with values between
2854        127 and 255 was not handled correctly if the compiled pattern was studied.
2855        In fixing this, I have also improved the studying algorithm for such
2856        classes (slightly).
2858     2. Three internal functions had redundant arguments passed to them. Removal
2859        might give a very teeny performance improvement.
2861     3. Documentation bug: the value of the capture_top field in a callout is *one
2862        more than* the number of the hightest numbered captured substring.
2864     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2865        in incorrectly linking with a previously installed version. They now link
2866        explicitly with libpcre.la.
2868     5. configure.in no longer needs to recognize Cygwin specially.
2870     6. A problem in pcre.in for Windows platforms is fixed.
2872     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2873        pcretest, it used to include the size of the study block as part of its
2874        output. Unfortunately, the structure contains a field that has a different
2875        size on different hardware architectures. This meant that the tests that
2876        showed this size failed. As the block is currently always of a fixed size,
2877        this information isn't actually particularly useful in pcretest output, so
2878        I have just removed it.
2880     8. Three pre-processor statements accidentally did not start in column 1.
2881        Sadly, there are *still* compilers around that complain, even though
2882        standard C has not required this for well over a decade. Sigh.
2884     9. In pcretest, the code for checking callouts passed small integers in the
2885        callout_data field, which is a void * field. However, some picky compilers
2886        complained about the casts involved for this on 64-bit systems. Now
2887        pcretest passes the address of the small integer instead, which should get
2888        rid of the warnings.
2890    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2891        both compile and run time, and gives an error if an invalid UTF-8 sequence
2892        is found. There is a option for disabling this check in cases where the
2893        string is known to be correct and/or the maximum performance is wanted.
2895    11. In response to a bug report, I changed one line in Makefile.in from
2897            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2898        to
2899            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2901        to look similar to other lines, but I have no way of telling whether this
2902        is the right thing to do, as I do not use Windows. No doubt I'll get told
2903        if it's wrong...
2906    Version 4.3 21-May-03
2907    ---------------------
2909    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2910       Makefile.
2912    2. Some refactoring to improve the quality of the code:
2914       (i)   The utf8_table... variables are now declared "const".
2916       (ii)  The code for \cx, which used the "case flipping" table to upper case
2917             lower case letters, now just substracts 32. This is ASCII-specific,
2918             but the whole concept of \cx is ASCII-specific, so it seems
2919             reasonable.
2921       (iii) PCRE was using its character types table to recognize decimal and
2922             hexadecimal digits in the pattern. This is silly, because it handles
2923             only 0-9, a-f, and A-F, but the character types table is locale-
2924             specific, which means strange things might happen. A private
2925             table is now used for this - though it costs 256 bytes, a table is
2926             much faster than multiple explicit tests. Of course, the standard
2927             character types table is still used for matching digits in subject
2928             strings against \d.
2930       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2931             ending in _t are). So I've renamed it as ESC_tee.
2933    3. The first argument for regexec() in the POSIX wrapper should have been
2934       defined as "const".
2936    4. Changed pcretest to use malloc() for its buffers so that they can be
2937       Electric Fenced for debugging.
2939    5. There were several places in the code where, in UTF-8 mode, PCRE would try
2940       to read one or more bytes before the start of the subject string. Often this
2941       had no effect on PCRE's behaviour, but in some circumstances it could
2942       provoke a segmentation fault.
2944    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2945       to try to read one or more bytes before the start of the subject string.
2947    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
2948       UTF-8 support could misbehave in various ways if the subject string
2949       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
2950       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
2951       back over UTF-8 characters.)
2954    Version 4.2 14-Apr-03
2955    ---------------------
2957    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
2959    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
2960         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
2961         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
2962         [WIN_PREFIX]: new variable, "cyg" for Cygwin
2963         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
2964           and BUILD_EXEEXT
2965         Note: automatic setting of the BUILD variables is not yet working
2966         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
2967           compile-time but not at link-time
2968         [LINK]: use for linking executables only
2969         make different versions for Windows and non-Windows
2970         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
2971           libraries
2972         [LINK_FOR_BUILD]: new variable
2973         [OBJEXT]: use throughout
2974         [EXEEXT]: use throughout
2975         <winshared>: new target
2976         <wininstall>: new target
2977         <dftables.o>: use native compiler
2978         <dftables>: use native linker
2979         <install>: handle Windows platform correctly
2980         <clean>: ditto
2981         <check>: ditto
2982         copy DLL to top builddir before testing
2984       As part of these changes, -no-undefined was removed again. This was reported
2985       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
2986       in any case.
