ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 77 by nigel, Sat Feb 24 21:40:45 2007 UTC revision 743 by ph10, Tue Nov 8 09:59:38 2011 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
4    Version 8.21
5    ------------
7    1.  Updating the JIT compiler.
9    2.  JIT compiler now supports OP_NCREF, OP_RREF and OP_NRREF. New test cases
10        are added as well.
12    3.  Fix cache-flush issue on PowerPC (It is still an experimental JIT port).
13        PCRE_EXTRA_TABLES is not suported by JIT, and should be checked before
14        calling _pcre_jit_exec. Some extra comments are added.
16    4.  Mark settings inside atomic groups that do not contain any capturing
17        parentheses, for example, (?>a(*:m)), were not being passed out. This bug
18        was introduced by change 18 for 8.20.
21    Version 8.20 21-Oct-2011
22    ------------------------
24    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
25        a POSIX class. After further experiments with Perl, which convinced me that
26        Perl has bugs and confusions, a closing square bracket is no longer allowed
27        in a POSIX name. This bug also affected patterns with classes that started
28        with full stops.
30    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
31        captured substring, but while checking the failing first alternative,
32        substring 1 is temporarily captured. If the output vector supplied to
33        pcre_exec() was not big enough for this capture, the yield of the function
34        was still zero ("insufficient space for captured substrings"). This cannot
35        be totally fixed without adding another stack variable, which seems a lot
36        of expense for a edge case. However, I have improved the situation in cases
37        such as /(a)(b)x|abc/ matched against "abc", where the return code
38        indicates that fewer than the maximum number of slots in the ovector have
39        been set.
41    3.  Related to (2) above: when there are more back references in a pattern than
42        slots in the output vector, pcre_exec() uses temporary memory during
43        matching, and copies in the captures as far as possible afterwards. It was
44        using the entire output vector, but this conflicts with the specification
45        that only 2/3 is used for passing back captured substrings. Now it uses
46        only the first 2/3, for compatibility. This is, of course, another edge
47        case.
49    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
50        main code base, and can be used by building with --enable-jit. When this is
51        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
52        runtime --no-jit option is given.
54    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
55        ovector, the return from the function was zero, implying that there were
56        other matches that did not fit. The correct "exactly full" value is now
57        returned.
59    6.  If a subpattern that was called recursively or as a subroutine contained
60        (*PRUNE) or any other control that caused it to give a non-standard return,
61        invalid errors such as "Error -26 (nested recursion at the same subject
62        position)" or even infinite loops could occur.
64    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
65        computing the minimum length on reaching *ACCEPT, and so ended up with the
66        wrong value of 1 rather than 0. Further investigation indicates that
67        computing a minimum subject length in the presence of *ACCEPT is difficult
68        (think back references, subroutine calls), and so I have changed the code
69        so that no minimum is registered for a pattern that contains *ACCEPT.
71    8.  If (*THEN) was present in the first (true) branch of a conditional group,
72        it was not handled as intended. [But see 16 below.]
74    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
75        Sheri Pierce.
77    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
78        the first byte in a match must be "a".
80    11. Change 17 for 8.13 increased the recursion depth for patterns like
81        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
82        pattern contains any instances of (*THEN). If it does not, the old
83        optimizations are restored. It would be nice to do this on a per-group
84        basis, but at the moment that is not feasible.
86    12. In some environments, the output of pcretest -C is CRLF terminated. This
87        broke RunTest's code that checks for the link size. A single white space
88        character after the value is now allowed for.
90    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
91        For "fr", it uses the Windows-specific input and output files.
93    14. If (*THEN) appeared in a group that was called recursively or as a
94        subroutine, it did not work as intended. [But see next item.]
96    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
97        pattern fragments (but not containing any | characters). If A and B are
98        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
99        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
100        D. In other words, Perl considers parentheses that do not contain any |
101        characters to be part of a surrounding alternative, whereas PCRE was
102        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
103        differently. PCRE now behaves in the same way as Perl, except in the case
104        of subroutine/recursion calls such as (?1) which have in any case always
105        been different (but PCRE had them first :-).
107    16. Related to 15 above: Perl does not treat the | in a conditional group as
108        creating alternatives. Such a group is treated in the same way as an
109        ordinary group without any | characters when processing (*THEN). PCRE has
110        been changed to match Perl's behaviour.
112    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
113        RunGrepTest script failed.
115    18. Change 22 for version 13 caused atomic groups to use more stack. This is
116        inevitable for groups that contain captures, but it can lead to a lot of
117        stack use in large patterns. The old behaviour has been restored for atomic
118        groups that do not contain any capturing parentheses.
120    19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not
121        suppress the check for a minimum subject length at run time. (If it was
122        given to pcre_exec() or pcre_dfa_exec() it did work.)
124    20. Fixed an ASCII-dependent infelicity in pcretest that would have made it
125        fail to work when decoding hex characters in data strings in EBCDIC
126        environments.
128    21. It appears that in at least one Mac OS environment, the isxdigit() function
129        is implemented as a macro that evaluates to its argument more than once,
130        contravening the C 90 Standard (I haven't checked a later standard). There
131        was an instance in pcretest which caused it to go wrong when processing
132        \x{...} escapes in subject strings. The has been rewritten to avoid using
133        things like p++ in the argument of isxdigit().
136    Version 8.13 16-Aug-2011
137    ------------------------
139    1.  The Unicode data tables have been updated to Unicode 6.0.0.
141    2.  Two minor typos in pcre_internal.h have been fixed.
143    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
144        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
145        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
147    4.  There were a number of related bugs in the code for matching backrefences
148        caselessly in UTF-8 mode when codes for the characters concerned were
149        different numbers of bytes. For example, U+023A and U+2C65 are an upper
150        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
151        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
152        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
153        2-byte code at the end of the subject (it thought there wasn't enough data
154        left).
156    5.  Comprehensive information about what went wrong is now returned by
157        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
158        as the output vector has at least 2 elements. The offset of the start of
159        the failing character and a reason code are placed in the vector.
161    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
162        now returned is for the first byte of the failing character, instead of the
163        last byte inspected. This is an incompatible change, but I hope it is small
164        enough not to be a problem. It makes the returned offset consistent with
165        pcre_exec() and pcre_dfa_exec().
167    7.  pcretest now gives a text phrase as well as the error number when
168        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
169        failure, the offset and reason code are output.
171    8.  When \R was used with a maximizing quantifier it failed to skip backwards
172        over a \r\n pair if the subsequent match failed. Instead, it just skipped
173        back over a single character (\n). This seems wrong (because it treated the
174        two characters as a single entity when going forwards), conflicts with the
175        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
176        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
177        has been changed.
179    9.  Some internal refactoring has changed the processing so that the handling
180        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
181        time (the PCRE_DOTALL option was changed this way some time ago: version
182        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
183        which was always a bit of a fudge. It also means that there is one less
184        argument for the match() function, which reduces its stack requirements
185        slightly. This change also fixes an incompatibility with Perl: the pattern
186        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
188    10. More internal refactoring has drastically reduced the number of recursive
189        calls to match() for possessively repeated groups such as (abc)++ when
190        using pcre_exec().
192    11. While implementing 10, a number of bugs in the handling of groups were
193        discovered and fixed:
195        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
196        (a|)*(?1) gave a compile-time internal error.
197        ((a|)+)+  did not notice that the outer group could match an empty string.
198        (^a|^)+   was not marked as anchored.
199        (.*a|.*)+ was not marked as matching at start or after a newline.
201    12. Yet more internal refactoring has removed another argument from the match()
202        function. Special calls to this function are now indicated by setting a
203        value in a variable in the "match data" data block.
205    13. Be more explicit in pcre_study() instead of relying on "default" for
206        opcodes that mean there is no starting character; this means that when new
207        ones are added and accidentally left out of pcre_study(), testing should
208        pick them up.
210    14. The -s option of pcretest has been documented for ages as being an old
211        synonym of -m (show memory usage). I have changed it to mean "force study
212        for every regex", that is, assume /S for every regex. This is similar to -i
213        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
214        using it. It makes it easier to run collections of tests with and without
215        study enabled, and thereby test pcre_study() more easily. All the standard
216        tests are now run with and without -s (but some patterns can be marked as
217        "never study" - see 20 below).
219    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
220        restoration of the capturing data to the outer values was not happening
221        correctly.
223    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
224        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
225        pattern had matched an empty string, and so incorrectly returned a no
226        match.
228    17. There was optimizing code for the last branch of non-capturing parentheses,
229        and also for the obeyed branch of a conditional subexpression, which used
230        tail recursion to cut down on stack usage. Unfortunately, now that there is
231        the possibility of (*THEN) occurring in these branches, tail recursion is
232        no longer possible because the return has to be checked for (*THEN). These
233        two optimizations have therefore been removed. [But see 8.20/11 above.]
235    18. If a pattern containing \R was studied, it was assumed that \R always
236        matched two bytes, thus causing the minimum subject length to be
237        incorrectly computed because \R can also match just one byte.
239    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
240        was incorrectly computed.
242    20. If /S is present twice on a test pattern in pcretest input, it now
243        *disables* studying, thereby overriding the use of -s on the command line
244        (see 14 above). This is necessary for one or two tests to keep the output
245        identical in both cases.
247    21. When (*ACCEPT) was used in an assertion that matched an empty string and
248        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
250    22. When an atomic group that contained a capturing parenthesis was
251        successfully matched, but the branch in which it appeared failed, the
252        capturing was not being forgotten if a higher numbered group was later
253        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
254        group 1 to "a", when in fact it should be unset. This applied to multi-
255        branched capturing and non-capturing groups, repeated or not, and also to
256        positive assertions (capturing in negative assertions does not happen
257        in PCRE) and also to nested atomic groups.
259    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
260        subject after a captured substring, to make it easier to tell which of a
261        number of identical substrings has been captured.
263    24. The way atomic groups are processed by pcre_exec() has been changed so that
264        if they are repeated, backtracking one repetition now resets captured
265        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
266        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
267        "aaa". Previously, it would have been "a". As part of this code
268        refactoring, the way recursive calls are handled has also been changed.
270    25. If an assertion condition captured any substrings, they were not passed
271        back unless some other capturing happened later. For example, if
272        (?(?=(a))a) was matched against "a", no capturing was returned.
274    26. When studying a pattern that contained subroutine calls or assertions,
275        the code for finding the minimum length of a possible match was handling
276        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
277        group 1 called group 2 while simultaneously a separate group 2 called group
278        1). A stack overflow occurred in this case. I have fixed this by limiting
279        the recursion depth to 10.
281    27. Updated RunTest.bat in the distribution to the version supplied by Tom
282        Fortmann. This supports explicit test numbers on the command line, and has
283        argument validation and error reporting.
285    28. An instance of \X with an unlimited repeat could fail if at any point the
286        first character it looked at was a mark character.
288    29. Some minor code refactoring concerning Unicode properties and scripts
289        should reduce the stack requirement of match() slightly.
291    30. Added the '=' option to pcretest to check the setting of unused capturing
292        slots at the end of the pattern, which are documented as being -1, but are
293        not included in the return count.
295    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
296        compiled something random. Now it gives a compile-time error (as does
297        Perl).
299    32. A *MARK encountered during the processing of a positive assertion is now
300        recorded and passed back (compatible with Perl).
302    33. If --only-matching or --colour was set on a pcregrep call whose pattern
303        had alternative anchored branches, the search for a second match in a line
304        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
305        matched the line "0102" twice. The same bug affected patterns that started
306        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
307        twice.
309    34. Previously, PCRE did not allow quantification of assertions. However, Perl
310        does, and because of capturing effects, quantifying parenthesized
311        assertions may at times be useful. Quantifiers are now allowed for
312        parenthesized assertions.
314    35. A minor code tidy in pcre_compile() when checking options for \R usage.
316    36. \g was being checked for fancy things in a character class, when it should
317        just be a literal "g".
319    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
320        appearance of a nested POSIX class supersedes an apparent external class.
321        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
322        unescaped square brackets may also appear as part of class names. For
323        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
324        more like Perl. (But see 8.20/1 above.)
326    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
327        was because it thought it was \N{name}, which is not supported).
