ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 35 by nigel, Sat Feb 24 21:39:05 2007 UTC revision 723 by ph10, Sat Oct 8 15:55:23 2011 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
4    Version 8.20 23-Sep-2011
5    ------------------------
7    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
8        a POSIX class. After further experiments with Perl, which convinced me that
9        Perl has bugs and confusions, a closing square bracket is no longer allowed
10        in a POSIX name. This bug also affected patterns with classes that started
11        with full stops.
13    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
14        captured substring, but while checking the failing first alternative,
15        substring 1 is temporarily captured. If the output vector supplied to
16        pcre_exec() was not big enough for this capture, the yield of the function
17        was still zero ("insufficient space for captured substrings"). This cannot
18        be totally fixed without adding another stack variable, which seems a lot
19        of expense for a edge case. However, I have improved the situation in cases
20        such as /(a)(b)x|abc/ matched against "abc", where the return code
21        indicates that fewer than the maximum number of slots in the ovector have
22        been set.
24    3.  Related to (2) above: when there are more back references in a pattern than
25        slots in the output vector, pcre_exec() uses temporary memory during
26        matching, and copies in the captures as far as possible afterwards. It was
27        using the entire output vector, but this conflicts with the specification
28        that only 2/3 is used for passing back captured substrings. Now it uses
29        only the first 2/3, for compatibility. This is, of course, another edge
30        case.
32    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
33        main code base, and can be used by building with --enable-jit. When this is
34        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
35        runtime --no-jit option is given.
37    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
38        ovector, the return from the function was zero, implying that there were
39        other matches that did not fit. The correct "exactly full" value is now
40        returned.
42    6.  If a subpattern that was called recursively or as a subroutine contained
43        (*PRUNE) or any other control that caused it to give a non-standard return,
44        invalid errors such as "Error -26 (nested recursion at the same subject
45        position)" or even infinite loops could occur.
47    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
48        computing the minimum length on reaching *ACCEPT, and so ended up with the
49        wrong value of 1 rather than 0. Further investigation indicates that
50        computing a minimum subject length in the presence of *ACCEPT is difficult
51        (think back references, subroutine calls), and so I have changed the code
52        so that no minimum is registered for a pattern that contains *ACCEPT.
54    8.  If (*THEN) was present in the first (true) branch of a conditional group,
55        it was not handled as intended. [But see 16 below.]
57    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
58        Sheri Pierce.
60    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
61        the first byte in a match must be "a".
63    11. Change 17 for 8.13 increased the recursion depth for patterns like
64        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
65        pattern contains any instances of (*THEN). If it does not, the old
66        optimizations are restored. It would be nice to do this on a per-group
67        basis, but at the moment that is not feasible.
69    12. In some environments, the output of pcretest -C is CRLF terminated. This
70        broke RunTest's code that checks for the link size. A single white space
71        after the value is now allowed for.
73    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
74        For "fr", it uses the Windows-specific input and output files.
76    14. If (*THEN) appeared in a group that was called recursively or as a
77        subroutine, it did not work as intended. [But see next item.]
79    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
80        pattern fragments (but not containing any | characters). If A and B are
81        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
82        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
83        D. In other words, Perl considers parentheses that do not contain any |
84        characters to be part of a surrounding alternative, whereas PCRE was
85        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
86        differently. PCRE now behaves in the same way as Perl, except in the case
87        of subroutine/recursion calls such as (?1) which have in any case always
88        been different (but PCRE had them first :-).
90    16. Related to 15 above: Perl does not treat the | in a conditional group as
91        creating alternatives. Such a group is treated in the same way as an
92        ordinary group without any | characters when processing (*THEN). PCRE has
93        been changed to match Perl's behaviour.
95    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
96        RunGrepTest script failed.
98    18. Change 22 for version 13 caused atomic groups to use more stack. This is
99        inevitable for groups that contain captures, but it can lead to a lot of
100        stack use in large patterns. The old behaviour has been restored for atomic
101        groups that do not contain any capturing parentheses.
104    Version 8.13 16-Aug-2011
105    ------------------------
107    1.  The Unicode data tables have been updated to Unicode 6.0.0.
109    2.  Two minor typos in pcre_internal.h have been fixed.
111    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
112        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
113        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
115    4.  There were a number of related bugs in the code for matching backrefences
116        caselessly in UTF-8 mode when codes for the characters concerned were
117        different numbers of bytes. For example, U+023A and U+2C65 are an upper
118        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
119        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
120        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
121        2-byte code at the end of the subject (it thought there wasn't enough data
122        left).
124    5.  Comprehensive information about what went wrong is now returned by
125        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
126        as the output vector has at least 2 elements. The offset of the start of
127        the failing character and a reason code are placed in the vector.
129    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
130        now returned is for the first byte of the failing character, instead of the
131        last byte inspected. This is an incompatible change, but I hope it is small
132        enough not to be a problem. It makes the returned offset consistent with
133        pcre_exec() and pcre_dfa_exec().
135    7.  pcretest now gives a text phrase as well as the error number when
136        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
137        failure, the offset and reason code are output.
139    8.  When \R was used with a maximizing quantifier it failed to skip backwards
140        over a \r\n pair if the subsequent match failed. Instead, it just skipped
141        back over a single character (\n). This seems wrong (because it treated the
142        two characters as a single entity when going forwards), conflicts with the
143        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
144        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
145        has been changed.
147    9.  Some internal refactoring has changed the processing so that the handling
148        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
149        time (the PCRE_DOTALL option was changed this way some time ago: version
150        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
151        which was always a bit of a fudge. It also means that there is one less
152        argument for the match() function, which reduces its stack requirements
153        slightly. This change also fixes an incompatibility with Perl: the pattern
154        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
156    10. More internal refactoring has drastically reduced the number of recursive
157        calls to match() for possessively repeated groups such as (abc)++ when
158        using pcre_exec().
160    11. While implementing 10, a number of bugs in the handling of groups were
161        discovered and fixed:
163        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
164        (a|)*(?1) gave a compile-time internal error.
165        ((a|)+)+  did not notice that the outer group could match an empty string.
166        (^a|^)+   was not marked as anchored.
167        (.*a|.*)+ was not marked as matching at start or after a newline.
169    12. Yet more internal refactoring has removed another argument from the match()
170        function. Special calls to this function are now indicated by setting a
171        value in a variable in the "match data" data block.
173    13. Be more explicit in pcre_study() instead of relying on "default" for
174        opcodes that mean there is no starting character; this means that when new
175        ones are added and accidentally left out of pcre_study(), testing should
176        pick them up.
178    14. The -s option of pcretest has been documented for ages as being an old
179        synonym of -m (show memory usage). I have changed it to mean "force study
180        for every regex", that is, assume /S for every regex. This is similar to -i
181        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
182        using it. It makes it easier to run collections of tests with and without
183        study enabled, and thereby test pcre_study() more easily. All the standard
184        tests are now run with and without -s (but some patterns can be marked as
185        "never study" - see 20 below).
187    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
188        restoration of the capturing data to the outer values was not happening
189        correctly.
191    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
192        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
193        pattern had matched an empty string, and so incorrectly returned a no
194        match.
196    17. There was optimizing code for the last branch of non-capturing parentheses,
197        and also for the obeyed branch of a conditional subexpression, which used
198        tail recursion to cut down on stack usage. Unfortunately, now that there is
199        the possibility of (*THEN) occurring in these branches, tail recursion is
200        no longer possible because the return has to be checked for (*THEN). These
201        two optimizations have therefore been removed. [But see 8.20/11 above.]
203    18. If a pattern containing \R was studied, it was assumed that \R always
204        matched two bytes, thus causing the minimum subject length to be
205        incorrectly computed because \R can also match just one byte.
207    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
208        was incorrectly computed.
210    20. If /S is present twice on a test pattern in pcretest input, it now
211        *disables* studying, thereby overriding the use of -s on the command line
212        (see 14 above). This is necessary for one or two tests to keep the output
213        identical in both cases.
215    21. When (*ACCEPT) was used in an assertion that matched an empty string and
216        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
218    22. When an atomic group that contained a capturing parenthesis was
219        successfully matched, but the branch in which it appeared failed, the
220        capturing was not being forgotten if a higher numbered group was later
221        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
222        group 1 to "a", when in fact it should be unset. This applied to multi-
223        branched capturing and non-capturing groups, repeated or not, and also to
224        positive assertions (capturing in negative assertions does not happen
225        in PCRE) and also to nested atomic groups.
227    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
228        subject after a captured substring, to make it easier to tell which of a
229        number of identical substrings has been captured.
231    24. The way atomic groups are processed by pcre_exec() has been changed so that
232        if they are repeated, backtracking one repetition now resets captured
233        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
234        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
235        "aaa". Previously, it would have been "a". As part of this code
236        refactoring, the way recursive calls are handled has also been changed.
238    25. If an assertion condition captured any substrings, they were not passed
239        back unless some other capturing happened later. For example, if
240        (?(?=(a))a) was matched against "a", no capturing was returned.
242    26. When studying a pattern that contained subroutine calls or assertions,
243        the code for finding the minimum length of a possible match was handling
244        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
245        group 1 called group 2 while simultaneously a separate group 2 called group
246        1). A stack overflow occurred in this case. I have fixed this by limiting
247        the recursion depth to 10.
249    27. Updated RunTest.bat in the distribution to the version supplied by Tom
250        Fortmann. This supports explicit test numbers on the command line, and has
251        argument validation and error reporting.
253    28. An instance of \X with an unlimited repeat could fail if at any point the
254        first character it looked at was a mark character.
256    29. Some minor code refactoring concerning Unicode properties and scripts
257        should reduce the stack requirement of match() slightly.
259    30. Added the '=' option to pcretest to check the setting of unused capturing
260        slots at the end of the pattern, which are documented as being -1, but are
261        not included in the return count.
263    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
264        compiled something random. Now it gives a compile-time error (as does
265        Perl).
267    32. A *MARK encountered during the processing of a positive assertion is now
268        recorded and passed back (compatible with Perl).
270    33. If --only-matching or --colour was set on a pcregrep call whose pattern
271        had alternative anchored branches, the search for a second match in a line
272        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
273        matched the line "0102" twice. The same bug affected patterns that started
274        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
275        twice.
277    34. Previously, PCRE did not allow quantification of assertions. However, Perl
278        does, and because of capturing effects, quantifying parenthesized
279        assertions may at times be useful. Quantifiers are now allowed for
280        parenthesized assertions.
282    35. A minor code tidy in pcre_compile() when checking options for \R usage.
284    36. \g was being checked for fancy things in a character class, when it should
285        just be a literal "g".
287    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
288        appearance of a nested POSIX class supersedes an apparent external class.
289        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
290        unescaped square brackets may also appear as part of class names. For
291        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
292        more like Perl. (But see 8.20/1 above.)
294    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
295        was because it thought it was \N{name}, which is not supported).
297    39. Add minix to OS list not supporting the -S option in pcretest.
299    40. PCRE tries to detect cases of infinite recursion at compile time, but it
300        cannot analyze patterns in sufficient detail to catch mutual recursions
301        such as ((?1))((?2)). There is now a runtime test that gives an error if a
302        subgroup is called recursively as a subpattern for a second time at the
303        same position in the subject string. In previous releases this might have
304        been caught by the recursion limit, or it might have run out of stack.
306    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
307        happen only once. PCRE was, however incorrectly giving a compile time error
308        "recursive call could loop indefinitely" because it cannot analyze the
309        pattern in sufficient detail. The compile time test no longer happens when
310        PCRE is compiling a conditional subpattern, but actual runaway loops are
311        now caught at runtime (see 40 above).
313    42. It seems that Perl allows any characters other than a closing parenthesis
314        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
315        has been changed to be the same.
317    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
318        as not to get warnings when autogen.sh is called. Also changed
319        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
321    44. To help people who use pcregrep to scan files containing exceedingly long
322        lines, the following changes have been made:
324        (a) The default value of the buffer size parameter has been increased from
325            8K to 20K. (The actual buffer used is three times this size.)
327        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
328            PCRE is built.
330        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
331            to be set at run time.
333        (d) Numerical values in pcregrep options can be followed by K or M, for
334            example --buffer-size=50K.
336        (e) If a line being scanned overflows pcregrep's buffer, an error is now
337            given and the return code is set to 2.
339    45. Add a pointer to the latest mark to the callout data block.
341    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
342        partial match of an empty string instead of no match. This was specific to
343        the use of ".".
345    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
346        complete match instead of a partial match. This bug was dependent on both
347        the PCRE_UTF8 and PCRE_DOTALL options being set.
349    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
350        starting byte set, because \b was not being ignored.
353    Version 8.12 15-Jan-2011
354    ------------------------
356    1.  Fixed some typos in the markup of the man pages, and wrote a script that
357        checks for such things as part of the documentation building process.
359    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
360        --match-limit and --recursion-limit options (added for 8.11). In
361        particular, this made one of the standard tests fail. (The integer value
362        went into the wrong half of a long int.)
364    3.  If the --colour option was given to pcregrep with -v (invert match), it
365        did strange things, either producing crazy output, or crashing. It should,
366        of course, ignore a request for colour when reporting lines that do not
367        match.
369    4.  Another pcregrep bug caused similar problems if --colour was specified with
370        -M (multiline) and the pattern match finished with a line ending.
372    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
373        matched in multiline mode, the following line was shown as part of the
374        match. This seems wrong, so I have changed it.
376    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
377        the check for further matches in the same line (so they could be coloured)
378        to overrun the end of the current line. If another match was found, it was
379        incorrectly shown (and then shown again when found in the next line).
381    7.  If pcregrep was compiled under Windows, there was a reference to the
382        function pcregrep_exit() before it was defined. I am assuming this was
383        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
384        reported by a user. I've moved the definition above the reference.
387    Version 8.11 10-Dec-2010
388    ------------------------
390    1.  (*THEN) was not working properly if there were untried alternatives prior
391        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
392        backtracked to try for "b" instead of moving to the next alternative branch
393        at the same level (in this case, to look for "c"). The Perl documentation
394        is clear that when (*THEN) is backtracked onto, it goes to the "next
395        alternative in the innermost enclosing group".
397    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
398        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
399        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
400        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
401        (*THEN).
403    3.  If \s appeared in a character class, it removed the VT character from
404        the class, even if it had been included by some previous item, for example
405        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
406        of \s, but is part of the POSIX "space" class.)
408    4.  A partial match never returns an empty string (because you can always
409        match an empty string at the end of the subject); however the checking for
410        an empty string was starting at the "start of match" point. This has been
411        changed to the "earliest inspected character" point, because the returned
412        data for a partial match starts at this character. This means that, for
413        example, /(?<=abc)def/ gives a partial match for the subject "abc"
414        (previously it gave "no match").
416    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
417        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
418        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
419        has an implication that the given string is incomplete (because a partial
420        match is preferred over a full match). For this reason, these items now
421        give a partial match in this situation. [Aside: previously, the one case
422        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
423        match rather than a full match, which was wrong by the old rules, but is
424        now correct.]
426    6.  There was a bug in the handling of #-introduced comments, recognized when
427        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
428        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
429        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
430        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
431        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
432        places in pcre_compile().
434    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
435        comments when looking ahead for named forward references to subpatterns,
436        the only newline sequence it recognized was NL. It now handles newlines
437        according to the set newline convention.
439    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
440        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
441        cater for a lack of strerror(). These oversights have been fixed.
443    9.  Added --match-limit and --recursion-limit to pcregrep.
445    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
447    11. When the -o option was used, pcregrep was setting a return code of 1, even
448        when matches were found, and --line-buffered was not being honoured.
450    12. Added an optional parentheses number to the -o and --only-matching options
451        of pcregrep.
453    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
454        can match an empty string. The code to do it in pcretest and pcredemo
455        needed fixing:
457        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
458            only one byte after an empty string match just before CRLF (this case
459            just got forgotten; "any" and "anycrlf" were OK).
