ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 61 by nigel, Sat Feb 24 21:39:58 2007 UTC revision 602 by ph10, Wed May 25 08:29:03 2011 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
4  Version 3.0 02-Jan-02  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.
57    Version 8.12 15-Jan-2011
58    ------------------------
60    1.  Fixed some typos in the markup of the man pages, and wrote a script that
61        checks for such things as part of the documentation building process.
63    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
64        --match-limit and --recursion-limit options (added for 8.11). In
65        particular, this made one of the standard tests fail. (The integer value
66        went into the wrong half of a long int.)
68    3.  If the --colour option was given to pcregrep with -v (invert match), it
69        did strange things, either producing crazy output, or crashing. It should,
70        of course, ignore a request for colour when reporting lines that do not
71        match.
73    4.  Another pcregrep bug caused similar problems if --colour was specified with
74        -M (multiline) and the pattern match finished with a line ending.
76    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
77        matched in multiline mode, the following line was shown as part of the
78        match. This seems wrong, so I have changed it.
80    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
81        the check for further matches in the same line (so they could be coloured)
82        to overrun the end of the current line. If another match was found, it was
83        incorrectly shown (and then shown again when found in the next line).
85    7.  If pcregrep was compiled under Windows, there was a reference to the
86        function pcregrep_exit() before it was defined. I am assuming this was
87        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
88        reported by a user. I've moved the definition above the reference.
91    Version 8.11 10-Dec-2010
92    ------------------------
94    1.  (*THEN) was not working properly if there were untried alternatives prior
95        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
96        backtracked to try for "b" instead of moving to the next alternative branch
97        at the same level (in this case, to look for "c"). The Perl documentation
98        is clear that when (*THEN) is backtracked onto, it goes to the "next
99        alternative in the innermost enclosing group".
101    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
102        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
103        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
104        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
105        (*THEN).
107    3.  If \s appeared in a character class, it removed the VT character from
108        the class, even if it had been included by some previous item, for example
109        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
110        of \s, but is part of the POSIX "space" class.)
112    4.  A partial match never returns an empty string (because you can always
113        match an empty string at the end of the subject); however the checking for
114        an empty string was starting at the "start of match" point. This has been
115        changed to the "earliest inspected character" point, because the returned
116        data for a partial match starts at this character. This means that, for
117        example, /(?<=abc)def/ gives a partial match for the subject "abc"
118        (previously it gave "no match").
120    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
121        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
122        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
123        has an implication that the given string is incomplete (because a partial
124        match is preferred over a full match). For this reason, these items now
125        give a partial match in this situation. [Aside: previously, the one case
126        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
127        match rather than a full match, which was wrong by the old rules, but is
128        now correct.]
130    6.  There was a bug in the handling of #-introduced comments, recognized when
131        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
132        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
133        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
134        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
135        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
136        places in pcre_compile().
138    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
139        comments when looking ahead for named forward references to subpatterns,
140        the only newline sequence it recognized was NL. It now handles newlines
141        according to the set newline convention.
143    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
144        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
145        cater for a lack of strerror(). These oversights have been fixed.
147    9.  Added --match-limit and --recursion-limit to pcregrep.
149    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
151    11. When the -o option was used, pcregrep was setting a return code of 1, even
152        when matches were found, and --line-buffered was not being honoured.
154    12. Added an optional parentheses number to the -o and --only-matching options
155        of pcregrep.
157    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
158        can match an empty string. The code to do it in pcretest and pcredemo
159        needed fixing:
161        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
162            only one byte after an empty string match just before CRLF (this case
163            just got forgotten; "any" and "anycrlf" were OK).
165        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
166            mode when an empty string match preceded an ASCII character followed by
167            a non-ASCII character. (The code for advancing by one character rather
168            than one byte was nonsense.)
170        (c) The pcredemo.c sample program did not have any code at all to handle
171            the cases when CRLF is a valid newline sequence.
173    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
174        as a starting offset was within the subject string. There is now a new
175        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
176        negative or greater than the length of the string. In order to test this,
177        pcretest is extended to allow the setting of negative starting offsets.
179    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
180        starting offset points to the beginning of a UTF-8 character was
181        unnecessarily clumsy. I tidied it up.
183    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
184        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
186    17. Nobody had reported that the --include_dir option, which was added in
187        release 7.7 should have been called --include-dir (hyphen, not underscore)
188        for compatibility with GNU grep. I have changed it to --include-dir, but
189        left --include_dir as an undocumented synonym, and the same for
190        --exclude-dir, though that is not available in GNU grep, at least as of
191        release 2.5.4.
193    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
194        characters from a string of bytes) have been redefined so as not to use
195        loops, in order to improve performance in some environments. At the same
196        time, I abstracted some of the common code into auxiliary macros to save
197        repetition (this should not affect the compiled code).
199    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
200        compile-time error is now given if \c is not followed by an ASCII
201        character, that is, a byte less than 128. (In EBCDIC mode, the code is
202        different, and any byte value is allowed.)
204    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
205        START_OPTIMIZE option, which is now allowed at compile time - but just
206        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
207        to pcregrep and other applications that have no direct access to PCRE
208        options. The new /Y option in pcretest sets this option when calling
209        pcre_compile().
211    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
212        back references. Groups containing recursive back references were forced to
213        be atomic by that change, but in the case of named groups, the amount of
214        memory required was incorrectly computed, leading to "Failed: internal
215        error: code overflow". This has been fixed.
217    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
218        pcretest.c, to avoid build problems in some Borland environments.
221    Version 8.10 25-Jun-2010
222    ------------------------
224    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
225        THEN.
227    2.  (*ACCEPT) was not working when inside an atomic group.
229    3.  Inside a character class, \B is treated as a literal by default, but
230        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
231        causes the error). The code is unchanged, but I tidied the documentation.
233    4.  Inside a character class, PCRE always treated \R and \X as literals,
234        whereas Perl faults them if its -w option is set. I have changed PCRE so
235        that it faults them when PCRE_EXTRA is set.
237    5.  Added support for \N, which always matches any character other than
238        newline. (It is the same as "." when PCRE_DOTALL is not set.)
240    6.  When compiling pcregrep with newer versions of gcc which may have
241        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
242        declared with attribute warn_unused_result" were given. Just casting the
243        result to (void) does not stop the warnings; a more elaborate fudge is
244        needed. I've used a macro to implement this.
246    7.  Minor change to pcretest.c to avoid a compiler warning.
248    8.  Added four artifical Unicode properties to help with an option to make
249        \s etc use properties (see next item). The new properties are: Xan
250        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
252    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
253        use Unicode properties. (*UCP) at the start of a pattern can be used to set
254        this option. Modified pcretest to add /W to test this facility. Added
255        REG_UCP to make it available via the POSIX interface.
257    10. Added --line-buffered to pcregrep.
259    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
260        studied, and the match started with a letter with a code point greater than
261        127 whose first byte was different to the first byte of the other case of
262        the letter, the other case of this starting letter was not recognized
263        (#976).
265    12. If a pattern that was studied started with a repeated Unicode property
266        test, for example, \p{Nd}+, there was the theoretical possibility of
267        setting up an incorrect bitmap of starting bytes, but fortunately it could
268        not have actually happened in practice until change 8 above was made (it
269        added property types that matched character-matching opcodes).
271    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
272        possible starting bytes for non-anchored patterns.
274    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
275        \R, and also a number of cases that involve Unicode properties, both
276        explicit and implicit when PCRE_UCP is set.
278    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
279        input, it could crash or give wrong results if characters with values
280        greater than 0xc0 were present in the subject string. (Detail: it assumed
281        UTF-8 input when processing these items.)
283    16. Added a lot of (int) casts to avoid compiler warnings in systems where
284        size_t is 64-bit (#991).
286    17. Added a check for running out of memory when PCRE is compiled with
287        --disable-stack-for-recursion (#990).
289    18. If the last data line in a file for pcretest does not have a newline on
290        the end, a newline was missing in the output.
292    19. The default pcre_chartables.c file recognizes only ASCII characters (values
293        less than 128) in its various bitmaps. However, there is a facility for
294        generating tables according to the current locale when PCRE is compiled. It
295        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
296        characters, are recognized by isspace() and therefore were getting set in
297        these tables, and indeed these tables seem to approximate to ISO 8859. This
298        caused a problem in UTF-8 mode when pcre_study() was used to create a list
299        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
300        which of course cannot start UTF-8 characters. I have changed the code so
301        that only real ASCII characters (less than 128) and the correct starting
302        bytes for UTF-8 encodings are set for characters greater than 127 when in
303        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
304        altogether.)
306    20. Added the /T option to pcretest so as to be able to run tests with non-
307        standard character tables, thus making it possible to include the tests
308        used for 19 above in the standard set of tests.
310    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
311        reference to a subpattern the other side of a comment that contains an
312        opening parenthesis caused either an internal compiling error, or a
313        reference to the wrong subpattern.
316    Version 8.02 19-Mar-2010
317    ------------------------
319    1.  The Unicode data tables have been updated to Unicode 5.2.0.
321    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
322        configured.
324    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
325        original author of that file, following a query about its status.
327    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
328        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
330    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
331        quantifier applied to a forward-referencing subroutine call, could compile
332        incorrect code or give the error "internal error: previously-checked
333        referenced subpattern not found".
335    6.  Both MS Visual Studio and Symbian OS have problems with initializing
336        variables to point to external functions. For these systems, therefore,
337        pcre_malloc etc. are now initialized to local functions that call the
338        relevant global functions.
340    7.  There were two entries missing in the vectors called coptable and poptable
341        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
342        I've fixed the data, and added a kludgy way of testing at compile time that
343        the lengths are correct (equal to the number of opcodes).
345    8.  Following on from 7, I added a similar kludge to check the length of the
346        eint vector in pcreposix.c.
348    9.  Error texts for pcre_compile() are held as one long string to avoid too
349        much relocation at load time. To find a text, the string is searched,
350        counting zeros. There was no check for running off the end of the string,
351        which could happen if a new error number was added without updating the
352        string.
354    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
356    11. \K was not working if it appeared in an atomic group or in a group that
357        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
358        \K is "not well defined" if used in an assertion. PCRE now accepts it if
359        the assertion is positive, but not if it is negative.
361    12. Change 11 fortuitously reduced the size of the stack frame used in the
362        "match()" function of pcre_exec.c by one pointer. Forthcoming
363        implementation of support for (*MARK) will need an extra pointer on the
364        stack; I have reserved it now, so that the stack frame size does not
365        decrease.
367    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
368        item in branch that calls a recursion is a subroutine call - as in the
369        second branch in the above example - was incorrectly given the compile-
370        time error "recursive call could loop indefinitely" because pcre_compile()
371        was not correctly checking the subroutine for matching a non-empty string.
373    14. The checks for overrunning compiling workspace could trigger after an
374        overrun had occurred. This is a "should never occur" error, but it can be
375        triggered by pathological patterns such as hundreds of nested parentheses.
376        The checks now trigger 100 bytes before the end of the workspace.
378    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
381    Version 8.01 19-Jan-2010
382    ------------------------
384    1.  If a pattern contained a conditional subpattern with only one branch (in
385        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
386        computed the wrong minimum data length (which is of course zero for such
387        subpatterns). This could cause incorrect "no match" results.
389    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
390        the pattern is reset in the first branch, pcre_compile() failed with
391        "internal error: code overflow at offset...". This happened only when
392        the reset was to the original external option setting. (An optimization
393        abstracts leading options settings into an external setting, which was the
394        cause of this.)
396    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
397        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
398        assertion pattern did not match (meaning that the assertion was true), it
399        was incorrectly treated as false if the SKIP had been reached during the
400        matching. This also applied to assertions used as conditions.
402    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
403        assertion subpattern, including such a pattern used as a condition,
404        unpredictable results occurred, instead of the error return
407    5.  The C++ GlobalReplace function was not working like Perl for the special
408        situation when an empty string is matched. It now does the fancy magic
409        stuff that is necessary.
411    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
412        removed. (These were left over from very, very early versions of PCRE.)
414    7.  Some cosmetic changes to the code to make life easier when compiling it
415        as part of something else:
417        (a) Change DEBUG to PCRE_DEBUG.
419        (b) In pcre_compile(), rename the member of the "branch_chain" structure
420            called "current" as "current_branch", to prevent a collision with the
421            Linux macro when compiled as a kernel module.
423        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
424            prevent a collision with the Linux macro when compiled as a kernel
425            module.
427    8.  In pcre_compile() there are some checks for integer overflows that used to
428        cast potentially large values to (double). This has been changed to that
429        when building, a check for int64_t is made, and if it is found, it is used
430        instead, thus avoiding the use of floating point arithmetic. (There is no
431        other use of FP in PCRE.) If int64_t is not found, the fallback is to
432        double.
434    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
435        2005 (difference between two addresses compared to an unsigned value).
437    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
438        custom one, because of the following reported problem in Windows:
440          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
441              under Win32.
442          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
443              therefore missing the function definition.
444          - The compiler thus generates a "C" signature for the test function.
445          - The linker fails to find the "C" function.
446          - PCRE fails to configure if asked to do so against libbz2.
448    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
449        messages were output:
451          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
452          rerunning libtoolize, to keep the correct libtool macros in-tree.
453          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
455        I have done both of these things.
457    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
458        most of the time, it *can* run out if it is given a pattern that contains a
459        runaway infinite recursion. I updated the discussion in the pcrestack man
460        page.
462    13. Now that we have gone to the x.xx style of version numbers, the minor
463        version may start with zero. Using 08 or 09 is a bad idea because users
464        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
465        interpreted as invalid octal numbers. I've updated the previous comment in
466        configure.ac, and also added a check that gives an error if 08 or 09 are
467        used.
469    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
470        causing partial matching to fail when the end of the subject matched \W
471        in a UTF-8 pattern where \W was quantified with a minimum of 3.
473    15. There were some discrepancies between the declarations in pcre_internal.h
474        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
475        their definitions. The declarations used "const uschar *" and the
476        definitions used USPTR. Even though USPTR is normally defined as "const
477        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
478        reported that: "This difference in casting confuses some C++ compilers, for
479        example, SunCC recognizes above declarations as different functions and
480        generates broken code for hbpcre." I have changed the declarations to use
481        USPTR.
483    16. GNU libtool is named differently on some systems. The autogen.sh script now
484        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
485        (FreeBSD).
487    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
488        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
489        comment: "Figure out how to create a longlong from a string: strtoll and
490        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
491        instance, but it only takes 2 args instead of 3!"
