ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 11 by nigel, Sat Feb 24 21:38:17 2007 UTC revision 750 by ph10, Fri Nov 18 11:07:14 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.
20    5.  Supporting of \x, \U and \u in JavaScript compatibility mode based on the
21        ECMA-262 standard.
23    6.  Lookbehinds such as (?<=a{2}b) that contained a fixed repetition were
24        erroneously being rejected as "not fixed length" if PCRE_CASELESS was set.
25        This bug was probably introduced by change 9 of 8.13.
27    7.  While fixing 6 above, I noticed that a number of other items were being
28        incorrectly rejected as "not fixed length". This arose partly because newer
29        opcodes had not been added to the fixed-length checking code. I have (a)
30        corrected the bug and added tests for these items, and (b) arranged for an
31        error to occur if an unknown opcode is encountered while checking for fixed
32        length instead of just assuming "not fixed length". The items that were
33        rejected were: (*ACCEPT), (*COMMIT), (*FAIL), (*MARK), (*PRUNE), (*SKIP),
34        (*THEN), \h, \H, \v, \V, and single character negative classes with fixed
35        repetitions, e.g. [^a]{3}, with and without PCRE_CASELESS.
37    8.  A possessively repeated conditional subpattern such as (?(?=c)c|d)++ was
38        being incorrectly compiled and would have given unpredicatble results.
40    9.  A possessively repeated subpattern with minimum repeat count greater than
41        one behaved incorrectly. For example, (A){2,}+ behaved as if it was
42        (A)(A)++ which meant that, after a subsequent mismatch, backtracking into
43        the first (A) could occur when it should not.
46    Version 8.20 21-Oct-2011
47    ------------------------
49    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
50        a POSIX class. After further experiments with Perl, which convinced me that
51        Perl has bugs and confusions, a closing square bracket is no longer allowed
52        in a POSIX name. This bug also affected patterns with classes that started
53        with full stops.
55    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
56        captured substring, but while checking the failing first alternative,
57        substring 1 is temporarily captured. If the output vector supplied to
58        pcre_exec() was not big enough for this capture, the yield of the function
59        was still zero ("insufficient space for captured substrings"). This cannot
60        be totally fixed without adding another stack variable, which seems a lot
61        of expense for a edge case. However, I have improved the situation in cases
62        such as /(a)(b)x|abc/ matched against "abc", where the return code
63        indicates that fewer than the maximum number of slots in the ovector have
64        been set.
66    3.  Related to (2) above: when there are more back references in a pattern than
67        slots in the output vector, pcre_exec() uses temporary memory during
68        matching, and copies in the captures as far as possible afterwards. It was
69        using the entire output vector, but this conflicts with the specification
70        that only 2/3 is used for passing back captured substrings. Now it uses
71        only the first 2/3, for compatibility. This is, of course, another edge
72        case.
74    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
75        main code base, and can be used by building with --enable-jit. When this is
76        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
77        runtime --no-jit option is given.
79    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
80        ovector, the return from the function was zero, implying that there were
81        other matches that did not fit. The correct "exactly full" value is now
82        returned.
84    6.  If a subpattern that was called recursively or as a subroutine contained
85        (*PRUNE) or any other control that caused it to give a non-standard return,
86        invalid errors such as "Error -26 (nested recursion at the same subject
87        position)" or even infinite loops could occur.
89    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
90        computing the minimum length on reaching *ACCEPT, and so ended up with the
91        wrong value of 1 rather than 0. Further investigation indicates that
92        computing a minimum subject length in the presence of *ACCEPT is difficult
93        (think back references, subroutine calls), and so I have changed the code
94        so that no minimum is registered for a pattern that contains *ACCEPT.
96    8.  If (*THEN) was present in the first (true) branch of a conditional group,
97        it was not handled as intended. [But see 16 below.]
99    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
100        Sheri Pierce.
102    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
103        the first byte in a match must be "a".
105    11. Change 17 for 8.13 increased the recursion depth for patterns like
106        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
107        pattern contains any instances of (*THEN). If it does not, the old
108        optimizations are restored. It would be nice to do this on a per-group
109        basis, but at the moment that is not feasible.
111    12. In some environments, the output of pcretest -C is CRLF terminated. This
112        broke RunTest's code that checks for the link size. A single white space
113        character after the value is now allowed for.
115    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
116        For "fr", it uses the Windows-specific input and output files.
118    14. If (*THEN) appeared in a group that was called recursively or as a
119        subroutine, it did not work as intended. [But see next item.]
121    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
122        pattern fragments (but not containing any | characters). If A and B are
123        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
124        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
125        D. In other words, Perl considers parentheses that do not contain any |
126        characters to be part of a surrounding alternative, whereas PCRE was
127        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
128        differently. PCRE now behaves in the same way as Perl, except in the case
129        of subroutine/recursion calls such as (?1) which have in any case always
130        been different (but PCRE had them first :-).
132    16. Related to 15 above: Perl does not treat the | in a conditional group as
133        creating alternatives. Such a group is treated in the same way as an
134        ordinary group without any | characters when processing (*THEN). PCRE has
135        been changed to match Perl's behaviour.
137    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
138        RunGrepTest script failed.
140    18. Change 22 for version 13 caused atomic groups to use more stack. This is
141        inevitable for groups that contain captures, but it can lead to a lot of
142        stack use in large patterns. The old behaviour has been restored for atomic
143        groups that do not contain any capturing parentheses.
145    19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not
146        suppress the check for a minimum subject length at run time. (If it was
147        given to pcre_exec() or pcre_dfa_exec() it did work.)
149    20. Fixed an ASCII-dependent infelicity in pcretest that would have made it
150        fail to work when decoding hex characters in data strings in EBCDIC
151        environments.
153    21. It appears that in at least one Mac OS environment, the isxdigit() function
154        is implemented as a macro that evaluates to its argument more than once,
155        contravening the C 90 Standard (I haven't checked a later standard). There
156        was an instance in pcretest which caused it to go wrong when processing
157        \x{...} escapes in subject strings. The has been rewritten to avoid using
158        things like p++ in the argument of isxdigit().
161    Version 8.13 16-Aug-2011
162    ------------------------
164    1.  The Unicode data tables have been updated to Unicode 6.0.0.
166    2.  Two minor typos in pcre_internal.h have been fixed.
168    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
169        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
170        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
172    4.  There were a number of related bugs in the code for matching backrefences
173        caselessly in UTF-8 mode when codes for the characters concerned were
174        different numbers of bytes. For example, U+023A and U+2C65 are an upper
175        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
176        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
177        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
178        2-byte code at the end of the subject (it thought there wasn't enough data
179        left).
181    5.  Comprehensive information about what went wrong is now returned by
182        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
183        as the output vector has at least 2 elements. The offset of the start of
184        the failing character and a reason code are placed in the vector.
186    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
187        now returned is for the first byte of the failing character, instead of the
188        last byte inspected. This is an incompatible change, but I hope it is small
189        enough not to be a problem. It makes the returned offset consistent with
190        pcre_exec() and pcre_dfa_exec().
192    7.  pcretest now gives a text phrase as well as the error number when
193        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
194        failure, the offset and reason code are output.
196    8.  When \R was used with a maximizing quantifier it failed to skip backwards
197        over a \r\n pair if the subsequent match failed. Instead, it just skipped
198        back over a single character (\n). This seems wrong (because it treated the
199        two characters as a single entity when going forwards), conflicts with the
200        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
201        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
202        has been changed.
204    9.  Some internal refactoring has changed the processing so that the handling
205        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
206        time (the PCRE_DOTALL option was changed this way some time ago: version
207        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
208        which was always a bit of a fudge. It also means that there is one less
209        argument for the match() function, which reduces its stack requirements
210        slightly. This change also fixes an incompatibility with Perl: the pattern
211        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
213    10. More internal refactoring has drastically reduced the number of recursive
214        calls to match() for possessively repeated groups such as (abc)++ when
215        using pcre_exec().
217    11. While implementing 10, a number of bugs in the handling of groups were
218        discovered and fixed:
220        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
221        (a|)*(?1) gave a compile-time internal error.
222        ((a|)+)+  did not notice that the outer group could match an empty string.
223        (^a|^)+   was not marked as anchored.
224        (.*a|.*)+ was not marked as matching at start or after a newline.