329    39. Add minix to OS list not supporting the -S option in pcretest.
331    40. PCRE tries to detect cases of infinite recursion at compile time, but it
332        cannot analyze patterns in sufficient detail to catch mutual recursions
333        such as ((?1))((?2)). There is now a runtime test that gives an error if a
334        subgroup is called recursively as a subpattern for a second time at the
335        same position in the subject string. In previous releases this might have
336        been caught by the recursion limit, or it might have run out of stack.
338    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
339        happen only once. PCRE was, however incorrectly giving a compile time error
340        "recursive call could loop indefinitely" because it cannot analyze the
341        pattern in sufficient detail. The compile time test no longer happens when
342        PCRE is compiling a conditional subpattern, but actual runaway loops are
343        now caught at runtime (see 40 above).
345    42. It seems that Perl allows any characters other than a closing parenthesis
346        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
347        has been changed to be the same.
349    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
350        as not to get warnings when autogen.sh is called. Also changed
351        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
353    44. To help people who use pcregrep to scan files containing exceedingly long
354        lines, the following changes have been made:
356        (a) The default value of the buffer size parameter has been increased from
357            8K to 20K. (The actual buffer used is three times this size.)
359        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
360            PCRE is built.
362        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
363            to be set at run time.
365        (d) Numerical values in pcregrep options can be followed by K or M, for
366            example --buffer-size=50K.
368        (e) If a line being scanned overflows pcregrep's buffer, an error is now
369            given and the return code is set to 2.
371    45. Add a pointer to the latest mark to the callout data block.
373    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
374        partial match of an empty string instead of no match. This was specific to
375        the use of ".".
377    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
378        complete match instead of a partial match. This bug was dependent on both
379        the PCRE_UTF8 and PCRE_DOTALL options being set.
381    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
382        starting byte set, because \b was not being ignored.
385    Version 8.12 15-Jan-2011
386    ------------------------
388    1.  Fixed some typos in the markup of the man pages, and wrote a script that
389        checks for such things as part of the documentation building process.
391    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
392        --match-limit and --recursion-limit options (added for 8.11). In
393        particular, this made one of the standard tests fail. (The integer value
394        went into the wrong half of a long int.)
396    3.  If the --colour option was given to pcregrep with -v (invert match), it
397        did strange things, either producing crazy output, or crashing. It should,
398        of course, ignore a request for colour when reporting lines that do not
399        match.
401    4.  Another pcregrep bug caused similar problems if --colour was specified with
402        -M (multiline) and the pattern match finished with a line ending.
404    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
405        matched in multiline mode, the following line was shown as part of the
406        match. This seems wrong, so I have changed it.
408    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
409        the check for further matches in the same line (so they could be coloured)
410        to overrun the end of the current line. If another match was found, it was
411        incorrectly shown (and then shown again when found in the next line).
413    7.  If pcregrep was compiled under Windows, there was a reference to the
414        function pcregrep_exit() before it was defined. I am assuming this was
415        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
416        reported by a user. I've moved the definition above the reference.
419    Version 8.11 10-Dec-2010
420    ------------------------
422    1.  (*THEN) was not working properly if there were untried alternatives prior
423        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
424        backtracked to try for "b" instead of moving to the next alternative branch
425        at the same level (in this case, to look for "c"). The Perl documentation
426        is clear that when (*THEN) is backtracked onto, it goes to the "next
427        alternative in the innermost enclosing group".
429    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
430        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
431        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
432        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
433        (*THEN).
435    3.  If \s appeared in a character class, it removed the VT character from
436        the class, even if it had been included by some previous item, for example
437        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
438        of \s, but is part of the POSIX "space" class.)
440    4.  A partial match never returns an empty string (because you can always
441        match an empty string at the end of the subject); however the checking for
442        an empty string was starting at the "start of match" point. This has been
443        changed to the "earliest inspected character" point, because the returned
444        data for a partial match starts at this character. This means that, for
445        example, /(?<=abc)def/ gives a partial match for the subject "abc"
446        (previously it gave "no match").
448    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
449        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
450        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
451        has an implication that the given string is incomplete (because a partial
452        match is preferred over a full match). For this reason, these items now
453        give a partial match in this situation. [Aside: previously, the one case
454        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
455        match rather than a full match, which was wrong by the old rules, but is
456        now correct.]
458    6.  There was a bug in the handling of #-introduced comments, recognized when
459        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
460        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
461        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
462        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
463        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
464        places in pcre_compile().
466    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
467        comments when looking ahead for named forward references to subpatterns,
468        the only newline sequence it recognized was NL. It now handles newlines
469        according to the set newline convention.
471    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
472        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
473        cater for a lack of strerror(). These oversights have been fixed.
475    9.  Added --match-limit and --recursion-limit to pcregrep.
477    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
479    11. When the -o option was used, pcregrep was setting a return code of 1, even
480        when matches were found, and --line-buffered was not being honoured.
482    12. Added an optional parentheses number to the -o and --only-matching options
483        of pcregrep.
485    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
486        can match an empty string. The code to do it in pcretest and pcredemo
487        needed fixing:
489        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
490            only one byte after an empty string match just before CRLF (this case
491            just got forgotten; "any" and "anycrlf" were OK).
493        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
494            mode when an empty string match preceded an ASCII character followed by
495            a non-ASCII character. (The code for advancing by one character rather
496            than one byte was nonsense.)
498        (c) The pcredemo.c sample program did not have any code at all to handle
499            the cases when CRLF is a valid newline sequence.
501    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
502        as a starting offset was within the subject string. There is now a new
503        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
504        negative or greater than the length of the string. In order to test this,
505        pcretest is extended to allow the setting of negative starting offsets.
507    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
508        starting offset points to the beginning of a UTF-8 character was
509        unnecessarily clumsy. I tidied it up.
511    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
512        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
514    17. Nobody had reported that the --include_dir option, which was added in
515        release 7.7 should have been called --include-dir (hyphen, not underscore)
516        for compatibility with GNU grep. I have changed it to --include-dir, but
517        left --include_dir as an undocumented synonym, and the same for
518        --exclude-dir, though that is not available in GNU grep, at least as of
519        release 2.5.4.
521    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
522        characters from a string of bytes) have been redefined so as not to use
523        loops, in order to improve performance in some environments. At the same
524        time, I abstracted some of the common code into auxiliary macros to save
525        repetition (this should not affect the compiled code).
527    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
528        compile-time error is now given if \c is not followed by an ASCII
529        character, that is, a byte less than 128. (In EBCDIC mode, the code is
530        different, and any byte value is allowed.)
532    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
533        START_OPTIMIZE option, which is now allowed at compile time - but just
534        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
535        to pcregrep and other applications that have no direct access to PCRE
536        options. The new /Y option in pcretest sets this option when calling
537        pcre_compile().
539    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
540        back references. Groups containing recursive back references were forced to
541        be atomic by that change, but in the case of named groups, the amount of
542        memory required was incorrectly computed, leading to "Failed: internal
543        error: code overflow". This has been fixed.
545    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
546        pcretest.c, to avoid build problems in some Borland environments.
549    Version 8.10 25-Jun-2010
550    ------------------------
552    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
553        THEN.
555    2.  (*ACCEPT) was not working when inside an atomic group.
557    3.  Inside a character class, \B is treated as a literal by default, but
558        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
559        causes the error). The code is unchanged, but I tidied the documentation.
561    4.  Inside a character class, PCRE always treated \R and \X as literals,
562        whereas Perl faults them if its -w option is set. I have changed PCRE so
563        that it faults them when PCRE_EXTRA is set.
565    5.  Added support for \N, which always matches any character other than
566        newline. (It is the same as "." when PCRE_DOTALL is not set.)
568    6.  When compiling pcregrep with newer versions of gcc which may have
569        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
570        declared with attribute warn_unused_result" were given. Just casting the
571        result to (void) does not stop the warnings; a more elaborate fudge is
572        needed. I've used a macro to implement this.
574    7.  Minor change to pcretest.c to avoid a compiler warning.
576    8.  Added four artifical Unicode properties to help with an option to make
577        \s etc use properties (see next item). The new properties are: Xan
578        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
580    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
581        use Unicode properties. (*UCP) at the start of a pattern can be used to set
582        this option. Modified pcretest to add /W to test this facility. Added
583        REG_UCP to make it available via the POSIX interface.
585    10. Added --line-buffered to pcregrep.
587    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
588        studied, and the match started with a letter with a code point greater than
589        127 whose first byte was different to the first byte of the other case of
590        the letter, the other case of this starting letter was not recognized
591        (#976).
593    12. If a pattern that was studied started with a repeated Unicode property
594        test, for example, \p{Nd}+, there was the theoretical possibility of
595        setting up an incorrect bitmap of starting bytes, but fortunately it could
596        not have actually happened in practice until change 8 above was made (it
597        added property types that matched character-matching opcodes).
599    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
600        possible starting bytes for non-anchored patterns.
602    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
603        \R, and also a number of cases that involve Unicode properties, both
604        explicit and implicit when PCRE_UCP is set.
606    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
607        input, it could crash or give wrong results if characters with values
608        greater than 0xc0 were present in the subject string. (Detail: it assumed
609        UTF-8 input when processing these items.)
611    16. Added a lot of (int) casts to avoid compiler warnings in systems where
612        size_t is 64-bit (#991).
614    17. Added a check for running out of memory when PCRE is compiled with
615        --disable-stack-for-recursion (#990).
617    18. If the last data line in a file for pcretest does not have a newline on
618        the end, a newline was missing in the output.
620    19. The default pcre_chartables.c file recognizes only ASCII characters (values
621        less than 128) in its various bitmaps. However, there is a facility for
622        generating tables according to the current locale when PCRE is compiled. It
623        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
624        characters, are recognized by isspace() and therefore were getting set in
625        these tables, and indeed these tables seem to approximate to ISO 8859. This
626        caused a problem in UTF-8 mode when pcre_study() was used to create a list
627        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
628        which of course cannot start UTF-8 characters. I have changed the code so
629        that only real ASCII characters (less than 128) and the correct starting
630        bytes for UTF-8 encodings are set for characters greater than 127 when in
631        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
632        altogether.)
634    20. Added the /T option to pcretest so as to be able to run tests with non-
635        standard character tables, thus making it possible to include the tests
636        used for 19 above in the standard set of tests.
638    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
639        reference to a subpattern the other side of a comment that contains an
640        opening parenthesis caused either an internal compiling error, or a
641        reference to the wrong subpattern.
644    Version 8.02 19-Mar-2010
645    ------------------------
647    1.  The Unicode data tables have been updated to Unicode 5.2.0.
649    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
650        configured.
652    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
653        original author of that file, following a query about its status.
655    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
656        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
658    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
659        quantifier applied to a forward-referencing subroutine call, could compile
660        incorrect code or give the error "internal error: previously-checked
661        referenced subpattern not found".
663    6.  Both MS Visual Studio and Symbian OS have problems with initializing
664        variables to point to external functions. For these systems, therefore,
665        pcre_malloc etc. are now initialized to local functions that call the
666        relevant global functions.
668    7.  There were two entries missing in the vectors called coptable and poptable
669        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
670        I've fixed the data, and added a kludgy way of testing at compile time that
671        the lengths are correct (equal to the number of opcodes).
673    8.  Following on from 7, I added a similar kludge to check the length of the
674        eint vector in pcreposix.c.
676    9.  Error texts for pcre_compile() are held as one long string to avoid too
677        much relocation at load time. To find a text, the string is searched,
678        counting zeros. There was no check for running off the end of the string,
679        which could happen if a new error number was added without updating the
680        string.
682    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
684    11. \K was not working if it appeared in an atomic group or in a group that
685        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
686        \K is "not well defined" if used in an assertion. PCRE now accepts it if
687        the assertion is positive, but not if it is negative.
689    12. Change 11 fortuitously reduced the size of the stack frame used in the
690        "match()" function of pcre_exec.c by one pointer. Forthcoming
691        implementation of support for (*MARK) will need an extra pointer on the
692        stack; I have reserved it now, so that the stack frame size does not
693        decrease.