493    18. A subtle bug concerned with back references has been fixed by a change of
494        specification, with a corresponding code fix. A pattern such as
495        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
496        refers, was giving matches when it shouldn't. For example, xa=xaaa would
497        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
498        same bug. Such groups have to be quantified to be useful, or contained
499        inside another quantified group. (If there's no repetition, the reference
500        can never match.) The problem arises because, having left the group and
501        moved on to the rest of the pattern, a later failure that backtracks into
502        the group uses the captured value from the final iteration of the group
503        rather than the correct earlier one. I have fixed this in PCRE by forcing
504        any group that contains a reference to itself to be an atomic group; that
505        is, there cannot be any backtracking into it once it has completed. This is
506        similar to recursive and subroutine calls.
509    Version 8.00 19-Oct-09
510    ----------------------
512    1.  The table for translating pcre_compile() error codes into POSIX error codes
513        was out-of-date, and there was no check on the pcre_compile() error code
514        being within the table. This could lead to an OK return being given in
515        error.
517    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
518        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
519        in a Windows environment.
521    3.  The pcregrep --count option prints the count for each file even when it is
522        zero, as does GNU grep. However, pcregrep was also printing all files when
523        --files-with-matches was added. Now, when both options are given, it prints
524        counts only for those files that have at least one match. (GNU grep just
525        prints the file name in this circumstance, but including the count seems
526        more useful - otherwise, why use --count?) Also ensured that the
527        combination -clh just lists non-zero counts, with no names.
529    4.  The long form of the pcregrep -F option was incorrectly implemented as
530        --fixed_strings instead of --fixed-strings. This is an incompatible change,
531        but it seems right to fix it, and I didn't think it was worth preserving
532        the old behaviour.
534    5.  The command line items --regex=pattern and --regexp=pattern were not
535        recognized by pcregrep, which required --regex pattern or --regexp pattern
536        (with a space rather than an '='). The man page documented the '=' forms,
537        which are compatible with GNU grep; these now work.
539    6.  No libpcreposix.pc file was created for pkg-config; there was just
540        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
542    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
543        when UCP support is not needed, by modifying the Python script that
544        generates it from Unicode data files. This should not matter if the module
545        is correctly used as a library, but I received one complaint about 50K of
546        unwanted data. My guess is that the person linked everything into his
547        program rather than using a library. Anyway, it does no harm.
549    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
550        was a minimum greater than 1 for a wide character in a possessive
551        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
552        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
553        character. Chaos in the form of incorrect output or a compiling loop could
554        result.
556    9.  The restrictions on what a pattern can contain when partial matching is
557        requested for pcre_exec() have been removed. All patterns can now be
558        partially matched by this function. In addition, if there are at least two
559        slots in the offset vector, the offset of the earliest inspected character
560        for the match and the offset of the end of the subject are set in them when
561        PCRE_ERROR_PARTIAL is returned.
563    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
564        synonymous with PCRE_PARTIAL, for backwards compatibility, and
565        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
566        and may be more useful for multi-segment matching.
568    11. Partial matching with pcre_exec() is now more intuitive. A partial match
569        used to be given if ever the end of the subject was reached; now it is
570        given only if matching could not proceed because another character was
571        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
572        string "Z", which now yields "no match" instead of "partial match". In the
573        case of pcre_dfa_exec(), "no match" is given if every matching path for the
574        final character ended with (*FAIL).
576    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
577        if the pattern had a "must contain" character that was already found in the
578        earlier partial match, unless partial matching was again requested. For
579        example, with the pattern /dog.(body)?/, the "must contain" character is
580        "g". If the first part-match was for the string "dog", restarting with
581        "sbody" failed. This bug has been fixed.
583    13. The string returned by pcre_dfa_exec() after a partial match has been
584        changed so that it starts at the first inspected character rather than the
585        first character of the match. This makes a difference only if the pattern
586        starts with a lookbehind assertion or \b or \B (\K is not supported by
587        pcre_dfa_exec()). It's an incompatible change, but it makes the two
588        matching functions compatible, and I think it's the right thing to do.
590    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
591        so that the demonstration program is easily available in environments where
592        PCRE has not been installed from source.
594    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
595        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
596        library.
598    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
599        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
600        is not the first non-POSIX option to be added. Clearly some people find
601        these options useful.
603    17. If a caller to the POSIX matching function regexec() passes a non-zero
604        value for nmatch with a NULL value for pmatch, the value of
605        nmatch is forced to zero.
607    18. RunGrepTest did not have a test for the availability of the -u option of
608        the diff command, as RunTest does. It now checks in the same way as
609        RunTest, and also checks for the -b option.
611    19. If an odd number of negated classes containing just a single character
612        interposed, within parentheses, between a forward reference to a named
613        subpattern and the definition of the subpattern, compilation crashed with
614        an internal error, complaining that it could not find the referenced
615        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
616        [The bug was that it was starting one character too far in when skipping
617        over the character class, thus treating the ] as data rather than
618        terminating the class. This meant it could skip too much.]
620    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
621        /g option in pcretest when the pattern contains \K, which makes it possible
622        to have an empty string match not at the start, even when the pattern is
623        anchored. Updated pcretest and pcredemo to use this option.
625    21. If the maximum number of capturing subpatterns in a recursion was greater
626        than the maximum at the outer level, the higher number was returned, but
627        with unset values at the outer level. The correct (outer level) value is
628        now given.
630    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
631        PCRE did not set those parentheses (unlike Perl). I have now found a way to
632        make it do so. The string so far is captured, making this feature
633        compatible with Perl.
635    23. The tests have been re-organized, adding tests 11 and 12, to make it
636        possible to check the Perl 5.10 features against Perl 5.10.
638    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
639        pattern matches a fixed length string. PCRE did not allow this; now it
640        does. Neither allows recursion.
642    25. I finally figured out how to implement a request to provide the minimum
643        length of subject string that was needed in order to match a given pattern.
644        (It was back references and recursion that I had previously got hung up
645        on.) This code has now been added to pcre_study(); it finds a lower bound
646        to the length of subject needed. It is not necessarily the greatest lower
647        bound, but using it to avoid searching strings that are too short does give
648        some useful speed-ups. The value is available to calling programs via
649        pcre_fullinfo().
651    26. While implementing 25, I discovered to my embarrassment that pcretest had
652        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
653        study optimizations had never been tested with that matching function.
654        Oops. What is worse, even when it was passed study data, there was a bug in
655        pcre_dfa_exec() that meant it never actually used it. Double oops. There
656        were also very few tests of studied patterns with pcre_dfa_exec().
658    27. If (?| is used to create subpatterns with duplicate numbers, they are now
659        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
660        on the other side of the coin, they are no longer allowed to have different
661        names, because these cannot be distinguished in PCRE, and this has caused
662        confusion. (This is a difference from Perl.)
664    28. When duplicate subpattern names are present (necessarily with different
665        numbers, as required by 27 above), and a test is made by name in a
666        conditional pattern, either for a subpattern having been matched, or for
667        recursion in such a pattern, all the associated numbered subpatterns are
668        tested, and the overall condition is true if the condition is true for any
669        one of them. This is the way Perl works, and is also more like the way
670        testing by number works.
673    Version 7.9 11-Apr-09
674    ---------------------
676    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
677        (pcretest), all targets were linked against these libraries. This included
678        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
679        libraries. This caused unwanted dependencies to be created. This problem
680        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
681        pcretest is linked with readline.
683    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
684        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
685        moved outside it again, because FALSE and TRUE are already defined in AIX,
686        but BOOL is not.
688    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
689        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
691    4.  The pcregrep documentation said spaces were inserted as well as colons (or
692        hyphens) following file names and line numbers when outputting matching
693        lines. This is not true; no spaces are inserted. I have also clarified the
694        wording for the --colour (or --color) option.
696    5.  In pcregrep, when --colour was used with -o, the list of matching strings
697        was not coloured; this is different to GNU grep, so I have changed it to be
698        the same.
700    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
701        each matching line was coloured. Now it goes on to look for further matches
702        of any of the test patterns, which is the same behaviour as GNU grep.
704    7.  A pattern that could match an empty string could cause pcregrep to loop; it
705        doesn't make sense to accept an empty string match in pcregrep, so I have
706        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
707        seems to be how GNU grep behaves.
709    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
710        start or after a newline", because the conditional assertion was not being
711        correctly handled. The rule now is that both the assertion and what follows
712        in the first alternative must satisfy the test.
714    9.  If auto-callout was enabled in a pattern with a conditional group whose
715        condition was an assertion, PCRE could crash during matching, both with
716        pcre_exec() and pcre_dfa_exec().
718    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
719        used for matching.
721    11. Unicode property support in character classes was not working for
722        characters (bytes) greater than 127 when not in UTF-8 mode.
724    12. Added the -M command line option to pcretest.
726    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
728    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
730    16. Added comments and documentation about mis-use of no_arg in the C++
731        wrapper.
733    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
734        from Martin Jerabek that uses macro names for all relevant character and
735        string constants.
737    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
738        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
739        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
740        these, but not everybody uses configure.
742    19. A conditional group that had only one branch was not being correctly
743        recognized as an item that could match an empty string. This meant that an
744        enclosing group might also not be so recognized, causing infinite looping
745        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
746        with the subject "ab", where knowledge that the repeated group can match
747        nothing is needed in order to break the loop.
749    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
750        exec(), but without supplying a callout function, matching went wrong.
752    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
753        leak if the size of the offset vector was greater than 30. When the vector
754        is smaller, the saved offsets during recursion go onto a local stack
755        vector, but for larger vectors malloc() is used. It was failing to free
756        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
757        error, in fact).
759    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
760        heapframe that is used only when UTF-8 support is enabled. This caused no
761        problem, but was untidy.
763    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
764        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
765        included within another project.
767    24. Steven Van Ingelgem's patches to add more options to the CMake support,
768        slightly modified by me:
770          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
771              not building pcregrep.
773          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
774              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
776    25. Forward references, both numeric and by name, in patterns that made use of
777        duplicate group numbers, could behave incorrectly or give incorrect errors,
778        because when scanning forward to find the reference group, PCRE was not
779        taking into account the duplicate group numbers. A pattern such as
780        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
782    26. Changed a few more instances of "const unsigned char *" to USPTR, making
783        the feature of a custom pointer more persuasive (as requested by a user).
785    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
786        pcretest.c, inside #ifndefs, because it seems they are sometimes already
787        pre-defined.
789    28. Added support for (*UTF8) at the start of a pattern.
791    29. Arrange for flags added by the "release type" setting in CMake to be shown
792        in the configuration summary.
795    Version 7.8 05-Sep-08
796    ---------------------
798    1.  Replaced UCP searching code with optimized version as implemented for Ad
799        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
800        stage table and inline lookup instead of a function, giving speed ups of 2
801        to 5 times on some simple patterns that I tested. Permission was given to
802        distribute the MultiStage2.py script that generates the tables (it's not in
803        the tarball, but is in the Subversion repository).
805    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
806        scripts.
808    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
809        a group with a zero qualifier. The result of the study could be incorrect,
810        or the function might crash, depending on the pattern.
812    4.  Caseless matching was not working for non-ASCII characters in back
813        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
814        It now works when Unicode Property Support is available.
816    5.  In pcretest, an escape such as \x{de} in the data was always generating
817        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
818        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
819        truncation.
821    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
823    7.  Added two (int) casts to pcregrep when printing the difference of two
824        pointers, in case they are 64-bit values.
826    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
827        test 2 if it fails.
829    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
830        and a #define of that name to empty if it is not externally set. This is to
831        allow users of MSVC to set it if necessary.
833    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
834        the convenience functions in the pcre_get.c source file.
836    11. An option change at the start of a pattern that had top-level alternatives
837        could cause overwriting and/or a crash. This command provoked a crash in
838        some environments:
840          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
842        This potential security problem was recorded as CVE-2008-2371.
844    12. For a pattern where the match had to start at the beginning or immediately
845        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
846        pcre_dfa_exec() could read past the end of the passed subject if there was
847        no match. To help with detecting such bugs (e.g. with valgrind), I modified
848        pcretest so that it places the subject at the end of its malloc-ed buffer.
850    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
851        exec() might read past the end of the data buffer in UTF-8 mode.
853    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
854        the data contained the byte 0x85 as part of a UTF-8 character within its
855        first line. This applied both to normal and DFA matching.
857    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
858        /^[^d]*?$/8 failed to match "abc".
860    16. Added a missing copyright notice to pcrecpp_internal.h.
862    17. Make it more clear in the documentation that values returned from
863        pcre_exec() in ovector are byte offsets, not character counts.
865    18. Tidied a few places to stop certain compilers from issuing warnings.
867    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
868        supplied by Stefan Weber. I made a further small update for 7.8 because
869        there is a change of source arrangements: the pcre_searchfuncs.c module is
870        replaced by pcre_ucd.c.
873    Version 7.7 07-May-08
874    ---------------------
876    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
877        a string to a long long, pretend we don't even have a long long." This is
878        done by checking for the strtoq, strtoll, and _strtoi64 functions.
880    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
881        pre-7.6 versions, which defined a global no_arg variable instead of putting
882        it in the RE class. (See also #8 below.)
884    3.  Remove a line of dead code, identified by coverity and reported by Nuno
885        Lopes.
887    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
889        (1) The include/exclude patterns were being applied to the whole pathnames
890            of files, instead of just to the final components.
892        (2) If there was more than one level of directory, the subdirectories were
893            skipped unless they satisfied the include/exclude conditions. This is
894            inconsistent with GNU grep (and could even be seen as contrary to the
895            pcregrep specification - which I improved to make it absolutely clear).
896            The action now is always to scan all levels of directory, and just
897            apply the include/exclude patterns to regular files.
899    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
900        --exclude_dir in the tests to avoid scanning .svn directories.
902    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
903        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
904        doesn't support NULs in patterns.
906    7.  Added some missing "const"s to declarations of static tables in
907        pcre_compile.c and pcre_dfa_exec.c.
909    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
910        caused by fix #2  above. (Subsequently also a second patch to fix the
911        first patch. And a third patch - this was a messy problem.)
913    9.  Applied Craig's patch to remove the use of push_back().
915    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
916        matching function regexec().
918    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
919        which, however, unlike Perl's \g{...}, are subroutine calls, not back
920        references. PCRE supports relative numbers with this syntax (I don't think
921        Oniguruma does).
923    12. Previously, a group with a zero repeat such as (...){0} was completely
924        omitted from the compiled regex. However, this means that if the group
925        was called as a subroutine from elsewhere in the pattern, things went wrong
926        (an internal error was given). Such groups are now left in the compiled
927        pattern, with a new opcode that causes them to be skipped at execution
928        time.
930    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
931        to the way PCRE behaves:
933        (a) A lone ] character is dis-allowed (Perl treats it as data).