461        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
462            mode when an empty string match preceded an ASCII character followed by
463            a non-ASCII character. (The code for advancing by one character rather
464            than one byte was nonsense.)
466        (c) The pcredemo.c sample program did not have any code at all to handle
467            the cases when CRLF is a valid newline sequence.
469    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
470        as a starting offset was within the subject string. There is now a new
471        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
472        negative or greater than the length of the string. In order to test this,
473        pcretest is extended to allow the setting of negative starting offsets.
475    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
476        starting offset points to the beginning of a UTF-8 character was
477        unnecessarily clumsy. I tidied it up.
479    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
480        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
482    17. Nobody had reported that the --include_dir option, which was added in
483        release 7.7 should have been called --include-dir (hyphen, not underscore)
484        for compatibility with GNU grep. I have changed it to --include-dir, but
485        left --include_dir as an undocumented synonym, and the same for
486        --exclude-dir, though that is not available in GNU grep, at least as of
487        release 2.5.4.
489    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
490        characters from a string of bytes) have been redefined so as not to use
491        loops, in order to improve performance in some environments. At the same
492        time, I abstracted some of the common code into auxiliary macros to save
493        repetition (this should not affect the compiled code).
495    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
496        compile-time error is now given if \c is not followed by an ASCII
497        character, that is, a byte less than 128. (In EBCDIC mode, the code is
498        different, and any byte value is allowed.)
500    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
501        START_OPTIMIZE option, which is now allowed at compile time - but just
502        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
503        to pcregrep and other applications that have no direct access to PCRE
504        options. The new /Y option in pcretest sets this option when calling
505        pcre_compile().
507    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
508        back references. Groups containing recursive back references were forced to
509        be atomic by that change, but in the case of named groups, the amount of
510        memory required was incorrectly computed, leading to "Failed: internal
511        error: code overflow". This has been fixed.
513    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
514        pcretest.c, to avoid build problems in some Borland environments.
517    Version 8.10 25-Jun-2010
518    ------------------------
520    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
521        THEN.
523    2.  (*ACCEPT) was not working when inside an atomic group.
525    3.  Inside a character class, \B is treated as a literal by default, but
526        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
527        causes the error). The code is unchanged, but I tidied the documentation.
529    4.  Inside a character class, PCRE always treated \R and \X as literals,
530        whereas Perl faults them if its -w option is set. I have changed PCRE so
531        that it faults them when PCRE_EXTRA is set.
533    5.  Added support for \N, which always matches any character other than
534        newline. (It is the same as "." when PCRE_DOTALL is not set.)
536    6.  When compiling pcregrep with newer versions of gcc which may have
537        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
538        declared with attribute warn_unused_result" were given. Just casting the
539        result to (void) does not stop the warnings; a more elaborate fudge is
540        needed. I've used a macro to implement this.
542    7.  Minor change to pcretest.c to avoid a compiler warning.
544    8.  Added four artifical Unicode properties to help with an option to make
545        \s etc use properties (see next item). The new properties are: Xan
546        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
548    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
549        use Unicode properties. (*UCP) at the start of a pattern can be used to set
550        this option. Modified pcretest to add /W to test this facility. Added
551        REG_UCP to make it available via the POSIX interface.
553    10. Added --line-buffered to pcregrep.
555    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
556        studied, and the match started with a letter with a code point greater than
557        127 whose first byte was different to the first byte of the other case of
558        the letter, the other case of this starting letter was not recognized
559        (#976).
561    12. If a pattern that was studied started with a repeated Unicode property
562        test, for example, \p{Nd}+, there was the theoretical possibility of
563        setting up an incorrect bitmap of starting bytes, but fortunately it could
564        not have actually happened in practice until change 8 above was made (it
565        added property types that matched character-matching opcodes).
567    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
568        possible starting bytes for non-anchored patterns.
570    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
571        \R, and also a number of cases that involve Unicode properties, both
572        explicit and implicit when PCRE_UCP is set.
574    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
575        input, it could crash or give wrong results if characters with values
576        greater than 0xc0 were present in the subject string. (Detail: it assumed
577        UTF-8 input when processing these items.)
579    16. Added a lot of (int) casts to avoid compiler warnings in systems where
580        size_t is 64-bit (#991).
582    17. Added a check for running out of memory when PCRE is compiled with
583        --disable-stack-for-recursion (#990).
585    18. If the last data line in a file for pcretest does not have a newline on
586        the end, a newline was missing in the output.
588    19. The default pcre_chartables.c file recognizes only ASCII characters (values
589        less than 128) in its various bitmaps. However, there is a facility for
590        generating tables according to the current locale when PCRE is compiled. It
591        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
592        characters, are recognized by isspace() and therefore were getting set in
593        these tables, and indeed these tables seem to approximate to ISO 8859. This
594        caused a problem in UTF-8 mode when pcre_study() was used to create a list
595        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
596        which of course cannot start UTF-8 characters. I have changed the code so
597        that only real ASCII characters (less than 128) and the correct starting
598        bytes for UTF-8 encodings are set for characters greater than 127 when in
599        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
600        altogether.)
602    20. Added the /T option to pcretest so as to be able to run tests with non-
603        standard character tables, thus making it possible to include the tests
604        used for 19 above in the standard set of tests.
606    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
607        reference to a subpattern the other side of a comment that contains an
608        opening parenthesis caused either an internal compiling error, or a
609        reference to the wrong subpattern.
612    Version 8.02 19-Mar-2010
613    ------------------------
615    1.  The Unicode data tables have been updated to Unicode 5.2.0.
617    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
618        configured.
620    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
621        original author of that file, following a query about its status.
623    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
624        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
626    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
627        quantifier applied to a forward-referencing subroutine call, could compile
628        incorrect code or give the error "internal error: previously-checked
629        referenced subpattern not found".
631    6.  Both MS Visual Studio and Symbian OS have problems with initializing
632        variables to point to external functions. For these systems, therefore,
633        pcre_malloc etc. are now initialized to local functions that call the
634        relevant global functions.
636    7.  There were two entries missing in the vectors called coptable and poptable
637        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
638        I've fixed the data, and added a kludgy way of testing at compile time that
639        the lengths are correct (equal to the number of opcodes).
641    8.  Following on from 7, I added a similar kludge to check the length of the
642        eint vector in pcreposix.c.
644    9.  Error texts for pcre_compile() are held as one long string to avoid too
645        much relocation at load time. To find a text, the string is searched,
646        counting zeros. There was no check for running off the end of the string,
647        which could happen if a new error number was added without updating the
648        string.
650    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
652    11. \K was not working if it appeared in an atomic group or in a group that
653        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
654        \K is "not well defined" if used in an assertion. PCRE now accepts it if
655        the assertion is positive, but not if it is negative.
657    12. Change 11 fortuitously reduced the size of the stack frame used in the
658        "match()" function of pcre_exec.c by one pointer. Forthcoming
659        implementation of support for (*MARK) will need an extra pointer on the
660        stack; I have reserved it now, so that the stack frame size does not
661        decrease.
663    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
664        item in branch that calls a recursion is a subroutine call - as in the
665        second branch in the above example - was incorrectly given the compile-
666        time error "recursive call could loop indefinitely" because pcre_compile()
667        was not correctly checking the subroutine for matching a non-empty string.
669    14. The checks for overrunning compiling workspace could trigger after an
670        overrun had occurred. This is a "should never occur" error, but it can be
671        triggered by pathological patterns such as hundreds of nested parentheses.
672        The checks now trigger 100 bytes before the end of the workspace.
674    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
677    Version 8.01 19-Jan-2010
678    ------------------------
680    1.  If a pattern contained a conditional subpattern with only one branch (in
681        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
682        computed the wrong minimum data length (which is of course zero for such
683        subpatterns). This could cause incorrect "no match" results.
685    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
686        the pattern is reset in the first branch, pcre_compile() failed with
687        "internal error: code overflow at offset...". This happened only when
688        the reset was to the original external option setting. (An optimization
689        abstracts leading options settings into an external setting, which was the
690        cause of this.)
692    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
693        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
694        assertion pattern did not match (meaning that the assertion was true), it
695        was incorrectly treated as false if the SKIP had been reached during the
696        matching. This also applied to assertions used as conditions.
698    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
699        assertion subpattern, including such a pattern used as a condition,
700        unpredictable results occurred, instead of the error return
703    5.  The C++ GlobalReplace function was not working like Perl for the special
704        situation when an empty string is matched. It now does the fancy magic
705        stuff that is necessary.
707    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
708        removed. (These were left over from very, very early versions of PCRE.)
710    7.  Some cosmetic changes to the code to make life easier when compiling it
711        as part of something else:
713        (a) Change DEBUG to PCRE_DEBUG.
715        (b) In pcre_compile(), rename the member of the "branch_chain" structure
716            called "current" as "current_branch", to prevent a collision with the
717            Linux macro when compiled as a kernel module.
719        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
720            prevent a collision with the Linux macro when compiled as a kernel
721            module.
723    8.  In pcre_compile() there are some checks for integer overflows that used to
724        cast potentially large values to (double). This has been changed to that
725        when building, a check for int64_t is made, and if it is found, it is used
726        instead, thus avoiding the use of floating point arithmetic. (There is no
727        other use of FP in PCRE.) If int64_t is not found, the fallback is to
728        double.
730    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
731        2005 (difference between two addresses compared to an unsigned value).
733    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
734        custom one, because of the following reported problem in Windows:
736          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
737              under Win32.
738          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
739              therefore missing the function definition.
740          - The compiler thus generates a "C" signature for the test function.
741          - The linker fails to find the "C" function.
742          - PCRE fails to configure if asked to do so against libbz2.
744    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
745        messages were output:
747          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
748          rerunning libtoolize, to keep the correct libtool macros in-tree.
749          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
751        I have done both of these things.
753    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
754        most of the time, it *can* run out if it is given a pattern that contains a
755        runaway infinite recursion. I updated the discussion in the pcrestack man
756        page.
758    13. Now that we have gone to the x.xx style of version numbers, the minor
759        version may start with zero. Using 08 or 09 is a bad idea because users
760        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
761        interpreted as invalid octal numbers. I've updated the previous comment in
762        configure.ac, and also added a check that gives an error if 08 or 09 are
763        used.
765    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
766        causing partial matching to fail when the end of the subject matched \W
767        in a UTF-8 pattern where \W was quantified with a minimum of 3.
769    15. There were some discrepancies between the declarations in pcre_internal.h
770        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
771        their definitions. The declarations used "const uschar *" and the
772        definitions used USPTR. Even though USPTR is normally defined as "const
773        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
774        reported that: "This difference in casting confuses some C++ compilers, for
775        example, SunCC recognizes above declarations as different functions and
776        generates broken code for hbpcre." I have changed the declarations to use
777        USPTR.
779    16. GNU libtool is named differently on some systems. The autogen.sh script now
780        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
781        (FreeBSD).
783    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
784        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
785        comment: "Figure out how to create a longlong from a string: strtoll and
786        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
787        instance, but it only takes 2 args instead of 3!"
789    18. A subtle bug concerned with back references has been fixed by a change of
790        specification, with a corresponding code fix. A pattern such as
791        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
792        refers, was giving matches when it shouldn't. For example, xa=xaaa would
793        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
794        same bug. Such groups have to be quantified to be useful, or contained
795        inside another quantified group. (If there's no repetition, the reference
796        can never match.) The problem arises because, having left the group and
797        moved on to the rest of the pattern, a later failure that backtracks into
798        the group uses the captured value from the final iteration of the group
799        rather than the correct earlier one. I have fixed this in PCRE by forcing
800        any group that contains a reference to itself to be an atomic group; that
801        is, there cannot be any backtracking into it once it has completed. This is
802        similar to recursive and subroutine calls.
805    Version 8.00 19-Oct-09
806    ----------------------
808    1.  The table for translating pcre_compile() error codes into POSIX error codes
809        was out-of-date, and there was no check on the pcre_compile() error code
810        being within the table. This could lead to an OK return being given in
811        error.
813    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
814        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
815        in a Windows environment.
817    3.  The pcregrep --count option prints the count for each file even when it is
818        zero, as does GNU grep. However, pcregrep was also printing all files when
819        --files-with-matches was added. Now, when both options are given, it prints
820        counts only for those files that have at least one match. (GNU grep just
821        prints the file name in this circumstance, but including the count seems
822        more useful - otherwise, why use --count?) Also ensured that the
823        combination -clh just lists non-zero counts, with no names.
825    4.  The long form of the pcregrep -F option was incorrectly implemented as
826        --fixed_strings instead of --fixed-strings. This is an incompatible change,
827        but it seems right to fix it, and I didn't think it was worth preserving
828        the old behaviour.
830    5.  The command line items --regex=pattern and --regexp=pattern were not
831        recognized by pcregrep, which required --regex pattern or --regexp pattern
832        (with a space rather than an '='). The man page documented the '=' forms,
833        which are compatible with GNU grep; these now work.
835    6.  No libpcreposix.pc file was created for pkg-config; there was just
836        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
838    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
839        when UCP support is not needed, by modifying the Python script that
840        generates it from Unicode data files. This should not matter if the module
841        is correctly used as a library, but I received one complaint about 50K of
842        unwanted data. My guess is that the person linked everything into his
843        program rather than using a library. Anyway, it does no harm.
845    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
846        was a minimum greater than 1 for a wide character in a possessive
847        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
848        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
849        character. Chaos in the form of incorrect output or a compiling loop could
850        result.
852    9.  The restrictions on what a pattern can contain when partial matching is
853        requested for pcre_exec() have been removed. All patterns can now be
854        partially matched by this function. In addition, if there are at least two
855        slots in the offset vector, the offset of the earliest inspected character
856        for the match and the offset of the end of the subject are set in them when
857        PCRE_ERROR_PARTIAL is returned.
859    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
860        synonymous with PCRE_PARTIAL, for backwards compatibility, and
861        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
862        and may be more useful for multi-segment matching.
864    11. Partial matching with pcre_exec() is now more intuitive. A partial match
865        used to be given if ever the end of the subject was reached; now it is
866        given only if matching could not proceed because another character was
867        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
868        string "Z", which now yields "no match" instead of "partial match". In the
869        case of pcre_dfa_exec(), "no match" is given if every matching path for the
870        final character ended with (*FAIL).
872    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
873        if the pattern had a "must contain" character that was already found in the
874        earlier partial match, unless partial matching was again requested. For
875        example, with the pattern /dog.(body)?/, the "must contain" character is
876        "g". If the first part-match was for the string "dog", restarting with
877        "sbody" failed. This bug has been fixed.
879    13. The string returned by pcre_dfa_exec() after a partial match has been
880        changed so that it starts at the first inspected character rather than the
881        first character of the match. This makes a difference only if the pattern
882        starts with a lookbehind assertion or \b or \B (\K is not supported by
883        pcre_dfa_exec()). It's an incompatible change, but it makes the two
884        matching functions compatible, and I think it's the right thing to do.
886    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
887        so that the demonstration program is easily available in environments where
888        PCRE has not been installed from source.
890    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
891        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
892        library.
894    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
895        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
896        is not the first non-POSIX option to be added. Clearly some people find
897        these options useful.
899    17. If a caller to the POSIX matching function regexec() passes a non-zero
900        value for nmatch with a NULL value for pmatch, the value of
901        nmatch is forced to zero.
903    18. RunGrepTest did not have a test for the availability of the -u option of
904        the diff command, as RunTest does. It now checks in the same way as
905        RunTest, and also checks for the -b option.
907    19. If an odd number of negated classes containing just a single character
908        interposed, within parentheses, between a forward reference to a named
909        subpattern and the definition of the subpattern, compilation crashed with
910        an internal error, complaining that it could not find the referenced
911        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
912        [The bug was that it was starting one character too far in when skipping
913        over the character class, thus treating the ] as data rather than
914        terminating the class. This meant it could skip too much.]
916    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
917        /g option in pcretest when the pattern contains \K, which makes it possible
918        to have an empty string match not at the start, even when the pattern is
919        anchored. Updated pcretest and pcredemo to use this option.