695    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
696        item in branch that calls a recursion is a subroutine call - as in the
697        second branch in the above example - was incorrectly given the compile-
698        time error "recursive call could loop indefinitely" because pcre_compile()
699        was not correctly checking the subroutine for matching a non-empty string.
701    14. The checks for overrunning compiling workspace could trigger after an
702        overrun had occurred. This is a "should never occur" error, but it can be
703        triggered by pathological patterns such as hundreds of nested parentheses.
704        The checks now trigger 100 bytes before the end of the workspace.
706    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
709    Version 8.01 19-Jan-2010
710    ------------------------
712    1.  If a pattern contained a conditional subpattern with only one branch (in
713        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
714        computed the wrong minimum data length (which is of course zero for such
715        subpatterns). This could cause incorrect "no match" results.
717    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
718        the pattern is reset in the first branch, pcre_compile() failed with
719        "internal error: code overflow at offset...". This happened only when
720        the reset was to the original external option setting. (An optimization
721        abstracts leading options settings into an external setting, which was the
722        cause of this.)
724    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
725        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
726        assertion pattern did not match (meaning that the assertion was true), it
727        was incorrectly treated as false if the SKIP had been reached during the
728        matching. This also applied to assertions used as conditions.
730    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
731        assertion subpattern, including such a pattern used as a condition,
732        unpredictable results occurred, instead of the error return
735    5.  The C++ GlobalReplace function was not working like Perl for the special
736        situation when an empty string is matched. It now does the fancy magic
737        stuff that is necessary.
739    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
740        removed. (These were left over from very, very early versions of PCRE.)
742    7.  Some cosmetic changes to the code to make life easier when compiling it
743        as part of something else:
745        (a) Change DEBUG to PCRE_DEBUG.
747        (b) In pcre_compile(), rename the member of the "branch_chain" structure
748            called "current" as "current_branch", to prevent a collision with the
749            Linux macro when compiled as a kernel module.
751        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
752            prevent a collision with the Linux macro when compiled as a kernel
753            module.
755    8.  In pcre_compile() there are some checks for integer overflows that used to
756        cast potentially large values to (double). This has been changed to that
757        when building, a check for int64_t is made, and if it is found, it is used
758        instead, thus avoiding the use of floating point arithmetic. (There is no
759        other use of FP in PCRE.) If int64_t is not found, the fallback is to
760        double.
762    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
763        2005 (difference between two addresses compared to an unsigned value).
765    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
766        custom one, because of the following reported problem in Windows:
768          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
769              under Win32.
770          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
771              therefore missing the function definition.
772          - The compiler thus generates a "C" signature for the test function.
773          - The linker fails to find the "C" function.
774          - PCRE fails to configure if asked to do so against libbz2.
776    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
777        messages were output:
779          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
780          rerunning libtoolize, to keep the correct libtool macros in-tree.
781          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
783        I have done both of these things.
785    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
786        most of the time, it *can* run out if it is given a pattern that contains a
787        runaway infinite recursion. I updated the discussion in the pcrestack man
788        page.
790    13. Now that we have gone to the x.xx style of version numbers, the minor
791        version may start with zero. Using 08 or 09 is a bad idea because users
792        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
793        interpreted as invalid octal numbers. I've updated the previous comment in
794        configure.ac, and also added a check that gives an error if 08 or 09 are
795        used.
797    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
798        causing partial matching to fail when the end of the subject matched \W
799        in a UTF-8 pattern where \W was quantified with a minimum of 3.
801    15. There were some discrepancies between the declarations in pcre_internal.h
802        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
803        their definitions. The declarations used "const uschar *" and the
804        definitions used USPTR. Even though USPTR is normally defined as "const
805        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
806        reported that: "This difference in casting confuses some C++ compilers, for
807        example, SunCC recognizes above declarations as different functions and
808        generates broken code for hbpcre." I have changed the declarations to use
809        USPTR.
811    16. GNU libtool is named differently on some systems. The autogen.sh script now
812        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
813        (FreeBSD).
815    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
816        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
817        comment: "Figure out how to create a longlong from a string: strtoll and
818        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
819        instance, but it only takes 2 args instead of 3!"
821    18. A subtle bug concerned with back references has been fixed by a change of
822        specification, with a corresponding code fix. A pattern such as
823        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
824        refers, was giving matches when it shouldn't. For example, xa=xaaa would
825        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
826        same bug. Such groups have to be quantified to be useful, or contained
827        inside another quantified group. (If there's no repetition, the reference
828        can never match.) The problem arises because, having left the group and
829        moved on to the rest of the pattern, a later failure that backtracks into
830        the group uses the captured value from the final iteration of the group
831        rather than the correct earlier one. I have fixed this in PCRE by forcing
832        any group that contains a reference to itself to be an atomic group; that
833        is, there cannot be any backtracking into it once it has completed. This is
834        similar to recursive and subroutine calls.
837    Version 8.00 19-Oct-09
838    ----------------------
840    1.  The table for translating pcre_compile() error codes into POSIX error codes
841        was out-of-date, and there was no check on the pcre_compile() error code
842        being within the table. This could lead to an OK return being given in
843        error.
845    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
846        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
847        in a Windows environment.
849    3.  The pcregrep --count option prints the count for each file even when it is
850        zero, as does GNU grep. However, pcregrep was also printing all files when
851        --files-with-matches was added. Now, when both options are given, it prints
852        counts only for those files that have at least one match. (GNU grep just
853        prints the file name in this circumstance, but including the count seems
854        more useful - otherwise, why use --count?) Also ensured that the
855        combination -clh just lists non-zero counts, with no names.
857    4.  The long form of the pcregrep -F option was incorrectly implemented as
858        --fixed_strings instead of --fixed-strings. This is an incompatible change,
859        but it seems right to fix it, and I didn't think it was worth preserving
860        the old behaviour.
862    5.  The command line items --regex=pattern and --regexp=pattern were not
863        recognized by pcregrep, which required --regex pattern or --regexp pattern
864        (with a space rather than an '='). The man page documented the '=' forms,
865        which are compatible with GNU grep; these now work.
867    6.  No libpcreposix.pc file was created for pkg-config; there was just
868        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
870    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
871        when UCP support is not needed, by modifying the Python script that
872        generates it from Unicode data files. This should not matter if the module
873        is correctly used as a library, but I received one complaint about 50K of
874        unwanted data. My guess is that the person linked everything into his
875        program rather than using a library. Anyway, it does no harm.
877    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
878        was a minimum greater than 1 for a wide character in a possessive
879        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
880        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
881        character. Chaos in the form of incorrect output or a compiling loop could
882        result.
884    9.  The restrictions on what a pattern can contain when partial matching is
885        requested for pcre_exec() have been removed. All patterns can now be
886        partially matched by this function. In addition, if there are at least two
887        slots in the offset vector, the offset of the earliest inspected character
888        for the match and the offset of the end of the subject are set in them when
889        PCRE_ERROR_PARTIAL is returned.
891    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
892        synonymous with PCRE_PARTIAL, for backwards compatibility, and
893        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
894        and may be more useful for multi-segment matching.
896    11. Partial matching with pcre_exec() is now more intuitive. A partial match
897        used to be given if ever the end of the subject was reached; now it is
898        given only if matching could not proceed because another character was
899        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
900        string "Z", which now yields "no match" instead of "partial match". In the
901        case of pcre_dfa_exec(), "no match" is given if every matching path for the
902        final character ended with (*FAIL).
904    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
905        if the pattern had a "must contain" character that was already found in the
906        earlier partial match, unless partial matching was again requested. For
907        example, with the pattern /dog.(body)?/, the "must contain" character is
908        "g". If the first part-match was for the string "dog", restarting with
909        "sbody" failed. This bug has been fixed.
911    13. The string returned by pcre_dfa_exec() after a partial match has been
912        changed so that it starts at the first inspected character rather than the
913        first character of the match. This makes a difference only if the pattern
914        starts with a lookbehind assertion or \b or \B (\K is not supported by
915        pcre_dfa_exec()). It's an incompatible change, but it makes the two
916        matching functions compatible, and I think it's the right thing to do.
918    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
919        so that the demonstration program is easily available in environments where
920        PCRE has not been installed from source.
922    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
923        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
924        library.
926    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
927        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
928        is not the first non-POSIX option to be added. Clearly some people find
929        these options useful.
931    17. If a caller to the POSIX matching function regexec() passes a non-zero
932        value for nmatch with a NULL value for pmatch, the value of
933        nmatch is forced to zero.
935    18. RunGrepTest did not have a test for the availability of the -u option of
936        the diff command, as RunTest does. It now checks in the same way as
937        RunTest, and also checks for the -b option.
939    19. If an odd number of negated classes containing just a single character
940        interposed, within parentheses, between a forward reference to a named
941        subpattern and the definition of the subpattern, compilation crashed with
942        an internal error, complaining that it could not find the referenced
943        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
944        [The bug was that it was starting one character too far in when skipping
945        over the character class, thus treating the ] as data rather than
946        terminating the class. This meant it could skip too much.]
948    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
949        /g option in pcretest when the pattern contains \K, which makes it possible
950        to have an empty string match not at the start, even when the pattern is
951        anchored. Updated pcretest and pcredemo to use this option.
953    21. If the maximum number of capturing subpatterns in a recursion was greater
954        than the maximum at the outer level, the higher number was returned, but
955        with unset values at the outer level. The correct (outer level) value is
956        now given.
958    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
959        PCRE did not set those parentheses (unlike Perl). I have now found a way to
960        make it do so. The string so far is captured, making this feature
961        compatible with Perl.
963    23. The tests have been re-organized, adding tests 11 and 12, to make it
964        possible to check the Perl 5.10 features against Perl 5.10.
966    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
967        pattern matches a fixed length string. PCRE did not allow this; now it
968        does. Neither allows recursion.
970    25. I finally figured out how to implement a request to provide the minimum
971        length of subject string that was needed in order to match a given pattern.
972        (It was back references and recursion that I had previously got hung up
973        on.) This code has now been added to pcre_study(); it finds a lower bound
974        to the length of subject needed. It is not necessarily the greatest lower
975        bound, but using it to avoid searching strings that are too short does give
976        some useful speed-ups. The value is available to calling programs via
977        pcre_fullinfo().
979    26. While implementing 25, I discovered to my embarrassment that pcretest had
980        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
981        study optimizations had never been tested with that matching function.
982        Oops. What is worse, even when it was passed study data, there was a bug in
983        pcre_dfa_exec() that meant it never actually used it. Double oops. There
984        were also very few tests of studied patterns with pcre_dfa_exec().
986    27. If (?| is used to create subpatterns with duplicate numbers, they are now
987        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
988        on the other side of the coin, they are no longer allowed to have different
989        names, because these cannot be distinguished in PCRE, and this has caused
990        confusion. (This is a difference from Perl.)
992    28. When duplicate subpattern names are present (necessarily with different
993        numbers, as required by 27 above), and a test is made by name in a
994        conditional pattern, either for a subpattern having been matched, or for
995        recursion in such a pattern, all the associated numbered subpatterns are
996        tested, and the overall condition is true if the condition is true for any
997        one of them. This is the way Perl works, and is also more like the way
998        testing by number works.
1001    Version 7.9 11-Apr-09
1002    ---------------------
1004    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
1005        (pcretest), all targets were linked against these libraries. This included
1006        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
1007        libraries. This caused unwanted dependencies to be created. This problem
1008        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
1009        pcretest is linked with readline.
1011    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
1012        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
1013        moved outside it again, because FALSE and TRUE are already defined in AIX,
1014        but BOOL is not.
1016    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
1017        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
1019    4.  The pcregrep documentation said spaces were inserted as well as colons (or
1020        hyphens) following file names and line numbers when outputting matching
1021        lines. This is not true; no spaces are inserted. I have also clarified the
1022        wording for the --colour (or --color) option.
1024    5.  In pcregrep, when --colour was used with -o, the list of matching strings
1025        was not coloured; this is different to GNU grep, so I have changed it to be
1026        the same.
1028    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
1029        each matching line was coloured. Now it goes on to look for further matches
1030        of any of the test patterns, which is the same behaviour as GNU grep.
1032    7.  A pattern that could match an empty string could cause pcregrep to loop; it
1033        doesn't make sense to accept an empty string match in pcregrep, so I have
1034        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
1035        seems to be how GNU grep behaves.