935        (b) A back reference to an unmatched subpattern matches an empty string
936            (Perl fails the current match path).
938        (c) A data ] in a character class must be notated as \] because if the
939            first data character in a class is ], it defines an empty class. (In
940            Perl it is not possible to have an empty class.) The empty class []
941            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
942            The negative empty class [^] matches any one character, independently
943            of the DOTALL setting.
945    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
946        non-existent subpattern following a character class starting with ']' and
947        containing () gave an internal compiling error instead of "reference to
948        non-existent subpattern". Fortunately, when the pattern did exist, the
949        compiled code was correct. (When scanning forwards to check for the
950        existencd of the subpattern, it was treating the data ']' as terminating
951        the class, so got the count wrong. When actually compiling, the reference
952        was subsequently set up correctly.)
954    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
955        it was being rejected as not supported by pcre_dfa_exec(), even though
956        other assertions are supported. I have made pcre_dfa_exec() support
957        (*FAIL).
959    16. The implementation of 13c above involved the invention of a new opcode,
960        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
961        cannot be changed at match time, I realized I could make a small
962        improvement to matching performance by compiling OP_ALLANY instead of
963        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
964        on the OP_ANY path.
966    17. Compiling pcretest on Windows with readline support failed without the
967        following two fixes: (1) Make the unistd.h include conditional on
968        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
970    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
971        ncurses library to be included for pcretest when ReadLine support is
972        requested, but also to allow for it to be overridden. This patch came from
973        Daniel Bergström.
975    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
976        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
977        any errors with the current Unicode tables. Thanks to Peter Kankowski for
978        spotting this.
981    Version 7.6 28-Jan-08
982    ---------------------
984    1.  A character class containing a very large number of characters with
985        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
986        overflow.
988    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
989        HAVE_LONG_LONG is not defined.
991    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
992        bring them up to date and include new features. This patch includes:
994        - Fixed PH's badly added libz and libbz2 support.
995        - Fixed a problem with static linking.
996        - Added pcredemo. [But later removed - see 7 below.]
997        - Fixed dftables problem and added an option.
998        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
999            HAVE_LONG_LONG.
1000        - Added readline support for pcretest.
1001        - Added an listing of the option settings after cmake has run.
1003    4.  A user submitted a patch to Makefile that makes it easy to create
1004        "pcre.dll" under mingw when using Configure/Make. I added stuff to
1005        Makefile.am that cause it to include this special target, without
1006        affecting anything else. Note that the same mingw target plus all
1007        the other distribution libraries and programs are now supported
1008        when configuring with CMake (see 6 below) instead of with
1009        Configure/Make.
1011    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
1012        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
1013        exported in the Windows port". It has not yet been confirmed that the patch
1014        solves the problem, but it does no harm.
1016    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
1017        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
1018        with CMake, and also correct the comment about stack recursion.
1020    7.  Remove the automatic building of pcredemo from the ./configure system and
1021        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
1022        of a program that users should build themselves after PCRE is installed, so
1023        building it automatically is not really right. What is more, it gave
1024        trouble in some build environments.
1026    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
1029    Version 7.5 10-Jan-08
1030    ---------------------
1032    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
1033        values in parens when parsing an RE using the C++ wrapper."
1035    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
1036        Characters greater than 255 were excluded from the class instead of being
1037        included.
1039    3.  The same bug as (2) above applied to negated POSIX classes such as
1040        [:^space:].
1042    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
1043        defined or documented. It seems to have been a typo for PCRE_STATIC, so
1044        I have changed it.
1046    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
1047        first named subpattern) and a construct such as (?&a) would reference the
1048        first named subpattern whose name started with "a" (in other words, the
1049        length check was missing). Both these problems are fixed. "Subpattern name
1050        expected" is now given for (?&) (a zero-length name), and this patch also
1051        makes it give the same error for \k'' (previously it complained that that
1052        was a reference to a non-existent subpattern).
1054    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
1055        this is right because (?- can be followed by option settings as well as by
1056        digits. I have, however, made the messages clearer.
1058    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
1059        than the number used in the conditional) now cause a compile-time error.
1060        This is actually not compatible with Perl, which accepts such patterns, but
1061        treats the conditional as always being FALSE (as PCRE used to), but it
1062        seems to me that giving a diagnostic is better.
1064    8.  Change "alphameric" to the more common word "alphanumeric" in comments
1065        and messages.
1067    9.  Fix two occurrences of "backslash" in comments that should have been
1068        "backspace".
1070    10. Remove two redundant lines of code that can never be obeyed (their function
1071        was moved elsewhere).
1073    11. The program that makes PCRE's Unicode character property table had a bug
1074        which caused it to generate incorrect table entries for sequences of
1075        characters that have the same character type, but are in different scripts.
1076        It amalgamated them into a single range, with the script of the first of
1077        them. In other words, some characters were in the wrong script. There were
1078        thirteen such cases, affecting characters in the following ranges:
1080          U+002b0 - U+002c1
1081          U+0060c - U+0060d
1082          U+0061e - U+00612
1083          U+0064b - U+0065e
1084          U+0074d - U+0076d
1085          U+01800 - U+01805
1086          U+01d00 - U+01d77
1087          U+01d9b - U+01dbf
1088          U+0200b - U+0200f
1089          U+030fc - U+030fe
1090          U+03260 - U+0327f
1091          U+0fb46 - U+0fbb1
1092          U+10450 - U+1049d
1094    12. The -o option (show only the matching part of a line) for pcregrep was not
1095        compatible with GNU grep in that, if there was more than one match in a
1096        line, it showed only the first of them. It now behaves in the same way as
1097        GNU grep.
1099    13. If the -o and -v options were combined for pcregrep, it printed a blank
1100        line for every non-matching line. GNU grep prints nothing, and pcregrep now
1101        does the same. The return code can be used to tell if there were any
1102        non-matching lines.
1104    14. Added --file-offsets and --line-offsets to pcregrep.
1106    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
1107        infinitely looping recursion. The bug was that positive lookaheads were not
1108        being skipped when checking for a possible empty match (negative lookaheads
1109        and both kinds of lookbehind were skipped).
1111    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
1112        inclusion of <windows.h> to before rather than after the definition of
1113        INVALID_FILE_ATTRIBUTES (patch from David Byron).
1115    17. Specifying a possessive quantifier with a specific limit for a Unicode
1116        character property caused pcre_compile() to compile bad code, which led at
1117        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
1118        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
1119        caused the error; without that there was no problem.
1121    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
1123    19. Added --enable-pcretest-libreadline.
1125    20. In pcrecpp.cc, the variable 'count' was incremented twice in
1126        RE::GlobalReplace(). As a result, the number of replacements returned was
1127        double what it should be. I removed one of the increments, but Craig sent a
1128        later patch that removed the other one (the right fix) and added unit tests
1129        that check the return values (which was not done before).
1131    21. Several CMake things:
1133        (1) Arranged that, when cmake is used on Unix, the libraries end up with
1134            the names libpcre and libpcreposix, not just pcre and pcreposix.
1136        (2) The above change means that pcretest and pcregrep are now correctly
1137            linked with the newly-built libraries, not previously installed ones.
1141    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
1142        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
1143        UTF-8 newline character). The key issue is that the pattern starts .*;
1144        this means that the match must be either at the beginning, or after a
1145        newline. The bug was in the code for advancing after a failed match and
1146        checking that the new position followed a newline. It was not taking
1147        account of UTF-8 characters correctly.
1149    23. PCRE was behaving differently from Perl in the way it recognized POSIX
1150        character classes. PCRE was not treating the sequence [:...:] as a
1151        character class unless the ... were all letters. Perl, however, seems to
1152        allow any characters between [: and :], though of course it rejects as
1153        unknown any "names" that contain non-letters, because all the known class
1154        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
1155        for example, whereas PCRE did not - it did not recognize a POSIX character
1156        class. This seemed a bit dangerous, so the code has been changed to be
1157        closer to Perl. The behaviour is not identical to Perl, because PCRE will
1158        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
1159        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
1160        Perl does, and where it didn't before.
1162    24. Rewrite so as to remove the single use of %n from pcregrep because in some
1163        Windows environments %n is disabled by default.
1166    Version 7.4 21-Sep-07
1167    ---------------------
1169    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
1170        means that a class such as [\s] counted as "explicit reference to CR or
1171        LF". That isn't really right - the whole point of the change was to try to
1172        help when there was an actual mention of one of the two characters. So now
1173        the change happens only if \r or \n (or a literal CR or LF) character is
1174        encountered.
1176    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
1177        of both had grown to the point where there were only 3 bits left.
1178        Fortunately, there was spare space in the data structure, and so I have
1179        moved the internal flags into a new 16-bit field to free up more option
1180        bits.
1182    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
1183        but did not set the internal JCHANGED flag - either of these is enough to
1184        control the way the "get" function works - but the PCRE_INFO_JCHANGED
1185        facility is supposed to tell if (?J) was ever used, so now (?J) at the
1186        start sets both bits.
1188    4.  Added options (at build time, compile time, exec time) to change \R from
1189        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
1191    5.  doc/pcresyntax.html was missing from the distribution.
1193    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
1194        compatibility, even though it is no longer used.
1196    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
1197        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
1198        windows.h file is present (where different names are used). [This was
1199        reversed later after testing - see 16 below.]
1201    8.  Changed all #include <config.h> to #include "config.h". There were also
1202        some further <pcre.h> cases that I changed to "pcre.h".
1204    9.  When pcregrep was used with the --colour option, it missed the line ending
1205        sequence off the lines that it output.
1207    10. It was pointed out to me that arrays of string pointers cause lots of
1208        relocations when a shared library is dynamically loaded. A technique of
1209        using a single long string with a table of offsets can drastically reduce
1210        these. I have refactored PCRE in four places to do this. The result is
1211        dramatic:
1213          Originally:                          290
1214          After changing UCP table:            187
1215          After changing error message table:   43
1216          After changing table of "verbs"       36
1217          After changing table of Posix names   22
1219        Thanks to the folks working on Gregex for glib for this insight.
1221    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
1222        unicode-properties was also set.
1224    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
1226    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
1227        checked only for CRLF.
1229    14. Added casts to pcretest.c to avoid compiler warnings.
1231    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
1233    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
1234        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
1235        entirely. This removes changes made in 7 above.
1237    17. The CMake files have been updated, and there is now more information about
1238        building with CMake in the NON-UNIX-USE document.
1241    Version 7.3 28-Aug-07
1242    ---------------------
1244     1. In the rejigging of the build system that eventually resulted in 7.1, the
1245        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
1246        brackets there is not right, since it causes compilers to look for an
1247        installed pcre.h, not the version that is in the source that is being
1248        compiled (which of course may be different). I have changed it back to:
1250          #include "pcre.h"
1252        I have a vague recollection that the change was concerned with compiling in
1253        different directories, but in the new build system, that is taken care of
1254        by the VPATH setting the Makefile.
1256     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
1257        when the subject happened to end in the byte 0x85 (e.g. if the last
1258        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
1259        characters but of course it shouldn't be taken as a newline when it is part
1260        of another character. The bug was that, for an unlimited repeat of . in
1261        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
1262        characters when looking for a newline.
1264     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
1266     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
1267        in debug output.
1269     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
1270        long printing in the pcrecpp unittest when running under MinGW.
1272     6. ESC_K was left out of the EBCDIC table.
1274     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
1275        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
1276        limit also applies to "virtual nesting" when a pattern is recursive, and in
1277        this case 1000 isn't so big. I have been able to remove this limit at the
1278        expense of backing off one optimization in certain circumstances. Normally,
1279        when pcre_exec() would call its internal match() function recursively and
1280        immediately return the result unconditionally, it uses a "tail recursion"
1281        feature to save stack. However, when a subpattern that can match an empty
1282        string has an unlimited repetition quantifier, it no longer makes this
1283        optimization. That gives it a stack frame in which to save the data for
1284        checking that an empty string has been matched. Previously this was taken
1285        from the 1000-entry workspace that had been reserved. So now there is no
1286        explicit limit, but more stack is used.
1288     8. Applied Daniel's patches to solve problems with the import/export magic
1289        syntax that is required for Windows, and which was going wrong for the
1290        pcreposix and pcrecpp parts of the library. These were overlooked when this
1291        problem was solved for the main library.
1293     9. There were some crude static tests to avoid integer overflow when computing
1294        the size of patterns that contain repeated groups with explicit upper
1295        limits. As the maximum quantifier is 65535, the maximum group length was
1296        set at 30,000 so that the product of these two numbers did not overflow a
1297        32-bit integer. However, it turns out that people want to use groups that
1298        are longer than 30,000 bytes (though not repeat them that many times).
1299        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1300        made it possible to implement the integer overflow checks in a much more
1301        dynamic way, which I have now done. The artificial limitation on group
1302        length has been removed - we now have only the limit on the total length of
1303        the compiled pattern, which depends on the LINK_SIZE setting.
1305    10. Fixed a bug in the documentation for get/copy named substring when
1306        duplicate names are permitted. If none of the named substrings are set, the
1307        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1308        empty string.
1310    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1311        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1312        because the ] is interpreted as the first data character and the
1313        terminating ] is not found. PCRE has been made compatible with Perl in this
1314        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1315        cause memory overwriting.
1317    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1318        string has been matched (to stop an infinite loop). It was not recognizing
1319        a conditional subpattern that could match an empty string if that
1320        subpattern was within another subpattern. For example, it looped when
1321        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
1322        condition was not nested. This bug has been fixed.
1324    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1325        past the start of the subject in the presence of bytes with the top bit
1326        set, for example "\x8aBCD".
1328    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1329        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1331    14. Optimized (?!) to (*FAIL).
1333    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1334        This restricts code points to be within the range 0 to 0x10FFFF, excluding
1335        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1336        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1337        does: it's just the validity check that is more restrictive.
1339    16. Inserted checks for integer overflows during escape sequence (backslash)
1340        processing, and also fixed erroneous offset values for syntax errors during
1341        backslash processing.
1343    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1344        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1346    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1347        caused an overrun.
1349    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1350        something other than just ASCII characters) inside a group that had an
1351        unlimited repeat caused a loop at compile time (while checking to see
1352        whether the group could match an empty string).
1354    20. Debugging a pattern containing \p or \P could cause a crash. For example,
1355        [\P{Any}] did so. (Error in the code for printing property names.)
1357    21. An orphan \E inside a character class could cause a crash.
1359    22. A repeated capturing bracket such as (A)? could cause a wild memory
1360        reference during compilation.