921    21. If the maximum number of capturing subpatterns in a recursion was greater
922        than the maximum at the outer level, the higher number was returned, but
923        with unset values at the outer level. The correct (outer level) value is
924        now given.
926    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
927        PCRE did not set those parentheses (unlike Perl). I have now found a way to
928        make it do so. The string so far is captured, making this feature
929        compatible with Perl.
931    23. The tests have been re-organized, adding tests 11 and 12, to make it
932        possible to check the Perl 5.10 features against Perl 5.10.
934    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
935        pattern matches a fixed length string. PCRE did not allow this; now it
936        does. Neither allows recursion.
938    25. I finally figured out how to implement a request to provide the minimum
939        length of subject string that was needed in order to match a given pattern.
940        (It was back references and recursion that I had previously got hung up
941        on.) This code has now been added to pcre_study(); it finds a lower bound
942        to the length of subject needed. It is not necessarily the greatest lower
943        bound, but using it to avoid searching strings that are too short does give
944        some useful speed-ups. The value is available to calling programs via
945        pcre_fullinfo().
947    26. While implementing 25, I discovered to my embarrassment that pcretest had
948        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
949        study optimizations had never been tested with that matching function.
950        Oops. What is worse, even when it was passed study data, there was a bug in
951        pcre_dfa_exec() that meant it never actually used it. Double oops. There
952        were also very few tests of studied patterns with pcre_dfa_exec().
954    27. If (?| is used to create subpatterns with duplicate numbers, they are now
955        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
956        on the other side of the coin, they are no longer allowed to have different
957        names, because these cannot be distinguished in PCRE, and this has caused
958        confusion. (This is a difference from Perl.)
960    28. When duplicate subpattern names are present (necessarily with different
961        numbers, as required by 27 above), and a test is made by name in a
962        conditional pattern, either for a subpattern having been matched, or for
963        recursion in such a pattern, all the associated numbered subpatterns are
964        tested, and the overall condition is true if the condition is true for any
965        one of them. This is the way Perl works, and is also more like the way
966        testing by number works.
969    Version 7.9 11-Apr-09
970    ---------------------
972    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
973        (pcretest), all targets were linked against these libraries. This included
974        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
975        libraries. This caused unwanted dependencies to be created. This problem
976        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
977        pcretest is linked with readline.
979    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
980        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
981        moved outside it again, because FALSE and TRUE are already defined in AIX,
982        but BOOL is not.
984    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
985        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
987    4.  The pcregrep documentation said spaces were inserted as well as colons (or
988        hyphens) following file names and line numbers when outputting matching
989        lines. This is not true; no spaces are inserted. I have also clarified the
990        wording for the --colour (or --color) option.
992    5.  In pcregrep, when --colour was used with -o, the list of matching strings
993        was not coloured; this is different to GNU grep, so I have changed it to be
994        the same.
996    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
997        each matching line was coloured. Now it goes on to look for further matches
998        of any of the test patterns, which is the same behaviour as GNU grep.
1000    7.  A pattern that could match an empty string could cause pcregrep to loop; it
1001        doesn't make sense to accept an empty string match in pcregrep, so I have
1002        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
1003        seems to be how GNU grep behaves.
1005    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
1006        start or after a newline", because the conditional assertion was not being
1007        correctly handled. The rule now is that both the assertion and what follows
1008        in the first alternative must satisfy the test.
1010    9.  If auto-callout was enabled in a pattern with a conditional group whose
1011        condition was an assertion, PCRE could crash during matching, both with
1012        pcre_exec() and pcre_dfa_exec().
1014    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
1015        used for matching.
1017    11. Unicode property support in character classes was not working for
1018        characters (bytes) greater than 127 when not in UTF-8 mode.
1020    12. Added the -M command line option to pcretest.
1022    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
1024    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
1026    16. Added comments and documentation about mis-use of no_arg in the C++
1027        wrapper.
1029    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
1030        from Martin Jerabek that uses macro names for all relevant character and
1031        string constants.
1033    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
1034        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
1035        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
1036        these, but not everybody uses configure.
1038    19. A conditional group that had only one branch was not being correctly
1039        recognized as an item that could match an empty string. This meant that an
1040        enclosing group might also not be so recognized, causing infinite looping
1041        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
1042        with the subject "ab", where knowledge that the repeated group can match
1043        nothing is needed in order to break the loop.
1045    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
1046        exec(), but without supplying a callout function, matching went wrong.
1048    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
1049        leak if the size of the offset vector was greater than 30. When the vector
1050        is smaller, the saved offsets during recursion go onto a local stack
1051        vector, but for larger vectors malloc() is used. It was failing to free
1052        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
1053        error, in fact).
1055    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
1056        heapframe that is used only when UTF-8 support is enabled. This caused no
1057        problem, but was untidy.
1059    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
1060        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
1061        included within another project.
1063    24. Steven Van Ingelgem's patches to add more options to the CMake support,
1064        slightly modified by me:
1066          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
1067              not building pcregrep.
1069          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
1070              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
1072    25. Forward references, both numeric and by name, in patterns that made use of
1073        duplicate group numbers, could behave incorrectly or give incorrect errors,
1074        because when scanning forward to find the reference group, PCRE was not
1075        taking into account the duplicate group numbers. A pattern such as
1076        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
1078    26. Changed a few more instances of "const unsigned char *" to USPTR, making
1079        the feature of a custom pointer more persuasive (as requested by a user).
1081    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
1082        pcretest.c, inside #ifndefs, because it seems they are sometimes already
1083        pre-defined.
1085    28. Added support for (*UTF8) at the start of a pattern.
1087    29. Arrange for flags added by the "release type" setting in CMake to be shown
1088        in the configuration summary.
1091    Version 7.8 05-Sep-08
1092    ---------------------
1094    1.  Replaced UCP searching code with optimized version as implemented for Ad
1095        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
1096        stage table and inline lookup instead of a function, giving speed ups of 2
1097        to 5 times on some simple patterns that I tested. Permission was given to
1098        distribute the MultiStage2.py script that generates the tables (it's not in
1099        the tarball, but is in the Subversion repository).
1101    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
1102        scripts.
1104    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
1105        a group with a zero qualifier. The result of the study could be incorrect,
1106        or the function might crash, depending on the pattern.
1108    4.  Caseless matching was not working for non-ASCII characters in back
1109        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
1110        It now works when Unicode Property Support is available.
1112    5.  In pcretest, an escape such as \x{de} in the data was always generating
1113        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
1114        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
1115        truncation.
1117    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
1119    7.  Added two (int) casts to pcregrep when printing the difference of two
1120        pointers, in case they are 64-bit values.
1122    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
1123        test 2 if it fails.
1125    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
1126        and a #define of that name to empty if it is not externally set. This is to
1127        allow users of MSVC to set it if necessary.
1129    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
1130        the convenience functions in the pcre_get.c source file.
1132    11. An option change at the start of a pattern that had top-level alternatives
1133        could cause overwriting and/or a crash. This command provoked a crash in
1134        some environments:
1136          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
1138        This potential security problem was recorded as CVE-2008-2371.
1140    12. For a pattern where the match had to start at the beginning or immediately
1141        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
1142        pcre_dfa_exec() could read past the end of the passed subject if there was
1143        no match. To help with detecting such bugs (e.g. with valgrind), I modified
1144        pcretest so that it places the subject at the end of its malloc-ed buffer.
1146    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
1147        exec() might read past the end of the data buffer in UTF-8 mode.
1149    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
1150        the data contained the byte 0x85 as part of a UTF-8 character within its
1151        first line. This applied both to normal and DFA matching.
1153    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
1154        /^[^d]*?$/8 failed to match "abc".
1156    16. Added a missing copyright notice to pcrecpp_internal.h.
1158    17. Make it more clear in the documentation that values returned from
1159        pcre_exec() in ovector are byte offsets, not character counts.
1161    18. Tidied a few places to stop certain compilers from issuing warnings.
1163    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
1164        supplied by Stefan Weber. I made a further small update for 7.8 because
1165        there is a change of source arrangements: the pcre_searchfuncs.c module is
1166        replaced by pcre_ucd.c.
1169    Version 7.7 07-May-08
1170    ---------------------
1172    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
1173        a string to a long long, pretend we don't even have a long long." This is
1174        done by checking for the strtoq, strtoll, and _strtoi64 functions.
1176    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
1177        pre-7.6 versions, which defined a global no_arg variable instead of putting
1178        it in the RE class. (See also #8 below.)
1180    3.  Remove a line of dead code, identified by coverity and reported by Nuno
1181        Lopes.
1183    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
1185        (1) The include/exclude patterns were being applied to the whole pathnames
1186            of files, instead of just to the final components.
1188        (2) If there was more than one level of directory, the subdirectories were
1189            skipped unless they satisfied the include/exclude conditions. This is
1190            inconsistent with GNU grep (and could even be seen as contrary to the
1191            pcregrep specification - which I improved to make it absolutely clear).
1192            The action now is always to scan all levels of directory, and just
1193            apply the include/exclude patterns to regular files.
1195    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
1196        --exclude_dir in the tests to avoid scanning .svn directories.
1198    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
1199        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
1200        doesn't support NULs in patterns.
1202    7.  Added some missing "const"s to declarations of static tables in
1203        pcre_compile.c and pcre_dfa_exec.c.
1205    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
1206        caused by fix #2  above. (Subsequently also a second patch to fix the
1207        first patch. And a third patch - this was a messy problem.)
1209    9.  Applied Craig's patch to remove the use of push_back().
1211    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
1212        matching function regexec().
1214    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
1215        which, however, unlike Perl's \g{...}, are subroutine calls, not back
1216        references. PCRE supports relative numbers with this syntax (I don't think
1217        Oniguruma does).
1219    12. Previously, a group with a zero repeat such as (...){0} was completely
1220        omitted from the compiled regex. However, this means that if the group
1221        was called as a subroutine from elsewhere in the pattern, things went wrong
1222        (an internal error was given). Such groups are now left in the compiled
1223        pattern, with a new opcode that causes them to be skipped at execution
1224        time.
1226    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
1227        to the way PCRE behaves:
1229        (a) A lone ] character is dis-allowed (Perl treats it as data).
1231        (b) A back reference to an unmatched subpattern matches an empty string
1232            (Perl fails the current match path).
1234        (c) A data ] in a character class must be notated as \] because if the
1235            first data character in a class is ], it defines an empty class. (In
1236            Perl it is not possible to have an empty class.) The empty class []
1237            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
1238            The negative empty class [^] matches any one character, independently
1239            of the DOTALL setting.
1241    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
1242        non-existent subpattern following a character class starting with ']' and
1243        containing () gave an internal compiling error instead of "reference to
1244        non-existent subpattern". Fortunately, when the pattern did exist, the
1245        compiled code was correct. (When scanning forwards to check for the
1246        existencd of the subpattern, it was treating the data ']' as terminating
1247        the class, so got the count wrong. When actually compiling, the reference
1248        was subsequently set up correctly.)
1250    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
1251        it was being rejected as not supported by pcre_dfa_exec(), even though
1252        other assertions are supported. I have made pcre_dfa_exec() support
1253        (*FAIL).
1255    16. The implementation of 13c above involved the invention of a new opcode,
1256        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
1257        cannot be changed at match time, I realized I could make a small
1258        improvement to matching performance by compiling OP_ALLANY instead of
1259        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
1260        on the OP_ANY path.
1262    17. Compiling pcretest on Windows with readline support failed without the
1263        following two fixes: (1) Make the unistd.h include conditional on
1264        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
1266    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
1267        ncurses library to be included for pcretest when ReadLine support is
1268        requested, but also to allow for it to be overridden. This patch came from
1269        Daniel Bergström.
1271    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
1272        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
1273        any errors with the current Unicode tables. Thanks to Peter Kankowski for
1274        spotting this.
1277    Version 7.6 28-Jan-08
1278    ---------------------
1280    1.  A character class containing a very large number of characters with
1281        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
1282        overflow.
1284    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
1285        HAVE_LONG_LONG is not defined.
1287    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
1288        bring them up to date and include new features. This patch includes:
1290        - Fixed PH's badly added libz and libbz2 support.
1291        - Fixed a problem with static linking.
1292        - Added pcredemo. [But later removed - see 7 below.]
1293        - Fixed dftables problem and added an option.
1294        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
1295            HAVE_LONG_LONG.
1296        - Added readline support for pcretest.
1297        - Added an listing of the option settings after cmake has run.
1299    4.  A user submitted a patch to Makefile that makes it easy to create
1300        "pcre.dll" under mingw when using Configure/Make. I added stuff to
1301        Makefile.am that cause it to include this special target, without
1302        affecting anything else. Note that the same mingw target plus all
1303        the other distribution libraries and programs are now supported
1304        when configuring with CMake (see 6 below) instead of with
1305        Configure/Make.
1307    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
1308        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
1309        exported in the Windows port". It has not yet been confirmed that the patch
1310        solves the problem, but it does no harm.
1312    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
1313        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
1314        with CMake, and also correct the comment about stack recursion.
1316    7.  Remove the automatic building of pcredemo from the ./configure system and
1317        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
1318        of a program that users should build themselves after PCRE is installed, so
1319        building it automatically is not really right. What is more, it gave
1320        trouble in some build environments.
1322    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
1325    Version 7.5 10-Jan-08
1326    ---------------------
1328    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
1329        values in parens when parsing an RE using the C++ wrapper."
1331    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
1332        Characters greater than 255 were excluded from the class instead of being
1333        included.
1335    3.  The same bug as (2) above applied to negated POSIX classes such as
1336        [:^space:].
1338    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
1339        defined or documented. It seems to have been a typo for PCRE_STATIC, so
1340        I have changed it.
1342    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
1343        first named subpattern) and a construct such as (?&a) would reference the
1344        first named subpattern whose name started with "a" (in other words, the
1345        length check was missing). Both these problems are fixed. "Subpattern name
1346        expected" is now given for (?&) (a zero-length name), and this patch also
1347        makes it give the same error for \k'' (previously it complained that that
1348        was a reference to a non-existent subpattern).
1350    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
1351        this is right because (?- can be followed by option settings as well as by
1352        digits. I have, however, made the messages clearer.
1354    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
1355        than the number used in the conditional) now cause a compile-time error.
1356        This is actually not compatible with Perl, which accepts such patterns, but
1357        treats the conditional as always being FALSE (as PCRE used to), but it
1358        seems to me that giving a diagnostic is better.
1360    8.  Change "alphameric" to the more common word "alphanumeric" in comments
1361        and messages.
1363    9.  Fix two occurrences of "backslash" in comments that should have been
1364        "backspace".
1366    10. Remove two redundant lines of code that can never be obeyed (their function
1367        was moved elsewhere).
1369    11. The program that makes PCRE's Unicode character property table had a bug
1370        which caused it to generate incorrect table entries for sequences of
1371        characters that have the same character type, but are in different scripts.
1372        It amalgamated them into a single range, with the script of the first of
1373        them. In other words, some characters were in the wrong script. There were
1374        thirteen such cases, affecting characters in the following ranges:
1376          U+002b0 - U+002c1
1377          U+0060c - U+0060d
1378          U+0061e - U+00612
1379          U+0064b - U+0065e
1380          U+0074d - U+0076d
1381          U+01800 - U+01805
1382          U+01d00 - U+01d77
1383          U+01d9b - U+01dbf
1384          U+0200b - U+0200f
1385          U+030fc - U+030fe
1386          U+03260 - U+0327f
1387          U+0fb46 - U+0fbb1
1388          U+10450 - U+1049d
1390    12. The -o option (show only the matching part of a line) for pcregrep was not
1391        compatible with GNU grep in that, if there was more than one match in a
1392        line, it showed only the first of them. It now behaves in the same way as
1393        GNU grep.
1395    13. If the -o and -v options were combined for pcregrep, it printed a blank
1396        line for every non-matching line. GNU grep prints nothing, and pcregrep now
1397        does the same. The return code can be used to tell if there were any
1398        non-matching lines.