2988    3. Some tidies to get rid of compiler warnings:
2990       . In the match_data structure, match_limit was an unsigned long int, whereas
2991         match_call_count was an int. I've made them both unsigned long ints.
2993       . In pcretest the fact that a const uschar * doesn't automatically cast to
2994         a void * provoked a warning.
2996       . Turning on some more compiler warnings threw up some "shadow" variables
2997         and a few more missing casts.
2999    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
3000       option, a class that contained a single character with a value between 128
3001       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
3003    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
3004       option, a class that contained several characters, but with at least one
3005       whose value was between 128 and 255 caused PCRE to crash.
3008    Version 4.1 12-Mar-03
3009    ---------------------
3011    1. Compiling with gcc -pedantic found a couple of places where casts were
3012    needed, and a string in dftables.c that was longer than standard compilers are
3013    required to support.
3015    2. Compiling with Sun's compiler found a few more places where the code could
3016    be tidied up in order to avoid warnings.
3018    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
3019    first of these names is deprecated in the latest Autoconf in favour of the name
3020    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
3021    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
3022    analogy I have changed it to CFLAGS_FOR_BUILD.
3024    4. Added -no-undefined to the linking command in the Makefile, because this is
3025    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
3026    linking step for the pcreposix library.
3028    5. PCRE was failing to diagnose the case of two named groups with the same
3029    name.
3031    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
3032    literal character that is needed in the subject for a match, and scans along to
3033    ensure that it is present before embarking on the full matching process. This
3034    saves time in cases of nested unlimited repeats that are never going to match.
3035    Problem: the scan can take a lot of time if the subject is very long (e.g.
3036    megabytes), thus penalizing straightforward matches. It is now done only if the
3037    amount of subject to be scanned is less than 1000 bytes.
3039    7. A lesser problem with the same optimization is that it was recording the
3040    first character of an anchored pattern as "needed", thus provoking a search
3041    right along the subject, even when the first match of the pattern was going to
3042    fail. The "needed" character is now not set for anchored patterns, unless it
3043    follows something in the pattern that is of non-fixed length. Thus, it still
3044    fulfils its original purpose of finding quick non-matches in cases of nested
3045    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
3048    Version 4.0 17-Feb-03
3049    ---------------------
3051    1. If a comment in an extended regex that started immediately after a meta-item
3052    extended to the end of string, PCRE compiled incorrect data. This could lead to
3053    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
3055    2. Moved to autoconf 2.53 and libtool 1.4.2.
3057    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
3058    the special perltest8 script is no longer needed - all the tests can be run
3059    from a single perltest script.
3061    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
3062    by \s. It has now been removed in PCRE. This means it isn't recognized as
3063    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
3064    class [:space:] *does* include VT, thereby creating a mess.
3066    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
3067    space and tab.
3069    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
3070    its new features into the main test script, reducing the number of scripts.
3072    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
3073    were backward compatible, and made the (?i) apply to the whole pattern, as if
3074    /i were given. Now it behaves more logically, and applies the option setting
3075    only to what follows. PCRE has been changed to follow suit. However, if it
3076    finds options settings right at the start of the pattern, it extracts them into
3077    the global options, as before. Thus, they show up in the info data.
3079    8. Added support for the \Q...\E escape sequence. Characters in between are
3080    treated as literals. This is slightly different from Perl in that $ and @ are
3081    also handled as literals inside the quotes. In Perl, they will cause variable
3082    interpolation. Note the following examples:
3084        Pattern            PCRE matches      Perl matches
3086        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
3087        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
3088        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
3090    For compatibility with Perl, \Q...\E sequences are recognized inside character
3091    classes as well as outside them.