1362    23. There are several functions in pcre_compile() that scan along a compiled
1363        expression for various reasons (e.g. to see if it's fixed length for look
1364        behind). There were bugs in these functions when a repeated \p or \P was
1365        present in the pattern. These operators have additional parameters compared
1366        with \d, etc, and these were not being taken into account when moving along
1367        the compiled data. Specifically:
1369        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1370            length.
1372        (b) An item such as \pL+ within a repeated group could cause crashes or
1373            loops.
1375        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1376            "reference to non-existent subpattern" error.
1378        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1380    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1381        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1383    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1385    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1386        character were causing crashes (broken optimization).
1388    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1389        \p or \P) caused a compile-time loop.
1391    28. More problems have arisen in unanchored patterns when CRLF is a valid line
1392        break. For example, the unstudied pattern [\r\n]A does not match the string
1393        "\r\nA" because change 7.0/46 below moves the current point on by two
1394        characters after failing to match at the start. However, the pattern \nA
1395        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
1396        the same is true. There doesn't seem any very clean way out of this, but
1397        what I have chosen to do makes the common cases work: PCRE now takes note
1398        of whether there can be an explicit match for \r or \n anywhere in the
1399        pattern, and if so, 7.0/46 no longer applies. As part of this change,
1400        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1401        pattern has explicit CR or LF references.
1403    29. Added (*CR) etc for changing newline setting at start of pattern.
1406    Version 7.2 19-Jun-07
1407    ---------------------
1409     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
1410        which is apparently normally available under Windows.
1412     2. Re-jig the pcregrep tests with different newline settings in an attempt
1413        to make them independent of the local environment's newline setting.
1415     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
1417     4. Some of the "internals" tests were previously cut out when the link size
1418        was not 2, because the output contained actual offsets. The recent new
1419        "Z" feature of pcretest means that these can be cut out, making the tests
1420        usable with all link sizes.
1422     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
1423        stack recursion. This gives a massive performance boost under BSD, but just
1424        a small improvement under Linux. However, it saves one field in the frame
1425        in all cases.
1427     6. Added more features from the forthcoming Perl 5.10:
1429        (a) (?-n) (where n is a string of digits) is a relative subroutine or
1430            recursion call. It refers to the nth most recently opened parentheses.
1432        (b) (?+n) is also a relative subroutine call; it refers to the nth next
1433            to be opened parentheses.
1435        (c) Conditions that refer to capturing parentheses can be specified
1436            relatively, for example, (?(-2)... or (?(+3)...
1438        (d) \K resets the start of the current match so that everything before
1439            is not part of it.
1441        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1443        (f) \g{name} is another synonym - part of Perl 5.10's unification of
1444            reference syntax.
1446        (g) (?| introduces a group in which the numbering of parentheses in each
1447            alternative starts with the same number.
1449        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
1451     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
1454     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
1455        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
1456        for detecting groups that can match an empty string.
1458     9. A pattern with a very large number of alternatives (more than several
1459        hundred) was running out of internal workspace during the pre-compile
1460        phase, where pcre_compile() figures out how much memory will be needed. A
1461        bit of new cunning has reduced the workspace needed for groups with
1462        alternatives. The 1000-alternative test pattern now uses 12 bytes of
1463        workspace instead of running out of the 4096 that are available.
1465    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
1467    11. Applied patch from Google to remove an optimization that didn't quite work.
1468        The report of the bug said:
1470          pcrecpp::RE("a*").FullMatch("aaa") matches, while
1471          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
1472          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
1474    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
1475        it matched the wrong number of bytes.
1478    Version 7.1 24-Apr-07
1479    ---------------------
1481     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
1482        that is more "standard", making use of automake and other Autotools. There
1483        is some re-arrangement of the files and adjustment of comments consequent
1484        on this.
1486     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
1487        for recursive directory scanning broke on some systems because the files
1488        are not scanned in any specific order and on different systems the order
1489        was different. A call to "sort" has been inserted into RunGrepTest for the
1490        approprate test as a short-term fix. In the longer term there may be an
1491        alternative.
1493     3. I had an email from Eric Raymond about problems translating some of PCRE's
1494        man pages to HTML (despite the fact that I distribute HTML pages, some
1495        people do their own conversions for various reasons). The problems
1496        concerned the use of low-level troff macros .br and .in. I have therefore
1497        removed all such uses from the man pages (some were redundant, some could
1498        be replaced by .nf/.fi pairs). The 132html script that I use to generate
1499        HTML has been updated to handle .nf/.fi and to complain if it encounters
1500        .br or .in.
1502     4. Updated comments in configure.ac that get placed in config.h.in and also
1503        arranged for config.h to be included in the distribution, with the name
1504        config.h.generic, for the benefit of those who have to compile without
1505        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
1507     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
1508        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
1509        makevp.bat for the current PCRE, using the additional files
1510        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
1512     6. A Windows user reported a minor discrepancy with test 2, which turned out
1513        to be caused by a trailing space on an input line that had got lost in his
1514        copy. The trailing space was an accident, so I've just removed it.
1516     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
1517        that is needed.
1519     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
1520        as "const" (a) because they are and (b) because it helps the PHP
1521        maintainers who have recently made a script to detect big data structures
1522        in the php code that should be moved to the .rodata section. I remembered
1523        to update Builducptable as well, so it won't revert if ucptable.h is ever
1524        re-created.
1526     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
1527        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
1528        order to be able to cut out the UTF-8 tables in the latter when UTF-8
1529        support is not required. This saves 1.5-2K of code, which is important in
1530        some applications.
1532        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
1533        so as not to refer to the tables, even though these functions will never be
1534        called when UTF-8 support is disabled. Otherwise there are problems with a
1535        shared library.
1537    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
1539        (a) It was defining its arguments as char * instead of void *.
1541        (b) It was assuming that all moves were upwards in memory; this was true
1542            a long time ago when I wrote it, but is no longer the case.
1544        The emulated memove() is provided for those environments that have neither
1545        memmove() nor bcopy(). I didn't think anyone used it these days, but that
1546        is clearly not the case, as these two bugs were recently reported.
1548    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
1549        and Detrail to create the HTML documentation, the .txt form of the man
1550        pages, and it removes trailing spaces from listed files. It also creates
1551        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
1552        case, it wraps all the #defines with #ifndefs. This script should be run
1553        before "make dist".
1555    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1556        with Unicode property support.
1558        (a) For a maximizing quantifier, if the two different cases of the
1559            character were of different lengths in their UTF-8 codings (there are
1560            some cases like this - I found 11), and the matching function had to
1561            back up over a mixture of the two cases, it incorrectly assumed they
1562            were both the same length.
1564        (b) When PCRE was configured to use the heap rather than the stack for
1565            recursion during matching, it was not correctly preserving the data for
1566            the other case of a UTF-8 character when checking ahead for a match
1567            while processing a minimizing repeat. If the check also involved
1568            matching a wide character, but failed, corruption could cause an
1569            erroneous result when trying to check for a repeat of the original
1570            character.
1572    13. Some tidying changes to the testing mechanism:
1574        (a) The RunTest script now detects the internal link size and whether there
1575            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1576            values substituted by "configure". (The RunGrepTest script already did
1577            this for UTF-8.) The configure.ac script no longer substitutes the
1578            relevant variables.
1580        (b) The debugging options /B and /D in pcretest show the compiled bytecode
1581            with length and offset values. This means that the output is different
1582            for different internal link sizes. Test 2 is skipped for link sizes
1583            other than 2 because of this, bypassing the problem. Unfortunately,
1584            there was also a test in test 3 (the locale tests) that used /B and
1585            failed for link sizes other than 2. Rather than cut the whole test out,
1586            I have added a new /Z option to pcretest that replaces the length and
1587            offset values with spaces. This is now used to make test 3 independent
1588            of link size. (Test 2 will be tidied up later.)
1590    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1591        segmentation fault instead of returning the appropriate error message.
1593    15. In multiline mode when the newline sequence was set to "any", the pattern
1594        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1595        This doesn't seem right; it now treats the CRLF combination as the line
1596        ending, and so does not match in that case. It's only a pattern such as ^$
1597        that would hit this one: something like ^ABC$ would have failed after \r
1598        and then tried again after \r\n.
1600    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1601        in an attempt to make files that differ only in their line terminators
1602        compare equal. This works on Linux.
1604    17. Under certain error circumstances pcregrep might try to free random memory
1605        as it exited. This is now fixed, thanks to valgrind.
1607    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1608        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1609        was because its rules for how to advance for /g after matching an empty
1610        string at the end of a line did not allow for this case. They now check for
1611        it specially.
1613    20. pcretest is supposed to handle patterns and data of any length, by
1614        extending its buffers when necessary. It was getting this wrong when the
1615        buffer for a data line had to be extended.
1617    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1618        CRLF as a newline sequence.
1620    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1621        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1622        I have nevertheless tidied it up.
1624    23. Added some casts to kill warnings from HP-UX ia64 compiler.
1626    24. Added a man page for pcre-config.
1629    Version 7.0 19-Dec-06
1630    ---------------------
1632     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1633        moving to gcc 4.1.1.
1635     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1636        sys/time.h, which is documented as needed for this function. It doesn't
1637        seem to matter on Linux, but it showed up on some releases of OS X.
1639     3. It seems that there are systems where bytes whose values are greater than
1640        127 match isprint() in the "C" locale. The "C" locale should be the
1641        default when a C program starts up. In most systems, only ASCII printing
1642        characters match isprint(). This difference caused the output from pcretest
1643        to vary, making some of the tests fail. I have changed pcretest so that:
1645        (a) When it is outputting text in the compiled version of a pattern, bytes
1646            other than 32-126 are always shown as hex escapes.
1648        (b) When it is outputting text that is a matched part of a subject string,
1649            it does the same, unless a different locale has been set for the match
1650            (using the /L modifier). In this case, it uses isprint() to decide.
1652     4. Fixed a major bug that caused incorrect computation of the amount of memory
1653        required for a compiled pattern when options that changed within the
1654        pattern affected the logic of the preliminary scan that determines the
1655        length. The relevant options are -x, and -i in UTF-8 mode. The result was
1656        that the computed length was too small. The symptoms of this bug were
1657        either the PCRE error "internal error: code overflow" from pcre_compile(),
1658        or a glibc crash with a message such as "pcretest: free(): invalid next
1659        size (fast)". Examples of patterns that provoked this bug (shown in
1660        pcretest format) are:
1662          /(?-x: )/x
1663          /(?x)(?-x: \s*#\s*)/
1664          /((?i)[\x{c0}])/8
1665          /(?i:[\x{c0}])/8
1667        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1668        is now done differently.
1670     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1671        wrapper classes; (b) implement a new function in the C++ scanner that is
1672        more efficient than the old way of doing things because it avoids levels of
1673        recursion in the regex matching; (c) add a paragraph to the documentation
1674        for the FullMatch() function.
1676     6. The escape sequence \n was being treated as whatever was defined as
1677        "newline". Not only was this contrary to the documentation, which states
1678        that \n is character 10 (hex 0A), but it also went horribly wrong when
1679        "newline" was defined as CRLF. This has been fixed.
1681     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1682        was being set to -1 for the "end of line" case (supposedly a value that no
1683        character can have). Though this value is never used (the check for end of
1684        line is "zero bytes in current character"), it caused compiler complaints.
1685        I've changed it to 0xffffffff.
1687     8. In pcre_version.c, the version string was being built by a sequence of
1688        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1689        string (as it is for production releases) called a macro with an empty
1690        argument. The C standard says the result of this is undefined. The gcc
1691        compiler treats it as an empty string (which was what was wanted) but it is
1692        reported that Visual C gives an error. The source has been hacked around to
1693        avoid this problem.
1695     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1696        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1697        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1698        of them did).
1700    10. Originally, pcretest opened its input and output without "b"; then I was
1701        told that "b" was needed in some environments, so it was added for release
1702        5.0 to both the input and output. (It makes no difference on Unix-like
1703        systems.) Later I was told that it is wrong for the input on Windows. I've
1704        now abstracted the modes into two macros, to make it easier to fiddle with
1705        them, and removed "b" from the input mode under Windows.
1707    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1709    12. Added -help and --help to pcretest as an official way of being reminded
1710        of the options.
1712    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1713        and pcrecpp.cc because they annoy compilers at high warning levels.
1715    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1717    15. Fixed an occurrence of == in configure.ac that should have been = (shell
1718        scripts are not C programs :-) and which was not noticed because it works
1719        on Linux.
1721    16. pcretest is supposed to handle any length of pattern and data line (as one
1722        line or as a continued sequence of lines) by extending its input buffer if
1723        necessary. This feature was broken for very long pattern lines, leading to
1724        a string of junk being passed to pcre_compile() if the pattern was longer
1725        than about 50K.
1727    17. I have done a major re-factoring of the way pcre_compile() computes the
1728        amount of memory needed for a compiled pattern. Previously, there was code
1729        that made a preliminary scan of the pattern in order to do this. That was
1730        OK when PCRE was new, but as the facilities have expanded, it has become
1731        harder and harder to keep it in step with the real compile phase, and there
1732        have been a number of bugs (see for example, 4 above). I have now found a
1733        cunning way of running the real compile function in a "fake" mode that
1734        enables it to compute how much memory it would need, while actually only
1735        ever using a few hundred bytes of working memory and without too many
1736        tests of the mode. This should make future maintenance and development
1737        easier. A side effect of this work is that the limit of 200 on the nesting
1738        depth of parentheses has been removed (though this was never a serious
1739        limitation, I suspect). However, there is a downside: pcre_compile() now
1740        runs more slowly than before (30% or more, depending on the pattern). I
1741        hope this isn't a big issue. There is no effect on runtime performance.
1743    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1744        newline (only possible for the last line of a file) and it was a
1745        pattern that set a locale (followed by /Lsomething), pcretest crashed.
1747    19. Added additional timing features to pcretest. (1) The -tm option now times
1748        matching only, not compiling. (2) Both -t and -tm can be followed, as a
1749        separate command line item, by a number that specifies the number of
1750        repeats to use when timing. The default is 50000; this gives better
1751        precision, but takes uncomfortably long for very large patterns.
1753    20. Extended pcre_study() to be more clever in cases where a branch of a
1754        subpattern has no definite first character. For example, (a*|b*)[cd] would
1755        previously give no result from pcre_study(). Now it recognizes that the
1756        first character must be a, b, c, or d.
1758    21. There was an incorrect error "recursive call could loop indefinitely" if
1759        a subpattern (or the entire pattern) that was being tested for matching an
1760        empty string contained only one non-empty item after a nested subpattern.
1761        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1762        incorrectly, because the \d was being skipped in the check.