1400    14. Added --file-offsets and --line-offsets to pcregrep.
1402    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
1403        infinitely looping recursion. The bug was that positive lookaheads were not
1404        being skipped when checking for a possible empty match (negative lookaheads
1405        and both kinds of lookbehind were skipped).
1407    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
1408        inclusion of <windows.h> to before rather than after the definition of
1409        INVALID_FILE_ATTRIBUTES (patch from David Byron).
1411    17. Specifying a possessive quantifier with a specific limit for a Unicode
1412        character property caused pcre_compile() to compile bad code, which led at
1413        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
1414        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
1415        caused the error; without that there was no problem.
1417    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
1419    19. Added --enable-pcretest-libreadline.
1421    20. In pcrecpp.cc, the variable 'count' was incremented twice in
1422        RE::GlobalReplace(). As a result, the number of replacements returned was
1423        double what it should be. I removed one of the increments, but Craig sent a
1424        later patch that removed the other one (the right fix) and added unit tests
1425        that check the return values (which was not done before).
1427    21. Several CMake things:
1429        (1) Arranged that, when cmake is used on Unix, the libraries end up with
1430            the names libpcre and libpcreposix, not just pcre and pcreposix.
1432        (2) The above change means that pcretest and pcregrep are now correctly
1433            linked with the newly-built libraries, not previously installed ones.
1437    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
1438        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
1439        UTF-8 newline character). The key issue is that the pattern starts .*;
1440        this means that the match must be either at the beginning, or after a
1441        newline. The bug was in the code for advancing after a failed match and
1442        checking that the new position followed a newline. It was not taking
1443        account of UTF-8 characters correctly.
1445    23. PCRE was behaving differently from Perl in the way it recognized POSIX
1446        character classes. PCRE was not treating the sequence [:...:] as a
1447        character class unless the ... were all letters. Perl, however, seems to
1448        allow any characters between [: and :], though of course it rejects as
1449        unknown any "names" that contain non-letters, because all the known class
1450        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
1451        for example, whereas PCRE did not - it did not recognize a POSIX character
1452        class. This seemed a bit dangerous, so the code has been changed to be
1453        closer to Perl. The behaviour is not identical to Perl, because PCRE will
1454        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
1455        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
1456        Perl does, and where it didn't before.
1458    24. Rewrite so as to remove the single use of %n from pcregrep because in some
1459        Windows environments %n is disabled by default.
1462    Version 7.4 21-Sep-07
1463    ---------------------
1465    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
1466        means that a class such as [\s] counted as "explicit reference to CR or
1467        LF". That isn't really right - the whole point of the change was to try to
1468        help when there was an actual mention of one of the two characters. So now
1469        the change happens only if \r or \n (or a literal CR or LF) character is
1470        encountered.
1472    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
1473        of both had grown to the point where there were only 3 bits left.
1474        Fortunately, there was spare space in the data structure, and so I have
1475        moved the internal flags into a new 16-bit field to free up more option
1476        bits.
1478    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
1479        but did not set the internal JCHANGED flag - either of these is enough to
1480        control the way the "get" function works - but the PCRE_INFO_JCHANGED
1481        facility is supposed to tell if (?J) was ever used, so now (?J) at the
1482        start sets both bits.
1484    4.  Added options (at build time, compile time, exec time) to change \R from
1485        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
1487    5.  doc/pcresyntax.html was missing from the distribution.
1489    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
1490        compatibility, even though it is no longer used.
1492    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
1493        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
1494        windows.h file is present (where different names are used). [This was
1495        reversed later after testing - see 16 below.]
1497    8.  Changed all #include <config.h> to #include "config.h". There were also
1498        some further <pcre.h> cases that I changed to "pcre.h".
1500    9.  When pcregrep was used with the --colour option, it missed the line ending
1501        sequence off the lines that it output.
1503    10. It was pointed out to me that arrays of string pointers cause lots of
1504        relocations when a shared library is dynamically loaded. A technique of
1505        using a single long string with a table of offsets can drastically reduce
1506        these. I have refactored PCRE in four places to do this. The result is
1507        dramatic:
1509          Originally:                          290
1510          After changing UCP table:            187
1511          After changing error message table:   43
1512          After changing table of "verbs"       36
1513          After changing table of Posix names   22
1515        Thanks to the folks working on Gregex for glib for this insight.
1517    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
1518        unicode-properties was also set.
1520    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
1522    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
1523        checked only for CRLF.
1525    14. Added casts to pcretest.c to avoid compiler warnings.
1527    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
1529    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
1530        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
1531        entirely. This removes changes made in 7 above.
1533    17. The CMake files have been updated, and there is now more information about
1534        building with CMake in the NON-UNIX-USE document.
1537    Version 7.3 28-Aug-07
1538    ---------------------
1540     1. In the rejigging of the build system that eventually resulted in 7.1, the
1541        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
1542        brackets there is not right, since it causes compilers to look for an
1543        installed pcre.h, not the version that is in the source that is being
1544        compiled (which of course may be different). I have changed it back to:
1546          #include "pcre.h"
1548        I have a vague recollection that the change was concerned with compiling in
1549        different directories, but in the new build system, that is taken care of
1550        by the VPATH setting the Makefile.
1552     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
1553        when the subject happened to end in the byte 0x85 (e.g. if the last
1554        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
1555        characters but of course it shouldn't be taken as a newline when it is part
1556        of another character. The bug was that, for an unlimited repeat of . in
1557        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
1558        characters when looking for a newline.
1560     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
1562     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
1563        in debug output.
1565     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
1566        long printing in the pcrecpp unittest when running under MinGW.
1568     6. ESC_K was left out of the EBCDIC table.
1570     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
1571        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
1572        limit also applies to "virtual nesting" when a pattern is recursive, and in
1573        this case 1000 isn't so big. I have been able to remove this limit at the
1574        expense of backing off one optimization in certain circumstances. Normally,
1575        when pcre_exec() would call its internal match() function recursively and
1576        immediately return the result unconditionally, it uses a "tail recursion"
1577        feature to save stack. However, when a subpattern that can match an empty
1578        string has an unlimited repetition quantifier, it no longer makes this
1579        optimization. That gives it a stack frame in which to save the data for
1580        checking that an empty string has been matched. Previously this was taken
1581        from the 1000-entry workspace that had been reserved. So now there is no
1582        explicit limit, but more stack is used.
1584     8. Applied Daniel's patches to solve problems with the import/export magic
1585        syntax that is required for Windows, and which was going wrong for the
1586        pcreposix and pcrecpp parts of the library. These were overlooked when this
1587        problem was solved for the main library.
1589     9. There were some crude static tests to avoid integer overflow when computing
1590        the size of patterns that contain repeated groups with explicit upper
1591        limits. As the maximum quantifier is 65535, the maximum group length was
1592        set at 30,000 so that the product of these two numbers did not overflow a
1593        32-bit integer. However, it turns out that people want to use groups that
1594        are longer than 30,000 bytes (though not repeat them that many times).
1595        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1596        made it possible to implement the integer overflow checks in a much more
1597        dynamic way, which I have now done. The artificial limitation on group
1598        length has been removed - we now have only the limit on the total length of
1599        the compiled pattern, which depends on the LINK_SIZE setting.
1601    10. Fixed a bug in the documentation for get/copy named substring when
1602        duplicate names are permitted. If none of the named substrings are set, the
1603        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1604        empty string.
1606    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1607        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1608        because the ] is interpreted as the first data character and the
1609        terminating ] is not found. PCRE has been made compatible with Perl in this
1610        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1611        cause memory overwriting.
1613    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1614        string has been matched (to stop an infinite loop). It was not recognizing
1615        a conditional subpattern that could match an empty string if that
1616        subpattern was within another subpattern. For example, it looped when
1617        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
1618        condition was not nested. This bug has been fixed.
1620    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1621        past the start of the subject in the presence of bytes with the top bit
1622        set, for example "\x8aBCD".
1624    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1625        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1627    14. Optimized (?!) to (*FAIL).
1629    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1630        This restricts code points to be within the range 0 to 0x10FFFF, excluding
1631        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1632        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1633        does: it's just the validity check that is more restrictive.
1635    16. Inserted checks for integer overflows during escape sequence (backslash)
1636        processing, and also fixed erroneous offset values for syntax errors during
1637        backslash processing.
1639    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1640        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1642    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1643        caused an overrun.
1645    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1646        something other than just ASCII characters) inside a group that had an
1647        unlimited repeat caused a loop at compile time (while checking to see
1648        whether the group could match an empty string).
1650    20. Debugging a pattern containing \p or \P could cause a crash. For example,
1651        [\P{Any}] did so. (Error in the code for printing property names.)
1653    21. An orphan \E inside a character class could cause a crash.
1655    22. A repeated capturing bracket such as (A)? could cause a wild memory
1656        reference during compilation.
1658    23. There are several functions in pcre_compile() that scan along a compiled
1659        expression for various reasons (e.g. to see if it's fixed length for look
1660        behind). There were bugs in these functions when a repeated \p or \P was
1661        present in the pattern. These operators have additional parameters compared
1662        with \d, etc, and these were not being taken into account when moving along
1663        the compiled data. Specifically:
1665        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1666            length.
1668        (b) An item such as \pL+ within a repeated group could cause crashes or
1669            loops.
1671        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1672            "reference to non-existent subpattern" error.
1674        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1676    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1677        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1679    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1681    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1682        character were causing crashes (broken optimization).
1684    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1685        \p or \P) caused a compile-time loop.
1687    28. More problems have arisen in unanchored patterns when CRLF is a valid line
1688        break. For example, the unstudied pattern [\r\n]A does not match the string
1689        "\r\nA" because change 7.0/46 below moves the current point on by two
1690        characters after failing to match at the start. However, the pattern \nA
1691        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
1692        the same is true. There doesn't seem any very clean way out of this, but
1693        what I have chosen to do makes the common cases work: PCRE now takes note
1694        of whether there can be an explicit match for \r or \n anywhere in the
1695        pattern, and if so, 7.0/46 no longer applies. As part of this change,
1696        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1697        pattern has explicit CR or LF references.
1699    29. Added (*CR) etc for changing newline setting at start of pattern.
1702    Version 7.2 19-Jun-07
1703    ---------------------
1705     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
1706        which is apparently normally available under Windows.
1708     2. Re-jig the pcregrep tests with different newline settings in an attempt
1709        to make them independent of the local environment's newline setting.
1711     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
1713     4. Some of the "internals" tests were previously cut out when the link size
1714        was not 2, because the output contained actual offsets. The recent new
1715        "Z" feature of pcretest means that these can be cut out, making the tests
1716        usable with all link sizes.
1718     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
1719        stack recursion. This gives a massive performance boost under BSD, but just
1720        a small improvement under Linux. However, it saves one field in the frame
1721        in all cases.
1723     6. Added more features from the forthcoming Perl 5.10:
1725        (a) (?-n) (where n is a string of digits) is a relative subroutine or
1726            recursion call. It refers to the nth most recently opened parentheses.
1728        (b) (?+n) is also a relative subroutine call; it refers to the nth next
1729            to be opened parentheses.
1731        (c) Conditions that refer to capturing parentheses can be specified
1732            relatively, for example, (?(-2)... or (?(+3)...
1734        (d) \K resets the start of the current match so that everything before
1735            is not part of it.
1737        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1739        (f) \g{name} is another synonym - part of Perl 5.10's unification of
1740            reference syntax.
1742        (g) (?| introduces a group in which the numbering of parentheses in each
1743            alternative starts with the same number.
1745        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
1747     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
1750     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
1751        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
1752        for detecting groups that can match an empty string.
1754     9. A pattern with a very large number of alternatives (more than several
1755        hundred) was running out of internal workspace during the pre-compile
1756        phase, where pcre_compile() figures out how much memory will be needed. A
1757        bit of new cunning has reduced the workspace needed for groups with
1758        alternatives. The 1000-alternative test pattern now uses 12 bytes of
1759        workspace instead of running out of the 4096 that are available.
1761    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
1763    11. Applied patch from Google to remove an optimization that didn't quite work.
1764        The report of the bug said:
1766          pcrecpp::RE("a*").FullMatch("aaa") matches, while
1767          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
1768          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
1770    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
1771        it matched the wrong number of bytes.
1774    Version 7.1 24-Apr-07
1775    ---------------------
1777     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
1778        that is more "standard", making use of automake and other Autotools. There
1779        is some re-arrangement of the files and adjustment of comments consequent
1780        on this.
1782     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
1783        for recursive directory scanning broke on some systems because the files
1784        are not scanned in any specific order and on different systems the order
1785        was different. A call to "sort" has been inserted into RunGrepTest for the
1786        approprate test as a short-term fix. In the longer term there may be an
1787        alternative.
1789     3. I had an email from Eric Raymond about problems translating some of PCRE's
1790        man pages to HTML (despite the fact that I distribute HTML pages, some
1791        people do their own conversions for various reasons). The problems
1792        concerned the use of low-level troff macros .br and .in. I have therefore
1793        removed all such uses from the man pages (some were redundant, some could
1794        be replaced by .nf/.fi pairs). The 132html script that I use to generate
1795        HTML has been updated to handle .nf/.fi and to complain if it encounters
1796        .br or .in.
1798     4. Updated comments in configure.ac that get placed in config.h.in and also
1799        arranged for config.h to be included in the distribution, with the name
1800        config.h.generic, for the benefit of those who have to compile without
1801        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
1803     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
1804        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
1805        makevp.bat for the current PCRE, using the additional files
1806        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
1808     6. A Windows user reported a minor discrepancy with test 2, which turned out
1809        to be caused by a trailing space on an input line that had got lost in his
1810        copy. The trailing space was an accident, so I've just removed it.
1812     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
1813        that is needed.
1815     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
1816        as "const" (a) because they are and (b) because it helps the PHP
1817        maintainers who have recently made a script to detect big data structures
1818        in the php code that should be moved to the .rodata section. I remembered
1819        to update Builducptable as well, so it won't revert if ucptable.h is ever
1820        re-created.
1822     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
1823        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
1824        order to be able to cut out the UTF-8 tables in the latter when UTF-8
1825        support is not required. This saves 1.5-2K of code, which is important in
1826        some applications.
1828        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
1829        so as not to refer to the tables, even though these functions will never be
1830        called when UTF-8 support is disabled. Otherwise there are problems with a
1831        shared library.
1833    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
1835        (a) It was defining its arguments as char * instead of void *.
1837        (b) It was assuming that all moves were upwards in memory; this was true
1838            a long time ago when I wrote it, but is no longer the case.
1840        The emulated memove() is provided for those environments that have neither
1841        memmove() nor bcopy(). I didn't think anyone used it these days, but that
1842        is clearly not the case, as these two bugs were recently reported.
1844    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
1845        and Detrail to create the HTML documentation, the .txt form of the man
1846        pages, and it removes trailing spaces from listed files. It also creates
1847        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
1848        case, it wraps all the #defines with #ifndefs. This script should be run
1849        before "make dist".
1851    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1852        with Unicode property support.
1854        (a) For a maximizing quantifier, if the two different cases of the
1855            character were of different lengths in their UTF-8 codings (there are
1856            some cases like this - I found 11), and the matching function had to
1857            back up over a mixture of the two cases, it incorrectly assumed they
1858            were both the same length.
1860        (b) When PCRE was configured to use the heap rather than the stack for
1861            recursion during matching, it was not correctly preserving the data for
1862            the other case of a UTF-8 character when checking ahead for a match
1863            while processing a minimizing repeat. If the check also involved
1864            matching a wide character, but failed, corruption could cause an
1865            erroneous result when trying to check for a repeat of the original
1866            character.
1868    13. Some tidying changes to the testing mechanism:
1870        (a) The RunTest script now detects the internal link size and whether there
1871            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1872            values substituted by "configure". (The RunGrepTest script already did
1873            this for UTF-8.) The configure.ac script no longer substitutes the
1874            relevant variables.