1037    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
1038        start or after a newline", because the conditional assertion was not being
1039        correctly handled. The rule now is that both the assertion and what follows
1040        in the first alternative must satisfy the test.
1042    9.  If auto-callout was enabled in a pattern with a conditional group whose
1043        condition was an assertion, PCRE could crash during matching, both with
1044        pcre_exec() and pcre_dfa_exec().
1046    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
1047        used for matching.
1049    11. Unicode property support in character classes was not working for
1050        characters (bytes) greater than 127 when not in UTF-8 mode.
1052    12. Added the -M command line option to pcretest.
1054    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
1056    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
1058    16. Added comments and documentation about mis-use of no_arg in the C++
1059        wrapper.
1061    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
1062        from Martin Jerabek that uses macro names for all relevant character and
1063        string constants.
1065    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
1066        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
1067        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
1068        these, but not everybody uses configure.
1070    19. A conditional group that had only one branch was not being correctly
1071        recognized as an item that could match an empty string. This meant that an
1072        enclosing group might also not be so recognized, causing infinite looping
1073        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
1074        with the subject "ab", where knowledge that the repeated group can match
1075        nothing is needed in order to break the loop.
1077    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
1078        exec(), but without supplying a callout function, matching went wrong.
1080    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
1081        leak if the size of the offset vector was greater than 30. When the vector
1082        is smaller, the saved offsets during recursion go onto a local stack
1083        vector, but for larger vectors malloc() is used. It was failing to free
1084        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
1085        error, in fact).
1087    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
1088        heapframe that is used only when UTF-8 support is enabled. This caused no
1089        problem, but was untidy.
1091    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
1092        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
1093        included within another project.
1095    24. Steven Van Ingelgem's patches to add more options to the CMake support,
1096        slightly modified by me:
1098          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
1099              not building pcregrep.
1101          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
1102              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
1104    25. Forward references, both numeric and by name, in patterns that made use of
1105        duplicate group numbers, could behave incorrectly or give incorrect errors,
1106        because when scanning forward to find the reference group, PCRE was not
1107        taking into account the duplicate group numbers. A pattern such as
1108        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
1110    26. Changed a few more instances of "const unsigned char *" to USPTR, making
1111        the feature of a custom pointer more persuasive (as requested by a user).
1113    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
1114        pcretest.c, inside #ifndefs, because it seems they are sometimes already
1115        pre-defined.
1117    28. Added support for (*UTF8) at the start of a pattern.
1119    29. Arrange for flags added by the "release type" setting in CMake to be shown
1120        in the configuration summary.
1123    Version 7.8 05-Sep-08
1124    ---------------------
1126    1.  Replaced UCP searching code with optimized version as implemented for Ad
1127        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
1128        stage table and inline lookup instead of a function, giving speed ups of 2
1129        to 5 times on some simple patterns that I tested. Permission was given to
1130        distribute the MultiStage2.py script that generates the tables (it's not in
1131        the tarball, but is in the Subversion repository).
1133    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
1134        scripts.
1136    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
1137        a group with a zero qualifier. The result of the study could be incorrect,
1138        or the function might crash, depending on the pattern.
1140    4.  Caseless matching was not working for non-ASCII characters in back
1141        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
1142        It now works when Unicode Property Support is available.
1144    5.  In pcretest, an escape such as \x{de} in the data was always generating
1145        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
1146        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
1147        truncation.
1149    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
1151    7.  Added two (int) casts to pcregrep when printing the difference of two
1152        pointers, in case they are 64-bit values.
1154    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
1155        test 2 if it fails.
1157    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
1158        and a #define of that name to empty if it is not externally set. This is to
1159        allow users of MSVC to set it if necessary.
1161    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
1162        the convenience functions in the pcre_get.c source file.
1164    11. An option change at the start of a pattern that had top-level alternatives
1165        could cause overwriting and/or a crash. This command provoked a crash in
1166        some environments:
1168          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
1170        This potential security problem was recorded as CVE-2008-2371.
1172    12. For a pattern where the match had to start at the beginning or immediately
1173        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
1174        pcre_dfa_exec() could read past the end of the passed subject if there was
1175        no match. To help with detecting such bugs (e.g. with valgrind), I modified
1176        pcretest so that it places the subject at the end of its malloc-ed buffer.
1178    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
1179        exec() might read past the end of the data buffer in UTF-8 mode.
1181    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
1182        the data contained the byte 0x85 as part of a UTF-8 character within its
1183        first line. This applied both to normal and DFA matching.
1185    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
1186        /^[^d]*?$/8 failed to match "abc".
1188    16. Added a missing copyright notice to pcrecpp_internal.h.
1190    17. Make it more clear in the documentation that values returned from
1191        pcre_exec() in ovector are byte offsets, not character counts.
1193    18. Tidied a few places to stop certain compilers from issuing warnings.
1195    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
1196        supplied by Stefan Weber. I made a further small update for 7.8 because
1197        there is a change of source arrangements: the pcre_searchfuncs.c module is
1198        replaced by pcre_ucd.c.
1201    Version 7.7 07-May-08
1202    ---------------------
1204    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
1205        a string to a long long, pretend we don't even have a long long." This is
1206        done by checking for the strtoq, strtoll, and _strtoi64 functions.
1208    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
1209        pre-7.6 versions, which defined a global no_arg variable instead of putting
1210        it in the RE class. (See also #8 below.)
1212    3.  Remove a line of dead code, identified by coverity and reported by Nuno
1213        Lopes.
1215    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
1217        (1) The include/exclude patterns were being applied to the whole pathnames
1218            of files, instead of just to the final components.
1220        (2) If there was more than one level of directory, the subdirectories were
1221            skipped unless they satisfied the include/exclude conditions. This is
1222            inconsistent with GNU grep (and could even be seen as contrary to the
1223            pcregrep specification - which I improved to make it absolutely clear).
1224            The action now is always to scan all levels of directory, and just
1225            apply the include/exclude patterns to regular files.
1227    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
1228        --exclude_dir in the tests to avoid scanning .svn directories.
1230    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
1231        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
1232        doesn't support NULs in patterns.
1234    7.  Added some missing "const"s to declarations of static tables in
1235        pcre_compile.c and pcre_dfa_exec.c.
1237    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
1238        caused by fix #2  above. (Subsequently also a second patch to fix the
1239        first patch. And a third patch - this was a messy problem.)
1241    9.  Applied Craig's patch to remove the use of push_back().
1243    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
1244        matching function regexec().
1246    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
1247        which, however, unlike Perl's \g{...}, are subroutine calls, not back
1248        references. PCRE supports relative numbers with this syntax (I don't think
1249        Oniguruma does).
1251    12. Previously, a group with a zero repeat such as (...){0} was completely
1252        omitted from the compiled regex. However, this means that if the group
1253        was called as a subroutine from elsewhere in the pattern, things went wrong
1254        (an internal error was given). Such groups are now left in the compiled
1255        pattern, with a new opcode that causes them to be skipped at execution
1256        time.
1258    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
1259        to the way PCRE behaves:
1261        (a) A lone ] character is dis-allowed (Perl treats it as data).
1263        (b) A back reference to an unmatched subpattern matches an empty string
1264            (Perl fails the current match path).
1266        (c) A data ] in a character class must be notated as \] because if the
1267            first data character in a class is ], it defines an empty class. (In
1268            Perl it is not possible to have an empty class.) The empty class []
1269            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
1270            The negative empty class [^] matches any one character, independently
1271            of the DOTALL setting.
1273    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
1274        non-existent subpattern following a character class starting with ']' and
1275        containing () gave an internal compiling error instead of "reference to
1276        non-existent subpattern". Fortunately, when the pattern did exist, the
1277        compiled code was correct. (When scanning forwards to check for the
1278        existencd of the subpattern, it was treating the data ']' as terminating
1279        the class, so got the count wrong. When actually compiling, the reference
1280        was subsequently set up correctly.)
1282    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
1283        it was being rejected as not supported by pcre_dfa_exec(), even though
1284        other assertions are supported. I have made pcre_dfa_exec() support
1285        (*FAIL).
1287    16. The implementation of 13c above involved the invention of a new opcode,
1288        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
1289        cannot be changed at match time, I realized I could make a small
1290        improvement to matching performance by compiling OP_ALLANY instead of
1291        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
1292        on the OP_ANY path.
1294    17. Compiling pcretest on Windows with readline support failed without the
1295        following two fixes: (1) Make the unistd.h include conditional on
1296        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
1298    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
1299        ncurses library to be included for pcretest when ReadLine support is
1300        requested, but also to allow for it to be overridden. This patch came from
1301        Daniel Bergström.
1303    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
1304        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
1305        any errors with the current Unicode tables. Thanks to Peter Kankowski for
1306        spotting this.
1309    Version 7.6 28-Jan-08
1310    ---------------------
1312    1.  A character class containing a very large number of characters with
1313        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
1314        overflow.
1316    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
1317        HAVE_LONG_LONG is not defined.
1319    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
1320        bring them up to date and include new features. This patch includes:
1322        - Fixed PH's badly added libz and libbz2 support.
1323        - Fixed a problem with static linking.
1324        - Added pcredemo. [But later removed - see 7 below.]
1325        - Fixed dftables problem and added an option.
1326        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
1327            HAVE_LONG_LONG.
1328        - Added readline support for pcretest.
1329        - Added an listing of the option settings after cmake has run.
1331    4.  A user submitted a patch to Makefile that makes it easy to create
1332        "pcre.dll" under mingw when using Configure/Make. I added stuff to
1333        Makefile.am that cause it to include this special target, without
1334        affecting anything else. Note that the same mingw target plus all
1335        the other distribution libraries and programs are now supported
1336        when configuring with CMake (see 6 below) instead of with
1337        Configure/Make.
1339    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
1340        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
1341        exported in the Windows port". It has not yet been confirmed that the patch
1342        solves the problem, but it does no harm.
1344    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
1345        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
1346        with CMake, and also correct the comment about stack recursion.
1348    7.  Remove the automatic building of pcredemo from the ./configure system and
1349        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
1350        of a program that users should build themselves after PCRE is installed, so
1351        building it automatically is not really right. What is more, it gave
1352        trouble in some build environments.
1354    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
1357    Version 7.5 10-Jan-08
1358    ---------------------
1360    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
1361        values in parens when parsing an RE using the C++ wrapper."
1363    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
1364        Characters greater than 255 were excluded from the class instead of being
1365        included.
1367    3.  The same bug as (2) above applied to negated POSIX classes such as
1368        [:^space:].
1370    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
1371        defined or documented. It seems to have been a typo for PCRE_STATIC, so
1372        I have changed it.
1374    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
1375        first named subpattern) and a construct such as (?&a) would reference the
1376        first named subpattern whose name started with "a" (in other words, the
1377        length check was missing). Both these problems are fixed. "Subpattern name
1378        expected" is now given for (?&) (a zero-length name), and this patch also
1379        makes it give the same error for \k'' (previously it complained that that
1380        was a reference to a non-existent subpattern).
1382    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
1383        this is right because (?- can be followed by option settings as well as by
1384        digits. I have, however, made the messages clearer.
1386    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
1387        than the number used in the conditional) now cause a compile-time error.
1388        This is actually not compatible with Perl, which accepts such patterns, but
1389        treats the conditional as always being FALSE (as PCRE used to), but it
1390        seems to me that giving a diagnostic is better.
1392    8.  Change "alphameric" to the more common word "alphanumeric" in comments
1393        and messages.
1395    9.  Fix two occurrences of "backslash" in comments that should have been
1396        "backspace".
1398    10. Remove two redundant lines of code that can never be obeyed (their function
1399        was moved elsewhere).
1401    11. The program that makes PCRE's Unicode character property table had a bug
1402        which caused it to generate incorrect table entries for sequences of
1403        characters that have the same character type, but are in different scripts.