1764    22. The pcretest program now has a new pattern option /B and a command line
1765        option -b, which is equivalent to adding /B to every pattern. This causes
1766        it to show the compiled bytecode, without the additional information that
1767        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1768        is the same as /B/I).
1770    23. A new optimization is now able automatically to treat some sequences such
1771        as a*b as a*+b. More specifically, if something simple (such as a character
1772        or a simple class like \d) has an unlimited quantifier, and is followed by
1773        something that cannot possibly match the quantified thing, the quantifier
1774        is automatically "possessified".
1776    24. A recursive reference to a subpattern whose number was greater than 39
1777        went wrong under certain circumstances in UTF-8 mode. This bug could also
1778        have affected the operation of pcre_study().
1780    25. Realized that a little bit of performance could be had by replacing
1781        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1783    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1785    27. Possessive quantifiers such as a++ were previously implemented by turning
1786        them into atomic groups such as ($>a+). Now they have their own opcodes,
1787        which improves performance. This includes the automatically created ones
1788        from 23 above.
1790    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1791        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1792        the first matched character to be a colon. This applied both to named and
1793        numbered groups.
1795    29. The ucpinternal.h header file was missing its idempotency #ifdef.
1797    30. I was sent a "project" file called libpcre.a.dev which I understand makes
1798        building PCRE on Windows easier, so I have included it in the distribution.
1800    31. There is now a check in pcretest against a ridiculously large number being
1801        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1802        loop, the loop is abandoned.
1804    32. Forward references to subpatterns in conditions such as (?(2)...) where
1805        subpattern 2 is defined later cause pcre_compile() to search forwards in
1806        the pattern for the relevant set of parentheses. This search went wrong
1807        when there were unescaped parentheses in a character class, parentheses
1808        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1810    33. "Subroutine" calls and backreferences were previously restricted to
1811        referencing subpatterns earlier in the regex. This restriction has now
1812        been removed.
1814    34. Added a number of extra features that are going to be in Perl 5.10. On the
1815        whole, these are just syntactic alternatives for features that PCRE had
1816        previously implemented using the Python syntax or my own invention. The
1817        other formats are all retained for compatibility.
1819        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1820            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1821            also .NET compatible.
1823        (b) A recursion or subroutine call to a named group can now be defined as
1824            (?&name) as well as (?P>name).
1826        (c) A backreference to a named group can now be defined as \k<name> or
1827            \k'name' as well as (?P=name). The new forms, as well as being in Perl
1828            5.10, are also .NET compatible.
1830        (d) A conditional reference to a named group can now use the syntax
1831            (?(<name>) or (?('name') as well as (?(name).
1833        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1834            groups (named and numbered) that are never evaluated inline, but can be
1835            called as "subroutines" from elsewhere. In effect, the DEFINE condition
1836            is always false. There may be only one alternative in such a group.
1838        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1839            as the simple (?(R). The condition is true only if the most recent
1840            recursion is that of the given number or name. It does not search out
1841            through the entire recursion stack.
1843        (g) The escape \gN or \g{N} has been added, where N is a positive or
1844            negative number, specifying an absolute or relative reference.
1846    35. Tidied to get rid of some further signed/unsigned compiler warnings and
1847        some "unreachable code" warnings.
1849    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1850        things, this adds five new scripts.
1852    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1853        There were also incompatibilities regarding the handling of \Q..\E inside
1854        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1855        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1857    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1858        matches an empty string, and forcibly breaks the loop. There were bugs in
1859        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
1860        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
1861        separate and independent bugs (that affected different cases) have been
1862        fixed.
1864    39. Refactored the code to abolish the use of different opcodes for small
1865        capturing bracket numbers. This is a tidy that I avoided doing when I
1866        removed the limit on the number of capturing brackets for 3.5 back in 2001.
1867        The new approach is not only tidier, it makes it possible to reduce the
1868        memory needed to fix the previous bug (38).
1870    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1871        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1872        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1873        mode.
1875    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1876        report.
1878    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1879        copy construction and assignment in the C++ wrapper.
1881    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1882        couple of bugs that could have given wrong results in the "--newline=crlf"
1883        case.
1885    44. Added a number of casts and did some reorganization of signed/unsigned int
1886        variables following suggestions from Dair Grant. Also renamed the variable
1887        "this" as "item" because it is a C++ keyword.
1889    45. Arranged for dftables to add
1891          #include "pcre_internal.h"
1893        to pcre_chartables.c because without it, gcc 4.x may remove the array
1894        definition from the final binary if PCRE is built into a static library and
1895        dead code stripping is activated.
1897    46. For an unanchored pattern, if a match attempt fails at the start of a
1898        newline sequence, and the newline setting is CRLF or ANY, and the next two
1899        characters are CRLF, advance by two characters instead of one.
1902    Version 6.7 04-Jul-06
1903    ---------------------
1905     1. In order to handle tests when input lines are enormously long, pcretest has
1906        been re-factored so that it automatically extends its buffers when
1907        necessary. The code is crude, but this _is_ just a test program. The
1908        default size has been increased from 32K to 50K.
1910     2. The code in pcre_study() was using the value of the re argument before
1911        testing it for NULL. (Of course, in any sensible call of the function, it
1912        won't be NULL.)
1914     3. The memmove() emulation function in pcre_internal.h, which is used on
1915        systems that lack both memmove() and bcopy() - that is, hardly ever -
1916        was missing a "static" storage class specifier.
1918     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1919        containing an extended class (one that cannot be represented by a bitmap
1920        because it contains high-valued characters or Unicode property items, e.g.
1921        [\pZ]). Almost always one would set UTF-8 mode when processing such a
1922        pattern, but PCRE should not loop if you do not (it no longer does).
1923        [Detail: two cases were found: (a) a repeated subpattern containing an
1924        extended class; (b) a recursive reference to a subpattern that followed a
1925        previous extended class. It wasn't skipping over the extended class
1926        correctly when UTF-8 mode was not set.]
1928     5. A negated single-character class was not being recognized as fixed-length
1929        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1930        compile error "lookbehind assertion is not fixed length".
1932     6. The RunPerlTest auxiliary script was showing an unexpected difference
1933        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1934        write a Perl script that can interpret lines of an input file either as
1935        byte characters or as UTF-8, which is what "perltest" was being required to
1936        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1937        can't do is switch easily at run time between having the "use utf8;" pragma
1938        or not. In the end, I fudged it by using the RunPerlTest script to insert
1939        "use utf8;" explicitly for the UTF-8 tests.
1941     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1942        the end of the subject string, contrary to the documentation and to what
1943        Perl does. This was true of both matching functions. Now it matches only at
1944        the start of the subject and immediately after *internal* newlines.
1946     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1947        a pointer to an int instead of a pointer to an unsigned long int. This
1948        caused problems on 64-bit systems.
1950     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1951        instance of the 'standard' template library not being so standard".
1953    10. There was no check on the number of named subpatterns nor the maximum
1954        length of a subpattern name. The product of these values is used to compute
1955        the size of the memory block for a compiled pattern. By supplying a very
1956        long subpattern name and a large number of named subpatterns, the size
1957        computation could be caused to overflow. This is now prevented by limiting
1958        the length of names to 32 characters, and the number of named subpatterns
1959        to 10,000.
1961    11. Subpatterns that are repeated with specific counts have to be replicated in
1962        the compiled pattern. The size of memory for this was computed from the
1963        length of the subpattern and the repeat count. The latter is limited to
1964        65535, but there was no limit on the former, meaning that integer overflow
1965        could in principle occur. The compiled length of a repeated subpattern is
1966        now limited to 30,000 bytes in order to prevent this.
1968    12. Added the optional facility to have named substrings with the same name.
1970    13. Added the ability to use a named substring as a condition, using the
1971        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1972        are numbers (not recommended). Forward references are permitted.
1974    14. Added forward references in named backreferences (if you see what I mean).
1976    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1977        pattern could run off the end of the subject. For example, the pattern
1978        "(?s)(.{1,5})"8 did this with the subject "ab".
1980    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1981        PCRE_CASELESS was set when matching characters that were quantified with ?
1982        or *.
1984    17. A character class other than a single negated character that had a minimum
1985        but no maximum quantifier - for example [ab]{6,} - was not handled
1986        correctly by pce_dfa_exec(). It would match only one character.
1988    18. A valid (though odd) pattern that looked like a POSIX character
1989        class but used an invalid character after [ (for example [[,abc,]]) caused
1990        pcre_compile() to give the error "Failed: internal error: code overflow" or
1991        in some cases to crash with a glibc free() error. This could even happen if
1992        the pattern terminated after [[ but there just happened to be a sequence of
1993        letters, a binary zero, and a closing ] in the memory that followed.
1995    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1996        over the years. Originally (before any Unicode support), just the bottom 8
1997        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1998        output from "man perlunicode" includes this:
2000          The regular expression compiler produces polymorphic opcodes.  That
2001          is, the pattern adapts to the data and automatically switches to
2002          the Unicode character scheme when presented with Unicode data--or
2003          instead uses a traditional byte scheme when presented with byte
2004          data.
2006        Sadly, a wide octal escape does not cause a switch, and in a string with
2007        no other multibyte characters, these octal escapes are treated as before.
2008        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
2009        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
2010        Unicode string.
2012        I have not perpetrated such confusion in PCRE. Up till now, it took just
2013        the bottom 8 bits, as in old Perl. I have now made octal escapes with
2014        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
2015        translate to the appropriate multibyte character.
2017    29. Applied some refactoring to reduce the number of warnings from Microsoft
2018        and Borland compilers. This has included removing the fudge introduced
2019        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
2020        a warning about an unused variable.
2022    21. PCRE has not included VT (character 0x0b) in the set of whitespace
2023        characters since release 4.0, because Perl (from release 5.004) does not.
2024        [Or at least, is documented not to: some releases seem to be in conflict
2025        with the documentation.] However, when a pattern was studied with
2026        pcre_study() and all its branches started with \s, PCRE still included VT
2027        as a possible starting character. Of course, this did no harm; it just
2028        caused an unnecessary match attempt.
2030    22. Removed a now-redundant internal flag bit that recorded the fact that case
2031        dependency changed within the pattern. This was once needed for "required
2032        byte" processing, but is no longer used. This recovers a now-scarce options
2033        bit. Also moved the least significant internal flag bit to the most-
2034        significant bit of the word, which was not previously used (hangover from
2035        the days when it was an int rather than a uint) to free up another bit for
2036        the future.
2038    23. Added support for CRLF line endings as well as CR and LF. As well as the
2039        default being selectable at build time, it can now be changed at runtime
2040        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
2041        specify that it is scanning data with non-default line endings.
2043    24. Changed the definition of CXXLINK to make it agree with the definition of
2044        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
2046    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
2047        recursions. This makes a big different to stack usage for some patterns.
2049    26. If a subpattern containing a named recursion or subroutine reference such
2050        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
2051        the space required for the compiled pattern went wrong and gave too small a
2052        value. Depending on the environment, this could lead to "Failed: internal
2053        error: code overflow at offset 49" or "glibc detected double free or
2054        corruption" errors.
2056    27. Applied patches from Google (a) to support the new newline modes and (b) to
2057        advance over multibyte UTF-8 characters in GlobalReplace.
2059    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
2060        difference for some implementation of PCRE in some Windows version.
2062    29. Added some extra testing facilities to pcretest:
2064        \q<number>   in a data line sets the "match limit" value
2065        \Q<number>   in a data line sets the "match recursion limt" value
2066        -S <number>  sets the stack size, where <number> is in megabytes
2068        The -S option isn't available for Windows.
2071    Version 6.6 06-Feb-06
2072    ---------------------
2074     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
2075        in pcreposix.h. I have copied the definition from pcre.h.
2077     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
2078        because pcre.h is no longer a built file.
2080     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
2081        not normally included in the compiled code.
2084    Version 6.5 01-Feb-06
2085    ---------------------
2087     1. When using the partial match feature with pcre_dfa_exec(), it was not
2088        anchoring the second and subsequent partial matches at the new starting
2089        point. This could lead to incorrect results. For example, with the pattern
2090        /1234/, partially matching against "123" and then "a4" gave a match.
2092     2. Changes to pcregrep:
2094        (a) All non-match returns from pcre_exec() were being treated as failures
2095            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
2096            error message is output. Some extra information is given for the
2097            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
2098            probably the only errors that are likely to be caused by users (by
2099            specifying a regex that has nested indefinite repeats, for instance).
2100            If there are more than 20 of these errors, pcregrep is abandoned.
2102        (b) A binary zero was treated as data while matching, but terminated the
2103            output line if it was written out. This has been fixed: binary zeroes
2104            are now no different to any other data bytes.
2106        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
2107            used to set a locale for matching. The --locale=xxxx long option has
2108            been added (no short equivalent) to specify a locale explicitly on the
2109            pcregrep command, overriding the environment variables.
2111        (d) When -B was used with -n, some line numbers in the output were one less
2112            than they should have been.
2114        (e) Added the -o (--only-matching) option.
2116        (f) If -A or -C was used with -c (count only), some lines of context were
2117            accidentally printed for the final match.
2119        (g) Added the -H (--with-filename) option.
2121        (h) The combination of options -rh failed to suppress file names for files
2122            that were found from directory arguments.
2124        (i) Added the -D (--devices) and -d (--directories) options.
2126        (j) Added the -F (--fixed-strings) option.
2128        (k) Allow "-" to be used as a file name for -f as well as for a data file.
2130        (l) Added the --colo(u)r option.
2132        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
2133            is not present by default.
2135     3. A nasty bug was discovered in the handling of recursive patterns, that is,
2136        items such as (?R) or (?1), when the recursion could match a number of
2137        alternatives. If it matched one of the alternatives, but subsequently,
2138        outside the recursion, there was a failure, the code tried to back up into
2139        the recursion. However, because of the way PCRE is implemented, this is not
2140        possible, and the result was an incorrect result from the match.
2142        In order to prevent this happening, the specification of recursion has
2143        been changed so that all such subpatterns are automatically treated as
2144        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
2146     4. I had overlooked the fact that, in some locales, there are characters for
2147        which isalpha() is true but neither isupper() nor islower() are true. In
2148        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
2149        and ordfeminine) are like this. This affected the treatment of \w and \W
2150        when they appeared in character classes, but not when they appeared outside
2151        a character class. The bit map for "word" characters is now created
2152        separately from the results of isalnum() instead of just taking it from the
2153        upper, lower, and digit maps. (Plus the underscore character, of course.)
2155     5. The above bug also affected the handling of POSIX character classes such as
2156        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
2157        permanent tables. Instead, the bit maps for such a class were previously
2158        created as the appropriate unions of the upper, lower, and digit bitmaps.