226    12. Yet more internal refactoring has removed another argument from the match()
227        function. Special calls to this function are now indicated by setting a
228        value in a variable in the "match data" data block.
230    13. Be more explicit in pcre_study() instead of relying on "default" for
231        opcodes that mean there is no starting character; this means that when new
232        ones are added and accidentally left out of pcre_study(), testing should
233        pick them up.
235    14. The -s option of pcretest has been documented for ages as being an old
236        synonym of -m (show memory usage). I have changed it to mean "force study
237        for every regex", that is, assume /S for every regex. This is similar to -i
238        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
239        using it. It makes it easier to run collections of tests with and without
240        study enabled, and thereby test pcre_study() more easily. All the standard
241        tests are now run with and without -s (but some patterns can be marked as
242        "never study" - see 20 below).
244    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
245        restoration of the capturing data to the outer values was not happening
246        correctly.
248    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
249        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
250        pattern had matched an empty string, and so incorrectly returned a no
251        match.
253    17. There was optimizing code for the last branch of non-capturing parentheses,
254        and also for the obeyed branch of a conditional subexpression, which used
255        tail recursion to cut down on stack usage. Unfortunately, now that there is
256        the possibility of (*THEN) occurring in these branches, tail recursion is
257        no longer possible because the return has to be checked for (*THEN). These
258        two optimizations have therefore been removed. [But see 8.20/11 above.]
260    18. If a pattern containing \R was studied, it was assumed that \R always
261        matched two bytes, thus causing the minimum subject length to be
262        incorrectly computed because \R can also match just one byte.
264    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
265        was incorrectly computed.
267    20. If /S is present twice on a test pattern in pcretest input, it now
268        *disables* studying, thereby overriding the use of -s on the command line
269        (see 14 above). This is necessary for one or two tests to keep the output
270        identical in both cases.
272    21. When (*ACCEPT) was used in an assertion that matched an empty string and
273        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
275    22. When an atomic group that contained a capturing parenthesis was
276        successfully matched, but the branch in which it appeared failed, the
277        capturing was not being forgotten if a higher numbered group was later
278        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
279        group 1 to "a", when in fact it should be unset. This applied to multi-
280        branched capturing and non-capturing groups, repeated or not, and also to
281        positive assertions (capturing in negative assertions does not happen
282        in PCRE) and also to nested atomic groups.
284    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
285        subject after a captured substring, to make it easier to tell which of a
286        number of identical substrings has been captured.
288    24. The way atomic groups are processed by pcre_exec() has been changed so that
289        if they are repeated, backtracking one repetition now resets captured
290        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
291        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
292        "aaa". Previously, it would have been "a". As part of this code
293        refactoring, the way recursive calls are handled has also been changed.
295    25. If an assertion condition captured any substrings, they were not passed
296        back unless some other capturing happened later. For example, if
297        (?(?=(a))a) was matched against "a", no capturing was returned.
299    26. When studying a pattern that contained subroutine calls or assertions,
300        the code for finding the minimum length of a possible match was handling
301        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
302        group 1 called group 2 while simultaneously a separate group 2 called group
303        1). A stack overflow occurred in this case. I have fixed this by limiting
304        the recursion depth to 10.
306    27. Updated RunTest.bat in the distribution to the version supplied by Tom
307        Fortmann. This supports explicit test numbers on the command line, and has
308        argument validation and error reporting.
310    28. An instance of \X with an unlimited repeat could fail if at any point the
311        first character it looked at was a mark character.
313    29. Some minor code refactoring concerning Unicode properties and scripts
314        should reduce the stack requirement of match() slightly.
316    30. Added the '=' option to pcretest to check the setting of unused capturing
317        slots at the end of the pattern, which are documented as being -1, but are
318        not included in the return count.
320    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
321        compiled something random. Now it gives a compile-time error (as does
322        Perl).
324    32. A *MARK encountered during the processing of a positive assertion is now
325        recorded and passed back (compatible with Perl).
327    33. If --only-matching or --colour was set on a pcregrep call whose pattern
328        had alternative anchored branches, the search for a second match in a line
329        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
330        matched the line "0102" twice. The same bug affected patterns that started
331        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
332        twice.
334    34. Previously, PCRE did not allow quantification of assertions. However, Perl
335        does, and because of capturing effects, quantifying parenthesized
336        assertions may at times be useful. Quantifiers are now allowed for
337        parenthesized assertions.
339    35. A minor code tidy in pcre_compile() when checking options for \R usage.
341    36. \g was being checked for fancy things in a character class, when it should
342        just be a literal "g".
344    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
345        appearance of a nested POSIX class supersedes an apparent external class.
346        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
347        unescaped square brackets may also appear as part of class names. For
348        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
349        more like Perl. (But see 8.20/1 above.)
351    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
352        was because it thought it was \N{name}, which is not supported).
354    39. Add minix to OS list not supporting the -S option in pcretest.
356    40. PCRE tries to detect cases of infinite recursion at compile time, but it
357        cannot analyze patterns in sufficient detail to catch mutual recursions
358        such as ((?1))((?2)). There is now a runtime test that gives an error if a
359        subgroup is called recursively as a subpattern for a second time at the
360        same position in the subject string. In previous releases this might have
361        been caught by the recursion limit, or it might have run out of stack.
363    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
364        happen only once. PCRE was, however incorrectly giving a compile time error
365        "recursive call could loop indefinitely" because it cannot analyze the
366        pattern in sufficient detail. The compile time test no longer happens when
367        PCRE is compiling a conditional subpattern, but actual runaway loops are
368        now caught at runtime (see 40 above).
370    42. It seems that Perl allows any characters other than a closing parenthesis
371        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
372        has been changed to be the same.
374    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
375        as not to get warnings when autogen.sh is called. Also changed
376        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
378    44. To help people who use pcregrep to scan files containing exceedingly long
379        lines, the following changes have been made:
381        (a) The default value of the buffer size parameter has been increased from
382            8K to 20K. (The actual buffer used is three times this size.)
384        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
385            PCRE is built.
387        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
388            to be set at run time.
390        (d) Numerical values in pcregrep options can be followed by K or M, for
391            example --buffer-size=50K.
393        (e) If a line being scanned overflows pcregrep's buffer, an error is now
394            given and the return code is set to 2.
396    45. Add a pointer to the latest mark to the callout data block.
398    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
399        partial match of an empty string instead of no match. This was specific to
400        the use of ".".
402    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
403        complete match instead of a partial match. This bug was dependent on both
404        the PCRE_UTF8 and PCRE_DOTALL options being set.
406    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
407        starting byte set, because \b was not being ignored.
410    Version 8.12 15-Jan-2011
411    ------------------------
413    1.  Fixed some typos in the markup of the man pages, and wrote a script that
414        checks for such things as part of the documentation building process.
416    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
417        --match-limit and --recursion-limit options (added for 8.11). In
418        particular, this made one of the standard tests fail. (The integer value
419        went into the wrong half of a long int.)
421    3.  If the --colour option was given to pcregrep with -v (invert match), it
422        did strange things, either producing crazy output, or crashing. It should,
423        of course, ignore a request for colour when reporting lines that do not
424        match.
426    4.  Another pcregrep bug caused similar problems if --colour was specified with
427        -M (multiline) and the pattern match finished with a line ending.
429    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
430        matched in multiline mode, the following line was shown as part of the
431        match. This seems wrong, so I have changed it.
433    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
434        the check for further matches in the same line (so they could be coloured)
435        to overrun the end of the current line. If another match was found, it was
436        incorrectly shown (and then shown again when found in the next line).
438    7.  If pcregrep was compiled under Windows, there was a reference to the
439        function pcregrep_exit() before it was defined. I am assuming this was
440        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
441        reported by a user. I've moved the definition above the reference.
444    Version 8.11 10-Dec-2010
445    ------------------------
447    1.  (*THEN) was not working properly if there were untried alternatives prior
448        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
449        backtracked to try for "b" instead of moving to the next alternative branch
450        at the same level (in this case, to look for "c"). The Perl documentation
451        is clear that when (*THEN) is backtracked onto, it goes to the "next
452        alternative in the innermost enclosing group".
454    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
455        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
456        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
457        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
458        (*THEN).