3093    9. Re-organized 3 code statements in pcretest to avoid "overflow in
3094    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
3095    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
3096    signed/unsigned warnings.
3098    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
3099    option for pcretest, so I've replaced it by a simple function that does just
3100    that job.
3102    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
3103    "pcregrep -".
3105    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
3106    Java package. This provides some syntactic sugar for simple cases of what my
3107    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
3108    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
3109    item, you can use this simplified notation. Note that only makes sense with
3110    greedy quantifiers. Consequently, the use of the possessive quantifier forces
3111    greediness, whatever the setting of the PCRE_UNGREEDY option.
3113    13. A change of greediness default within a pattern was not taking effect at
3114    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
3115    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
3116    was abstracted outside.
3118    14. PCRE now supports the \G assertion. It is true when the current matching
3119    position is at the start point of the match. This differs from \A when the
3120    starting offset is non-zero. Used with the /g option of pcretest (or similar
3121    code), it works in the same way as it does for Perl's /g option. If all
3122    alternatives of a regex begin with \G, the expression is anchored to the start
3123    match position, and the "anchored" flag is set in the compiled expression.
3125    15. Some bugs concerning the handling of certain option changes within patterns
3126    have been fixed. These applied to options other than (?ims). For example,
3127    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
3128    been the other way round. Some of this was related to change 7 above.
3130    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
3131    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
3132    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
3133    POSIX classes only within a class (e.g. /[[:alpha:]]/).
3135    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
3136    mode. Unlike ".", it always matches newline, whatever the setting of
3137    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
3138    assertions. Perl allows it, but it doesn't (in general) work because it can't
3139    calculate the length of the lookbehind. At least, that's the case for Perl
3140    5.8.0 - I've been told they are going to document that it doesn't work in
3141    future.
3143    18. Added an error diagnosis for escapes that PCRE does not support: these are
3144    \L, \l, \N, \P, \p, \U, \u, and \X.
3146    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
3147    reading past the end of the pattern in cases such as /[abcd/.
3149    20. PCRE was getting more memory than necessary for patterns with classes that
3150    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
3152    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
3153    compiling PCRE for use with Virtual Pascal.
3155    22. Small fix to the Makefile to make it work properly if the build is done
3156    outside the source tree.
3158    23. Added a new extension: a condition to go with recursion. If a conditional
3159    subpattern starts with (?(R) the "true" branch is used if recursion has
3160    happened, whereas the "false" branch is used only at the top level.
3162    24. When there was a very long string of literal characters (over 255 bytes
3163    without UTF support, over 250 bytes with UTF support), the computation of how
3164    much memory was required could be incorrect, leading to segfaults or other
3165    strange effects.
3167    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
3168    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
3169    there was a subsequent back reference to those brackets. This meant that, for
3170    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
3171    possible to check for precisely this case. All we can do is abandon the
3172    optimization if .* occurs inside capturing brackets when there are any back
3173    references whatsoever. (See below for a better fix that came later.)
3175    26. The handling of the optimization for finding the first character of a
3176    non-anchored pattern, and for finding a character that is required later in the
3177    match were failing in some cases. This didn't break the matching; it just
3178    failed to optimize when it could. The way this is done has been re-implemented.
3180    27. Fixed typo in error message for invalid (?R item (it said "(?p").
3182    28. Added a new feature that provides some of the functionality that Perl
3183    provides with (?{...}). The facility is termed a "callout". The way it is done
3184    in PCRE is for the caller to provide an optional function, by setting
3185    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
3186    global variable. By default it is unset, which disables all calling out. To get
3187    the function called, the regex must include (?C) at appropriate points. This
3188    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
3189    This provides a means of identifying different callout points. When PCRE
3190    reaches such a point in the regex, if pcre_callout has been set, the external
3191    function is called. It is provided with data in a structure called
3192    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
3193    matching continues; if it returns a non-zero value, the match at the current
3194    point fails. However, backtracking will occur if possible. [This was changed
3195    later and other features added - see item 49 below.]