1404        It amalgamated them into a single range, with the script of the first of
1405        them. In other words, some characters were in the wrong script. There were
1406        thirteen such cases, affecting characters in the following ranges:
1408          U+002b0 - U+002c1
1409          U+0060c - U+0060d
1410          U+0061e - U+00612
1411          U+0064b - U+0065e
1412          U+0074d - U+0076d
1413          U+01800 - U+01805
1414          U+01d00 - U+01d77
1415          U+01d9b - U+01dbf
1416          U+0200b - U+0200f
1417          U+030fc - U+030fe
1418          U+03260 - U+0327f
1419          U+0fb46 - U+0fbb1
1420          U+10450 - U+1049d
1422    12. The -o option (show only the matching part of a line) for pcregrep was not
1423        compatible with GNU grep in that, if there was more than one match in a
1424        line, it showed only the first of them. It now behaves in the same way as
1425        GNU grep.
1427    13. If the -o and -v options were combined for pcregrep, it printed a blank
1428        line for every non-matching line. GNU grep prints nothing, and pcregrep now
1429        does the same. The return code can be used to tell if there were any
1430        non-matching lines.
1432    14. Added --file-offsets and --line-offsets to pcregrep.
1434    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
1435        infinitely looping recursion. The bug was that positive lookaheads were not
1436        being skipped when checking for a possible empty match (negative lookaheads
1437        and both kinds of lookbehind were skipped).
1439    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
1440        inclusion of <windows.h> to before rather than after the definition of
1441        INVALID_FILE_ATTRIBUTES (patch from David Byron).
1443    17. Specifying a possessive quantifier with a specific limit for a Unicode
1444        character property caused pcre_compile() to compile bad code, which led at
1445        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
1446        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
1447        caused the error; without that there was no problem.
1449    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
1451    19. Added --enable-pcretest-libreadline.
1453    20. In pcrecpp.cc, the variable 'count' was incremented twice in
1454        RE::GlobalReplace(). As a result, the number of replacements returned was
1455        double what it should be. I removed one of the increments, but Craig sent a
1456        later patch that removed the other one (the right fix) and added unit tests
1457        that check the return values (which was not done before).
1459    21. Several CMake things:
1461        (1) Arranged that, when cmake is used on Unix, the libraries end up with
1462            the names libpcre and libpcreposix, not just pcre and pcreposix.
1464        (2) The above change means that pcretest and pcregrep are now correctly
1465            linked with the newly-built libraries, not previously installed ones.
1469    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
1470        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
1471        UTF-8 newline character). The key issue is that the pattern starts .*;
1472        this means that the match must be either at the beginning, or after a
1473        newline. The bug was in the code for advancing after a failed match and
1474        checking that the new position followed a newline. It was not taking
1475        account of UTF-8 characters correctly.
1477    23. PCRE was behaving differently from Perl in the way it recognized POSIX
1478        character classes. PCRE was not treating the sequence [:...:] as a
1479        character class unless the ... were all letters. Perl, however, seems to
1480        allow any characters between [: and :], though of course it rejects as
1481        unknown any "names" that contain non-letters, because all the known class
1482        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
1483        for example, whereas PCRE did not - it did not recognize a POSIX character
1484        class. This seemed a bit dangerous, so the code has been changed to be
1485        closer to Perl. The behaviour is not identical to Perl, because PCRE will
1486        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
1487        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
1488        Perl does, and where it didn't before.
1490    24. Rewrite so as to remove the single use of %n from pcregrep because in some
1491        Windows environments %n is disabled by default.
1494    Version 7.4 21-Sep-07
1495    ---------------------
1497    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
1498        means that a class such as [\s] counted as "explicit reference to CR or
1499        LF". That isn't really right - the whole point of the change was to try to
1500        help when there was an actual mention of one of the two characters. So now
1501        the change happens only if \r or \n (or a literal CR or LF) character is
1502        encountered.
1504    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
1505        of both had grown to the point where there were only 3 bits left.
1506        Fortunately, there was spare space in the data structure, and so I have
1507        moved the internal flags into a new 16-bit field to free up more option
1508        bits.
1510    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
1511        but did not set the internal JCHANGED flag - either of these is enough to
1512        control the way the "get" function works - but the PCRE_INFO_JCHANGED
1513        facility is supposed to tell if (?J) was ever used, so now (?J) at the
1514        start sets both bits.
1516    4.  Added options (at build time, compile time, exec time) to change \R from
1517        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
1519    5.  doc/pcresyntax.html was missing from the distribution.
1521    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
1522        compatibility, even though it is no longer used.
1524    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
1525        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
1526        windows.h file is present (where different names are used). [This was
1527        reversed later after testing - see 16 below.]
1529    8.  Changed all #include <config.h> to #include "config.h". There were also
1530        some further <pcre.h> cases that I changed to "pcre.h".
1532    9.  When pcregrep was used with the --colour option, it missed the line ending
1533        sequence off the lines that it output.
1535    10. It was pointed out to me that arrays of string pointers cause lots of
1536        relocations when a shared library is dynamically loaded. A technique of
1537        using a single long string with a table of offsets can drastically reduce
1538        these. I have refactored PCRE in four places to do this. The result is
1539        dramatic:
1541          Originally:                          290
1542          After changing UCP table:            187
1543          After changing error message table:   43
1544          After changing table of "verbs"       36
1545          After changing table of Posix names   22
1547        Thanks to the folks working on Gregex for glib for this insight.
1549    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
1550        unicode-properties was also set.
1552    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
1554    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
1555        checked only for CRLF.
1557    14. Added casts to pcretest.c to avoid compiler warnings.
1559    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
1561    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
1562        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
1563        entirely. This removes changes made in 7 above.
1565    17. The CMake files have been updated, and there is now more information about
1566        building with CMake in the NON-UNIX-USE document.
1569    Version 7.3 28-Aug-07
1570    ---------------------
1572     1. In the rejigging of the build system that eventually resulted in 7.1, the
1573        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
1574        brackets there is not right, since it causes compilers to look for an
1575        installed pcre.h, not the version that is in the source that is being
1576        compiled (which of course may be different). I have changed it back to:
1578          #include "pcre.h"
1580        I have a vague recollection that the change was concerned with compiling in
1581        different directories, but in the new build system, that is taken care of
1582        by the VPATH setting the Makefile.
1584     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
1585        when the subject happened to end in the byte 0x85 (e.g. if the last
1586        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
1587        characters but of course it shouldn't be taken as a newline when it is part
1588        of another character. The bug was that, for an unlimited repeat of . in
1589        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
1590        characters when looking for a newline.
1592     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
1594     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
1595        in debug output.
1597     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
1598        long printing in the pcrecpp unittest when running under MinGW.
1600     6. ESC_K was left out of the EBCDIC table.
1602     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
1603        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
1604        limit also applies to "virtual nesting" when a pattern is recursive, and in
1605        this case 1000 isn't so big. I have been able to remove this limit at the
1606        expense of backing off one optimization in certain circumstances. Normally,
1607        when pcre_exec() would call its internal match() function recursively and
1608        immediately return the result unconditionally, it uses a "tail recursion"
1609        feature to save stack. However, when a subpattern that can match an empty
1610        string has an unlimited repetition quantifier, it no longer makes this
1611        optimization. That gives it a stack frame in which to save the data for
1612        checking that an empty string has been matched. Previously this was taken
1613        from the 1000-entry workspace that had been reserved. So now there is no
1614        explicit limit, but more stack is used.
1616     8. Applied Daniel's patches to solve problems with the import/export magic
1617        syntax that is required for Windows, and which was going wrong for the
1618        pcreposix and pcrecpp parts of the library. These were overlooked when this
1619        problem was solved for the main library.
1621     9. There were some crude static tests to avoid integer overflow when computing
1622        the size of patterns that contain repeated groups with explicit upper
1623        limits. As the maximum quantifier is 65535, the maximum group length was
1624        set at 30,000 so that the product of these two numbers did not overflow a
1625        32-bit integer. However, it turns out that people want to use groups that
1626        are longer than 30,000 bytes (though not repeat them that many times).
1627        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1628        made it possible to implement the integer overflow checks in a much more
1629        dynamic way, which I have now done. The artificial limitation on group
1630        length has been removed - we now have only the limit on the total length of
1631        the compiled pattern, which depends on the LINK_SIZE setting.
1633    10. Fixed a bug in the documentation for get/copy named substring when
1634        duplicate names are permitted. If none of the named substrings are set, the
1635        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1636        empty string.
1638    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1639        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1640        because the ] is interpreted as the first data character and the
1641        terminating ] is not found. PCRE has been made compatible with Perl in this
1642        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1643        cause memory overwriting.
1645    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1646        string has been matched (to stop an infinite loop). It was not recognizing
1647        a conditional subpattern that could match an empty string if that
1648        subpattern was within another subpattern. For example, it looped when
1649        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
1650        condition was not nested. This bug has been fixed.
1652    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1653        past the start of the subject in the presence of bytes with the top bit
1654        set, for example "\x8aBCD".
1656    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1657        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1659    14. Optimized (?!) to (*FAIL).
1661    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1662        This restricts code points to be within the range 0 to 0x10FFFF, excluding
1663        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1664        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1665        does: it's just the validity check that is more restrictive.
1667    16. Inserted checks for integer overflows during escape sequence (backslash)
1668        processing, and also fixed erroneous offset values for syntax errors during
1669        backslash processing.
1671    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1672        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1674    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1675        caused an overrun.
1677    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1678        something other than just ASCII characters) inside a group that had an
1679        unlimited repeat caused a loop at compile time (while checking to see
1680        whether the group could match an empty string).
1682    20. Debugging a pattern containing \p or \P could cause a crash. For example,
1683        [\P{Any}] did so. (Error in the code for printing property names.)
1685    21. An orphan \E inside a character class could cause a crash.
1687    22. A repeated capturing bracket such as (A)? could cause a wild memory
1688        reference during compilation.
1690    23. There are several functions in pcre_compile() that scan along a compiled
1691        expression for various reasons (e.g. to see if it's fixed length for look
1692        behind). There were bugs in these functions when a repeated \p or \P was
1693        present in the pattern. These operators have additional parameters compared
1694        with \d, etc, and these were not being taken into account when moving along
1695        the compiled data. Specifically:
1697        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1698            length.
1700        (b) An item such as \pL+ within a repeated group could cause crashes or
1701            loops.
1703        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1704            "reference to non-existent subpattern" error.
1706        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1708    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1709        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1711    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1713    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1714        character were causing crashes (broken optimization).
1716    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1717        \p or \P) caused a compile-time loop.
1719    28. More problems have arisen in unanchored patterns when CRLF is a valid line
1720        break. For example, the unstudied pattern [\r\n]A does not match the string
1721        "\r\nA" because change 7.0/46 below moves the current point on by two
1722        characters after failing to match at the start. However, the pattern \nA
1723        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
1724        the same is true. There doesn't seem any very clean way out of this, but
1725        what I have chosen to do makes the common cases work: PCRE now takes note
1726        of whether there can be an explicit match for \r or \n anywhere in the
1727        pattern, and if so, 7.0/46 no longer applies. As part of this change,
1728        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1729        pattern has explicit CR or LF references.
1731    29. Added (*CR) etc for changing newline setting at start of pattern.
1734    Version 7.2 19-Jun-07
1735    ---------------------
1737     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
1738        which is apparently normally available under Windows.
1740     2. Re-jig the pcregrep tests with different newline settings in an attempt
1741        to make them independent of the local environment's newline setting.
1743     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
1745     4. Some of the "internals" tests were previously cut out when the link size
1746        was not 2, because the output contained actual offsets. The recent new
1747        "Z" feature of pcretest means that these can be cut out, making the tests
1748        usable with all link sizes.
1750     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
1751        stack recursion. This gives a massive performance boost under BSD, but just
1752        a small improvement under Linux. However, it saves one field in the frame
1753        in all cases.
1755     6. Added more features from the forthcoming Perl 5.10:
1757        (a) (?-n) (where n is a string of digits) is a relative subroutine or
1758            recursion call. It refers to the nth most recently opened parentheses.
1760        (b) (?+n) is also a relative subroutine call; it refers to the nth next
1761            to be opened parentheses.
1763        (c) Conditions that refer to capturing parentheses can be specified
1764            relatively, for example, (?(-2)... or (?(+3)...
1766        (d) \K resets the start of the current match so that everything before
1767            is not part of it.