1876        (b) The debugging options /B and /D in pcretest show the compiled bytecode
1877            with length and offset values. This means that the output is different
1878            for different internal link sizes. Test 2 is skipped for link sizes
1879            other than 2 because of this, bypassing the problem. Unfortunately,
1880            there was also a test in test 3 (the locale tests) that used /B and
1881            failed for link sizes other than 2. Rather than cut the whole test out,
1882            I have added a new /Z option to pcretest that replaces the length and
1883            offset values with spaces. This is now used to make test 3 independent
1884            of link size. (Test 2 will be tidied up later.)
1886    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1887        segmentation fault instead of returning the appropriate error message.
1889    15. In multiline mode when the newline sequence was set to "any", the pattern
1890        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1891        This doesn't seem right; it now treats the CRLF combination as the line
1892        ending, and so does not match in that case. It's only a pattern such as ^$
1893        that would hit this one: something like ^ABC$ would have failed after \r
1894        and then tried again after \r\n.
1896    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1897        in an attempt to make files that differ only in their line terminators
1898        compare equal. This works on Linux.
1900    17. Under certain error circumstances pcregrep might try to free random memory
1901        as it exited. This is now fixed, thanks to valgrind.
1903    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1904        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1905        was because its rules for how to advance for /g after matching an empty
1906        string at the end of a line did not allow for this case. They now check for
1907        it specially.
1909    20. pcretest is supposed to handle patterns and data of any length, by
1910        extending its buffers when necessary. It was getting this wrong when the
1911        buffer for a data line had to be extended.
1913    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1914        CRLF as a newline sequence.
1916    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1917        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1918        I have nevertheless tidied it up.
1920    23. Added some casts to kill warnings from HP-UX ia64 compiler.
1922    24. Added a man page for pcre-config.
1925    Version 7.0 19-Dec-06
1926    ---------------------
1928     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1929        moving to gcc 4.1.1.
1931     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1932        sys/time.h, which is documented as needed for this function. It doesn't
1933        seem to matter on Linux, but it showed up on some releases of OS X.
1935     3. It seems that there are systems where bytes whose values are greater than
1936        127 match isprint() in the "C" locale. The "C" locale should be the
1937        default when a C program starts up. In most systems, only ASCII printing
1938        characters match isprint(). This difference caused the output from pcretest
1939        to vary, making some of the tests fail. I have changed pcretest so that:
1941        (a) When it is outputting text in the compiled version of a pattern, bytes
1942            other than 32-126 are always shown as hex escapes.
1944        (b) When it is outputting text that is a matched part of a subject string,
1945            it does the same, unless a different locale has been set for the match
1946            (using the /L modifier). In this case, it uses isprint() to decide.
1948     4. Fixed a major bug that caused incorrect computation of the amount of memory
1949        required for a compiled pattern when options that changed within the
1950        pattern affected the logic of the preliminary scan that determines the
1951        length. The relevant options are -x, and -i in UTF-8 mode. The result was
1952        that the computed length was too small. The symptoms of this bug were
1953        either the PCRE error "internal error: code overflow" from pcre_compile(),
1954        or a glibc crash with a message such as "pcretest: free(): invalid next
1955        size (fast)". Examples of patterns that provoked this bug (shown in
1956        pcretest format) are:
1958          /(?-x: )/x
1959          /(?x)(?-x: \s*#\s*)/
1960          /((?i)[\x{c0}])/8
1961          /(?i:[\x{c0}])/8
1963        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1964        is now done differently.
1966     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1967        wrapper classes; (b) implement a new function in the C++ scanner that is
1968        more efficient than the old way of doing things because it avoids levels of
1969        recursion in the regex matching; (c) add a paragraph to the documentation
1970        for the FullMatch() function.
1972     6. The escape sequence \n was being treated as whatever was defined as
1973        "newline". Not only was this contrary to the documentation, which states
1974        that \n is character 10 (hex 0A), but it also went horribly wrong when
1975        "newline" was defined as CRLF. This has been fixed.
1977     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1978        was being set to -1 for the "end of line" case (supposedly a value that no
1979        character can have). Though this value is never used (the check for end of
1980        line is "zero bytes in current character"), it caused compiler complaints.
1981        I've changed it to 0xffffffff.
1983     8. In pcre_version.c, the version string was being built by a sequence of
1984        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1985        string (as it is for production releases) called a macro with an empty
1986        argument. The C standard says the result of this is undefined. The gcc
1987        compiler treats it as an empty string (which was what was wanted) but it is
1988        reported that Visual C gives an error. The source has been hacked around to
1989        avoid this problem.
1991     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1992        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1993        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1994        of them did).
1996    10. Originally, pcretest opened its input and output without "b"; then I was
1997        told that "b" was needed in some environments, so it was added for release
1998        5.0 to both the input and output. (It makes no difference on Unix-like
1999        systems.) Later I was told that it is wrong for the input on Windows. I've
2000        now abstracted the modes into two macros, to make it easier to fiddle with
2001        them, and removed "b" from the input mode under Windows.
2003    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
2005    12. Added -help and --help to pcretest as an official way of being reminded
2006        of the options.
2008    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
2009        and pcrecpp.cc because they annoy compilers at high warning levels.
2011    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
2013    15. Fixed an occurrence of == in configure.ac that should have been = (shell
2014        scripts are not C programs :-) and which was not noticed because it works
2015        on Linux.
2017    16. pcretest is supposed to handle any length of pattern and data line (as one
2018        line or as a continued sequence of lines) by extending its input buffer if
2019        necessary. This feature was broken for very long pattern lines, leading to
2020        a string of junk being passed to pcre_compile() if the pattern was longer
2021        than about 50K.
2023    17. I have done a major re-factoring of the way pcre_compile() computes the
2024        amount of memory needed for a compiled pattern. Previously, there was code
2025        that made a preliminary scan of the pattern in order to do this. That was
2026        OK when PCRE was new, but as the facilities have expanded, it has become
2027        harder and harder to keep it in step with the real compile phase, and there
2028        have been a number of bugs (see for example, 4 above). I have now found a
2029        cunning way of running the real compile function in a "fake" mode that
2030        enables it to compute how much memory it would need, while actually only
2031        ever using a few hundred bytes of working memory and without too many
2032        tests of the mode. This should make future maintenance and development
2033        easier. A side effect of this work is that the limit of 200 on the nesting
2034        depth of parentheses has been removed (though this was never a serious
2035        limitation, I suspect). However, there is a downside: pcre_compile() now
2036        runs more slowly than before (30% or more, depending on the pattern). I
2037        hope this isn't a big issue. There is no effect on runtime performance.
2039    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
2040        newline (only possible for the last line of a file) and it was a
2041        pattern that set a locale (followed by /Lsomething), pcretest crashed.
2043    19. Added additional timing features to pcretest. (1) The -tm option now times
2044        matching only, not compiling. (2) Both -t and -tm can be followed, as a
2045        separate command line item, by a number that specifies the number of
2046        repeats to use when timing. The default is 50000; this gives better
2047        precision, but takes uncomfortably long for very large patterns.
2049    20. Extended pcre_study() to be more clever in cases where a branch of a
2050        subpattern has no definite first character. For example, (a*|b*)[cd] would
2051        previously give no result from pcre_study(). Now it recognizes that the
2052        first character must be a, b, c, or d.
2054    21. There was an incorrect error "recursive call could loop indefinitely" if
2055        a subpattern (or the entire pattern) that was being tested for matching an
2056        empty string contained only one non-empty item after a nested subpattern.
2057        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
2058        incorrectly, because the \d was being skipped in the check.
2060    22. The pcretest program now has a new pattern option /B and a command line
2061        option -b, which is equivalent to adding /B to every pattern. This causes
2062        it to show the compiled bytecode, without the additional information that
2063        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
2064        is the same as /B/I).
2066    23. A new optimization is now able automatically to treat some sequences such
2067        as a*b as a*+b. More specifically, if something simple (such as a character
2068        or a simple class like \d) has an unlimited quantifier, and is followed by
2069        something that cannot possibly match the quantified thing, the quantifier
2070        is automatically "possessified".
2072    24. A recursive reference to a subpattern whose number was greater than 39
2073        went wrong under certain circumstances in UTF-8 mode. This bug could also
2074        have affected the operation of pcre_study().
2076    25. Realized that a little bit of performance could be had by replacing
2077        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
2079    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
2081    27. Possessive quantifiers such as a++ were previously implemented by turning
2082        them into atomic groups such as ($>a+). Now they have their own opcodes,
2083        which improves performance. This includes the automatically created ones
2084        from 23 above.
2086    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
2087        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
2088        the first matched character to be a colon. This applied both to named and
2089        numbered groups.
2091    29. The ucpinternal.h header file was missing its idempotency #ifdef.
2093    30. I was sent a "project" file called libpcre.a.dev which I understand makes
2094        building PCRE on Windows easier, so I have included it in the distribution.
2096    31. There is now a check in pcretest against a ridiculously large number being
2097        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
2098        loop, the loop is abandoned.
2100    32. Forward references to subpatterns in conditions such as (?(2)...) where
2101        subpattern 2 is defined later cause pcre_compile() to search forwards in
2102        the pattern for the relevant set of parentheses. This search went wrong
2103        when there were unescaped parentheses in a character class, parentheses
2104        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
2106    33. "Subroutine" calls and backreferences were previously restricted to
2107        referencing subpatterns earlier in the regex. This restriction has now
2108        been removed.
2110    34. Added a number of extra features that are going to be in Perl 5.10. On the
2111        whole, these are just syntactic alternatives for features that PCRE had
2112        previously implemented using the Python syntax or my own invention. The
2113        other formats are all retained for compatibility.
2115        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
2116            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
2117            also .NET compatible.
2119        (b) A recursion or subroutine call to a named group can now be defined as
2120            (?&name) as well as (?P>name).
2122        (c) A backreference to a named group can now be defined as \k<name> or
2123            \k'name' as well as (?P=name). The new forms, as well as being in Perl
2124            5.10, are also .NET compatible.
2126        (d) A conditional reference to a named group can now use the syntax
2127            (?(<name>) or (?('name') as well as (?(name).
2129        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
2130            groups (named and numbered) that are never evaluated inline, but can be
2131            called as "subroutines" from elsewhere. In effect, the DEFINE condition
2132            is always false. There may be only one alternative in such a group.
2134        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
2135            as the simple (?(R). The condition is true only if the most recent
2136            recursion is that of the given number or name. It does not search out
2137            through the entire recursion stack.
2139        (g) The escape \gN or \g{N} has been added, where N is a positive or
2140            negative number, specifying an absolute or relative reference.
2142    35. Tidied to get rid of some further signed/unsigned compiler warnings and
2143        some "unreachable code" warnings.
2145    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
2146        things, this adds five new scripts.
2148    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
2149        There were also incompatibilities regarding the handling of \Q..\E inside
2150        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
2151        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
2153    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
2154        matches an empty string, and forcibly breaks the loop. There were bugs in
2155        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
2156        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
2157        separate and independent bugs (that affected different cases) have been
2158        fixed.
2160    39. Refactored the code to abolish the use of different opcodes for small
2161        capturing bracket numbers. This is a tidy that I avoided doing when I
2162        removed the limit on the number of capturing brackets for 3.5 back in 2001.
2163        The new approach is not only tidier, it makes it possible to reduce the
2164        memory needed to fix the previous bug (38).
2166    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
2167        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
2168        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
2169        mode.
2171    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
2172        report.
2174    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
2175        copy construction and assignment in the C++ wrapper.
2177    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
2178        couple of bugs that could have given wrong results in the "--newline=crlf"
2179        case.
2181    44. Added a number of casts and did some reorganization of signed/unsigned int
2182        variables following suggestions from Dair Grant. Also renamed the variable
2183        "this" as "item" because it is a C++ keyword.
2185    45. Arranged for dftables to add
2187          #include "pcre_internal.h"
2189        to pcre_chartables.c because without it, gcc 4.x may remove the array
2190        definition from the final binary if PCRE is built into a static library and
2191        dead code stripping is activated.
2193    46. For an unanchored pattern, if a match attempt fails at the start of a
2194        newline sequence, and the newline setting is CRLF or ANY, and the next two
2195        characters are CRLF, advance by two characters instead of one.
2198    Version 6.7 04-Jul-06
2199    ---------------------
2201     1. In order to handle tests when input lines are enormously long, pcretest has
2202        been re-factored so that it automatically extends its buffers when
2203        necessary. The code is crude, but this _is_ just a test program. The
2204        default size has been increased from 32K to 50K.
2206     2. The code in pcre_study() was using the value of the re argument before
2207        testing it for NULL. (Of course, in any sensible call of the function, it
2208        won't be NULL.)
2210     3. The memmove() emulation function in pcre_internal.h, which is used on
2211        systems that lack both memmove() and bcopy() - that is, hardly ever -
2212        was missing a "static" storage class specifier.
2214     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
2215        containing an extended class (one that cannot be represented by a bitmap
2216        because it contains high-valued characters or Unicode property items, e.g.
2217        [\pZ]). Almost always one would set UTF-8 mode when processing such a
2218        pattern, but PCRE should not loop if you do not (it no longer does).
2219        [Detail: two cases were found: (a) a repeated subpattern containing an
2220        extended class; (b) a recursive reference to a subpattern that followed a
2221        previous extended class. It wasn't skipping over the extended class
2222        correctly when UTF-8 mode was not set.]
2224     5. A negated single-character class was not being recognized as fixed-length
2225        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
2226        compile error "lookbehind assertion is not fixed length".
2228     6. The RunPerlTest auxiliary script was showing an unexpected difference
2229        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
2230        write a Perl script that can interpret lines of an input file either as
2231        byte characters or as UTF-8, which is what "perltest" was being required to
2232        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
2233        can't do is switch easily at run time between having the "use utf8;" pragma
2234        or not. In the end, I fudged it by using the RunPerlTest script to insert
2235        "use utf8;" explicitly for the UTF-8 tests.
2237     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
2238        the end of the subject string, contrary to the documentation and to what
2239        Perl does. This was true of both matching functions. Now it matches only at
2240        the start of the subject and immediately after *internal* newlines.
2242     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
2243        a pointer to an int instead of a pointer to an unsigned long int. This
2244        caused problems on 64-bit systems.
2246     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
2247        instance of the 'standard' template library not being so standard".
2249    10. There was no check on the number of named subpatterns nor the maximum
2250        length of a subpattern name. The product of these values is used to compute
2251        the size of the memory block for a compiled pattern. By supplying a very
2252        long subpattern name and a large number of named subpatterns, the size
2253        computation could be caused to overflow. This is now prevented by limiting
2254        the length of names to 32 characters, and the number of named subpatterns
2255        to 10,000.
2257    11. Subpatterns that are repeated with specific counts have to be replicated in
2258        the compiled pattern. The size of memory for this was computed from the
2259        length of the subpattern and the repeat count. The latter is limited to
2260        65535, but there was no limit on the former, meaning that integer overflow
2261        could in principle occur. The compiled length of a repeated subpattern is
2262        now limited to 30,000 bytes in order to prevent this.
2264    12. Added the optional facility to have named substrings with the same name.
2266    13. Added the ability to use a named substring as a condition, using the
2267        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
2268        are numbers (not recommended). Forward references are permitted.
2270    14. Added forward references in named backreferences (if you see what I mean).
2272    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
2273        pattern could run off the end of the subject. For example, the pattern
2274        "(?s)(.{1,5})"8 did this with the subject "ab".
2276    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
2277        PCRE_CASELESS was set when matching characters that were quantified with ?
2278        or *.
2280    17. A character class other than a single negated character that had a minimum
2281        but no maximum quantifier - for example [ab]{6,} - was not handled
2282        correctly by pce_dfa_exec(). It would match only one character.