460    3.  If \s appeared in a character class, it removed the VT character from
461        the class, even if it had been included by some previous item, for example
462        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
463        of \s, but is part of the POSIX "space" class.)
465    4.  A partial match never returns an empty string (because you can always
466        match an empty string at the end of the subject); however the checking for
467        an empty string was starting at the "start of match" point. This has been
468        changed to the "earliest inspected character" point, because the returned
469        data for a partial match starts at this character. This means that, for
470        example, /(?<=abc)def/ gives a partial match for the subject "abc"
471        (previously it gave "no match").
473    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
474        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
475        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
476        has an implication that the given string is incomplete (because a partial
477        match is preferred over a full match). For this reason, these items now
478        give a partial match in this situation. [Aside: previously, the one case
479        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
480        match rather than a full match, which was wrong by the old rules, but is
481        now correct.]
483    6.  There was a bug in the handling of #-introduced comments, recognized when
484        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
485        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
486        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
487        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
488        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
489        places in pcre_compile().
491    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
492        comments when looking ahead for named forward references to subpatterns,
493        the only newline sequence it recognized was NL. It now handles newlines
494        according to the set newline convention.
496    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
497        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
498        cater for a lack of strerror(). These oversights have been fixed.
500    9.  Added --match-limit and --recursion-limit to pcregrep.
502    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
504    11. When the -o option was used, pcregrep was setting a return code of 1, even
505        when matches were found, and --line-buffered was not being honoured.
507    12. Added an optional parentheses number to the -o and --only-matching options
508        of pcregrep.
510    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
511        can match an empty string. The code to do it in pcretest and pcredemo
512        needed fixing:
514        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
515            only one byte after an empty string match just before CRLF (this case
516            just got forgotten; "any" and "anycrlf" were OK).
518        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
519            mode when an empty string match preceded an ASCII character followed by
520            a non-ASCII character. (The code for advancing by one character rather
521            than one byte was nonsense.)
523        (c) The pcredemo.c sample program did not have any code at all to handle
524            the cases when CRLF is a valid newline sequence.
526    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
527        as a starting offset was within the subject string. There is now a new
528        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
529        negative or greater than the length of the string. In order to test this,
530        pcretest is extended to allow the setting of negative starting offsets.
532    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
533        starting offset points to the beginning of a UTF-8 character was
534        unnecessarily clumsy. I tidied it up.
536    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
537        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
539    17. Nobody had reported that the --include_dir option, which was added in
540        release 7.7 should have been called --include-dir (hyphen, not underscore)
541        for compatibility with GNU grep. I have changed it to --include-dir, but
542        left --include_dir as an undocumented synonym, and the same for
543        --exclude-dir, though that is not available in GNU grep, at least as of
544        release 2.5.4.
546    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
547        characters from a string of bytes) have been redefined so as not to use
548        loops, in order to improve performance in some environments. At the same
549        time, I abstracted some of the common code into auxiliary macros to save
550        repetition (this should not affect the compiled code).
552    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
553        compile-time error is now given if \c is not followed by an ASCII
554        character, that is, a byte less than 128. (In EBCDIC mode, the code is
555        different, and any byte value is allowed.)
557    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
558        START_OPTIMIZE option, which is now allowed at compile time - but just
559        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
560        to pcregrep and other applications that have no direct access to PCRE
561        options. The new /Y option in pcretest sets this option when calling
562        pcre_compile().
564    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
565        back references. Groups containing recursive back references were forced to
566        be atomic by that change, but in the case of named groups, the amount of
567        memory required was incorrectly computed, leading to "Failed: internal
568        error: code overflow". This has been fixed.
570    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
571        pcretest.c, to avoid build problems in some Borland environments.
574    Version 8.10 25-Jun-2010
575    ------------------------
577    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
578        THEN.
580    2.  (*ACCEPT) was not working when inside an atomic group.
582    3.  Inside a character class, \B is treated as a literal by default, but
583        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
584        causes the error). The code is unchanged, but I tidied the documentation.
586    4.  Inside a character class, PCRE always treated \R and \X as literals,
587        whereas Perl faults them if its -w option is set. I have changed PCRE so
588        that it faults them when PCRE_EXTRA is set.
590    5.  Added support for \N, which always matches any character other than
591        newline. (It is the same as "." when PCRE_DOTALL is not set.)
593    6.  When compiling pcregrep with newer versions of gcc which may have
594        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
595        declared with attribute warn_unused_result" were given. Just casting the
596        result to (void) does not stop the warnings; a more elaborate fudge is
597        needed. I've used a macro to implement this.
599    7.  Minor change to pcretest.c to avoid a compiler warning.
601    8.  Added four artifical Unicode properties to help with an option to make
602        \s etc use properties (see next item). The new properties are: Xan
603        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
605    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
606        use Unicode properties. (*UCP) at the start of a pattern can be used to set
607        this option. Modified pcretest to add /W to test this facility. Added
608        REG_UCP to make it available via the POSIX interface.
610    10. Added --line-buffered to pcregrep.
612    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
613        studied, and the match started with a letter with a code point greater than
614        127 whose first byte was different to the first byte of the other case of
615        the letter, the other case of this starting letter was not recognized
616        (#976).
618    12. If a pattern that was studied started with a repeated Unicode property
619        test, for example, \p{Nd}+, there was the theoretical possibility of
620        setting up an incorrect bitmap of starting bytes, but fortunately it could
621        not have actually happened in practice until change 8 above was made (it
622        added property types that matched character-matching opcodes).
624    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
625        possible starting bytes for non-anchored patterns.
627    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
628        \R, and also a number of cases that involve Unicode properties, both
629        explicit and implicit when PCRE_UCP is set.
631    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
632        input, it could crash or give wrong results if characters with values
633        greater than 0xc0 were present in the subject string. (Detail: it assumed
634        UTF-8 input when processing these items.)
636    16. Added a lot of (int) casts to avoid compiler warnings in systems where
637        size_t is 64-bit (#991).
639    17. Added a check for running out of memory when PCRE is compiled with
640        --disable-stack-for-recursion (#990).
642    18. If the last data line in a file for pcretest does not have a newline on
643        the end, a newline was missing in the output.
645    19. The default pcre_chartables.c file recognizes only ASCII characters (values
646        less than 128) in its various bitmaps. However, there is a facility for
647        generating tables according to the current locale when PCRE is compiled. It
648        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
649        characters, are recognized by isspace() and therefore were getting set in
650        these tables, and indeed these tables seem to approximate to ISO 8859. This
651        caused a problem in UTF-8 mode when pcre_study() was used to create a list
652        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
653        which of course cannot start UTF-8 characters. I have changed the code so
654        that only real ASCII characters (less than 128) and the correct starting
655        bytes for UTF-8 encodings are set for characters greater than 127 when in
656        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
657        altogether.)
659    20. Added the /T option to pcretest so as to be able to run tests with non-
660        standard character tables, thus making it possible to include the tests
661        used for 19 above in the standard set of tests.
663    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
664        reference to a subpattern the other side of a comment that contains an
665        opening parenthesis caused either an internal compiling error, or a
666        reference to the wrong subpattern.
669    Version 8.02 19-Mar-2010
670    ------------------------
672    1.  The Unicode data tables have been updated to Unicode 5.2.0.
674    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
675        configured.
677    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
678        original author of that file, following a query about its status.
680    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
681        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
683    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
684        quantifier applied to a forward-referencing subroutine call, could compile
685        incorrect code or give the error "internal error: previously-checked
686        referenced subpattern not found".
688    6.  Both MS Visual Studio and Symbian OS have problems with initializing
689        variables to point to external functions. For these systems, therefore,
690        pcre_malloc etc. are now initialized to local functions that call the
691        relevant global functions.
693    7.  There were two entries missing in the vectors called coptable and poptable
694        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
695        I've fixed the data, and added a kludgy way of testing at compile time that
696        the lengths are correct (equal to the number of opcodes).
698    8.  Following on from 7, I added a similar kludge to check the length of the
699        eint vector in pcreposix.c.
701    9.  Error texts for pcre_compile() are held as one long string to avoid too
702        much relocation at load time. To find a text, the string is searched,
703        counting zeros. There was no check for running off the end of the string,
704        which could happen if a new error number was added without updating the
705        string.