1769        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1771        (f) \g{name} is another synonym - part of Perl 5.10's unification of
1772            reference syntax.
1774        (g) (?| introduces a group in which the numbering of parentheses in each
1775            alternative starts with the same number.
1777        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
1779     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
1782     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
1783        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
1784        for detecting groups that can match an empty string.
1786     9. A pattern with a very large number of alternatives (more than several
1787        hundred) was running out of internal workspace during the pre-compile
1788        phase, where pcre_compile() figures out how much memory will be needed. A
1789        bit of new cunning has reduced the workspace needed for groups with
1790        alternatives. The 1000-alternative test pattern now uses 12 bytes of
1791        workspace instead of running out of the 4096 that are available.
1793    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
1795    11. Applied patch from Google to remove an optimization that didn't quite work.
1796        The report of the bug said:
1798          pcrecpp::RE("a*").FullMatch("aaa") matches, while
1799          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
1800          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
1802    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
1803        it matched the wrong number of bytes.
1806    Version 7.1 24-Apr-07
1807    ---------------------
1809     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
1810        that is more "standard", making use of automake and other Autotools. There
1811        is some re-arrangement of the files and adjustment of comments consequent
1812        on this.
1814     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
1815        for recursive directory scanning broke on some systems because the files
1816        are not scanned in any specific order and on different systems the order
1817        was different. A call to "sort" has been inserted into RunGrepTest for the
1818        approprate test as a short-term fix. In the longer term there may be an
1819        alternative.
1821     3. I had an email from Eric Raymond about problems translating some of PCRE's
1822        man pages to HTML (despite the fact that I distribute HTML pages, some
1823        people do their own conversions for various reasons). The problems
1824        concerned the use of low-level troff macros .br and .in. I have therefore
1825        removed all such uses from the man pages (some were redundant, some could
1826        be replaced by .nf/.fi pairs). The 132html script that I use to generate
1827        HTML has been updated to handle .nf/.fi and to complain if it encounters
1828        .br or .in.
1830     4. Updated comments in configure.ac that get placed in config.h.in and also
1831        arranged for config.h to be included in the distribution, with the name
1832        config.h.generic, for the benefit of those who have to compile without
1833        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
1835     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
1836        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
1837        makevp.bat for the current PCRE, using the additional files
1838        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
1840     6. A Windows user reported a minor discrepancy with test 2, which turned out
1841        to be caused by a trailing space on an input line that had got lost in his
1842        copy. The trailing space was an accident, so I've just removed it.
1844     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
1845        that is needed.
1847     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
1848        as "const" (a) because they are and (b) because it helps the PHP
1849        maintainers who have recently made a script to detect big data structures
1850        in the php code that should be moved to the .rodata section. I remembered
1851        to update Builducptable as well, so it won't revert if ucptable.h is ever
1852        re-created.
1854     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
1855        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
1856        order to be able to cut out the UTF-8 tables in the latter when UTF-8
1857        support is not required. This saves 1.5-2K of code, which is important in
1858        some applications.
1860        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
1861        so as not to refer to the tables, even though these functions will never be
1862        called when UTF-8 support is disabled. Otherwise there are problems with a
1863        shared library.
1865    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
1867        (a) It was defining its arguments as char * instead of void *.
1869        (b) It was assuming that all moves were upwards in memory; this was true
1870            a long time ago when I wrote it, but is no longer the case.
1872        The emulated memove() is provided for those environments that have neither
1873        memmove() nor bcopy(). I didn't think anyone used it these days, but that
1874        is clearly not the case, as these two bugs were recently reported.
1876    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
1877        and Detrail to create the HTML documentation, the .txt form of the man
1878        pages, and it removes trailing spaces from listed files. It also creates
1879        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
1880        case, it wraps all the #defines with #ifndefs. This script should be run
1881        before "make dist".
1883    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1884        with Unicode property support.
1886        (a) For a maximizing quantifier, if the two different cases of the
1887            character were of different lengths in their UTF-8 codings (there are
1888            some cases like this - I found 11), and the matching function had to
1889            back up over a mixture of the two cases, it incorrectly assumed they
1890            were both the same length.
1892        (b) When PCRE was configured to use the heap rather than the stack for
1893            recursion during matching, it was not correctly preserving the data for
1894            the other case of a UTF-8 character when checking ahead for a match
1895            while processing a minimizing repeat. If the check also involved
1896            matching a wide character, but failed, corruption could cause an
1897            erroneous result when trying to check for a repeat of the original
1898            character.
1900    13. Some tidying changes to the testing mechanism:
1902        (a) The RunTest script now detects the internal link size and whether there
1903            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1904            values substituted by "configure". (The RunGrepTest script already did
1905            this for UTF-8.) The configure.ac script no longer substitutes the
1906            relevant variables.
1908        (b) The debugging options /B and /D in pcretest show the compiled bytecode
1909            with length and offset values. This means that the output is different
1910            for different internal link sizes. Test 2 is skipped for link sizes
1911            other than 2 because of this, bypassing the problem. Unfortunately,
1912            there was also a test in test 3 (the locale tests) that used /B and
1913            failed for link sizes other than 2. Rather than cut the whole test out,
1914            I have added a new /Z option to pcretest that replaces the length and
1915            offset values with spaces. This is now used to make test 3 independent
1916            of link size. (Test 2 will be tidied up later.)
1918    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1919        segmentation fault instead of returning the appropriate error message.
1921    15. In multiline mode when the newline sequence was set to "any", the pattern
1922        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1923        This doesn't seem right; it now treats the CRLF combination as the line
1924        ending, and so does not match in that case. It's only a pattern such as ^$
1925        that would hit this one: something like ^ABC$ would have failed after \r
1926        and then tried again after \r\n.
1928    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1929        in an attempt to make files that differ only in their line terminators
1930        compare equal. This works on Linux.
1932    17. Under certain error circumstances pcregrep might try to free random memory
1933        as it exited. This is now fixed, thanks to valgrind.
1935    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1936        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1937        was because its rules for how to advance for /g after matching an empty
1938        string at the end of a line did not allow for this case. They now check for
1939        it specially.
1941    20. pcretest is supposed to handle patterns and data of any length, by
1942        extending its buffers when necessary. It was getting this wrong when the
1943        buffer for a data line had to be extended.
1945    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1946        CRLF as a newline sequence.
1948    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1949        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1950        I have nevertheless tidied it up.
1952    23. Added some casts to kill warnings from HP-UX ia64 compiler.
1954    24. Added a man page for pcre-config.
1957    Version 7.0 19-Dec-06
1958    ---------------------
1960     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1961        moving to gcc 4.1.1.
1963     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1964        sys/time.h, which is documented as needed for this function. It doesn't
1965        seem to matter on Linux, but it showed up on some releases of OS X.
1967     3. It seems that there are systems where bytes whose values are greater than
1968        127 match isprint() in the "C" locale. The "C" locale should be the
1969        default when a C program starts up. In most systems, only ASCII printing
1970        characters match isprint(). This difference caused the output from pcretest
1971        to vary, making some of the tests fail. I have changed pcretest so that:
1973        (a) When it is outputting text in the compiled version of a pattern, bytes
1974            other than 32-126 are always shown as hex escapes.
1976        (b) When it is outputting text that is a matched part of a subject string,
1977            it does the same, unless a different locale has been set for the match
1978            (using the /L modifier). In this case, it uses isprint() to decide.
1980     4. Fixed a major bug that caused incorrect computation of the amount of memory
1981        required for a compiled pattern when options that changed within the
1982        pattern affected the logic of the preliminary scan that determines the
1983        length. The relevant options are -x, and -i in UTF-8 mode. The result was
1984        that the computed length was too small. The symptoms of this bug were
1985        either the PCRE error "internal error: code overflow" from pcre_compile(),
1986        or a glibc crash with a message such as "pcretest: free(): invalid next
1987        size (fast)". Examples of patterns that provoked this bug (shown in
1988        pcretest format) are:
1990          /(?-x: )/x
1991          /(?x)(?-x: \s*#\s*)/
1992          /((?i)[\x{c0}])/8
1993          /(?i:[\x{c0}])/8
1995        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1996        is now done differently.
1998     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1999        wrapper classes; (b) implement a new function in the C++ scanner that is
2000        more efficient than the old way of doing things because it avoids levels of
2001        recursion in the regex matching; (c) add a paragraph to the documentation
2002        for the FullMatch() function.
2004     6. The escape sequence \n was being treated as whatever was defined as
2005        "newline". Not only was this contrary to the documentation, which states
2006        that \n is character 10 (hex 0A), but it also went horribly wrong when
2007        "newline" was defined as CRLF. This has been fixed.
2009     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
2010        was being set to -1 for the "end of line" case (supposedly a value that no
2011        character can have). Though this value is never used (the check for end of
2012        line is "zero bytes in current character"), it caused compiler complaints.
2013        I've changed it to 0xffffffff.
2015     8. In pcre_version.c, the version string was being built by a sequence of
2016        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
2017        string (as it is for production releases) called a macro with an empty
2018        argument. The C standard says the result of this is undefined. The gcc
2019        compiler treats it as an empty string (which was what was wanted) but it is
2020        reported that Visual C gives an error. The source has been hacked around to
2021        avoid this problem.
2023     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
2024        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
2025        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
2026        of them did).
2028    10. Originally, pcretest opened its input and output without "b"; then I was
2029        told that "b" was needed in some environments, so it was added for release
2030        5.0 to both the input and output. (It makes no difference on Unix-like
2031        systems.) Later I was told that it is wrong for the input on Windows. I've
2032        now abstracted the modes into two macros, to make it easier to fiddle with
2033        them, and removed "b" from the input mode under Windows.
2035    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
2037    12. Added -help and --help to pcretest as an official way of being reminded
2038        of the options.
2040    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
2041        and pcrecpp.cc because they annoy compilers at high warning levels.
2043    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
2045    15. Fixed an occurrence of == in configure.ac that should have been = (shell
2046        scripts are not C programs :-) and which was not noticed because it works
2047        on Linux.
2049    16. pcretest is supposed to handle any length of pattern and data line (as one
2050        line or as a continued sequence of lines) by extending its input buffer if
2051        necessary. This feature was broken for very long pattern lines, leading to
2052        a string of junk being passed to pcre_compile() if the pattern was longer
2053        than about 50K.
2055    17. I have done a major re-factoring of the way pcre_compile() computes the
2056        amount of memory needed for a compiled pattern. Previously, there was code
2057        that made a preliminary scan of the pattern in order to do this. That was
2058        OK when PCRE was new, but as the facilities have expanded, it has become
2059        harder and harder to keep it in step with the real compile phase, and there
2060        have been a number of bugs (see for example, 4 above). I have now found a
2061        cunning way of running the real compile function in a "fake" mode that
2062        enables it to compute how much memory it would need, while actually only
2063        ever using a few hundred bytes of working memory and without too many
2064        tests of the mode. This should make future maintenance and development
2065        easier. A side effect of this work is that the limit of 200 on the nesting
2066        depth of parentheses has been removed (though this was never a serious
2067        limitation, I suspect). However, there is a downside: pcre_compile() now
2068        runs more slowly than before (30% or more, depending on the pattern). I
2069        hope this isn't a big issue. There is no effect on runtime performance.
2071    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
2072        newline (only possible for the last line of a file) and it was a
2073        pattern that set a locale (followed by /Lsomething), pcretest crashed.
2075    19. Added additional timing features to pcretest. (1) The -tm option now times
2076        matching only, not compiling. (2) Both -t and -tm can be followed, as a
2077        separate command line item, by a number that specifies the number of
2078        repeats to use when timing. The default is 50000; this gives better
2079        precision, but takes uncomfortably long for very large patterns.
2081    20. Extended pcre_study() to be more clever in cases where a branch of a
2082        subpattern has no definite first character. For example, (a*|b*)[cd] would
2083        previously give no result from pcre_study(). Now it recognizes that the
2084        first character must be a, b, c, or d.
2086    21. There was an incorrect error "recursive call could loop indefinitely" if
2087        a subpattern (or the entire pattern) that was being tested for matching an
2088        empty string contained only one non-empty item after a nested subpattern.
2089        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
2090        incorrectly, because the \d was being skipped in the check.