2284    18. A valid (though odd) pattern that looked like a POSIX character
2285        class but used an invalid character after [ (for example [[,abc,]]) caused
2286        pcre_compile() to give the error "Failed: internal error: code overflow" or
2287        in some cases to crash with a glibc free() error. This could even happen if
2288        the pattern terminated after [[ but there just happened to be a sequence of
2289        letters, a binary zero, and a closing ] in the memory that followed.
2291    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
2292        over the years. Originally (before any Unicode support), just the bottom 8
2293        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
2294        output from "man perlunicode" includes this:
2296          The regular expression compiler produces polymorphic opcodes.  That
2297          is, the pattern adapts to the data and automatically switches to
2298          the Unicode character scheme when presented with Unicode data--or
2299          instead uses a traditional byte scheme when presented with byte
2300          data.
2302        Sadly, a wide octal escape does not cause a switch, and in a string with
2303        no other multibyte characters, these octal escapes are treated as before.
2304        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
2305        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
2306        Unicode string.
2308        I have not perpetrated such confusion in PCRE. Up till now, it took just
2309        the bottom 8 bits, as in old Perl. I have now made octal escapes with
2310        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
2311        translate to the appropriate multibyte character.
2313    29. Applied some refactoring to reduce the number of warnings from Microsoft
2314        and Borland compilers. This has included removing the fudge introduced
2315        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
2316        a warning about an unused variable.
2318    21. PCRE has not included VT (character 0x0b) in the set of whitespace
2319        characters since release 4.0, because Perl (from release 5.004) does not.
2320        [Or at least, is documented not to: some releases seem to be in conflict
2321        with the documentation.] However, when a pattern was studied with
2322        pcre_study() and all its branches started with \s, PCRE still included VT
2323        as a possible starting character. Of course, this did no harm; it just
2324        caused an unnecessary match attempt.
2326    22. Removed a now-redundant internal flag bit that recorded the fact that case
2327        dependency changed within the pattern. This was once needed for "required
2328        byte" processing, but is no longer used. This recovers a now-scarce options
2329        bit. Also moved the least significant internal flag bit to the most-
2330        significant bit of the word, which was not previously used (hangover from
2331        the days when it was an int rather than a uint) to free up another bit for
2332        the future.
2334    23. Added support for CRLF line endings as well as CR and LF. As well as the
2335        default being selectable at build time, it can now be changed at runtime
2336        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
2337        specify that it is scanning data with non-default line endings.
2339    24. Changed the definition of CXXLINK to make it agree with the definition of
2340        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
2342    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
2343        recursions. This makes a big different to stack usage for some patterns.
2345    26. If a subpattern containing a named recursion or subroutine reference such
2346        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
2347        the space required for the compiled pattern went wrong and gave too small a
2348        value. Depending on the environment, this could lead to "Failed: internal
2349        error: code overflow at offset 49" or "glibc detected double free or
2350        corruption" errors.
2352    27. Applied patches from Google (a) to support the new newline modes and (b) to
2353        advance over multibyte UTF-8 characters in GlobalReplace.
2355    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
2356        difference for some implementation of PCRE in some Windows version.
2358    29. Added some extra testing facilities to pcretest:
2360        \q<number>   in a data line sets the "match limit" value
2361        \Q<number>   in a data line sets the "match recursion limt" value
2362        -S <number>  sets the stack size, where <number> is in megabytes
2364        The -S option isn't available for Windows.
2367    Version 6.6 06-Feb-06
2368    ---------------------
2370     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
2371        in pcreposix.h. I have copied the definition from pcre.h.
2373     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
2374        because pcre.h is no longer a built file.
2376     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
2377        not normally included in the compiled code.
2380    Version 6.5 01-Feb-06
2381    ---------------------
2383     1. When using the partial match feature with pcre_dfa_exec(), it was not
2384        anchoring the second and subsequent partial matches at the new starting
2385        point. This could lead to incorrect results. For example, with the pattern
2386        /1234/, partially matching against "123" and then "a4" gave a match.
2388     2. Changes to pcregrep:
2390        (a) All non-match returns from pcre_exec() were being treated as failures
2391            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
2392            error message is output. Some extra information is given for the
2393            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
2394            probably the only errors that are likely to be caused by users (by
2395            specifying a regex that has nested indefinite repeats, for instance).
2396            If there are more than 20 of these errors, pcregrep is abandoned.
2398        (b) A binary zero was treated as data while matching, but terminated the
2399            output line if it was written out. This has been fixed: binary zeroes
2400            are now no different to any other data bytes.
2402        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
2403            used to set a locale for matching. The --locale=xxxx long option has
2404            been added (no short equivalent) to specify a locale explicitly on the
2405            pcregrep command, overriding the environment variables.
2407        (d) When -B was used with -n, some line numbers in the output were one less
2408            than they should have been.
2410        (e) Added the -o (--only-matching) option.
2412        (f) If -A or -C was used with -c (count only), some lines of context were
2413            accidentally printed for the final match.
2415        (g) Added the -H (--with-filename) option.
2417        (h) The combination of options -rh failed to suppress file names for files
2418            that were found from directory arguments.
2420        (i) Added the -D (--devices) and -d (--directories) options.
2422        (j) Added the -F (--fixed-strings) option.
2424        (k) Allow "-" to be used as a file name for -f as well as for a data file.
2426        (l) Added the --colo(u)r option.
2428        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
2429            is not present by default.
2431     3. A nasty bug was discovered in the handling of recursive patterns, that is,
2432        items such as (?R) or (?1), when the recursion could match a number of
2433        alternatives. If it matched one of the alternatives, but subsequently,
2434        outside the recursion, there was a failure, the code tried to back up into
2435        the recursion. However, because of the way PCRE is implemented, this is not
2436        possible, and the result was an incorrect result from the match.
2438        In order to prevent this happening, the specification of recursion has
2439        been changed so that all such subpatterns are automatically treated as
2440        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
2442     4. I had overlooked the fact that, in some locales, there are characters for
2443        which isalpha() is true but neither isupper() nor islower() are true. In
2444        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
2445        and ordfeminine) are like this. This affected the treatment of \w and \W
2446        when they appeared in character classes, but not when they appeared outside
2447        a character class. The bit map for "word" characters is now created
2448        separately from the results of isalnum() instead of just taking it from the
2449        upper, lower, and digit maps. (Plus the underscore character, of course.)
2451     5. The above bug also affected the handling of POSIX character classes such as
2452        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
2453        permanent tables. Instead, the bit maps for such a class were previously
2454        created as the appropriate unions of the upper, lower, and digit bitmaps.
2455        Now they are created by subtraction from the [[:word:]] class, which has
2456        its own bitmap.
2458     6. The [[:blank:]] character class matches horizontal, but not vertical space.
2459        It is created by subtracting the vertical space characters (\x09, \x0a,
2460        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
2461        subtraction was done in the overall bitmap for a character class, meaning
2462        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
2463        be recognized. This bug has been fixed.
2465     7. Patches from the folks at Google:
2467          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
2468          real life, but is still worth protecting against".
2470          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
2471          regular expressions".
2473          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
2474          have it.
2476          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
2477          "configure" and the latter not, in order to fix a problem somebody had
2478          with compiling the Arg class on HP-UX.
2480          (e) Improve the error-handling of the C++ wrapper a little bit.
2482          (f) New tests for checking recursion limiting.
2484     8. The pcre_memmove() function, which is used only if the environment does not
2485        have a standard memmove() function (and is therefore rarely compiled),
2486        contained two bugs: (a) use of int instead of size_t, and (b) it was not
2487        returning a result (though PCRE never actually uses the result).
2489     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
2490        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
2491        returned instead of calling malloc() with an overflowing number that would
2492        most likely cause subsequent chaos.
2494    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
2496    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
2497        with this option is matched, the nmatch and pmatch options of regexec() are
2498        ignored.
2500    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
2501        provided in case anyone wants to the the POSIX interface with UTF-8
2502        strings.
2504    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
2505        C++ linking (needed for some HP-UX environments).
2507    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
2508        (unused parameter) and in the pcre_printint() function (omitted "default"
2509        switch label when the default is to do nothing).
2511    15. Added some code to make it possible, when PCRE is compiled as a C++
2512        library, to replace subject pointers for pcre_exec() with a smart pointer
2513        class, thus making it possible to process discontinuous strings.
2515    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
2516        much the same function. They were added by different people who were trying
2517        to make PCRE easy to compile on non-Unix systems. It has been suggested
2518        that PCRE_EXPORT be abolished now that there is more automatic apparatus
2519        for compiling on Windows systems. I have therefore replaced it with
2520        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
2521        defaults to "extern" for C or "extern C" for C++, which works fine on
2522        Unix-like systems. It is now possible to override the value of PCRE_DATA_
2523        SCOPE with something explicit in config.h. In addition:
2525        (a) pcreposix.h still had just "extern" instead of either of these macros;
2526            I have replaced it with PCRE_DATA_SCOPE.
2528        (b) Functions such as _pcre_xclass(), which are internal to the library,
2529            but external in the C sense, all had PCRE_EXPORT in their definitions.
2530            This is apparently wrong for the Windows case, so I have removed it.
2531            (It makes no difference on Unix-like systems.)
2533    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
2534        of recursive calls to match(). This is different to MATCH_LIMIT because
2535        that limits the total number of calls to match(), not all of which increase
2536        the depth of recursion. Limiting the recursion depth limits the amount of
2537        stack (or heap if NO_RECURSE is set) that is used. The default can be set
2538        when PCRE is compiled, and changed at run time. A patch from Google adds
2539        this functionality to the C++ interface.
2541    18. Changes to the handling of Unicode character properties:
2543        (a) Updated the table to Unicode 4.1.0.
2545        (b) Recognize characters that are not in the table as "Cn" (undefined).
2547        (c) I revised the way the table is implemented to a much improved format
2548            which includes recognition of ranges. It now supports the ranges that
2549            are defined in UnicodeData.txt, and it also amalgamates other
2550            characters into ranges. This has reduced the number of entries in the
2551            table from around 16,000 to around 3,000, thus reducing its size
2552            considerably. I realized I did not need to use a tree structure after
2553            all - a binary chop search is just as efficient. Having reduced the
2554            number of entries, I extended their size from 6 bytes to 8 bytes to
2555            allow for more data.
2557        (d) Added support for Unicode script names via properties such as \p{Han}.
2559    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
2560        matching that character.
2562    20. When matching a repeated Unicode property with a minimum greater than zero,
2563        (for example \pL{2,}), PCRE could look past the end of the subject if it
2564        reached it while seeking the minimum number of characters. This could
2565        happen only if some of the characters were more than one byte long, because
2566        there is a check for at least the minimum number of bytes.
2568    21. Refactored the implementation of \p and \P so as to be more general, to
2569        allow for more different types of property in future. This has changed the
2570        compiled form incompatibly. Anybody with saved compiled patterns that use
2571        \p or \P will have to recompile them.
2573    22. Added "Any" and "L&" to the supported property types.
2575    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
2576        but give a compile time error if the value is greater than 0xff.
2578    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
2579        accidentally not being installed or uninstalled.
2581    25. The pcre.h file was built from pcre.h.in, but the only changes that were
2582        made were to insert the current release number. This seemed silly, because
2583        it made things harder for people building PCRE on systems that don't run
2584        "configure". I have turned pcre.h into a distributed file, no longer built
2585        by "configure", with the version identification directly included. There is
2586        no longer a pcre.h.in file.
2588        However, this change necessitated a change to the pcre-config script as
2589        well. It is built from pcre-config.in, and one of the substitutions was the
2590        release number. I have updated configure.ac so that ./configure now finds
2591        the release number by grepping pcre.h.
2593    26. Added the ability to run the tests under valgrind.
2596    Version 6.4 05-Sep-05
2597    ---------------------
2599     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
2600        "--" to be printed when multiple files were scanned, even when none of the
2601        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
2602        consider it to be a bug, and have restored the previous behaviour.
2604     2. A couple of code tidies to get rid of compiler warnings.
2606     3. The pcretest program used to cheat by referring to symbols in the library
2607        whose names begin with _pcre_. These are internal symbols that are not
2608        really supposed to be visible externally, and in some environments it is
2609        possible to suppress them. The cheating is now confined to including
2610        certain files from the library's source, which is a bit cleaner.
2612     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
2613        file's purpose clearer.
2615     5. Reorganized pcre_ucp_findchar().
2618    Version 6.3 15-Aug-05
2619    ---------------------
2621     1. The file libpcre.pc.in did not have general read permission in the tarball.
2623     2. There were some problems when building without C++ support:
2625        (a) If C++ support was not built, "make install" and "make test" still
2626            tried to test it.
2628        (b) There were problems when the value of CXX was explicitly set. Some
2629            changes have been made to try to fix these, and ...
2631        (c) --disable-cpp can now be used to explicitly disable C++ support.
2633        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
2634            backslash in a target when C++ was disabled. This confuses some
2635            versions of "make", apparently. Using an intermediate variable solves
2636            this. (Same for CPP_LOBJ.)
2638     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
2639        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
2640        necessary on certain architectures.
2642     4. Added a setting of -export-symbols-regex to the link command to remove
2643        those symbols that are exported in the C sense, but actually are local
2644        within the library, and not documented. Their names all begin with
2645        "_pcre_". This is not a perfect job, because (a) we have to except some
2646        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
2647        available (and never for static libraries). I have made a note to try to
2648        find a way round (a) in the future.
2651    Version 6.2 01-Aug-05
2652    ---------------------
2654     1. There was no test for integer overflow of quantifier values. A construction
2655        such as {1111111111111111} would give undefined results. What is worse, if
2656        a minimum quantifier for a parenthesized subpattern overflowed and became
2657        negative, the calculation of the memory size went wrong. This could have
2658        led to memory overwriting.
2660     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
2662     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
2663        operating environments where this matters.
2665     4. Applied Giuseppe Maxia's patch to add additional features for controlling
2666        PCRE options from within the C++ wrapper.
2668     5. Named capturing subpatterns were not being correctly counted when a pattern
2669        was compiled. This caused two problems: (a) If there were more than 100
2670        such subpatterns, the calculation of the memory needed for the whole
2671        compiled pattern went wrong, leading to an overflow error. (b) Numerical
2672        back references of the form \12, where the number was greater than 9, were
2673        not recognized as back references, even though there were sufficient
2674        previous subpatterns.
2676     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
2677        versions of gcc, e.g. 2.95.4.
2680    Version 6.1 21-Jun-05
2681    ---------------------
2683     1. There was one reference to the variable "posix" in pcretest.c that was not
2684        surrounded by "#if !defined NOPOSIX".
2686     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
2687        the cross-check on the old pcre_info() function, for the benefit of the
2688        cut-down version of PCRE that is currently imported into Exim.
2690     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
2691        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
2692        patterns that start (?i) but I don't think that matters. The use of (?i) is
2693        just an example; this all applies to the other options as well.
2695     4. Since libtool seems to echo the compile commands it is issuing, the output
2696        from "make" can be reduced a bit by putting "@" in front of each libtool
2697        compile command.
2699     5. Patch from the folks at Google for configure.in to be a bit more thorough
2700        in checking for a suitable C++ installation before trying to compile the
2701        C++ stuff. This should fix a reported problem when a compiler was present,
2702        but no suitable headers.
2704     6. The man pages all had just "PCRE" as their title. I have changed them to
2705        be the relevant file name. I have also arranged that these names are
2706        retained in the file doc/pcre.txt, which is a concatenation in text format
2707        of all the man pages except the little individual ones for each function.
2709     7. The NON-UNIX-USE file had not been updated for the different set of source
2710        files that come with release 6. I also added a few comments about the C++
2711        wrapper.
2714    Version 6.0 07-Jun-05
2715    ---------------------
2717     1. Some minor internal re-organization to help with my DFA experiments.
2719     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
2720        didn't matter for the library itself when fully configured, but did matter
2721        when compiling without UCP support, or within Exim, where the ucp files are
2722        not imported.