707    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
709    11. \K was not working if it appeared in an atomic group or in a group that
710        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
711        \K is "not well defined" if used in an assertion. PCRE now accepts it if
712        the assertion is positive, but not if it is negative.
714    12. Change 11 fortuitously reduced the size of the stack frame used in the
715        "match()" function of pcre_exec.c by one pointer. Forthcoming
716        implementation of support for (*MARK) will need an extra pointer on the
717        stack; I have reserved it now, so that the stack frame size does not
718        decrease.
720    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
721        item in branch that calls a recursion is a subroutine call - as in the
722        second branch in the above example - was incorrectly given the compile-
723        time error "recursive call could loop indefinitely" because pcre_compile()
724        was not correctly checking the subroutine for matching a non-empty string.
726    14. The checks for overrunning compiling workspace could trigger after an
727        overrun had occurred. This is a "should never occur" error, but it can be
728        triggered by pathological patterns such as hundreds of nested parentheses.
729        The checks now trigger 100 bytes before the end of the workspace.
731    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
734    Version 8.01 19-Jan-2010
735    ------------------------
737    1.  If a pattern contained a conditional subpattern with only one branch (in
738        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
739        computed the wrong minimum data length (which is of course zero for such
740        subpatterns). This could cause incorrect "no match" results.
742    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
743        the pattern is reset in the first branch, pcre_compile() failed with
744        "internal error: code overflow at offset...". This happened only when
745        the reset was to the original external option setting. (An optimization
746        abstracts leading options settings into an external setting, which was the
747        cause of this.)
749    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
750        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
751        assertion pattern did not match (meaning that the assertion was true), it
752        was incorrectly treated as false if the SKIP had been reached during the
753        matching. This also applied to assertions used as conditions.
755    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
756        assertion subpattern, including such a pattern used as a condition,
757        unpredictable results occurred, instead of the error return
760    5.  The C++ GlobalReplace function was not working like Perl for the special
761        situation when an empty string is matched. It now does the fancy magic
762        stuff that is necessary.
764    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
765        removed. (These were left over from very, very early versions of PCRE.)
767    7.  Some cosmetic changes to the code to make life easier when compiling it
768        as part of something else:
770        (a) Change DEBUG to PCRE_DEBUG.
772        (b) In pcre_compile(), rename the member of the "branch_chain" structure
773            called "current" as "current_branch", to prevent a collision with the
774            Linux macro when compiled as a kernel module.
776        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
777            prevent a collision with the Linux macro when compiled as a kernel
778            module.
780    8.  In pcre_compile() there are some checks for integer overflows that used to
781        cast potentially large values to (double). This has been changed to that
782        when building, a check for int64_t is made, and if it is found, it is used
783        instead, thus avoiding the use of floating point arithmetic. (There is no
784        other use of FP in PCRE.) If int64_t is not found, the fallback is to
785        double.
787    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
788        2005 (difference between two addresses compared to an unsigned value).
790    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
791        custom one, because of the following reported problem in Windows:
793          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
794              under Win32.
795          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
796              therefore missing the function definition.
797          - The compiler thus generates a "C" signature for the test function.
798          - The linker fails to find the "C" function.
799          - PCRE fails to configure if asked to do so against libbz2.
801    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
802        messages were output:
804          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
805          rerunning libtoolize, to keep the correct libtool macros in-tree.
806          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
808        I have done both of these things.
810    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
811        most of the time, it *can* run out if it is given a pattern that contains a
812        runaway infinite recursion. I updated the discussion in the pcrestack man
813        page.
815    13. Now that we have gone to the x.xx style of version numbers, the minor
816        version may start with zero. Using 08 or 09 is a bad idea because users
817        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
818        interpreted as invalid octal numbers. I've updated the previous comment in
819        configure.ac, and also added a check that gives an error if 08 or 09 are
820        used.
822    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
823        causing partial matching to fail when the end of the subject matched \W
824        in a UTF-8 pattern where \W was quantified with a minimum of 3.
826    15. There were some discrepancies between the declarations in pcre_internal.h
827        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
828        their definitions. The declarations used "const uschar *" and the
829        definitions used USPTR. Even though USPTR is normally defined as "const
830        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
831        reported that: "This difference in casting confuses some C++ compilers, for
832        example, SunCC recognizes above declarations as different functions and
833        generates broken code for hbpcre." I have changed the declarations to use
834        USPTR.
836    16. GNU libtool is named differently on some systems. The autogen.sh script now
837        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
838        (FreeBSD).
840    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
841        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
842        comment: "Figure out how to create a longlong from a string: strtoll and
843        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
844        instance, but it only takes 2 args instead of 3!"
846    18. A subtle bug concerned with back references has been fixed by a change of
847        specification, with a corresponding code fix. A pattern such as
848        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
849        refers, was giving matches when it shouldn't. For example, xa=xaaa would
850        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
851        same bug. Such groups have to be quantified to be useful, or contained
852        inside another quantified group. (If there's no repetition, the reference
853        can never match.) The problem arises because, having left the group and
854        moved on to the rest of the pattern, a later failure that backtracks into
855        the group uses the captured value from the final iteration of the group
856        rather than the correct earlier one. I have fixed this in PCRE by forcing
857        any group that contains a reference to itself to be an atomic group; that
858        is, there cannot be any backtracking into it once it has completed. This is
859        similar to recursive and subroutine calls.
862    Version 8.00 19-Oct-09
863    ----------------------
865    1.  The table for translating pcre_compile() error codes into POSIX error codes
866        was out-of-date, and there was no check on the pcre_compile() error code
867        being within the table. This could lead to an OK return being given in
868        error.
870    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
871        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
872        in a Windows environment.
874    3.  The pcregrep --count option prints the count for each file even when it is
875        zero, as does GNU grep. However, pcregrep was also printing all files when
876        --files-with-matches was added. Now, when both options are given, it prints
877        counts only for those files that have at least one match. (GNU grep just
878        prints the file name in this circumstance, but including the count seems
879        more useful - otherwise, why use --count?) Also ensured that the
880        combination -clh just lists non-zero counts, with no names.
882    4.  The long form of the pcregrep -F option was incorrectly implemented as
883        --fixed_strings instead of --fixed-strings. This is an incompatible change,
884        but it seems right to fix it, and I didn't think it was worth preserving
885        the old behaviour.
887    5.  The command line items --regex=pattern and --regexp=pattern were not
888        recognized by pcregrep, which required --regex pattern or --regexp pattern
889        (with a space rather than an '='). The man page documented the '=' forms,
890        which are compatible with GNU grep; these now work.
892    6.  No libpcreposix.pc file was created for pkg-config; there was just
893        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
895    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
896        when UCP support is not needed, by modifying the Python script that
897        generates it from Unicode data files. This should not matter if the module
898        is correctly used as a library, but I received one complaint about 50K of
899        unwanted data. My guess is that the person linked everything into his
900        program rather than using a library. Anyway, it does no harm.
902    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
903        was a minimum greater than 1 for a wide character in a possessive
904        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
905        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
906        character. Chaos in the form of incorrect output or a compiling loop could
907        result.
909    9.  The restrictions on what a pattern can contain when partial matching is
910        requested for pcre_exec() have been removed. All patterns can now be
911        partially matched by this function. In addition, if there are at least two
912        slots in the offset vector, the offset of the earliest inspected character
913        for the match and the offset of the end of the subject are set in them when
914        PCRE_ERROR_PARTIAL is returned.
916    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
917        synonymous with PCRE_PARTIAL, for backwards compatibility, and
918        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
919        and may be more useful for multi-segment matching.
921    11. Partial matching with pcre_exec() is now more intuitive. A partial match
922        used to be given if ever the end of the subject was reached; now it is
923        given only if matching could not proceed because another character was
924        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
925        string "Z", which now yields "no match" instead of "partial match". In the
926        case of pcre_dfa_exec(), "no match" is given if every matching path for the
927        final character ended with (*FAIL).
929    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
930        if the pattern had a "must contain" character that was already found in the
931        earlier partial match, unless partial matching was again requested. For
932        example, with the pattern /dog.(body)?/, the "must contain" character is
933        "g". If the first part-match was for the string "dog", restarting with
934        "sbody" failed. This bug has been fixed.