2159        Now they are created by subtraction from the [[:word:]] class, which has
2160        its own bitmap.
2162     6. The [[:blank:]] character class matches horizontal, but not vertical space.
2163        It is created by subtracting the vertical space characters (\x09, \x0a,
2164        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
2165        subtraction was done in the overall bitmap for a character class, meaning
2166        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
2167        be recognized. This bug has been fixed.
2169     7. Patches from the folks at Google:
2171          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
2172          real life, but is still worth protecting against".
2174          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
2175          regular expressions".
2177          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
2178          have it.
2180          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
2181          "configure" and the latter not, in order to fix a problem somebody had
2182          with compiling the Arg class on HP-UX.
2184          (e) Improve the error-handling of the C++ wrapper a little bit.
2186          (f) New tests for checking recursion limiting.
2188     8. The pcre_memmove() function, which is used only if the environment does not
2189        have a standard memmove() function (and is therefore rarely compiled),
2190        contained two bugs: (a) use of int instead of size_t, and (b) it was not
2191        returning a result (though PCRE never actually uses the result).
2193     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
2194        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
2195        returned instead of calling malloc() with an overflowing number that would
2196        most likely cause subsequent chaos.
2198    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
2200    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
2201        with this option is matched, the nmatch and pmatch options of regexec() are
2202        ignored.
2204    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
2205        provided in case anyone wants to the the POSIX interface with UTF-8
2206        strings.
2208    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
2209        C++ linking (needed for some HP-UX environments).
2211    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
2212        (unused parameter) and in the pcre_printint() function (omitted "default"
2213        switch label when the default is to do nothing).
2215    15. Added some code to make it possible, when PCRE is compiled as a C++
2216        library, to replace subject pointers for pcre_exec() with a smart pointer
2217        class, thus making it possible to process discontinuous strings.
2219    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
2220        much the same function. They were added by different people who were trying
2221        to make PCRE easy to compile on non-Unix systems. It has been suggested
2222        that PCRE_EXPORT be abolished now that there is more automatic apparatus
2223        for compiling on Windows systems. I have therefore replaced it with
2224        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
2225        defaults to "extern" for C or "extern C" for C++, which works fine on
2226        Unix-like systems. It is now possible to override the value of PCRE_DATA_
2227        SCOPE with something explicit in config.h. In addition:
2229        (a) pcreposix.h still had just "extern" instead of either of these macros;
2230            I have replaced it with PCRE_DATA_SCOPE.
2232        (b) Functions such as _pcre_xclass(), which are internal to the library,
2233            but external in the C sense, all had PCRE_EXPORT in their definitions.
2234            This is apparently wrong for the Windows case, so I have removed it.
2235            (It makes no difference on Unix-like systems.)
2237    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
2238        of recursive calls to match(). This is different to MATCH_LIMIT because
2239        that limits the total number of calls to match(), not all of which increase
2240        the depth of recursion. Limiting the recursion depth limits the amount of
2241        stack (or heap if NO_RECURSE is set) that is used. The default can be set
2242        when PCRE is compiled, and changed at run time. A patch from Google adds
2243        this functionality to the C++ interface.
2245    18. Changes to the handling of Unicode character properties:
2247        (a) Updated the table to Unicode 4.1.0.
2249        (b) Recognize characters that are not in the table as "Cn" (undefined).
2251        (c) I revised the way the table is implemented to a much improved format
2252            which includes recognition of ranges. It now supports the ranges that
2253            are defined in UnicodeData.txt, and it also amalgamates other
2254            characters into ranges. This has reduced the number of entries in the
2255            table from around 16,000 to around 3,000, thus reducing its size
2256            considerably. I realized I did not need to use a tree structure after
2257            all - a binary chop search is just as efficient. Having reduced the
2258            number of entries, I extended their size from 6 bytes to 8 bytes to
2259            allow for more data.
2261        (d) Added support for Unicode script names via properties such as \p{Han}.
2263    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
2264        matching that character.
2266    20. When matching a repeated Unicode property with a minimum greater than zero,
2267        (for example \pL{2,}), PCRE could look past the end of the subject if it
2268        reached it while seeking the minimum number of characters. This could
2269        happen only if some of the characters were more than one byte long, because
2270        there is a check for at least the minimum number of bytes.
2272    21. Refactored the implementation of \p and \P so as to be more general, to
2273        allow for more different types of property in future. This has changed the
2274        compiled form incompatibly. Anybody with saved compiled patterns that use
2275        \p or \P will have to recompile them.
2277    22. Added "Any" and "L&" to the supported property types.
2279    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
2280        but give a compile time error if the value is greater than 0xff.
2282    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
2283        accidentally not being installed or uninstalled.
2285    25. The pcre.h file was built from pcre.h.in, but the only changes that were
2286        made were to insert the current release number. This seemed silly, because
2287        it made things harder for people building PCRE on systems that don't run
2288        "configure". I have turned pcre.h into a distributed file, no longer built
2289        by "configure", with the version identification directly included. There is
2290        no longer a pcre.h.in file.
2292        However, this change necessitated a change to the pcre-config script as
2293        well. It is built from pcre-config.in, and one of the substitutions was the
2294        release number. I have updated configure.ac so that ./configure now finds
2295        the release number by grepping pcre.h.
2297    26. Added the ability to run the tests under valgrind.
2300    Version 6.4 05-Sep-05
2301    ---------------------
2303     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
2304        "--" to be printed when multiple files were scanned, even when none of the
2305        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
2306        consider it to be a bug, and have restored the previous behaviour.
2308     2. A couple of code tidies to get rid of compiler warnings.
2310     3. The pcretest program used to cheat by referring to symbols in the library
2311        whose names begin with _pcre_. These are internal symbols that are not
2312        really supposed to be visible externally, and in some environments it is
2313        possible to suppress them. The cheating is now confined to including
2314        certain files from the library's source, which is a bit cleaner.
2316     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
2317        file's purpose clearer.
2319     5. Reorganized pcre_ucp_findchar().
2322    Version 6.3 15-Aug-05
2323    ---------------------
2325     1. The file libpcre.pc.in did not have general read permission in the tarball.
2327     2. There were some problems when building without C++ support:
2329        (a) If C++ support was not built, "make install" and "make test" still
2330            tried to test it.
2332        (b) There were problems when the value of CXX was explicitly set. Some
2333            changes have been made to try to fix these, and ...
2335        (c) --disable-cpp can now be used to explicitly disable C++ support.
2337        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
2338            backslash in a target when C++ was disabled. This confuses some
2339            versions of "make", apparently. Using an intermediate variable solves
2340            this. (Same for CPP_LOBJ.)
2342     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
2343        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
2344        necessary on certain architectures.
2346     4. Added a setting of -export-symbols-regex to the link command to remove
2347        those symbols that are exported in the C sense, but actually are local
2348        within the library, and not documented. Their names all begin with
2349        "_pcre_". This is not a perfect job, because (a) we have to except some
2350        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
2351        available (and never for static libraries). I have made a note to try to
2352        find a way round (a) in the future.
2355    Version 6.2 01-Aug-05
2356    ---------------------
2358     1. There was no test for integer overflow of quantifier values. A construction
2359        such as {1111111111111111} would give undefined results. What is worse, if
2360        a minimum quantifier for a parenthesized subpattern overflowed and became
2361        negative, the calculation of the memory size went wrong. This could have
2362        led to memory overwriting.
2364     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
2366     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
2367        operating environments where this matters.
2369     4. Applied Giuseppe Maxia's patch to add additional features for controlling
2370        PCRE options from within the C++ wrapper.
2372     5. Named capturing subpatterns were not being correctly counted when a pattern
2373        was compiled. This caused two problems: (a) If there were more than 100
2374        such subpatterns, the calculation of the memory needed for the whole
2375        compiled pattern went wrong, leading to an overflow error. (b) Numerical
2376        back references of the form \12, where the number was greater than 9, were
2377        not recognized as back references, even though there were sufficient
2378        previous subpatterns.
2380     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
2381        versions of gcc, e.g. 2.95.4.
2384    Version 6.1 21-Jun-05
2385    ---------------------
2387     1. There was one reference to the variable "posix" in pcretest.c that was not
2388        surrounded by "#if !defined NOPOSIX".
2390     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
2391        the cross-check on the old pcre_info() function, for the benefit of the
2392        cut-down version of PCRE that is currently imported into Exim.
2394     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
2395        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
2396        patterns that start (?i) but I don't think that matters. The use of (?i) is
2397        just an example; this all applies to the other options as well.
2399     4. Since libtool seems to echo the compile commands it is issuing, the output
2400        from "make" can be reduced a bit by putting "@" in front of each libtool
2401        compile command.
2403     5. Patch from the folks at Google for configure.in to be a bit more thorough
2404        in checking for a suitable C++ installation before trying to compile the
2405        C++ stuff. This should fix a reported problem when a compiler was present,
2406        but no suitable headers.
2408     6. The man pages all had just "PCRE" as their title. I have changed them to
2409        be the relevant file name. I have also arranged that these names are
2410        retained in the file doc/pcre.txt, which is a concatenation in text format
2411        of all the man pages except the little individual ones for each function.
2413     7. The NON-UNIX-USE file had not been updated for the different set of source
2414        files that come with release 6. I also added a few comments about the C++
2415        wrapper.
2418    Version 6.0 07-Jun-05
2419    ---------------------
2421     1. Some minor internal re-organization to help with my DFA experiments.
2423     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
2424        didn't matter for the library itself when fully configured, but did matter
2425        when compiling without UCP support, or within Exim, where the ucp files are
2426        not imported.
2428     3. Refactoring of the library code to split up the various functions into
2429        different source modules. The addition of the new DFA matching code (see
2430        below) to a single monolithic source would have made it really too
2431        unwieldy, quite apart from causing all the code to be include in a
2432        statically linked application, when only some functions are used. This is
2433        relevant even without the DFA addition now that patterns can be compiled in
2434        one application and matched in another.
2436        The downside of splitting up is that there have to be some external
2437        functions and data tables that are used internally in different modules of
2438        the library but which are not part of the API. These have all had their
2439        names changed to start with "_pcre_" so that they are unlikely to clash
2440        with other external names.
2442     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
2443        a different (DFA) algorithm. Although it is slower than the original
2444        function, it does have some advantages for certain types of matching
2445        problem.
2447     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
2448        including restarting after a partial match.
2450     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
2451        defined when compiling for Windows was sent to me. I have put it into the
2452        code, though I have no means of testing or verifying it.
2454     7. Added the pcre_refcount() auxiliary function.
2456     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
2457        match before or at the first newline in the subject string. In pcretest,
2458        the /f option on a pattern can be used to set this.
2460     9. A repeated \w when used in UTF-8 mode with characters greater than 256
2461        would behave wrongly. This has been present in PCRE since release 4.0.
2463    10. A number of changes to the pcregrep command:
2465        (a) Refactored how -x works; insert ^(...)$ instead of setting
2466            PCRE_ANCHORED and checking the length, in preparation for adding
2467            something similar for -w.
2469        (b) Added the -w (match as a word) option.
2471        (c) Refactored the way lines are read and buffered so as to have more
2472            than one at a time available.
2474        (d) Implemented a pcregrep test script.
2476        (e) Added the -M (multiline match) option. This allows patterns to match
2477            over several lines of the subject. The buffering ensures that at least
2478            8K, or the rest of the document (whichever is the shorter) is available
2479            for matching (and similarly the previous 8K for lookbehind assertions).
2481        (f) Changed the --help output so that it now says
2483              -w, --word-regex(p)
2485            instead of two lines, one with "regex" and the other with "regexp"
2486            because that confused at least one person since the short forms are the
2487            same. (This required a bit of code, as the output is generated
2488            automatically from a table. It wasn't just a text change.)
2490        (g) -- can be used to terminate pcregrep options if the next thing isn't an
2491            option but starts with a hyphen. Could be a pattern or a path name
2492            starting with a hyphen, for instance.
2494        (h) "-" can be given as a file name to represent stdin.
2496        (i) When file names are being printed, "(standard input)" is used for
2497            the standard input, for compatibility with GNU grep. Previously
2498            "<stdin>" was used.
2500        (j) The option --label=xxx can be used to supply a name to be used for
2501            stdin when file names are being printed. There is no short form.
2503        (k) Re-factored the options decoding logic because we are going to add
2504            two more options that take data. Such options can now be given in four
2505            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
2507        (l) Added the -A, -B, and -C options for requesting that lines of context
2508            around matches be printed.
2510        (m) Added the -L option to print the names of files that do not contain
2511            any matching lines, that is, the complement of -l.
2513        (n) The return code is 2 if any file cannot be opened, but pcregrep does
2514            continue to scan other files.
2516        (o) The -s option was incorrectly implemented. For compatibility with other
2517            greps, it now suppresses the error message for a non-existent or non-
2518            accessible file (but not the return code). There is a new option called
2519            -q that suppresses the output of matching lines, which was what -s was
2520            previously doing.
2522        (p) Added --include and --exclude options to specify files for inclusion
2523            and exclusion when recursing.
2525    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
2526        Hopefully, it now does.
2528    12. Missing cast in pcre_study().
2530    13. Added an "uninstall" target to the makefile.
2532    14. Replaced "extern" in the function prototypes in Makefile.in with
2533        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
2534        world, but is set differently for Windows.
2536    15. Added a second compiling function called pcre_compile2(). The only
2537        difference is that it has an extra argument, which is a pointer to an
2538        integer error code. When there is a compile-time failure, this is set
2539        non-zero, in addition to the error test pointer being set to point to an
2540        error message. The new argument may be NULL if no error number is required
2541        (but then you may as well call pcre_compile(), which is now just a
2542        wrapper). This facility is provided because some applications need a
2543        numeric error indication, but it has also enabled me to tidy up the way
2544        compile-time errors are handled in the POSIX wrapper.
2546    16. Added VPATH=.libs to the makefile; this should help when building with one
2547        prefix path and installing with another. (Or so I'm told by someone who
2548        knows more about this stuff than I do.)
2550    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
2551        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
2552        match everything, including newlines. This is not POSIX-compatible, but
2553        somebody wanted the feature. From pcretest it can be activated by using
2554        both the P and the s flags.
2556    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
2558    19. libpcre.pc was being incorrectly installed as executable.
2560    20. A couple of places in pcretest check for end-of-line by looking for '\n';
2561        it now also looks for '\r' so that it will work unmodified on Windows.
2563    21. Added Google's contributed C++ wrapper to the distribution.