2724     3. Refactoring of the library code to split up the various functions into
2725        different source modules. The addition of the new DFA matching code (see
2726        below) to a single monolithic source would have made it really too
2727        unwieldy, quite apart from causing all the code to be include in a
2728        statically linked application, when only some functions are used. This is
2729        relevant even without the DFA addition now that patterns can be compiled in
2730        one application and matched in another.
2732        The downside of splitting up is that there have to be some external
2733        functions and data tables that are used internally in different modules of
2734        the library but which are not part of the API. These have all had their
2735        names changed to start with "_pcre_" so that they are unlikely to clash
2736        with other external names.
2738     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
2739        a different (DFA) algorithm. Although it is slower than the original
2740        function, it does have some advantages for certain types of matching
2741        problem.
2743     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
2744        including restarting after a partial match.
2746     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
2747        defined when compiling for Windows was sent to me. I have put it into the
2748        code, though I have no means of testing or verifying it.
2750     7. Added the pcre_refcount() auxiliary function.
2752     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
2753        match before or at the first newline in the subject string. In pcretest,
2754        the /f option on a pattern can be used to set this.
2756     9. A repeated \w when used in UTF-8 mode with characters greater than 256
2757        would behave wrongly. This has been present in PCRE since release 4.0.
2759    10. A number of changes to the pcregrep command:
2761        (a) Refactored how -x works; insert ^(...)$ instead of setting
2762            PCRE_ANCHORED and checking the length, in preparation for adding
2763            something similar for -w.
2765        (b) Added the -w (match as a word) option.
2767        (c) Refactored the way lines are read and buffered so as to have more
2768            than one at a time available.
2770        (d) Implemented a pcregrep test script.
2772        (e) Added the -M (multiline match) option. This allows patterns to match
2773            over several lines of the subject. The buffering ensures that at least
2774            8K, or the rest of the document (whichever is the shorter) is available
2775            for matching (and similarly the previous 8K for lookbehind assertions).
2777        (f) Changed the --help output so that it now says
2779              -w, --word-regex(p)
2781            instead of two lines, one with "regex" and the other with "regexp"
2782            because that confused at least one person since the short forms are the
2783            same. (This required a bit of code, as the output is generated
2784            automatically from a table. It wasn't just a text change.)
2786        (g) -- can be used to terminate pcregrep options if the next thing isn't an
2787            option but starts with a hyphen. Could be a pattern or a path name
2788            starting with a hyphen, for instance.
2790        (h) "-" can be given as a file name to represent stdin.
2792        (i) When file names are being printed, "(standard input)" is used for
2793            the standard input, for compatibility with GNU grep. Previously
2794            "<stdin>" was used.
2796        (j) The option --label=xxx can be used to supply a name to be used for
2797            stdin when file names are being printed. There is no short form.
2799        (k) Re-factored the options decoding logic because we are going to add
2800            two more options that take data. Such options can now be given in four
2801            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
2803        (l) Added the -A, -B, and -C options for requesting that lines of context
2804            around matches be printed.
2806        (m) Added the -L option to print the names of files that do not contain
2807            any matching lines, that is, the complement of -l.
2809        (n) The return code is 2 if any file cannot be opened, but pcregrep does
2810            continue to scan other files.
2812        (o) The -s option was incorrectly implemented. For compatibility with other
2813            greps, it now suppresses the error message for a non-existent or non-
2814            accessible file (but not the return code). There is a new option called
2815            -q that suppresses the output of matching lines, which was what -s was
2816            previously doing.
2818        (p) Added --include and --exclude options to specify files for inclusion
2819            and exclusion when recursing.
2821    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
2822        Hopefully, it now does.
2824    12. Missing cast in pcre_study().
2826    13. Added an "uninstall" target to the makefile.
2828    14. Replaced "extern" in the function prototypes in Makefile.in with
2829        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
2830        world, but is set differently for Windows.
2832    15. Added a second compiling function called pcre_compile2(). The only
2833        difference is that it has an extra argument, which is a pointer to an
2834        integer error code. When there is a compile-time failure, this is set
2835        non-zero, in addition to the error test pointer being set to point to an
2836        error message. The new argument may be NULL if no error number is required
2837        (but then you may as well call pcre_compile(), which is now just a
2838        wrapper). This facility is provided because some applications need a
2839        numeric error indication, but it has also enabled me to tidy up the way
2840        compile-time errors are handled in the POSIX wrapper.
2842    16. Added VPATH=.libs to the makefile; this should help when building with one
2843        prefix path and installing with another. (Or so I'm told by someone who
2844        knows more about this stuff than I do.)
2846    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
2847        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
2848        match everything, including newlines. This is not POSIX-compatible, but
2849        somebody wanted the feature. From pcretest it can be activated by using
2850        both the P and the s flags.
2852    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
2854    19. libpcre.pc was being incorrectly installed as executable.
2856    20. A couple of places in pcretest check for end-of-line by looking for '\n';
2857        it now also looks for '\r' so that it will work unmodified on Windows.
2859    21. Added Google's contributed C++ wrapper to the distribution.
2861    22. Added some untidy missing memory free() calls in pcretest, to keep
2862        Electric Fence happy when testing.
2866    Version 5.0 13-Sep-04
2867    ---------------------
2869     1. Internal change: literal characters are no longer packed up into items
2870        containing multiple characters in a single byte-string. Each character
2871        is now matched using a separate opcode. However, there may be more than one
2872        byte in the character in UTF-8 mode.
2874     2. The pcre_callout_block structure has two new fields: pattern_position and
2875        next_item_length. These contain the offset in the pattern to the next match
2876        item, and its length, respectively.
2878     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
2879        insertion of callouts before each pattern item. Added the /C option to
2880        pcretest to make use of this.
2882     4. On the advice of a Windows user, the lines
2884          #if defined(_WIN32) || defined(WIN32)
2885          _setmode( _fileno( stdout ), 0x8000 );
2886          #endif  /* defined(_WIN32) || defined(WIN32) */
2888        have been added to the source of pcretest. This apparently does useful
2889        magic in relation to line terminators.
2891     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
2892        for the benefit of those environments where the "b" makes a difference.
2894     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
2895        to know about it. I have put a hack into configure.in that adds in code
2896        to set GCC=yes if CC=icc. This seems to end up at a point in the
2897        generated configure script that is early enough to affect the setting of
2898        compiler options, which is what is needed, but I have no means of testing
2899        whether it really works. (The user who reported this had patched the
2900        generated configure script, which of course I cannot do.)
2902        LATER: After change 22 below (new libtool files), the configure script
2903        seems to know about icc (and also ecc). Therefore, I have commented out
2904        this hack in configure.in.
2906     7. Added support for pkg-config (2 patches were sent in).
2908     8. Negated POSIX character classes that used a combination of internal tables
2909        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2910        [[:^ascii]]. Typically, they would match almost any characters. The other
2911        POSIX classes were not broken in this way.
2913     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2914        to find the match, as PCRE was deluded into thinking that the match had to
2915        start at the start point or following a newline. The same bug applied to
2916        patterns with negative forward assertions or any backward assertions
2917        preceding ".*" at the start, unless the pattern required a fixed first
2918        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2920    10. In UTF-8 mode, when moving forwards in the subject after a failed match
2921        starting at the last subject character, bytes beyond the end of the subject
2922        string were read.
2924    11. Renamed the variable "class" as "classbits" to make life easier for C++
2925        users. (Previously there was a macro definition, but it apparently wasn't
2926        enough.)
2928    12. Added the new field "tables" to the extra data so that tables can be passed
2929        in at exec time, or the internal tables can be re-selected. This allows
2930        a compiled regex to be saved and re-used at a later time by a different
2931        program that might have everything at different addresses.
2933    13. Modified the pcre-config script so that, when run on Solaris, it shows a
2934        -R library as well as a -L library.
2936    14. The debugging options of pcretest (-d on the command line or D on a
2937        pattern) showed incorrect output for anything following an extended class
2938        that contained multibyte characters and which was followed by a quantifier.
2940    15. Added optional support for general category Unicode character properties
2941        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2942        support. It adds about 90K to the size of the library. The meanings of the
2943        inbuilt class escapes such as \d and \s have NOT been changed.
2945    16. Updated pcredemo.c to include calls to free() to release the memory for the
2946        compiled pattern.
2948    17. The generated file chartables.c was being created in the source directory
2949        instead of in the building directory. This caused the build to fail if the
2950        source directory was different from the building directory, and was
2951        read-only.
2953    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2954        file. No doubt somebody will tell me if they don't make sense... Also added
2955        Dan Mooney's comments about building on OpenVMS.
2957    19. Added support for partial matching via the PCRE_PARTIAL option for
2958        pcre_exec() and the \P data escape in pcretest.
2960    20. Extended pcretest with 3 new pattern features:
2962        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
2963              write the compiled pattern to the file whose name is "rest-of-line".
2964              This is a straight binary dump of the data, with the saved pointer to
2965              the character tables forced to be NULL. The study data, if any, is
2966              written too. After writing, pcretest reads a new pattern.
2968        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2969              compiled pattern from the given file. There must not be any
2970              occurrences of "<" in the file name (pretty unlikely); if there are,
2971              pcretest will instead treat the initial "<" as a pattern delimiter.
2972              After reading in the pattern, pcretest goes on to read data lines as
2973              usual.
2975        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2976              and 16-bit fields in a compiled pattern, to simulate a pattern that
2977              was compiled on a host of opposite endianness.
2979    21. The pcre-exec() function can now cope with patterns that were compiled on
2980        hosts of opposite endianness, with this restriction:
2982          As for any compiled expression that is saved and used later, the tables
2983          pointer field cannot be preserved; the extra_data field in the arguments
2984          to pcre_exec() should be used to pass in a tables address if a value
2985          other than the default internal tables were used at compile time.
2987    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2988        now diagnosed as an error. Previously, most of the time, a negative number
2989        would have been treated as zero, but if in addition "ovector" was passed as
2990        NULL, a crash could occur.
2992    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2993        new versions from the libtool 1.5 distribution (the last one is a copy of
2994        a file called libtool.m4). This seems to have fixed the need to patch
2995        "configure" to support Darwin 1.3 (which I used to do). However, I still
2996        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2997        workstation).
2999    24. Changed the PCRE licence to be the more standard "BSD" licence.
3002    Version 4.5 01-Dec-03
3003    ---------------------
3005     1. There has been some re-arrangement of the code for the match() function so
3006        that it can be compiled in a version that does not call itself recursively.
3007        Instead, it keeps those local variables that need separate instances for
3008        each "recursion" in a frame on the heap, and gets/frees frames whenever it
3009        needs to "recurse". Keeping track of where control must go is done by means
3010        of setjmp/longjmp. The whole thing is implemented by a set of macros that
3011        hide most of the details from the main code, and operates only if
3012        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
3013        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
3014        operating.
3016        To make it easier for callers to provide specially tailored get/free
3017        functions for this usage, two new functions, pcre_stack_malloc, and
3018        pcre_stack_free, are used. They are always called in strict stacking order,
3019        and the size of block requested is always the same.
3021        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
3022        PCRE has been compiled to use the stack or the heap for recursion. The
3023        -C option of pcretest uses this to show which version is compiled.
3025        A new data escape \S, is added to pcretest; it causes the amounts of store
3026        obtained and freed by both kinds of malloc/free at match time to be added
3027        to the output.
3029     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
3030        what's available on my current Linux desktop machine.
3032     3. When matching a UTF-8 string, the test for a valid string at the start has
3033        been extended. If start_offset is not zero, PCRE now checks that it points
3034        to a byte that is the start of a UTF-8 character. If not, it returns
3035        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
3036        this is necessary because there may be backward assertions in the pattern.
3037        When matching the same subject several times, it may save resources to use
3038        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
3040     4. The code for checking the validity of UTF-8 strings has been tightened so
3041        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
3042        containing "overlong sequences".
3044     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
3045        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
3046        should have been "&", but it just so happened that all the cases this let
3047        through by mistake were picked up later in the function.
3049     6. I had used a variable called "isblank" - this is a C99 function, causing
3050        some compilers to warn. To avoid this, I renamed it (as "blankclass").
3052     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
3053        prompting; (b) run "./pcretest /dev/null" at the start of the test script
3054        so the version is shown; (c) stop "make test" echoing "./RunTest".
3056     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
3058     9. The prototype for memmove() for systems that don't have it was using
3059        size_t, but the inclusion of the header that defines size_t was later. I've
3060        moved the #includes for the C headers earlier to avoid this.
3062    10. Added some adjustments to the code to make it easier to compiler on certain
3063        special systems:
3065          (a) Some "const" qualifiers were missing.
3066          (b) Added the macro EXPORT before all exported functions; by default this
3067              is defined to be empty.
3068          (c) Changed the dftables auxiliary program (that builds chartables.c) so
3069              that it reads its output file name as an argument instead of writing
3070              to the standard output and assuming this can be redirected.
3072    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
3073        class containing characters with values greater than 255, PCRE compilation
3074        went into a loop.
3076    12. A recursive reference to a subpattern that was within another subpattern
3077        that had a minimum quantifier of zero caused PCRE to crash. For example,
3078        (x(y(?2))z)? provoked this bug with a subject that got as far as the
3079        recursion. If the recursively-called subpattern itself had a zero repeat,
3080        that was OK.
3082    13. In pcretest, the buffer for reading a data line was set at 30K, but the
3083        buffer into which it was copied (for escape processing) was still set at
3084        1024, so long lines caused crashes.
3086    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
3087        "internal error: code overflow...". This applied to any character class
3088        that was followed by a possessive quantifier.
3090    15. Modified the Makefile to add libpcre.la as a prerequisite for
3091        libpcreposix.la because I was told this is needed for a parallel build to
3092        work.
3094    16. If a pattern that contained .* following optional items at the start was
3095        studied, the wrong optimizing data was generated, leading to matching
3096        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
3097        matching string must start with a or b or c. The correct conclusion for
3098        this pattern is that a match can start with any character.
3101    Version 4.4 13-Aug-03
3102    ---------------------
3104     1. In UTF-8 mode, a character class containing characters with values between
3105        127 and 255 was not handled correctly if the compiled pattern was studied.
3106        In fixing this, I have also improved the studying algorithm for such
3107        classes (slightly).
3109     2. Three internal functions had redundant arguments passed to them. Removal
3110        might give a very teeny performance improvement.
3112     3. Documentation bug: the value of the capture_top field in a callout is *one
3113        more than* the number of the hightest numbered captured substring.
3115     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
3116        in incorrectly linking with a previously installed version. They now link
3117        explicitly with libpcre.la.
3119     5. configure.in no longer needs to recognize Cygwin specially.
3121     6. A problem in pcre.in for Windows platforms is fixed.
3123     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
3124        pcretest, it used to include the size of the study block as part of its
3125        output. Unfortunately, the structure contains a field that has a different
3126        size on different hardware architectures. This meant that the tests that
3127        showed this size failed. As the block is currently always of a fixed size,
3128        this information isn't actually particularly useful in pcretest output, so
3129        I have just removed it.
3131     8. Three pre-processor statements accidentally did not start in column 1.
3132        Sadly, there are *still* compilers around that complain, even though
3133        standard C has not required this for well over a decade. Sigh.
3135     9. In pcretest, the code for checking callouts passed small integers in the
3136        callout_data field, which is a void * field. However, some picky compilers
3137        complained about the casts involved for this on 64-bit systems. Now
3138        pcretest passes the address of the small integer instead, which should get
3139        rid of the warnings.
3141    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
3142        both compile and run time, and gives an error if an invalid UTF-8 sequence
3143        is found. There is a option for disabling this check in cases where the
3144        string is known to be correct and/or the maximum performance is wanted.
3146    11. In response to a bug report, I changed one line in Makefile.in from
3148            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
3149        to
3150            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
3152        to look similar to other lines, but I have no way of telling whether this
3153        is the right thing to do, as I do not use Windows. No doubt I'll get told
3154        if it's wrong...