2092    22. The pcretest program now has a new pattern option /B and a command line
2093        option -b, which is equivalent to adding /B to every pattern. This causes
2094        it to show the compiled bytecode, without the additional information that
2095        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
2096        is the same as /B/I).
2098    23. A new optimization is now able automatically to treat some sequences such
2099        as a*b as a*+b. More specifically, if something simple (such as a character
2100        or a simple class like \d) has an unlimited quantifier, and is followed by
2101        something that cannot possibly match the quantified thing, the quantifier
2102        is automatically "possessified".
2104    24. A recursive reference to a subpattern whose number was greater than 39
2105        went wrong under certain circumstances in UTF-8 mode. This bug could also
2106        have affected the operation of pcre_study().
2108    25. Realized that a little bit of performance could be had by replacing
2109        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
2111    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
2113    27. Possessive quantifiers such as a++ were previously implemented by turning
2114        them into atomic groups such as ($>a+). Now they have their own opcodes,
2115        which improves performance. This includes the automatically created ones
2116        from 23 above.
2118    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
2119        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
2120        the first matched character to be a colon. This applied both to named and
2121        numbered groups.
2123    29. The ucpinternal.h header file was missing its idempotency #ifdef.
2125    30. I was sent a "project" file called libpcre.a.dev which I understand makes
2126        building PCRE on Windows easier, so I have included it in the distribution.
2128    31. There is now a check in pcretest against a ridiculously large number being
2129        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
2130        loop, the loop is abandoned.
2132    32. Forward references to subpatterns in conditions such as (?(2)...) where
2133        subpattern 2 is defined later cause pcre_compile() to search forwards in
2134        the pattern for the relevant set of parentheses. This search went wrong
2135        when there were unescaped parentheses in a character class, parentheses
2136        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
2138    33. "Subroutine" calls and backreferences were previously restricted to
2139        referencing subpatterns earlier in the regex. This restriction has now
2140        been removed.
2142    34. Added a number of extra features that are going to be in Perl 5.10. On the
2143        whole, these are just syntactic alternatives for features that PCRE had
2144        previously implemented using the Python syntax or my own invention. The
2145        other formats are all retained for compatibility.
2147        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
2148            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
2149            also .NET compatible.
2151        (b) A recursion or subroutine call to a named group can now be defined as
2152            (?&name) as well as (?P>name).
2154        (c) A backreference to a named group can now be defined as \k<name> or
2155            \k'name' as well as (?P=name). The new forms, as well as being in Perl
2156            5.10, are also .NET compatible.
2158        (d) A conditional reference to a named group can now use the syntax
2159            (?(<name>) or (?('name') as well as (?(name).
2161        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
2162            groups (named and numbered) that are never evaluated inline, but can be
2163            called as "subroutines" from elsewhere. In effect, the DEFINE condition
2164            is always false. There may be only one alternative in such a group.
2166        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
2167            as the simple (?(R). The condition is true only if the most recent
2168            recursion is that of the given number or name. It does not search out
2169            through the entire recursion stack.
2171        (g) The escape \gN or \g{N} has been added, where N is a positive or
2172            negative number, specifying an absolute or relative reference.
2174    35. Tidied to get rid of some further signed/unsigned compiler warnings and
2175        some "unreachable code" warnings.
2177    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
2178        things, this adds five new scripts.
2180    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
2181        There were also incompatibilities regarding the handling of \Q..\E inside
2182        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
2183        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
2185    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
2186        matches an empty string, and forcibly breaks the loop. There were bugs in
2187        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
2188        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
2189        separate and independent bugs (that affected different cases) have been
2190        fixed.
2192    39. Refactored the code to abolish the use of different opcodes for small
2193        capturing bracket numbers. This is a tidy that I avoided doing when I
2194        removed the limit on the number of capturing brackets for 3.5 back in 2001.
2195        The new approach is not only tidier, it makes it possible to reduce the
2196        memory needed to fix the previous bug (38).
2198    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
2199        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
2200        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
2201        mode.
2203    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
2204        report.
2206    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
2207        copy construction and assignment in the C++ wrapper.
2209    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
2210        couple of bugs that could have given wrong results in the "--newline=crlf"
2211        case.
2213    44. Added a number of casts and did some reorganization of signed/unsigned int
2214        variables following suggestions from Dair Grant. Also renamed the variable
2215        "this" as "item" because it is a C++ keyword.
2217    45. Arranged for dftables to add
2219          #include "pcre_internal.h"
2221        to pcre_chartables.c because without it, gcc 4.x may remove the array
2222        definition from the final binary if PCRE is built into a static library and
2223        dead code stripping is activated.
2225    46. For an unanchored pattern, if a match attempt fails at the start of a
2226        newline sequence, and the newline setting is CRLF or ANY, and the next two
2227        characters are CRLF, advance by two characters instead of one.
2230    Version 6.7 04-Jul-06
2231    ---------------------
2233     1. In order to handle tests when input lines are enormously long, pcretest has
2234        been re-factored so that it automatically extends its buffers when
2235        necessary. The code is crude, but this _is_ just a test program. The
2236        default size has been increased from 32K to 50K.
2238     2. The code in pcre_study() was using the value of the re argument before
2239        testing it for NULL. (Of course, in any sensible call of the function, it
2240        won't be NULL.)
2242     3. The memmove() emulation function in pcre_internal.h, which is used on
2243        systems that lack both memmove() and bcopy() - that is, hardly ever -
2244        was missing a "static" storage class specifier.
2246     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
2247        containing an extended class (one that cannot be represented by a bitmap
2248        because it contains high-valued characters or Unicode property items, e.g.
2249        [\pZ]). Almost always one would set UTF-8 mode when processing such a
2250        pattern, but PCRE should not loop if you do not (it no longer does).
2251        [Detail: two cases were found: (a) a repeated subpattern containing an
2252        extended class; (b) a recursive reference to a subpattern that followed a
2253        previous extended class. It wasn't skipping over the extended class
2254        correctly when UTF-8 mode was not set.]
2256     5. A negated single-character class was not being recognized as fixed-length
2257        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
2258        compile error "lookbehind assertion is not fixed length".
2260     6. The RunPerlTest auxiliary script was showing an unexpected difference
2261        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
2262        write a Perl script that can interpret lines of an input file either as
2263        byte characters or as UTF-8, which is what "perltest" was being required to
2264        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
2265        can't do is switch easily at run time between having the "use utf8;" pragma
2266        or not. In the end, I fudged it by using the RunPerlTest script to insert
2267        "use utf8;" explicitly for the UTF-8 tests.
2269     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
2270        the end of the subject string, contrary to the documentation and to what
2271        Perl does. This was true of both matching functions. Now it matches only at
2272        the start of the subject and immediately after *internal* newlines.
2274     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
2275        a pointer to an int instead of a pointer to an unsigned long int. This
2276        caused problems on 64-bit systems.
2278     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
2279        instance of the 'standard' template library not being so standard".
2281    10. There was no check on the number of named subpatterns nor the maximum
2282        length of a subpattern name. The product of these values is used to compute
2283        the size of the memory block for a compiled pattern. By supplying a very
2284        long subpattern name and a large number of named subpatterns, the size
2285        computation could be caused to overflow. This is now prevented by limiting
2286        the length of names to 32 characters, and the number of named subpatterns
2287        to 10,000.
2289    11. Subpatterns that are repeated with specific counts have to be replicated in
2290        the compiled pattern. The size of memory for this was computed from the
2291        length of the subpattern and the repeat count. The latter is limited to
2292        65535, but there was no limit on the former, meaning that integer overflow
2293        could in principle occur. The compiled length of a repeated subpattern is
2294        now limited to 30,000 bytes in order to prevent this.
2296    12. Added the optional facility to have named substrings with the same name.
2298    13. Added the ability to use a named substring as a condition, using the
2299        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
2300        are numbers (not recommended). Forward references are permitted.
2302    14. Added forward references in named backreferences (if you see what I mean).
2304    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
2305        pattern could run off the end of the subject. For example, the pattern
2306        "(?s)(.{1,5})"8 did this with the subject "ab".
2308    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
2309        PCRE_CASELESS was set when matching characters that were quantified with ?
2310        or *.
2312    17. A character class other than a single negated character that had a minimum
2313        but no maximum quantifier - for example [ab]{6,} - was not handled
2314        correctly by pce_dfa_exec(). It would match only one character.
2316    18. A valid (though odd) pattern that looked like a POSIX character
2317        class but used an invalid character after [ (for example [[,abc,]]) caused
2318        pcre_compile() to give the error "Failed: internal error: code overflow" or
2319        in some cases to crash with a glibc free() error. This could even happen if
2320        the pattern terminated after [[ but there just happened to be a sequence of
2321        letters, a binary zero, and a closing ] in the memory that followed.
2323    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
2324        over the years. Originally (before any Unicode support), just the bottom 8
2325        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
2326        output from "man perlunicode" includes this:
2328          The regular expression compiler produces polymorphic opcodes.  That
2329          is, the pattern adapts to the data and automatically switches to
2330          the Unicode character scheme when presented with Unicode data--or
2331          instead uses a traditional byte scheme when presented with byte
2332          data.
2334        Sadly, a wide octal escape does not cause a switch, and in a string with
2335        no other multibyte characters, these octal escapes are treated as before.
2336        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
2337        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
2338        Unicode string.
2340        I have not perpetrated such confusion in PCRE. Up till now, it took just
2341        the bottom 8 bits, as in old Perl. I have now made octal escapes with
2342        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
2343        translate to the appropriate multibyte character.
2345    29. Applied some refactoring to reduce the number of warnings from Microsoft
2346        and Borland compilers. This has included removing the fudge introduced
2347        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
2348        a warning about an unused variable.
2350    21. PCRE has not included VT (character 0x0b) in the set of whitespace
2351        characters since release 4.0, because Perl (from release 5.004) does not.
2352        [Or at least, is documented not to: some releases seem to be in conflict
2353        with the documentation.] However, when a pattern was studied with
2354        pcre_study() and all its branches started with \s, PCRE still included VT
2355        as a possible starting character. Of course, this did no harm; it just
2356        caused an unnecessary match attempt.
2358    22. Removed a now-redundant internal flag bit that recorded the fact that case
2359        dependency changed within the pattern. This was once needed for "required
2360        byte" processing, but is no longer used. This recovers a now-scarce options
2361        bit. Also moved the least significant internal flag bit to the most-
2362        significant bit of the word, which was not previously used (hangover from
2363        the days when it was an int rather than a uint) to free up another bit for
2364        the future.
2366    23. Added support for CRLF line endings as well as CR and LF. As well as the
2367        default being selectable at build time, it can now be changed at runtime
2368        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
2369        specify that it is scanning data with non-default line endings.
2371    24. Changed the definition of CXXLINK to make it agree with the definition of
2372        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
2374    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
2375        recursions. This makes a big different to stack usage for some patterns.
2377    26. If a subpattern containing a named recursion or subroutine reference such
2378        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
2379        the space required for the compiled pattern went wrong and gave too small a
2380        value. Depending on the environment, this could lead to "Failed: internal
2381        error: code overflow at offset 49" or "glibc detected double free or
2382        corruption" errors.
2384    27. Applied patches from Google (a) to support the new newline modes and (b) to
2385        advance over multibyte UTF-8 characters in GlobalReplace.
2387    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
2388        difference for some implementation of PCRE in some Windows version.
2390    29. Added some extra testing facilities to pcretest:
2392        \q<number>   in a data line sets the "match limit" value
2393        \Q<number>   in a data line sets the "match recursion limt" value
2394        -S <number>  sets the stack size, where <number> is in megabytes
2396        The -S option isn't available for Windows.
2399    Version 6.6 06-Feb-06
2400    ---------------------
2402     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
2403        in pcreposix.h. I have copied the definition from pcre.h.
2405     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
2406        because pcre.h is no longer a built file.
2408     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
2409        not normally included in the compiled code.
2412    Version 6.5 01-Feb-06
2413    ---------------------
2415     1. When using the partial match feature with pcre_dfa_exec(), it was not
2416        anchoring the second and subsequent partial matches at the new starting
2417        point. This could lead to incorrect results. For example, with the pattern
2418        /1234/, partially matching against "123" and then "a4" gave a match.