936    13. The string returned by pcre_dfa_exec() after a partial match has been
937        changed so that it starts at the first inspected character rather than the
938        first character of the match. This makes a difference only if the pattern
939        starts with a lookbehind assertion or \b or \B (\K is not supported by
940        pcre_dfa_exec()). It's an incompatible change, but it makes the two
941        matching functions compatible, and I think it's the right thing to do.
943    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
944        so that the demonstration program is easily available in environments where
945        PCRE has not been installed from source.
947    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
948        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
949        library.
951    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
952        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
953        is not the first non-POSIX option to be added. Clearly some people find
954        these options useful.
956    17. If a caller to the POSIX matching function regexec() passes a non-zero
957        value for nmatch with a NULL value for pmatch, the value of
958        nmatch is forced to zero.
960    18. RunGrepTest did not have a test for the availability of the -u option of
961        the diff command, as RunTest does. It now checks in the same way as
962        RunTest, and also checks for the -b option.
964    19. If an odd number of negated classes containing just a single character
965        interposed, within parentheses, between a forward reference to a named
966        subpattern and the definition of the subpattern, compilation crashed with
967        an internal error, complaining that it could not find the referenced
968        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
969        [The bug was that it was starting one character too far in when skipping
970        over the character class, thus treating the ] as data rather than
971        terminating the class. This meant it could skip too much.]
973    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
974        /g option in pcretest when the pattern contains \K, which makes it possible
975        to have an empty string match not at the start, even when the pattern is
976        anchored. Updated pcretest and pcredemo to use this option.
978    21. If the maximum number of capturing subpatterns in a recursion was greater
979        than the maximum at the outer level, the higher number was returned, but
980        with unset values at the outer level. The correct (outer level) value is
981        now given.
983    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
984        PCRE did not set those parentheses (unlike Perl). I have now found a way to
985        make it do so. The string so far is captured, making this feature
986        compatible with Perl.
988    23. The tests have been re-organized, adding tests 11 and 12, to make it
989        possible to check the Perl 5.10 features against Perl 5.10.
991    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
992        pattern matches a fixed length string. PCRE did not allow this; now it
993        does. Neither allows recursion.
995    25. I finally figured out how to implement a request to provide the minimum
996        length of subject string that was needed in order to match a given pattern.
997        (It was back references and recursion that I had previously got hung up
998        on.) This code has now been added to pcre_study(); it finds a lower bound
999        to the length of subject needed. It is not necessarily the greatest lower
1000        bound, but using it to avoid searching strings that are too short does give
1001        some useful speed-ups. The value is available to calling programs via
1002        pcre_fullinfo().
1004    26. While implementing 25, I discovered to my embarrassment that pcretest had
1005        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
1006        study optimizations had never been tested with that matching function.
1007        Oops. What is worse, even when it was passed study data, there was a bug in
1008        pcre_dfa_exec() that meant it never actually used it. Double oops. There
1009        were also very few tests of studied patterns with pcre_dfa_exec().
1011    27. If (?| is used to create subpatterns with duplicate numbers, they are now
1012        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
1013        on the other side of the coin, they are no longer allowed to have different
1014        names, because these cannot be distinguished in PCRE, and this has caused
1015        confusion. (This is a difference from Perl.)
1017    28. When duplicate subpattern names are present (necessarily with different
1018        numbers, as required by 27 above), and a test is made by name in a
1019        conditional pattern, either for a subpattern having been matched, or for
1020        recursion in such a pattern, all the associated numbered subpatterns are
1021        tested, and the overall condition is true if the condition is true for any
1022        one of them. This is the way Perl works, and is also more like the way
1023        testing by number works.
1026    Version 7.9 11-Apr-09
1027    ---------------------
1029    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
1030        (pcretest), all targets were linked against these libraries. This included
1031        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
1032        libraries. This caused unwanted dependencies to be created. This problem
1033        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
1034        pcretest is linked with readline.
1036    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
1037        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
1038        moved outside it again, because FALSE and TRUE are already defined in AIX,
1039        but BOOL is not.
1041    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
1042        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
1044    4.  The pcregrep documentation said spaces were inserted as well as colons (or
1045        hyphens) following file names and line numbers when outputting matching
1046        lines. This is not true; no spaces are inserted. I have also clarified the
1047        wording for the --colour (or --color) option.
1049    5.  In pcregrep, when --colour was used with -o, the list of matching strings
1050        was not coloured; this is different to GNU grep, so I have changed it to be
1051        the same.
1053    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
1054        each matching line was coloured. Now it goes on to look for further matches
1055        of any of the test patterns, which is the same behaviour as GNU grep.
1057    7.  A pattern that could match an empty string could cause pcregrep to loop; it
1058        doesn't make sense to accept an empty string match in pcregrep, so I have
1059        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
1060        seems to be how GNU grep behaves.
1062    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
1063        start or after a newline", because the conditional assertion was not being
1064        correctly handled. The rule now is that both the assertion and what follows
1065        in the first alternative must satisfy the test.
1067    9.  If auto-callout was enabled in a pattern with a conditional group whose
1068        condition was an assertion, PCRE could crash during matching, both with
1069        pcre_exec() and pcre_dfa_exec().
1071    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
1072        used for matching.
1074    11. Unicode property support in character classes was not working for
1075        characters (bytes) greater than 127 when not in UTF-8 mode.
1077    12. Added the -M command line option to pcretest.
1079    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
1081    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
1083    16. Added comments and documentation about mis-use of no_arg in the C++
1084        wrapper.
1086    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
1087        from Martin Jerabek that uses macro names for all relevant character and
1088        string constants.
1090    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
1091        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
1092        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
1093        these, but not everybody uses configure.
1095    19. A conditional group that had only one branch was not being correctly
1096        recognized as an item that could match an empty string. This meant that an
1097        enclosing group might also not be so recognized, causing infinite looping
1098        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
1099        with the subject "ab", where knowledge that the repeated group can match
1100        nothing is needed in order to break the loop.
1102    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
1103        exec(), but without supplying a callout function, matching went wrong.
1105    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
1106        leak if the size of the offset vector was greater than 30. When the vector
1107        is smaller, the saved offsets during recursion go onto a local stack
1108        vector, but for larger vectors malloc() is used. It was failing to free
1109        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
1110        error, in fact).
1112    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
1113        heapframe that is used only when UTF-8 support is enabled. This caused no
1114        problem, but was untidy.
1116    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
1117        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
1118        included within another project.
1120    24. Steven Van Ingelgem's patches to add more options to the CMake support,
1121        slightly modified by me:
1123          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
1124              not building pcregrep.
1126          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
1127              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
1129    25. Forward references, both numeric and by name, in patterns that made use of
1130        duplicate group numbers, could behave incorrectly or give incorrect errors,
1131        because when scanning forward to find the reference group, PCRE was not
1132        taking into account the duplicate group numbers. A pattern such as
1133        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
1135    26. Changed a few more instances of "const unsigned char *" to USPTR, making
1136        the feature of a custom pointer more persuasive (as requested by a user).
1138    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
1139        pcretest.c, inside #ifndefs, because it seems they are sometimes already
1140        pre-defined.
1142    28. Added support for (*UTF8) at the start of a pattern.
1144    29. Arrange for flags added by the "release type" setting in CMake to be shown
1145        in the configuration summary.
1148    Version 7.8 05-Sep-08
1149    ---------------------
1151    1.  Replaced UCP searching code with optimized version as implemented for Ad
1152        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
1153        stage table and inline lookup instead of a function, giving speed ups of 2
1154        to 5 times on some simple patterns that I tested. Permission was given to
1155        distribute the MultiStage2.py script that generates the tables (it's not in
1156        the tarball, but is in the Subversion repository).
1158    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
1159        scripts.
1161    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
1162        a group with a zero qualifier. The result of the study could be incorrect,
1163        or the function might crash, depending on the pattern.
1165    4.  Caseless matching was not working for non-ASCII characters in back
1166        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
1167        It now works when Unicode Property Support is available.
1169    5.  In pcretest, an escape such as \x{de} in the data was always generating
1170        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
1171        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
1172        truncation.
1174    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
1176    7.  Added two (int) casts to pcregrep when printing the difference of two
1177        pointers, in case they are 64-bit values.
1179    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
1180        test 2 if it fails.
1182    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
1183        and a #define of that name to empty if it is not externally set. This is to
1184        allow users of MSVC to set it if necessary.