3157    Version 4.3 21-May-03
3158    ---------------------
3160    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
3161       Makefile.
3163    2. Some refactoring to improve the quality of the code:
3165       (i)   The utf8_table... variables are now declared "const".
3167       (ii)  The code for \cx, which used the "case flipping" table to upper case
3168             lower case letters, now just substracts 32. This is ASCII-specific,
3169             but the whole concept of \cx is ASCII-specific, so it seems
3170             reasonable.
3172       (iii) PCRE was using its character types table to recognize decimal and
3173             hexadecimal digits in the pattern. This is silly, because it handles
3174             only 0-9, a-f, and A-F, but the character types table is locale-
3175             specific, which means strange things might happen. A private
3176             table is now used for this - though it costs 256 bytes, a table is
3177             much faster than multiple explicit tests. Of course, the standard
3178             character types table is still used for matching digits in subject
3179             strings against \d.
3181       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
3182             ending in _t are). So I've renamed it as ESC_tee.
3184    3. The first argument for regexec() in the POSIX wrapper should have been
3185       defined as "const".
3187    4. Changed pcretest to use malloc() for its buffers so that they can be
3188       Electric Fenced for debugging.
3190    5. There were several places in the code where, in UTF-8 mode, PCRE would try
3191       to read one or more bytes before the start of the subject string. Often this
3192       had no effect on PCRE's behaviour, but in some circumstances it could
3193       provoke a segmentation fault.
3195    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
3196       to try to read one or more bytes before the start of the subject string.
3198    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
3199       UTF-8 support could misbehave in various ways if the subject string
3200       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
3201       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
3202       back over UTF-8 characters.)
3205    Version 4.2 14-Apr-03
3206    ---------------------
3208    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
3210    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
3211         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
3212         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
3213         [WIN_PREFIX]: new variable, "cyg" for Cygwin
3214         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
3215           and BUILD_EXEEXT
3216         Note: automatic setting of the BUILD variables is not yet working
3217         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
3218           compile-time but not at link-time
3219         [LINK]: use for linking executables only
3220         make different versions for Windows and non-Windows
3221         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
3222           libraries
3223         [LINK_FOR_BUILD]: new variable
3224         [OBJEXT]: use throughout
3225         [EXEEXT]: use throughout
3226         <winshared>: new target
3227         <wininstall>: new target
3228         <dftables.o>: use native compiler
3229         <dftables>: use native linker
3230         <install>: handle Windows platform correctly
3231         <clean>: ditto
3232         <check>: ditto
3233         copy DLL to top builddir before testing
3235       As part of these changes, -no-undefined was removed again. This was reported
3236       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
3237       in any case.
3239    3. Some tidies to get rid of compiler warnings:
3241       . In the match_data structure, match_limit was an unsigned long int, whereas
3242         match_call_count was an int. I've made them both unsigned long ints.
3244       . In pcretest the fact that a const uschar * doesn't automatically cast to
3245         a void * provoked a warning.
3247       . Turning on some more compiler warnings threw up some "shadow" variables
3248         and a few more missing casts.
3250    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
3251       option, a class that contained a single character with a value between 128
3252       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
3254    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
3255       option, a class that contained several characters, but with at least one
3256       whose value was between 128 and 255 caused PCRE to crash.
3259    Version 4.1 12-Mar-03
3260    ---------------------
3262    1. Compiling with gcc -pedantic found a couple of places where casts were
3263    needed, and a string in dftables.c that was longer than standard compilers are
3264    required to support.
3266    2. Compiling with Sun's compiler found a few more places where the code could
3267    be tidied up in order to avoid warnings.
3269    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
3270    first of these names is deprecated in the latest Autoconf in favour of the name
3271    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
3272    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
3273    analogy I have changed it to CFLAGS_FOR_BUILD.
3275    4. Added -no-undefined to the linking command in the Makefile, because this is
3276    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
3277    linking step for the pcreposix library.
3279    5. PCRE was failing to diagnose the case of two named groups with the same
3280    name.
3282    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
3283    literal character that is needed in the subject for a match, and scans along to
3284    ensure that it is present before embarking on the full matching process. This
3285    saves time in cases of nested unlimited repeats that are never going to match.
3286    Problem: the scan can take a lot of time if the subject is very long (e.g.
3287    megabytes), thus penalizing straightforward matches. It is now done only if the
3288    amount of subject to be scanned is less than 1000 bytes.
3290    7. A lesser problem with the same optimization is that it was recording the
3291    first character of an anchored pattern as "needed", thus provoking a search
3292    right along the subject, even when the first match of the pattern was going to
3293    fail. The "needed" character is now not set for anchored patterns, unless it
3294    follows something in the pattern that is of non-fixed length. Thus, it still
3295    fulfils its original purpose of finding quick non-matches in cases of nested
3296    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
3299    Version 4.0 17-Feb-03
3300    ---------------------
3302    1. If a comment in an extended regex that started immediately after a meta-item
3303    extended to the end of string, PCRE compiled incorrect data. This could lead to
3304    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
3306    2. Moved to autoconf 2.53 and libtool 1.4.2.
3308    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
3309    the special perltest8 script is no longer needed - all the tests can be run
3310    from a single perltest script.
3312    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
3313    by \s. It has now been removed in PCRE. This means it isn't recognized as
3314    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
3315    class [:space:] *does* include VT, thereby creating a mess.
3317    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
3318    space and tab.
3320    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
3321    its new features into the main test script, reducing the number of scripts.
3323    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
3324    were backward compatible, and made the (?i) apply to the whole pattern, as if
3325    /i were given. Now it behaves more logically, and applies the option setting
3326    only to what follows. PCRE has been changed to follow suit. However, if it
3327    finds options settings right at the start of the pattern, it extracts them into
3328    the global options, as before. Thus, they show up in the info data.
3330    8. Added support for the \Q...\E escape sequence. Characters in between are
3331    treated as literals. This is slightly different from Perl in that $ and @ are
3332    also handled as literals inside the quotes. In Perl, they will cause variable
3333    interpolation. Note the following examples:
3335        Pattern            PCRE matches      Perl matches
3337        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
3338        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
3339        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
3341    For compatibility with Perl, \Q...\E sequences are recognized inside character
3342    classes as well as outside them.
3344    9. Re-organized 3 code statements in pcretest to avoid "overflow in
3345    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
3346    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
3347    signed/unsigned warnings.
3349    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
3350    option for pcretest, so I've replaced it by a simple function that does just
3351    that job.
3353    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
3354    "pcregrep -".
3356    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
3357    Java package. This provides some syntactic sugar for simple cases of what my
3358    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
3359    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
3360    item, you can use this simplified notation. Note that only makes sense with
3361    greedy quantifiers. Consequently, the use of the possessive quantifier forces
3362    greediness, whatever the setting of the PCRE_UNGREEDY option.
3364    13. A change of greediness default within a pattern was not taking effect at
3365    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
3366    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
3367    was abstracted outside.
3369    14. PCRE now supports the \G assertion. It is true when the current matching
3370    position is at the start point of the match. This differs from \A when the
3371    starting offset is non-zero. Used with the /g option of pcretest (or similar
3372    code), it works in the same way as it does for Perl's /g option. If all
3373    alternatives of a regex begin with \G, the expression is anchored to the start
3374    match position, and the "anchored" flag is set in the compiled expression.
3376    15. Some bugs concerning the handling of certain option changes within patterns
3377    have been fixed. These applied to options other than (?ims). For example,
3378    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
3379    been the other way round. Some of this was related to change 7 above.
3381    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
3382    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
3383    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
3384    POSIX classes only within a class (e.g. /[[:alpha:]]/).
3386    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
3387    mode. Unlike ".", it always matches newline, whatever the setting of
3388    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
3389    assertions. Perl allows it, but it doesn't (in general) work because it can't
3390    calculate the length of the lookbehind. At least, that's the case for Perl
3391    5.8.0 - I've been told they are going to document that it doesn't work in
3392    future.
3394    18. Added an error diagnosis for escapes that PCRE does not support: these are
3395    \L, \l, \N, \P, \p, \U, \u, and \X.
3397    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
3398    reading past the end of the pattern in cases such as /[abcd/.
3400    20. PCRE was getting more memory than necessary for patterns with classes that
3401    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
3403    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
3404    compiling PCRE for use with Virtual Pascal.
3406    22. Small fix to the Makefile to make it work properly if the build is done
3407    outside the source tree.
3409    23. Added a new extension: a condition to go with recursion. If a conditional
3410    subpattern starts with (?(R) the "true" branch is used if recursion has
3411    happened, whereas the "false" branch is used only at the top level.
3413    24. When there was a very long string of literal characters (over 255 bytes
3414    without UTF support, over 250 bytes with UTF support), the computation of how
3415    much memory was required could be incorrect, leading to segfaults or other
3416    strange effects.
3418    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
3419    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
3420    there was a subsequent back reference to those brackets. This meant that, for
3421    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
3422    possible to check for precisely this case. All we can do is abandon the
3423    optimization if .* occurs inside capturing brackets when there are any back
3424    references whatsoever. (See below for a better fix that came later.)
3426    26. The handling of the optimization for finding the first character of a
3427    non-anchored pattern, and for finding a character that is required later in the
3428    match were failing in some cases. This didn't break the matching; it just
3429    failed to optimize when it could. The way this is done has been re-implemented.
3431    27. Fixed typo in error message for invalid (?R item (it said "(?p").
3433    28. Added a new feature that provides some of the functionality that Perl
3434    provides with (?{...}). The facility is termed a "callout". The way it is done
3435    in PCRE is for the caller to provide an optional function, by setting
3436    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
3437    global variable. By default it is unset, which disables all calling out. To get
3438    the function called, the regex must include (?C) at appropriate points. This
3439    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
3440    This provides a means of identifying different callout points. When PCRE
3441    reaches such a point in the regex, if pcre_callout has been set, the external
3442    function is called. It is provided with data in a structure called
3443    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
3444    matching continues; if it returns a non-zero value, the match at the current
3445    point fails. However, backtracking will occur if possible. [This was changed
3446    later and other features added - see item 49 below.]
3448    29. pcretest is upgraded to test the callout functionality. It provides a
3449    callout function that displays information. By default, it shows the start of
3450    the match and the current position in the text. There are some new data escapes
3451    to vary what happens:
3453        \C+         in addition, show current contents of captured substrings
3454        \C-         do not supply a callout function
3455        \C!n        return 1 when callout number n is reached
3456        \C!n!m      return 1 when callout number n is reached for the mth time
3458    30. If pcregrep was called with the -l option and just a single file name, it
3459    output "<stdin>" if a match was found, instead of the file name.
3461    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
3462    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
3463    pcre_exec(). This saves a malloc/free per call. The default value of
3464    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
3465    when configuring.
3467    32. The default maximum size of a compiled pattern is 64K. There have been a
3468    few cases of people hitting this limit. The code now uses macros to handle the
3469    storing of links as offsets within the compiled pattern. It defaults to 2-byte
3470    links, but this can be changed to 3 or 4 bytes by --with-link-size when
3471    configuring. Tests 2 and 5 work only with 2-byte links because they output
3472    debugging information about compiled patterns.
3474    33. Internal code re-arrangements:
3476    (a) Moved the debugging function for printing out a compiled regex into
3477        its own source file (printint.c) and used #include to pull it into
3478        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
3479        separate copies.
3481    (b) Defined the list of op-code names for debugging as a macro in
3482        internal.h so that it is next to the definition of the opcodes.
3484    (c) Defined a table of op-code lengths for simpler skipping along compiled
3485        code. This is again a macro in internal.h so that it is next to the
3486        definition of the opcodes.
3488    34. Added support for recursive calls to individual subpatterns, along the
3489    lines of Robin Houston's patch (but implemented somewhat differently).
3491    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
3492    allow it to read and process whole directories in Win32. This code was
3493    contributed by Lionel Fourquaux; it has not been tested by me.
3495    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
3496    used to name a group. Names consist of alphanumerics and underscores, and must
3497    be unique. Back references use the syntax (?P=name) and recursive calls use
3498    (?P>name) which is a PCRE extension to the Python extension. Groups still have
3499    numbers. The function pcre_fullinfo() can be used after compilation to extract
3500    a name/number map. There are three relevant calls:
3502      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
3503      PCRE_INFO_NAMECOUNT            yields the number of entries
3504      PCRE_INFO_NAMETABLE            yields a pointer to the map.
3506    The map is a vector of fixed-size entries. The size of each entry depends on
3507    the length of the longest name used. The first two bytes of each entry are the
3508    group number, most significant byte first. There follows the corresponding
3509    name, zero terminated. The names are in alphabetical order.
3511    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
3512    case instead of 255. Making it the same both with and without UTF-8 support
3513    means that the same test output works with both.
3515    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
3516    calling malloc() with a zero argument.
3518    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
3519    optimization. I've improved things by keeping a bitmap of backreferences with
3520    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
3521    fact referenced, the optimization can be applied. It is unlikely that a
3522    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
3523    the match to follow \n) will appear inside brackets with a number greater than
3524    31, but if it does, any back reference > 31 suppresses the optimization.
3526    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
3527    of disabling numbered capturing parentheses. Any opening parenthesis that is
3528    not followed by ? behaves as if it were followed by ?: but named parentheses
3529    can still be used for capturing (and they will acquire numbers in the usual
3530    way).
3532    41. Redesigned the return codes from the match() function into yes/no/error so
3533    that errors can be passed back from deep inside the nested calls. A malloc
3534    failure while inside a recursive subpattern call now causes the
3535    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
3537    42. It is now possible to set a limit on the number of times the match()
3538    function is called in a call to pcre_exec(). This facility makes it possible to
3539    limit the amount of recursion and backtracking, though not in a directly
3540    obvious way, because the match() function is used in a number of different
3541    circumstances. The count starts from zero for each position in the subject
3542    string (for non-anchored patterns). The default limit is, for compatibility, a
3543    large number, namely 10 000 000. You can change this in two ways:
3545    (a) When configuring PCRE before making, you can use --with-match-limit=n
3546        to set a default value for the compiled library.
3548    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
3549        a different value is set. See 45 below.
3551    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
3553    43. Added a new function pcre_config(int, void *) to enable run-time extraction
3554    of things that can be changed at compile time. The first argument specifies
3555    what is wanted and the second points to where the information is to be placed.
3556    The current list of available information is:
3558      PCRE_CONFIG_UTF8
3560    The output is an integer that is set to one if UTF-8 support is available;
3561    otherwise it is set to zero.
3565    The output is an integer that it set to the value of the code that is used for
3566    newline. It is either LF (10) or CR (13).
3570    The output is an integer that contains the number of bytes used for internal
3571    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
3575    The output is an integer that contains the threshold above which the POSIX
3576    interface uses malloc() for output vectors. See item 31 above.
3580    The output is an unsigned integer that contains the default limit of the number
3581    of match() calls in a pcre_exec() execution. See 42 above.
3583    44. pcretest has been upgraded by the addition of the -C option. This causes it
3584    to extract all the available output from the new pcre_config() function, and to
3585    output it. The program then exits immediately.
3587    45. A need has arisen to pass over additional data with calls to pcre_exec() in
3588    order to support additional features. One way would have been to define
3589    pcre_exec2() (for example) with extra arguments, but this would not have been
3590    extensible, and would also have required all calls to the original function to
3591    be mapped to the new one. Instead, I have chosen to extend the mechanism that
3592    is used for passing in "extra" data from pcre_study().
3594    The pcre_extra structure is now exposed and defined in pcre.h. It currently
3595    contains the following fields:
3597      flags         a bitmap indicating which of the following fields are set
3598      study_data    opaque data from pcre_study()
3599      match_limit   a way of specifying a limit on match() calls for a specific
3600                      call to pcre_exec()
3601      callout_data  data for callouts (see 49 below)
3603    The flag bits are also defined in pcre.h, and are
3609    The pcre_study() function now returns one of these new pcre_extra blocks, with
3610    the actual study data pointed to by the study_data field, and the
3611    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as