3197    29. pcretest is upgraded to test the callout functionality. It provides a
3198    callout function that displays information. By default, it shows the start of
3199    the match and the current position in the text. There are some new data escapes
3200    to vary what happens:
3202        \C+         in addition, show current contents of captured substrings
3203        \C-         do not supply a callout function
3204        \C!n        return 1 when callout number n is reached
3205        \C!n!m      return 1 when callout number n is reached for the mth time
3207    30. If pcregrep was called with the -l option and just a single file name, it
3208    output "<stdin>" if a match was found, instead of the file name.
3210    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
3211    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
3212    pcre_exec(). This saves a malloc/free per call. The default value of
3213    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
3214    when configuring.
3216    32. The default maximum size of a compiled pattern is 64K. There have been a
3217    few cases of people hitting this limit. The code now uses macros to handle the
3218    storing of links as offsets within the compiled pattern. It defaults to 2-byte
3219    links, but this can be changed to 3 or 4 bytes by --with-link-size when
3220    configuring. Tests 2 and 5 work only with 2-byte links because they output
3221    debugging information about compiled patterns.
3223    33. Internal code re-arrangements:
3225    (a) Moved the debugging function for printing out a compiled regex into
3226        its own source file (printint.c) and used #include to pull it into
3227        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
3228        separate copies.
3230    (b) Defined the list of op-code names for debugging as a macro in
3231        internal.h so that it is next to the definition of the opcodes.
3233    (c) Defined a table of op-code lengths for simpler skipping along compiled
3234        code. This is again a macro in internal.h so that it is next to the
3235        definition of the opcodes.
3237    34. Added support for recursive calls to individual subpatterns, along the
3238    lines of Robin Houston's patch (but implemented somewhat differently).
3240    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
3241    allow it to read and process whole directories in Win32. This code was
3242    contributed by Lionel Fourquaux; it has not been tested by me.
3244    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
3245    used to name a group. Names consist of alphanumerics and underscores, and must
3246    be unique. Back references use the syntax (?P=name) and recursive calls use
3247    (?P>name) which is a PCRE extension to the Python extension. Groups still have
3248    numbers. The function pcre_fullinfo() can be used after compilation to extract
3249    a name/number map. There are three relevant calls:
3251      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
3252      PCRE_INFO_NAMECOUNT            yields the number of entries
3253      PCRE_INFO_NAMETABLE            yields a pointer to the map.
3255    The map is a vector of fixed-size entries. The size of each entry depends on
3256    the length of the longest name used. The first two bytes of each entry are the
3257    group number, most significant byte first. There follows the corresponding
3258    name, zero terminated. The names are in alphabetical order.
3260    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
3261    case instead of 255. Making it the same both with and without UTF-8 support
3262    means that the same test output works with both.
3264    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
3265    calling malloc() with a zero argument.
3267    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
3268    optimization. I've improved things by keeping a bitmap of backreferences with
3269    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
3270    fact referenced, the optimization can be applied. It is unlikely that a
3271    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
3272    the match to follow \n) will appear inside brackets with a number greater than
3273    31, but if it does, any back reference > 31 suppresses the optimization.
3275    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
3276    of disabling numbered capturing parentheses. Any opening parenthesis that is
3277    not followed by ? behaves as if it were followed by ?: but named parentheses
3278    can still be used for capturing (and they will acquire numbers in the usual
3279    way).
3281    41. Redesigned the return codes from the match() function into yes/no/error so
3282    that errors can be passed back from deep inside the nested calls. A malloc
3283    failure while inside a recursive subpattern call now causes the
3284    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
3286    42. It is now possible to set a limit on the number of times the match()
3287    function is called in a call to pcre_exec(). This facility makes it possible to
3288    limit the amount of recursion and backtracking, though not in a directly
3289    obvious way, because the match() function is used in a number of different
3290    circumstances. The count starts from zero for each position in the subject
3291    string (for non-anchored patterns). The default limit is, for compatibility, a
3292    large number, namely 10 000 000. You can change this in two ways:
3294    (a) When configuring PCRE before making, you can use --with-match-limit=n
3295        to set a default value for the compiled library.