1186    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
1187        the convenience functions in the pcre_get.c source file.
1189    11. An option change at the start of a pattern that had top-level alternatives
1190        could cause overwriting and/or a crash. This command provoked a crash in
1191        some environments:
1193          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
1195        This potential security problem was recorded as CVE-2008-2371.
1197    12. For a pattern where the match had to start at the beginning or immediately
1198        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
1199        pcre_dfa_exec() could read past the end of the passed subject if there was
1200        no match. To help with detecting such bugs (e.g. with valgrind), I modified
1201        pcretest so that it places the subject at the end of its malloc-ed buffer.
1203    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
1204        exec() might read past the end of the data buffer in UTF-8 mode.
1206    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
1207        the data contained the byte 0x85 as part of a UTF-8 character within its
1208        first line. This applied both to normal and DFA matching.
1210    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
1211        /^[^d]*?$/8 failed to match "abc".
1213    16. Added a missing copyright notice to pcrecpp_internal.h.
1215    17. Make it more clear in the documentation that values returned from
1216        pcre_exec() in ovector are byte offsets, not character counts.
1218    18. Tidied a few places to stop certain compilers from issuing warnings.
1220    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
1221        supplied by Stefan Weber. I made a further small update for 7.8 because
1222        there is a change of source arrangements: the pcre_searchfuncs.c module is
1223        replaced by pcre_ucd.c.
1226    Version 7.7 07-May-08
1227    ---------------------
1229    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
1230        a string to a long long, pretend we don't even have a long long." This is
1231        done by checking for the strtoq, strtoll, and _strtoi64 functions.
1233    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
1234        pre-7.6 versions, which defined a global no_arg variable instead of putting
1235        it in the RE class. (See also #8 below.)
1237    3.  Remove a line of dead code, identified by coverity and reported by Nuno
1238        Lopes.
1240    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
1242        (1) The include/exclude patterns were being applied to the whole pathnames
1243            of files, instead of just to the final components.
1245        (2) If there was more than one level of directory, the subdirectories were
1246            skipped unless they satisfied the include/exclude conditions. This is
1247            inconsistent with GNU grep (and could even be seen as contrary to the
1248            pcregrep specification - which I improved to make it absolutely clear).
1249            The action now is always to scan all levels of directory, and just
1250            apply the include/exclude patterns to regular files.
1252    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
1253        --exclude_dir in the tests to avoid scanning .svn directories.
1255    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
1256        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
1257        doesn't support NULs in patterns.
1259    7.  Added some missing "const"s to declarations of static tables in
1260        pcre_compile.c and pcre_dfa_exec.c.
1262    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
1263        caused by fix #2  above. (Subsequently also a second patch to fix the
1264        first patch. And a third patch - this was a messy problem.)
1266    9.  Applied Craig's patch to remove the use of push_back().
1268    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
1269        matching function regexec().
1271    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
1272        which, however, unlike Perl's \g{...}, are subroutine calls, not back
1273        references. PCRE supports relative numbers with this syntax (I don't think
1274        Oniguruma does).
1276    12. Previously, a group with a zero repeat such as (...){0} was completely
1277        omitted from the compiled regex. However, this means that if the group
1278        was called as a subroutine from elsewhere in the pattern, things went wrong
1279        (an internal error was given). Such groups are now left in the compiled
1280        pattern, with a new opcode that causes them to be skipped at execution
1281        time.
1283    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
1284        to the way PCRE behaves:
1286        (a) A lone ] character is dis-allowed (Perl treats it as data).
1288        (b) A back reference to an unmatched subpattern matches an empty string
1289            (Perl fails the current match path).
1291        (c) A data ] in a character class must be notated as \] because if the
1292            first data character in a class is ], it defines an empty class. (In
1293            Perl it is not possible to have an empty class.) The empty class []
1294            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
1295            The negative empty class [^] matches any one character, independently
1296            of the DOTALL setting.
1298    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
1299        non-existent subpattern following a character class starting with ']' and
1300        containing () gave an internal compiling error instead of "reference to
1301        non-existent subpattern". Fortunately, when the pattern did exist, the
1302        compiled code was correct. (When scanning forwards to check for the
1303        existencd of the subpattern, it was treating the data ']' as terminating
1304        the class, so got the count wrong. When actually compiling, the reference
1305        was subsequently set up correctly.)
1307    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
1308        it was being rejected as not supported by pcre_dfa_exec(), even though
1309        other assertions are supported. I have made pcre_dfa_exec() support
1310        (*FAIL).
1312    16. The implementation of 13c above involved the invention of a new opcode,
1313        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
1314        cannot be changed at match time, I realized I could make a small
1315        improvement to matching performance by compiling OP_ALLANY instead of
1316        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
1317        on the OP_ANY path.
1319    17. Compiling pcretest on Windows with readline support failed without the
1320        following two fixes: (1) Make the unistd.h include conditional on
1321        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
1323    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
1324        ncurses library to be included for pcretest when ReadLine support is
1325        requested, but also to allow for it to be overridden. This patch came from
1326        Daniel Bergström.
1328    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
1329        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
1330        any errors with the current Unicode tables. Thanks to Peter Kankowski for
1331        spotting this.
1334    Version 7.6 28-Jan-08
1335    ---------------------
1337    1.  A character class containing a very large number of characters with
1338        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
1339        overflow.
1341    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
1342        HAVE_LONG_LONG is not defined.
1344    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
1345        bring them up to date and include new features. This patch includes:
1347        - Fixed PH's badly added libz and libbz2 support.
1348        - Fixed a problem with static linking.
1349        - Added pcredemo. [But later removed - see 7 below.]
1350        - Fixed dftables problem and added an option.
1351        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
1352            HAVE_LONG_LONG.
1353        - Added readline support for pcretest.
1354        - Added an listing of the option settings after cmake has run.
1356    4.  A user submitted a patch to Makefile that makes it easy to create
1357        "pcre.dll" under mingw when using Configure/Make. I added stuff to
1358        Makefile.am that cause it to include this special target, without
1359        affecting anything else. Note that the same mingw target plus all
1360        the other distribution libraries and programs are now supported
1361        when configuring with CMake (see 6 below) instead of with
1362        Configure/Make.
1364    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
1365        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
1366        exported in the Windows port". It has not yet been confirmed that the patch
1367        solves the problem, but it does no harm.
1369    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
1370        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
1371        with CMake, and also correct the comment about stack recursion.
1373    7.  Remove the automatic building of pcredemo from the ./configure system and
1374        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
1375        of a program that users should build themselves after PCRE is installed, so
1376        building it automatically is not really right. What is more, it gave
1377        trouble in some build environments.
1379    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
1382    Version 7.5 10-Jan-08
1383    ---------------------
1385    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
1386        values in parens when parsing an RE using the C++ wrapper."
1388    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
1389        Characters greater than 255 were excluded from the class instead of being
1390        included.
1392    3.  The same bug as (2) above applied to negated POSIX classes such as
1393        [:^space:].
1395    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
1396        defined or documented. It seems to have been a typo for PCRE_STATIC, so
1397        I have changed it.
1399    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
1400        first named subpattern) and a construct such as (?&a) would reference the
1401        first named subpattern whose name started with "a" (in other words, the
1402        length check was missing). Both these problems are fixed. "Subpattern name
1403        expected" is now given for (?&) (a zero-length name), and this patch also
1404        makes it give the same error for \k'' (previously it complained that that
1405        was a reference to a non-existent subpattern).
1407    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
1408        this is right because (?- can be followed by option settings as well as by
1409        digits. I have, however, made the messages clearer.
1411    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
1412        than the number used in the conditional) now cause a compile-time error.
1413        This is actually not compatible with Perl, which accepts such patterns, but
1414        treats the conditional as always being FALSE (as PCRE used to), but it
1415        seems to me that giving a diagnostic is better.
1417    8.  Change "alphameric" to the more common word "alphanumeric" in comments
1418        and messages.
1420    9.  Fix two occurrences of "backslash" in comments that should have been
1421        "backspace".
1423    10. Remove two redundant lines of code that can never be obeyed (their function
1424        was moved elsewhere).
1426    11. The program that makes PCRE's Unicode character property table had a bug
1427        which caused it to generate incorrect table entries for sequences of
1428        characters that have the same character type, but are in different scripts.