2565    22. Added some untidy missing memory free() calls in pcretest, to keep
2566        Electric Fence happy when testing.
2570    Version 5.0 13-Sep-04
2571    ---------------------
2573     1. Internal change: literal characters are no longer packed up into items
2574        containing multiple characters in a single byte-string. Each character
2575        is now matched using a separate opcode. However, there may be more than one
2576        byte in the character in UTF-8 mode.
2578     2. The pcre_callout_block structure has two new fields: pattern_position and
2579        next_item_length. These contain the offset in the pattern to the next match
2580        item, and its length, respectively.
2582     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
2583        insertion of callouts before each pattern item. Added the /C option to
2584        pcretest to make use of this.
2586     4. On the advice of a Windows user, the lines
2588          #if defined(_WIN32) || defined(WIN32)
2589          _setmode( _fileno( stdout ), 0x8000 );
2590          #endif  /* defined(_WIN32) || defined(WIN32) */
2592        have been added to the source of pcretest. This apparently does useful
2593        magic in relation to line terminators.
2595     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
2596        for the benefit of those environments where the "b" makes a difference.
2598     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
2599        to know about it. I have put a hack into configure.in that adds in code
2600        to set GCC=yes if CC=icc. This seems to end up at a point in the
2601        generated configure script that is early enough to affect the setting of
2602        compiler options, which is what is needed, but I have no means of testing
2603        whether it really works. (The user who reported this had patched the
2604        generated configure script, which of course I cannot do.)
2606        LATER: After change 22 below (new libtool files), the configure script
2607        seems to know about icc (and also ecc). Therefore, I have commented out
2608        this hack in configure.in.
2610     7. Added support for pkg-config (2 patches were sent in).
2612     8. Negated POSIX character classes that used a combination of internal tables
2613        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2614        [[:^ascii]]. Typically, they would match almost any characters. The other
2615        POSIX classes were not broken in this way.
2617     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2618        to find the match, as PCRE was deluded into thinking that the match had to
2619        start at the start point or following a newline. The same bug applied to
2620        patterns with negative forward assertions or any backward assertions
2621        preceding ".*" at the start, unless the pattern required a fixed first
2622        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2624    10. In UTF-8 mode, when moving forwards in the subject after a failed match
2625        starting at the last subject character, bytes beyond the end of the subject
2626        string were read.
2628    11. Renamed the variable "class" as "classbits" to make life easier for C++
2629        users. (Previously there was a macro definition, but it apparently wasn't
2630        enough.)
2632    12. Added the new field "tables" to the extra data so that tables can be passed
2633        in at exec time, or the internal tables can be re-selected. This allows
2634        a compiled regex to be saved and re-used at a later time by a different
2635        program that might have everything at different addresses.
2637    13. Modified the pcre-config script so that, when run on Solaris, it shows a
2638        -R library as well as a -L library.
2640    14. The debugging options of pcretest (-d on the command line or D on a
2641        pattern) showed incorrect output for anything following an extended class
2642        that contained multibyte characters and which was followed by a quantifier.
2644    15. Added optional support for general category Unicode character properties
2645        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2646        support. It adds about 90K to the size of the library. The meanings of the
2647        inbuilt class escapes such as \d and \s have NOT been changed.
2649    16. Updated pcredemo.c to include calls to free() to release the memory for the
2650        compiled pattern.
2652    17. The generated file chartables.c was being created in the source directory
2653        instead of in the building directory. This caused the build to fail if the
2654        source directory was different from the building directory, and was
2655        read-only.
2657    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2658        file. No doubt somebody will tell me if they don't make sense... Also added
2659        Dan Mooney's comments about building on OpenVMS.
2661    19. Added support for partial matching via the PCRE_PARTIAL option for
2662        pcre_exec() and the \P data escape in pcretest.
2664    20. Extended pcretest with 3 new pattern features:
2666        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
2667              write the compiled pattern to the file whose name is "rest-of-line".
2668              This is a straight binary dump of the data, with the saved pointer to
2669              the character tables forced to be NULL. The study data, if any, is
2670              written too. After writing, pcretest reads a new pattern.
2672        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2673              compiled pattern from the given file. There must not be any
2674              occurrences of "<" in the file name (pretty unlikely); if there are,
2675              pcretest will instead treat the initial "<" as a pattern delimiter.
2676              After reading in the pattern, pcretest goes on to read data lines as
2677              usual.
2679        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2680              and 16-bit fields in a compiled pattern, to simulate a pattern that
2681              was compiled on a host of opposite endianness.
2683    21. The pcre-exec() function can now cope with patterns that were compiled on
2684        hosts of opposite endianness, with this restriction:
2686          As for any compiled expression that is saved and used later, the tables
2687          pointer field cannot be preserved; the extra_data field in the arguments
2688          to pcre_exec() should be used to pass in a tables address if a value
2689          other than the default internal tables were used at compile time.
2691    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2692        now diagnosed as an error. Previously, most of the time, a negative number
2693        would have been treated as zero, but if in addition "ovector" was passed as
2694        NULL, a crash could occur.
2696    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2697        new versions from the libtool 1.5 distribution (the last one is a copy of
2698        a file called libtool.m4). This seems to have fixed the need to patch
2699        "configure" to support Darwin 1.3 (which I used to do). However, I still
2700        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2701        workstation).
2703    24. Changed the PCRE licence to be the more standard "BSD" licence.
2706    Version 4.5 01-Dec-03
2707    ---------------------
2709     1. There has been some re-arrangement of the code for the match() function so
2710        that it can be compiled in a version that does not call itself recursively.
2711        Instead, it keeps those local variables that need separate instances for
2712        each "recursion" in a frame on the heap, and gets/frees frames whenever it
2713        needs to "recurse". Keeping track of where control must go is done by means
2714        of setjmp/longjmp. The whole thing is implemented by a set of macros that
2715        hide most of the details from the main code, and operates only if
2716        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2717        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2718        operating.
2720        To make it easier for callers to provide specially tailored get/free
2721        functions for this usage, two new functions, pcre_stack_malloc, and
2722        pcre_stack_free, are used. They are always called in strict stacking order,
2723        and the size of block requested is always the same.
2725        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2726        PCRE has been compiled to use the stack or the heap for recursion. The
2727        -C option of pcretest uses this to show which version is compiled.
2729        A new data escape \S, is added to pcretest; it causes the amounts of store
2730        obtained and freed by both kinds of malloc/free at match time to be added
2731        to the output.
2733     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2734        what's available on my current Linux desktop machine.
2736     3. When matching a UTF-8 string, the test for a valid string at the start has
2737        been extended. If start_offset is not zero, PCRE now checks that it points
2738        to a byte that is the start of a UTF-8 character. If not, it returns
2739        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2740        this is necessary because there may be backward assertions in the pattern.
2741        When matching the same subject several times, it may save resources to use
2742        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2744     4. The code for checking the validity of UTF-8 strings has been tightened so
2745        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2746        containing "overlong sequences".
2748     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2749        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2750        should have been "&", but it just so happened that all the cases this let
2751        through by mistake were picked up later in the function.
2753     6. I had used a variable called "isblank" - this is a C99 function, causing
2754        some compilers to warn. To avoid this, I renamed it (as "blankclass").
2756     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2757        prompting; (b) run "./pcretest /dev/null" at the start of the test script
2758        so the version is shown; (c) stop "make test" echoing "./RunTest".
2760     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2762     9. The prototype for memmove() for systems that don't have it was using
2763        size_t, but the inclusion of the header that defines size_t was later. I've
2764        moved the #includes for the C headers earlier to avoid this.
2766    10. Added some adjustments to the code to make it easier to compiler on certain
2767        special systems:
2769          (a) Some "const" qualifiers were missing.
2770          (b) Added the macro EXPORT before all exported functions; by default this
2771              is defined to be empty.
2772          (c) Changed the dftables auxiliary program (that builds chartables.c) so
2773              that it reads its output file name as an argument instead of writing
2774              to the standard output and assuming this can be redirected.
2776    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2777        class containing characters with values greater than 255, PCRE compilation
2778        went into a loop.
2780    12. A recursive reference to a subpattern that was within another subpattern
2781        that had a minimum quantifier of zero caused PCRE to crash. For example,
2782        (x(y(?2))z)? provoked this bug with a subject that got as far as the
2783        recursion. If the recursively-called subpattern itself had a zero repeat,
2784        that was OK.
2786    13. In pcretest, the buffer for reading a data line was set at 30K, but the
2787        buffer into which it was copied (for escape processing) was still set at
2788        1024, so long lines caused crashes.
2790    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2791        "internal error: code overflow...". This applied to any character class
2792        that was followed by a possessive quantifier.
2794    15. Modified the Makefile to add libpcre.la as a prerequisite for
2795        libpcreposix.la because I was told this is needed for a parallel build to
2796        work.
2798    16. If a pattern that contained .* following optional items at the start was
2799        studied, the wrong optimizing data was generated, leading to matching
2800        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2801        matching string must start with a or b or c. The correct conclusion for
2802        this pattern is that a match can start with any character.
2805    Version 4.4 13-Aug-03
2806    ---------------------
2808     1. In UTF-8 mode, a character class containing characters with values between
2809        127 and 255 was not handled correctly if the compiled pattern was studied.
2810        In fixing this, I have also improved the studying algorithm for such
2811        classes (slightly).
2813     2. Three internal functions had redundant arguments passed to them. Removal
2814        might give a very teeny performance improvement.
2816     3. Documentation bug: the value of the capture_top field in a callout is *one
2817        more than* the number of the hightest numbered captured substring.
2819     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2820        in incorrectly linking with a previously installed version. They now link
2821        explicitly with libpcre.la.
2823     5. configure.in no longer needs to recognize Cygwin specially.
2825     6. A problem in pcre.in for Windows platforms is fixed.
2827     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2828        pcretest, it used to include the size of the study block as part of its
2829        output. Unfortunately, the structure contains a field that has a different
2830        size on different hardware architectures. This meant that the tests that
2831        showed this size failed. As the block is currently always of a fixed size,
2832        this information isn't actually particularly useful in pcretest output, so
2833        I have just removed it.
2835     8. Three pre-processor statements accidentally did not start in column 1.
2836        Sadly, there are *still* compilers around that complain, even though
2837        standard C has not required this for well over a decade. Sigh.
2839     9. In pcretest, the code for checking callouts passed small integers in the
2840        callout_data field, which is a void * field. However, some picky compilers
2841        complained about the casts involved for this on 64-bit systems. Now
2842        pcretest passes the address of the small integer instead, which should get
2843        rid of the warnings.
2845    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2846        both compile and run time, and gives an error if an invalid UTF-8 sequence
2847        is found. There is a option for disabling this check in cases where the
2848        string is known to be correct and/or the maximum performance is wanted.
2850    11. In response to a bug report, I changed one line in Makefile.in from
2852            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2853        to
2854            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2856        to look similar to other lines, but I have no way of telling whether this
2857        is the right thing to do, as I do not use Windows. No doubt I'll get told
2858        if it's wrong...
2861    Version 4.3 21-May-03
2862    ---------------------
2864    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2865       Makefile.
2867    2. Some refactoring to improve the quality of the code:
2869       (i)   The utf8_table... variables are now declared "const".
2871       (ii)  The code for \cx, which used the "case flipping" table to upper case
2872             lower case letters, now just substracts 32. This is ASCII-specific,
2873             but the whole concept of \cx is ASCII-specific, so it seems
2874             reasonable.
2876       (iii) PCRE was using its character types table to recognize decimal and
2877             hexadecimal digits in the pattern. This is silly, because it handles
2878             only 0-9, a-f, and A-F, but the character types table is locale-
2879             specific, which means strange things might happen. A private
2880             table is now used for this - though it costs 256 bytes, a table is
2881             much faster than multiple explicit tests. Of course, the standard
2882             character types table is still used for matching digits in subject
2883             strings against \d.
2885       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2886             ending in _t are). So I've renamed it as ESC_tee.
2888    3. The first argument for regexec() in the POSIX wrapper should have been
2889       defined as "const".
2891    4. Changed pcretest to use malloc() for its buffers so that they can be
2892       Electric Fenced for debugging.
2894    5. There were several places in the code where, in UTF-8 mode, PCRE would try
2895       to read one or more bytes before the start of the subject string. Often this
2896       had no effect on PCRE's behaviour, but in some circumstances it could
2897       provoke a segmentation fault.
2899    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2900       to try to read one or more bytes before the start of the subject string.
2902    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
2903       UTF-8 support could misbehave in various ways if the subject string
2904       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
2905       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
2906       back over UTF-8 characters.)
2909    Version 4.2 14-Apr-03
2910    ---------------------
2912    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
2914    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
2915         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
2916         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
2917         [WIN_PREFIX]: new variable, "cyg" for Cygwin
2918         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
2919           and BUILD_EXEEXT
2920         Note: automatic setting of the BUILD variables is not yet working
2921         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
2922           compile-time but not at link-time
2923         [LINK]: use for linking executables only
2924         make different versions for Windows and non-Windows
2925         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
2926           libraries
2927         [LINK_FOR_BUILD]: new variable
2928         [OBJEXT]: use throughout
2929         [EXEEXT]: use throughout
2930         <winshared>: new target
2931         <wininstall>: new target
2932         <dftables.o>: use native compiler
2933         <dftables>: use native linker
2934         <install>: handle Windows platform correctly
2935         <clean>: ditto
2936         <check>: ditto
2937         copy DLL to top builddir before testing
2939       As part of these changes, -no-undefined was removed again. This was reported
2940       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
2941       in any case.
2943    3. Some tidies to get rid of compiler warnings:
2945       . In the match_data structure, match_limit was an unsigned long int, whereas
2946         match_call_count was an int. I've made them both unsigned long ints.
2948       . In pcretest the fact that a const uschar * doesn't automatically cast to
2949         a void * provoked a warning.
2951       . Turning on some more compiler warnings threw up some "shadow" variables
2952         and a few more missing casts.
2954    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
2955       option, a class that contained a single character with a value between 128
2956       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
2958    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
2959       option, a class that contained several characters, but with at least one
2960       whose value was between 128 and 255 caused PCRE to crash.
2963    Version 4.1 12-Mar-03
2964    ---------------------
2966    1. Compiling with gcc -pedantic found a couple of places where casts were
2967    needed, and a string in dftables.c that was longer than standard compilers are
2968    required to support.
2970    2. Compiling with Sun's compiler found a few more places where the code could
2971    be tidied up in order to avoid warnings.
2973    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
2974    first of these names is deprecated in the latest Autoconf in favour of the name
2975    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
2976    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2977    analogy I have changed it to CFLAGS_FOR_BUILD.
2979    4. Added -no-undefined to the linking command in the Makefile, because this is
2980    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2981    linking step for the pcreposix library.