3297    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
3298        a different value is set. See 45 below.
3300    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
3302    43. Added a new function pcre_config(int, void *) to enable run-time extraction
3303    of things that can be changed at compile time. The first argument specifies
3304    what is wanted and the second points to where the information is to be placed.
3305    The current list of available information is:
3307      PCRE_CONFIG_UTF8
3309    The output is an integer that is set to one if UTF-8 support is available;
3310    otherwise it is set to zero.
3314    The output is an integer that it set to the value of the code that is used for
3315    newline. It is either LF (10) or CR (13).
3319    The output is an integer that contains the number of bytes used for internal
3320    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
3324    The output is an integer that contains the threshold above which the POSIX
3325    interface uses malloc() for output vectors. See item 31 above.
3329    The output is an unsigned integer that contains the default limit of the number
3330    of match() calls in a pcre_exec() execution. See 42 above.
3332    44. pcretest has been upgraded by the addition of the -C option. This causes it
3333    to extract all the available output from the new pcre_config() function, and to
3334    output it. The program then exits immediately.
3336    45. A need has arisen to pass over additional data with calls to pcre_exec() in
3337    order to support additional features. One way would have been to define
3338    pcre_exec2() (for example) with extra arguments, but this would not have been
3339    extensible, and would also have required all calls to the original function to
3340    be mapped to the new one. Instead, I have chosen to extend the mechanism that
3341    is used for passing in "extra" data from pcre_study().
3343    The pcre_extra structure is now exposed and defined in pcre.h. It currently
3344    contains the following fields:
3346      flags         a bitmap indicating which of the following fields are set
3347      study_data    opaque data from pcre_study()
3348      match_limit   a way of specifying a limit on match() calls for a specific
3349                      call to pcre_exec()
3350      callout_data  data for callouts (see 49 below)
3352    The flag bits are also defined in pcre.h, and are
3358    The pcre_study() function now returns one of these new pcre_extra blocks, with
3359    the actual study data pointed to by the study_data field, and the
3360    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
3361    before. That is, this change is entirely upwards-compatible and requires no
3362    change to existing code.
3364    If you want to pass in additional data to pcre_exec(), you can either place it
3365    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
3366    block.
3368    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
3369    data string contains the escape sequence \M, pcretest calls pcre_exec() several
3370    times with different match limits, until it finds the minimum value needed for
3371    pcre_exec() to complete. The value is then output. This can be instructive; for
3372    most simple matches the number is quite small, but for pathological cases it
3373    gets very large very quickly.
3375    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
3376    returns the size of the data block pointed to by the study_data field in a
3377    pcre_extra block, that is, the value that was passed as the argument to
3378    pcre_malloc() when PCRE was getting memory in which to place the information
3379    created by pcre_study(). The fourth argument should point to a size_t variable.
3380    pcretest has been extended so that this information is shown after a successful
3381    pcre_study() call when information about the compiled regex is being displayed.
3383    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
3384    because what follows is always an absolute path. (Later: it turns out that this
3385    is more than cosmetic for MinGW, because it doesn't like empty path
3386    components.)
3388    49. Some changes have been made to the callout feature (see 28 above):
3390    (i)  A callout function now has three choices for what it returns:
3392           0  =>  success, carry on matching
3393         > 0  =>  failure at this point, but backtrack if possible
3394         < 0  =>  serious error, return this value from pcre_exec()
3396         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
3397         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
3398         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
3399         use by callout functions. It will never be used by PCRE itself.
3401    (ii) The pcre_extra structure (see 45 above) has a void * field called
3402         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
3403         pcre_callout_block structure has a field of the same name. The contents of
3404         the field passed in the pcre_extra structure are passed to the callout
3405         function in the corresponding field in the callout block. This makes it
3406         easier to use the same callout-containing regex from multiple threads. For
3407         testing, the pcretest program has a new data escape
3409           \C*n        pass the number n (may be negative) as callout_data
3411         If the callout function in pcretest receives a non-zero value as
3412         callout_data, it returns that value.