1429        It amalgamated them into a single range, with the script of the first of
1430        them. In other words, some characters were in the wrong script. There were
1431        thirteen such cases, affecting characters in the following ranges:
1433          U+002b0 - U+002c1
1434          U+0060c - U+0060d
1435          U+0061e - U+00612
1436          U+0064b - U+0065e
1437          U+0074d - U+0076d
1438          U+01800 - U+01805
1439          U+01d00 - U+01d77
1440          U+01d9b - U+01dbf
1441          U+0200b - U+0200f
1442          U+030fc - U+030fe
1443          U+03260 - U+0327f
1444          U+0fb46 - U+0fbb1
1445          U+10450 - U+1049d
1447    12. The -o option (show only the matching part of a line) for pcregrep was not
1448        compatible with GNU grep in that, if there was more than one match in a
1449        line, it showed only the first of them. It now behaves in the same way as
1450        GNU grep.
1452    13. If the -o and -v options were combined for pcregrep, it printed a blank
1453        line for every non-matching line. GNU grep prints nothing, and pcregrep now
1454        does the same. The return code can be used to tell if there were any
1455        non-matching lines.
1457    14. Added --file-offsets and --line-offsets to pcregrep.
1459    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
1460        infinitely looping recursion. The bug was that positive lookaheads were not
1461        being skipped when checking for a possible empty match (negative lookaheads
1462        and both kinds of lookbehind were skipped).
1464    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
1465        inclusion of <windows.h> to before rather than after the definition of
1466        INVALID_FILE_ATTRIBUTES (patch from David Byron).
1468    17. Specifying a possessive quantifier with a specific limit for a Unicode
1469        character property caused pcre_compile() to compile bad code, which led at
1470        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
1471        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
1472        caused the error; without that there was no problem.
1474    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
1476    19. Added --enable-pcretest-libreadline.
1478    20. In pcrecpp.cc, the variable 'count' was incremented twice in
1479        RE::GlobalReplace(). As a result, the number of replacements returned was
1480        double what it should be. I removed one of the increments, but Craig sent a
1481        later patch that removed the other one (the right fix) and added unit tests
1482        that check the return values (which was not done before).
1484    21. Several CMake things:
1486        (1) Arranged that, when cmake is used on Unix, the libraries end up with
1487            the names libpcre and libpcreposix, not just pcre and pcreposix.
1489        (2) The above change means that pcretest and pcregrep are now correctly
1490            linked with the newly-built libraries, not previously installed ones.
1494    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
1495        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
1496        UTF-8 newline character). The key issue is that the pattern starts .*;
1497        this means that the match must be either at the beginning, or after a
1498        newline. The bug was in the code for advancing after a failed match and
1499        checking that the new position followed a newline. It was not taking
1500        account of UTF-8 characters correctly.
1502    23. PCRE was behaving differently from Perl in the way it recognized POSIX
1503        character classes. PCRE was not treating the sequence [:...:] as a
1504        character class unless the ... were all letters. Perl, however, seems to
1505        allow any characters between [: and :], though of course it rejects as
1506        unknown any "names" that contain non-letters, because all the known class
1507        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
1508        for example, whereas PCRE did not - it did not recognize a POSIX character
1509        class. This seemed a bit dangerous, so the code has been changed to be
1510        closer to Perl. The behaviour is not identical to Perl, because PCRE will
1511        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
1512        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
1513        Perl does, and where it didn't before.
1515    24. Rewrite so as to remove the single use of %n from pcregrep because in some
1516        Windows environments %n is disabled by default.
1519    Version 7.4 21-Sep-07
1520    ---------------------
1522    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
1523        means that a class such as [\s] counted as "explicit reference to CR or
1524        LF". That isn't really right - the whole point of the change was to try to
1525        help when there was an actual mention of one of the two characters. So now
1526        the change happens only if \r or \n (or a literal CR or LF) character is
1527        encountered.
1529    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
1530        of both had grown to the point where there were only 3 bits left.
1531        Fortunately, there was spare space in the data structure, and so I have
1532        moved the internal flags into a new 16-bit field to free up more option
1533        bits.
1535    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
1536        but did not set the internal JCHANGED flag - either of these is enough to
1537        control the way the "get" function works - but the PCRE_INFO_JCHANGED
1538        facility is supposed to tell if (?J) was ever used, so now (?J) at the
1539        start sets both bits.
1541    4.  Added options (at build time, compile time, exec time) to change \R from
1542        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
1544    5.  doc/pcresyntax.html was missing from the distribution.
1546    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
1547        compatibility, even though it is no longer used.
1549    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
1550        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
1551        windows.h file is present (where different names are used). [This was
1552        reversed later after testing - see 16 below.]
1554    8.  Changed all #include <config.h> to #include "config.h". There were also
1555        some further <pcre.h> cases that I changed to "pcre.h".
1557    9.  When pcregrep was used with the --colour option, it missed the line ending
1558        sequence off the lines that it output.
1560    10. It was pointed out to me that arrays of string pointers cause lots of
1561        relocations when a shared library is dynamically loaded. A technique of
1562        using a single long string with a table of offsets can drastically reduce
1563        these. I have refactored PCRE in four places to do this. The result is
1564        dramatic:
1566          Originally:                          290
1567          After changing UCP table:            187
1568          After changing error message table:   43
1569          After changing table of "verbs"       36
1570          After changing table of Posix names   22
1572        Thanks to the folks working on Gregex for glib for this insight.
1574    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
1575        unicode-properties was also set.
1577    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
1579    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
1580        checked only for CRLF.
1582    14. Added casts to pcretest.c to avoid compiler warnings.
1584    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
1586    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
1587        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
1588        entirely. This removes changes made in 7 above.
1590    17. The CMake files have been updated, and there is now more information about
1591        building with CMake in the NON-UNIX-USE document.
1594    Version 7.3 28-Aug-07
1595    ---------------------
1597     1. In the rejigging of the build system that eventually resulted in 7.1, the
1598        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
1599        brackets there is not right, since it causes compilers to look for an
1600        installed pcre.h, not the version that is in the source that is being
1601        compiled (which of course may be different). I have changed it back to:
1603          #include "pcre.h"
1605        I have a vague recollection that the change was concerned with compiling in
1606        different directories, but in the new build system, that is taken care of
1607        by the VPATH setting the Makefile.
1609     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
1610        when the subject happened to end in the byte 0x85 (e.g. if the last
1611        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
1612        characters but of course it shouldn't be taken as a newline when it is part
1613        of another character. The bug was that, for an unlimited repeat of . in
1614        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
1615        characters when looking for a newline.
1617     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
1619     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
1620        in debug output.
1622     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
1623        long printing in the pcrecpp unittest when running under MinGW.
1625     6. ESC_K was left out of the EBCDIC table.
1627     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
1628        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
1629        limit also applies to "virtual nesting" when a pattern is recursive, and in
1630        this case 1000 isn't so big. I have been able to remove this limit at the
1631        expense of backing off one optimization in certain circumstances. Normally,
1632        when pcre_exec() would call its internal match() function recursively and
1633        immediately return the result unconditionally, it uses a "tail recursion"
1634        feature to save stack. However, when a subpattern that can match an empty
1635        string has an unlimited repetition quantifier, it no longer makes this
1636        optimization. That gives it a stack frame in which to save the data for
1637        checking that an empty string has been matched. Previously this was taken
1638        from the 1000-entry workspace that had been reserved. So now there is no
1639        explicit limit, but more stack is used.
1641     8. Applied Daniel's patches to solve problems with the import/export magic
1642        syntax that is required for Windows, and which was going wrong for the
1643        pcreposix and pcrecpp parts of the library. These were overlooked when this
1644        problem was solved for the main library.
1646     9. There were some crude static tests to avoid integer overflow when computing
1647        the size of patterns that contain repeated groups with explicit upper
1648        limits. As the maximum quantifier is 65535, the maximum group length was
1649        set at 30,000 so that the product of these two numbers did not overflow a
1650        32-bit integer. However, it turns out that people want to use groups that
1651        are longer than 30,000 bytes (though not repeat them that many times).