2983    5. PCRE was failing to diagnose the case of two named groups with the same
2984    name.
2986    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2987    literal character that is needed in the subject for a match, and scans along to
2988    ensure that it is present before embarking on the full matching process. This
2989    saves time in cases of nested unlimited repeats that are never going to match.
2990    Problem: the scan can take a lot of time if the subject is very long (e.g.
2991    megabytes), thus penalizing straightforward matches. It is now done only if the
2992    amount of subject to be scanned is less than 1000 bytes.
2994    7. A lesser problem with the same optimization is that it was recording the
2995    first character of an anchored pattern as "needed", thus provoking a search
2996    right along the subject, even when the first match of the pattern was going to
2997    fail. The "needed" character is now not set for anchored patterns, unless it
2998    follows something in the pattern that is of non-fixed length. Thus, it still
2999    fulfils its original purpose of finding quick non-matches in cases of nested
3000    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
3003    Version 4.0 17-Feb-03
3004    ---------------------
3006    1. If a comment in an extended regex that started immediately after a meta-item
3007    extended to the end of string, PCRE compiled incorrect data. This could lead to
3008    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
3010    2. Moved to autoconf 2.53 and libtool 1.4.2.
3012    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
3013    the special perltest8 script is no longer needed - all the tests can be run
3014    from a single perltest script.
3016    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
3017    by \s. It has now been removed in PCRE. This means it isn't recognized as
3018    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
3019    class [:space:] *does* include VT, thereby creating a mess.
3021    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
3022    space and tab.
3024    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
3025    its new features into the main test script, reducing the number of scripts.
3027    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
3028    were backward compatible, and made the (?i) apply to the whole pattern, as if
3029    /i were given. Now it behaves more logically, and applies the option setting
3030    only to what follows. PCRE has been changed to follow suit. However, if it
3031    finds options settings right at the start of the pattern, it extracts them into
3032    the global options, as before. Thus, they show up in the info data.
3034    8. Added support for the \Q...\E escape sequence. Characters in between are
3035    treated as literals. This is slightly different from Perl in that $ and @ are
3036    also handled as literals inside the quotes. In Perl, they will cause variable
3037    interpolation. Note the following examples:
3039        Pattern            PCRE matches      Perl matches
3041        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
3042        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
3043        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
3045    For compatibility with Perl, \Q...\E sequences are recognized inside character
3046    classes as well as outside them.
3048    9. Re-organized 3 code statements in pcretest to avoid "overflow in
3049    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
3050    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
3051    signed/unsigned warnings.
3053    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
3054    option for pcretest, so I've replaced it by a simple function that does just
3055    that job.
3057    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
3058    "pcregrep -".
3060    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
3061    Java package. This provides some syntactic sugar for simple cases of what my
3062    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
3063    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
3064    item, you can use this simplified notation. Note that only makes sense with
3065    greedy quantifiers. Consequently, the use of the possessive quantifier forces
3066    greediness, whatever the setting of the PCRE_UNGREEDY option.
3068    13. A change of greediness default within a pattern was not taking effect at
3069    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
3070    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
3071    was abstracted outside.
3073    14. PCRE now supports the \G assertion. It is true when the current matching
3074    position is at the start point of the match. This differs from \A when the
3075    starting offset is non-zero. Used with the /g option of pcretest (or similar
3076    code), it works in the same way as it does for Perl's /g option. If all
3077    alternatives of a regex begin with \G, the expression is anchored to the start
3078    match position, and the "anchored" flag is set in the compiled expression.
3080    15. Some bugs concerning the handling of certain option changes within patterns
3081    have been fixed. These applied to options other than (?ims). For example,
3082    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
3083    been the other way round. Some of this was related to change 7 above.
3085    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
3086    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
3087    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
3088    POSIX classes only within a class (e.g. /[[:alpha:]]/).
3090    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
3091    mode. Unlike ".", it always matches newline, whatever the setting of
3092    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
3093    assertions. Perl allows it, but it doesn't (in general) work because it can't
3094    calculate the length of the lookbehind. At least, that's the case for Perl
3095    5.8.0 - I've been told they are going to document that it doesn't work in
3096    future.
3098    18. Added an error diagnosis for escapes that PCRE does not support: these are
3099    \L, \l, \N, \P, \p, \U, \u, and \X.
3101    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
3102    reading past the end of the pattern in cases such as /[abcd/.
3104    20. PCRE was getting more memory than necessary for patterns with classes that
3105    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
3107    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
3108    compiling PCRE for use with Virtual Pascal.
3110    22. Small fix to the Makefile to make it work properly if the build is done
3111    outside the source tree.
3113    23. Added a new extension: a condition to go with recursion. If a conditional
3114    subpattern starts with (?(R) the "true" branch is used if recursion has
3115    happened, whereas the "false" branch is used only at the top level.
3117    24. When there was a very long string of literal characters (over 255 bytes
3118    without UTF support, over 250 bytes with UTF support), the computation of how
3119    much memory was required could be incorrect, leading to segfaults or other
3120    strange effects.
3122    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
3123    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
3124    there was a subsequent back reference to those brackets. This meant that, for
3125    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
3126    possible to check for precisely this case. All we can do is abandon the
3127    optimization if .* occurs inside capturing brackets when there are any back
3128    references whatsoever. (See below for a better fix that came later.)
3130    26. The handling of the optimization for finding the first character of a
3131    non-anchored pattern, and for finding a character that is required later in the
3132    match were failing in some cases. This didn't break the matching; it just
3133    failed to optimize when it could. The way this is done has been re-implemented.
3135    27. Fixed typo in error message for invalid (?R item (it said "(?p").
3137    28. Added a new feature that provides some of the functionality that Perl
3138    provides with (?{...}). The facility is termed a "callout". The way it is done
3139    in PCRE is for the caller to provide an optional function, by setting
3140    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
3141    global variable. By default it is unset, which disables all calling out. To get
3142    the function called, the regex must include (?C) at appropriate points. This
3143    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
3144    This provides a means of identifying different callout points. When PCRE
3145    reaches such a point in the regex, if pcre_callout has been set, the external
3146    function is called. It is provided with data in a structure called
3147    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
3148    matching continues; if it returns a non-zero value, the match at the current
3149    point fails. However, backtracking will occur if possible. [This was changed
3150    later and other features added - see item 49 below.]
3152    29. pcretest is upgraded to test the callout functionality. It provides a
3153    callout function that displays information. By default, it shows the start of
3154    the match and the current position in the text. There are some new data escapes
3155    to vary what happens:
3157        \C+         in addition, show current contents of captured substrings
3158        \C-         do not supply a callout function
3159        \C!n        return 1 when callout number n is reached
3160        \C!n!m      return 1 when callout number n is reached for the mth time
3162    30. If pcregrep was called with the -l option and just a single file name, it
3163    output "<stdin>" if a match was found, instead of the file name.
3165    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
3166    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
3167    pcre_exec(). This saves a malloc/free per call. The default value of
3168    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
3169    when configuring.
3171    32. The default maximum size of a compiled pattern is 64K. There have been a
3172    few cases of people hitting this limit. The code now uses macros to handle the
3173    storing of links as offsets within the compiled pattern. It defaults to 2-byte
3174    links, but this can be changed to 3 or 4 bytes by --with-link-size when
3175    configuring. Tests 2 and 5 work only with 2-byte links because they output
3176    debugging information about compiled patterns.
3178    33. Internal code re-arrangements:
3180    (a) Moved the debugging function for printing out a compiled regex into
3181        its own source file (printint.c) and used #include to pull it into
3182        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
3183        separate copies.
3185    (b) Defined the list of op-code names for debugging as a macro in
3186        internal.h so that it is next to the definition of the opcodes.
3188    (c) Defined a table of op-code lengths for simpler skipping along compiled
3189        code. This is again a macro in internal.h so that it is next to the
3190        definition of the opcodes.
3192    34. Added support for recursive calls to individual subpatterns, along the
3193    lines of Robin Houston's patch (but implemented somewhat differently).
3195    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
3196    allow it to read and process whole directories in Win32. This code was
3197    contributed by Lionel Fourquaux; it has not been tested by me.
3199    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
3200    used to name a group. Names consist of alphanumerics and underscores, and must
3201    be unique. Back references use the syntax (?P=name) and recursive calls use
3202    (?P>name) which is a PCRE extension to the Python extension. Groups still have
3203    numbers. The function pcre_fullinfo() can be used after compilation to extract
3204    a name/number map. There are three relevant calls:
3206      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
3207      PCRE_INFO_NAMECOUNT            yields the number of entries
3208      PCRE_INFO_NAMETABLE            yields a pointer to the map.
3210    The map is a vector of fixed-size entries. The size of each entry depends on
3211    the length of the longest name used. The first two bytes of each entry are the
3212    group number, most significant byte first. There follows the corresponding
3213    name, zero terminated. The names are in alphabetical order.
3215    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
3216    case instead of 255. Making it the same both with and without UTF-8 support
3217    means that the same test output works with both.
3219    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
3220    calling malloc() with a zero argument.
3222    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
3223    optimization. I've improved things by keeping a bitmap of backreferences with
3224    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
3225    fact referenced, the optimization can be applied. It is unlikely that a
3226    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
3227    the match to follow \n) will appear inside brackets with a number greater than
3228    31, but if it does, any back reference > 31 suppresses the optimization.
3230    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
3231    of disabling numbered capturing parentheses. Any opening parenthesis that is
3232    not followed by ? behaves as if it were followed by ?: but named parentheses
3233    can still be used for capturing (and they will acquire numbers in the usual
3234    way).
3236    41. Redesigned the return codes from the match() function into yes/no/error so
3237    that errors can be passed back from deep inside the nested calls. A malloc
3238    failure while inside a recursive subpattern call now causes the
3239    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
3241    42. It is now possible to set a limit on the number of times the match()
3242    function is called in a call to pcre_exec(). This facility makes it possible to
3243    limit the amount of recursion and backtracking, though not in a directly
3244    obvious way, because the match() function is used in a number of different
3245    circumstances. The count starts from zero for each position in the subject
3246    string (for non-anchored patterns). The default limit is, for compatibility, a
3247    large number, namely 10 000 000. You can change this in two ways:
3249    (a) When configuring PCRE before making, you can use --with-match-limit=n
3250        to set a default value for the compiled library.
3252    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
3253        a different value is set. See 45 below.
3255    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
3257    43. Added a new function pcre_config(int, void *) to enable run-time extraction
3258    of things that can be changed at compile time. The first argument specifies
3259    what is wanted and the second points to where the information is to be placed.
3260    The current list of available information is:
3262      PCRE_CONFIG_UTF8
3264    The output is an integer that is set to one if UTF-8 support is available;
3265    otherwise it is set to zero.
3269    The output is an integer that it set to the value of the code that is used for
3270    newline. It is either LF (10) or CR (13).
3274    The output is an integer that contains the number of bytes used for internal
3275    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
3279    The output is an integer that contains the threshold above which the POSIX
3280    interface uses malloc() for output vectors. See item 31 above.
3284    The output is an unsigned integer that contains the default limit of the number
3285    of match() calls in a pcre_exec() execution. See 42 above.
3287    44. pcretest has been upgraded by the addition of the -C option. This causes it
3288    to extract all the available output from the new pcre_config() function, and to
3289    output it. The program then exits immediately.
3291    45. A need has arisen to pass over additional data with calls to pcre_exec() in
3292    order to support additional features. One way would have been to define
3293    pcre_exec2() (for example) with extra arguments, but this would not have been
3294    extensible, and would also have required all calls to the original function to
3295    be mapped to the new one. Instead, I have chosen to extend the mechanism that
3296    is used for passing in "extra" data from pcre_study().
3298    The pcre_extra structure is now exposed and defined in pcre.h. It currently
3299    contains the following fields:
3301      flags         a bitmap indicating which of the following fields are set
3302      study_data    opaque data from pcre_study()
3303      match_limit   a way of specifying a limit on match() calls for a specific
3304                      call to pcre_exec()
3305      callout_data  data for callouts (see 49 below)
3307    The flag bits are also defined in pcre.h, and are
3313    The pcre_study() function now returns one of these new pcre_extra blocks, with
3314    the actual study data pointed to by the study_data field, and the
3315    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
3316    before. That is, this change is entirely upwards-compatible and requires no
3317    change to existing code.
3319    If you want to pass in additional data to pcre_exec(), you can either place it
3320    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
3321    block.
3323    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
3324    data string contains the escape sequence \M, pcretest calls pcre_exec() several
3325    times with different match limits, until it finds the minimum value needed for
3326    pcre_exec() to complete. The value is then output. This can be instructive; for
3327    most simple matches the number is quite small, but for pathological cases it
3328    gets very large very quickly.
3330    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
3331    returns the size of the data block pointed to by the study_data field in a
3332    pcre_extra block, that is, the value that was passed as the argument to
3333    pcre_malloc() when PCRE was getting memory in which to place the information
3334    created by pcre_study(). The fourth argument should point to a size_t variable.
3335    pcretest has been extended so that this information is shown after a successful
3336    pcre_study() call when information about the compiled regex is being displayed.
3338    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
3339    because what follows is always an absolute path. (Later: it turns out that this
3340    is more than cosmetic for MinGW, because it doesn't like empty path
3341    components.)
3343    49. Some changes have been made to the callout feature (see 28 above):
3345    (i)  A callout function now has three choices for what it returns:
3347           0  =>  success, carry on matching
3348         > 0  =>  failure at this point, but backtrack if possible
3349         < 0  =>  serious error, return this value from pcre_exec()
3351         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
3352         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
3353         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
3354         use by callout functions. It will never be used by PCRE itself.
3356    (ii) The pcre_extra structure (see 45 above) has a void * field called
3357         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
3358         pcre_callout_block structure has a field of the same name. The contents of
3359         the field passed in the pcre_extra structure are passed to the callout
3360         function in the corresponding field in the callout block. This makes it
3361         easier to use the same callout-containing regex from multiple threads. For
3362         testing, the pcretest program has a new data escape
3364           \C*n        pass the number n (may be negative) as callout_data
3366         If the callout function in pcretest receives a non-zero value as
3367         callout_data, it returns that value.
3369    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
3370    there were some redundant $(CFLAGS) in commands that are now specified as
3371    $(LINK), which already includes $(CFLAGS).
3373    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
3374    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
3375    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
3376    one-byte characters throughout. Note that case-insensitive matching applies
3377    only to characters whose values are less than 256. PCRE doesn't support the
3378    notion of cases for higher-valued characters.
3380    (i)   A character class whose characters are all within 0-255 is handled as
3381          a bit map, a