2420     2. Changes to pcregrep:
2422        (a) All non-match returns from pcre_exec() were being treated as failures
2423            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
2424            error message is output. Some extra information is given for the
2425            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
2426            probably the only errors that are likely to be caused by users (by
2427            specifying a regex that has nested indefinite repeats, for instance).
2428            If there are more than 20 of these errors, pcregrep is abandoned.
2430        (b) A binary zero was treated as data while matching, but terminated the
2431            output line if it was written out. This has been fixed: binary zeroes
2432            are now no different to any other data bytes.
2434        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
2435            used to set a locale for matching. The --locale=xxxx long option has
2436            been added (no short equivalent) to specify a locale explicitly on the
2437            pcregrep command, overriding the environment variables.
2439        (d) When -B was used with -n, some line numbers in the output were one less
2440            than they should have been.
2442        (e) Added the -o (--only-matching) option.
2444        (f) If -A or -C was used with -c (count only), some lines of context were
2445            accidentally printed for the final match.
2447        (g) Added the -H (--with-filename) option.
2449        (h) The combination of options -rh failed to suppress file names for files
2450            that were found from directory arguments.
2452        (i) Added the -D (--devices) and -d (--directories) options.
2454        (j) Added the -F (--fixed-strings) option.
2456        (k) Allow "-" to be used as a file name for -f as well as for a data file.
2458        (l) Added the --colo(u)r option.
2460        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
2461            is not present by default.
2463     3. A nasty bug was discovered in the handling of recursive patterns, that is,
2464        items such as (?R) or (?1), when the recursion could match a number of
2465        alternatives. If it matched one of the alternatives, but subsequently,
2466        outside the recursion, there was a failure, the code tried to back up into
2467        the recursion. However, because of the way PCRE is implemented, this is not
2468        possible, and the result was an incorrect result from the match.
2470        In order to prevent this happening, the specification of recursion has
2471        been changed so that all such subpatterns are automatically treated as
2472        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
2474     4. I had overlooked the fact that, in some locales, there are characters for
2475        which isalpha() is true but neither isupper() nor islower() are true. In
2476        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
2477        and ordfeminine) are like this. This affected the treatment of \w and \W
2478        when they appeared in character classes, but not when they appeared outside
2479        a character class. The bit map for "word" characters is now created
2480        separately from the results of isalnum() instead of just taking it from the
2481        upper, lower, and digit maps. (Plus the underscore character, of course.)
2483     5. The above bug also affected the handling of POSIX character classes such as
2484        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
2485        permanent tables. Instead, the bit maps for such a class were previously
2486        created as the appropriate unions of the upper, lower, and digit bitmaps.
2487        Now they are created by subtraction from the [[:word:]] class, which has
2488        its own bitmap.
2490     6. The [[:blank:]] character class matches horizontal, but not vertical space.
2491        It is created by subtracting the vertical space characters (\x09, \x0a,
2492        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
2493        subtraction was done in the overall bitmap for a character class, meaning
2494        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
2495        be recognized. This bug has been fixed.
2497     7. Patches from the folks at Google:
2499          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
2500          real life, but is still worth protecting against".
2502          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
2503          regular expressions".
2505          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
2506          have it.
2508          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
2509          "configure" and the latter not, in order to fix a problem somebody had
2510          with compiling the Arg class on HP-UX.
2512          (e) Improve the error-handling of the C++ wrapper a little bit.
2514          (f) New tests for checking recursion limiting.
2516     8. The pcre_memmove() function, which is used only if the environment does not
2517        have a standard memmove() function (and is therefore rarely compiled),
2518        contained two bugs: (a) use of int instead of size_t, and (b) it was not
2519        returning a result (though PCRE never actually uses the result).
2521     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
2522        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
2523        returned instead of calling malloc() with an overflowing number that would
2524        most likely cause subsequent chaos.
2526    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
2528    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
2529        with this option is matched, the nmatch and pmatch options of regexec() are
2530        ignored.
2532    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
2533        provided in case anyone wants to the the POSIX interface with UTF-8
2534        strings.
2536    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
2537        C++ linking (needed for some HP-UX environments).
2539    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
2540        (unused parameter) and in the pcre_printint() function (omitted "default"
2541        switch label when the default is to do nothing).
2543    15. Added some code to make it possible, when PCRE is compiled as a C++
2544        library, to replace subject pointers for pcre_exec() with a smart pointer
2545        class, thus making it possible to process discontinuous strings.
2547    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
2548        much the same function. They were added by different people who were trying
2549        to make PCRE easy to compile on non-Unix systems. It has been suggested
2550        that PCRE_EXPORT be abolished now that there is more automatic apparatus
2551        for compiling on Windows systems. I have therefore replaced it with
2552        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
2553        defaults to "extern" for C or "extern C" for C++, which works fine on
2554        Unix-like systems. It is now possible to override the value of PCRE_DATA_
2555        SCOPE with something explicit in config.h. In addition:
2557        (a) pcreposix.h still had just "extern" instead of either of these macros;
2558            I have replaced it with PCRE_DATA_SCOPE.
2560        (b) Functions such as _pcre_xclass(), which are internal to the library,
2561            but external in the C sense, all had PCRE_EXPORT in their definitions.
2562            This is apparently wrong for the Windows case, so I have removed it.
2563            (It makes no difference on Unix-like systems.)
2565    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
2566        of recursive calls to match(). This is different to MATCH_LIMIT because
2567        that limits the total number of calls to match(), not all of which increase
2568        the depth of recursion. Limiting the recursion depth limits the amount of
2569        stack (or heap if NO_RECURSE is set) that is used. The default can be set
2570        when PCRE is compiled, and changed at run time. A patch from Google adds
2571        this functionality to the C++ interface.
2573    18. Changes to the handling of Unicode character properties:
2575        (a) Updated the table to Unicode 4.1.0.
2577        (b) Recognize characters that are not in the table as "Cn" (undefined).
2579        (c) I revised the way the table is implemented to a much improved format
2580            which includes recognition of ranges. It now supports the ranges that
2581            are defined in UnicodeData.txt, and it also amalgamates other
2582            characters into ranges. This has reduced the number of entries in the
2583            table from around 16,000 to around 3,000, thus reducing its size
2584            considerably. I realized I did not need to use a tree structure after
2585            all - a binary chop search is just as efficient. Having reduced the
2586            number of entries, I extended their size from 6 bytes to 8 bytes to
2587            allow for more data.
2589        (d) Added support for Unicode script names via properties such as \p{Han}.
2591    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
2592        matching that character.
2594    20. When matching a repeated Unicode property with a minimum greater than zero,
2595        (for example \pL{2,}), PCRE could look past the end of the subject if it
2596        reached it while seeking the minimum number of characters. This could
2597        happen only if some of the characters were more than one byte long, because
2598        there is a check for at least the minimum number of bytes.
2600    21. Refactored the implementation of \p and \P so as to be more general, to
2601        allow for more different types of property in future. This has changed the
2602        compiled form incompatibly. Anybody with saved compiled patterns that use
2603        \p or \P will have to recompile them.
2605    22. Added "Any" and "L&" to the supported property types.
2607    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
2608        but give a compile time error if the value is greater than 0xff.
2610    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
2611        accidentally not being installed or uninstalled.
2613    25. The pcre.h file was built from pcre.h.in, but the only changes that were
2614        made were to insert the current release number. This seemed silly, because
2615        it made things harder for people building PCRE on systems that don't run
2616        "configure". I have turned pcre.h into a distributed file, no longer built
2617        by "configure", with the version identification directly included. There is
2618        no longer a pcre.h.in file.
2620        However, this change necessitated a change to the pcre-config script as
2621        well. It is built from pcre-config.in, and one of the substitutions was the
2622        release number. I have updated configure.ac so that ./configure now finds
2623        the release number by grepping pcre.h.
2625    26. Added the ability to run the tests under valgrind.
2628    Version 6.4 05-Sep-05
2629    ---------------------
2631     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
2632        "--" to be printed when multiple files were scanned, even when none of the
2633        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
2634        consider it to be a bug, and have restored the previous behaviour.
2636     2. A couple of code tidies to get rid of compiler warnings.
2638     3. The pcretest program used to cheat by referring to symbols in the library
2639        whose names begin with _pcre_. These are internal symbols that are not
2640        really supposed to be visible externally, and in some environments it is
2641        possible to suppress them. The cheating is now confined to including
2642        certain files from the library's source, which is a bit cleaner.
2644     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
2645        file's purpose clearer.
2647     5. Reorganized pcre_ucp_findchar().
2650    Version 6.3 15-Aug-05
2651    ---------------------
2653     1. The file libpcre.pc.in did not have general read permission in the tarball.
2655     2. There were some problems when building without C++ support:
2657        (a) If C++ support was not built, "make install" and "make test" still
2658            tried to test it.
2660        (b) There were problems when the value of CXX was explicitly set. Some
2661            changes have been made to try to fix these, and ...
2663        (c) --disable-cpp can now be used to explicitly disable C++ support.
2665        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
2666            backslash in a target when C++ was disabled. This confuses some
2667            versions of "make", apparently. Using an intermediate variable solves
2668            this. (Same for CPP_LOBJ.)
2670     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
2671        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
2672        necessary on certain architectures.
2674     4. Added a setting of -export-symbols-regex to the link command to remove
2675        those symbols that are exported in the C sense, but actually are local
2676        within the library, and not documented. Their names all begin with
2677        "_pcre_". This is not a perfect job, because (a) we have to except some
2678        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
2679        available (and never for static libraries). I have made a note to try to
2680        find a way round (a) in the future.
2683    Version 6.2 01-Aug-05
2684    ---------------------
2686     1. There was no test for integer overflow of quantifier values. A construction
2687        such as {1111111111111111} would give undefined results. What is worse, if
2688        a minimum quantifier for a parenthesized subpattern overflowed and became
2689        negative, the calculation of the memory size went wrong. This could have
2690        led to memory overwriting.
2692     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
2694     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
2695        operating environments where this matters.
2697     4. Applied Giuseppe Maxia's patch to add additional features for controlling
2698        PCRE options from within the C++ wrapper.
2700     5. Named capturing subpatterns were not being correctly counted when a pattern
2701        was compiled. This caused two problems: (a) If there were more than 100
2702        such subpatterns, the calculation of the memory needed for the whole
2703        compiled pattern went wrong, leading to an overflow error. (b) Numerical
2704        back references of the form \12, where the number was greater than 9, were
2705        not recognized as back references, even though there were sufficient
2706        previous subpatterns.
2708     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
2709        versions of gcc, e.g. 2.95.4.
2712    Version 6.1 21-Jun-05
2713    ---------------------
2715     1. There was one reference to the variable "posix" in pcretest.c that was not
2716        surrounded by "#if !defined NOPOSIX".
2718     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
2719        the cross-check on the old pcre_info() function, for the benefit of the
2720        cut-down version of PCRE that is currently imported into Exim.
2722     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
2723        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
2724        patterns that start (?i) but I don't think that matters. The use of (?i) is
2725        just an example; this all applies to the other options as well.
2727     4. Since libtool seems to echo the compile commands it is issuing, the output
2728        from "make" can be reduced a bit by putting "@" in front of each libtool
2729        compile command.
2731     5. Patch from the folks at Google for configure.in to be a bit more thorough
2732        in checking for a suitable C++ installation before trying to compile the
2733        C++ stuff. This should fix a reported problem when a compiler was present,
2734        but no suitable headers.
2736     6. The man pages all had just "PCRE" as their title. I have changed them to
2737        be the relevant file name. I have also arranged that these names are
2738        retained in the file doc/pcre.txt, which is a concatenation in text format
2739        of all the man pages except the little individual ones for each function.
2741     7. The NON-UNIX-USE file had not been updated for the different set of source
2742        files that come with release 6. I also added a few comments about the C++
2743        wrapper.
2746  Version 6.0 07-Jun-05  Version 6.0 07-Jun-05
2747  ---------------------  ---------------------

Removed from v.77  
changed lines
  Added in v.743

  ViewVC Help
Powered by ViewVC 1.1.5