1652        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1653        made it possible to implement the integer overflow checks in a much more
1654        dynamic way, which I have now done. The artificial limitation on group
1655        length has been removed - we now have only the limit on the total length of
1656        the compiled pattern, which depends on the LINK_SIZE setting.
1658    10. Fixed a bug in the documentation for get/copy named substring when
1659        duplicate names are permitted. If none of the named substrings are set, the
1660        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1661        empty string.
1663    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1664        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1665        because the ] is interpreted as the first data character and the
1666        terminating ] is not found. PCRE has been made compatible with Perl in this
1667        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1668        cause memory overwriting.
1670    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1671        string has been matched (to stop an infinite loop). It was not recognizing
1672        a conditional subpattern that could match an empty string if that
1673        subpattern was within another subpattern. For example, it looped when
1674        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
1675        condition was not nested. This bug has been fixed.
1677    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1678        past the start of the subject in the presence of bytes with the top bit
1679        set, for example "\x8aBCD".
1681    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1682        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1684    14. Optimized (?!) to (*FAIL).
1686    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1687        This restricts code points to be within the range 0 to 0x10FFFF, excluding
1688        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1689        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1690        does: it's just the validity check that is more restrictive.
1692    16. Inserted checks for integer overflows during escape sequence (backslash)
1693        processing, and also fixed erroneous offset values for syntax errors during
1694        backslash processing.
1696    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1697        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1699    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1700        caused an overrun.
1702    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1703        something other than just ASCII characters) inside a group that had an
1704        unlimited repeat caused a loop at compile time (while checking to see
1705        whether the group could match an empty string).
1707    20. Debugging a pattern containing \p or \P could cause a crash. For example,
1708        [\P{Any}] did so. (Error in the code for printing property names.)
1710    21. An orphan \E inside a character class could cause a crash.
1712    22. A repeated capturing bracket such as (A)? could cause a wild memory
1713        reference during compilation.
1715    23. There are several functions in pcre_compile() that scan along a compiled
1716        expression for various reasons (e.g. to see if it's fixed length for look
1717        behind). There were bugs in these functions when a repeated \p or \P was
1718        present in the pattern. These operators have additional parameters compared
1719        with \d, etc, and these were not being taken into account when moving along
1720        the compiled data. Specifically:
1722        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1723            length.
1725        (b) An item such as \pL+ within a repeated group could cause crashes or
1726            loops.
1728        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1729            "reference to non-existent subpattern" error.
1731        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1733    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1734        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1736    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1738    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1739        character were causing crashes (broken optimization).
1741    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1742        \p or \P) caused a compile-time loop.
1744    28. More problems have arisen in unanchored patterns when CRLF is a valid line
1745        break. For example, the unstudied pattern [\r\n]A does not match the string
1746        "\r\nA" because change 7.0/46 below moves the current point on by two
1747        characters after failing to match at the start. However, the pattern \nA
1748        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
1749        the same is true. There doesn't seem any very clean way out of this, but
1750        what I have chosen to do makes the common cases work: PCRE now takes note
1751        of whether there can be an explicit match for \r or \n anywhere in the
1752        pattern, and if so, 7.0/46 no longer applies. As part of this change,
1753        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1754        pattern has explicit CR or LF references.
1756    29. Added (*CR) etc for changing newline setting at start of pattern.
1759    Version 7.2 19-Jun-07
1760    ---------------------
1762     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
1763        which is apparently normally available under Windows.
1765     2. Re-jig the pcregrep tests with different newline settings in an attempt
1766        to make them independent of the local environment's newline setting.
1768     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
1770     4. Some of the "internals" tests were previously cut out when the link size
1771        was not 2, because the output contained actual offsets. The recent new
1772        "Z" feature of pcretest means that these can be cut out, making the tests
1773        usable with all link sizes.
1775     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
1776        stack recursion. This gives a massive performance boost under BSD, but just
1777        a small improvement under Linux. However, it saves one field in the frame
1778        in all cases.
1780     6. Added more features from the forthcoming Perl 5.10:
1782        (a) (?-n) (where n is a string of digits) is a relative subroutine or
1783            recursion call. It refers to the nth most recently opened parentheses.
1785        (b) (?+n) is also a relative subroutine call; it refers to the nth next
1786            to be opened parentheses.
1788        (c) Conditions that refer to capturing parentheses can be specified
1789            relatively, for example, (?(-2)... or (?(+3)...
1791        (d) \K resets the start of the current match so that everything before
1792            is not part of it.
1794        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1796        (f) \g{name} is another synonym - part of Perl 5.10's unification of
1797            reference syntax.
1799        (g) (?| introduces a group in which the numbering of parentheses in each
1800            alternative starts with the same number.
1802        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
1804     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
1807     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
1808        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
1809        for detecting groups that can match an empty string.
1811     9. A pattern with a very large number of alternatives (more than several
1812        hundred) was running out of internal workspace during the pre-compile
1813        phase, where pcre_compile() figures out how much memory will be needed. A
1814        bit of new cunning has reduced the workspace needed for groups with
1815        alternatives. The 1000-alternative test pattern now uses 12 bytes of
1816        workspace instead of running out of the 4096 that are available.
1818    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
1820    11. Applied patch from Google to remove an optimization that didn't quite work.
1821        The report of the bug said:
1823          pcrecpp::RE("a*").FullMatch("aaa") matches, while
1824          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
1825          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
1827    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
1828        it matched the wrong number of bytes.
1831    Version 7.1 24-Apr-07
1832    ---------------------
1834     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
1835        that is more "standard", making use of automake and other Autotools. There
1836        is some re-arrangement of the files and adjustment of comments consequent
1837        on this.
1839     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
1840        for recursive directory scanning broke on some systems because the files
1841        are not scanned in any specific order and on different systems the order
1842        was different. A call to "sort" has been inserted into RunGrepTest for the
1843        approprate test as a short-term fix. In the longer term there may be an
1844        alternative.
1846     3. I had an email from Eric Raymond about problems translating some of PCRE's
1847        man pages to HTML (despite the fact that I distribute HTML pages, some
1848        people do their own conversions for various reasons). The problems
1849        concerned the use of low-level troff macros .br and .in. I have therefore
1850        removed all such uses from the man pages (some were redundant, some could
1851        be replaced by .nf/.fi pairs). The 132html script that I use to generate
1852        HTML has been updated to handle .nf/.fi and to complain if it encounters
1853        .br or .in.
1855     4. Updated comments in configure.ac that get placed in config.h.in and also
1856        arranged for config.h to be included in the distribution, with the name
1857        config.h.generic, for the benefit of those who have to compile without
1858        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
1860     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
1861        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
1862        makevp.bat for the current PCRE, using the additional files
1863        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
1865     6. A Windows user reported a minor discrepancy with test 2, which turned out
1866        to be caused by a trailing space on an input line that had got lost in his
1867        copy. The trailing space was an accident, so I've just removed it.
1869     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
1870        that is needed.
1872     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
1873        as "const" (a) because they are and (b) because it helps the PHP
1874        maintainers who have recently made a script to detect big data structures
1875        in the php code that should be moved to the .rodata section. I remembered
1876        to update Builducptable as well, so it won't revert if ucptable.h is ever
1877        re-created.
1879     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
1880        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
1881        order to be able to cut out the UTF-8 tables in the latter when UTF-8
1882        support is not required. This saves 1.5-2K of code, which is important in
1883        some applications.
1885        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
1886        so as not to refer to the tables, even though these functions will never be
1887        called when UTF-8 support is disabled. Otherwise there are problems with a
1888        shared library.
1890    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
1892        (a) It was defining its arguments as char * instead of void *.
1894        (b) It was assuming that all moves were upwards in memory; this was true
1895            a long time ago when I wrote it, but is no longer the case.
1897        The emulated memove() is provided for those environments that have neither
1898        memmove() nor bcopy(). I didn't think anyone used it these days, but that
1899        is clearly not the case, as these two bugs were recently reported.
1901    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
1902        and Detrail to create the HTML documentation, the .txt form of the man
1903        pages, and it removes trailing spaces from listed files. It also creates
1904        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
1905        case, it wraps all the #defines with #ifndefs. This script should be run
1906        before "make dist".
1908    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1909        with Unicode property support.
1911        (a) For a maximizing quantifier, if the two different cases of the
1912            character were of different lengths in their UTF-8 codings (there are
1913            some cases like this - I found 11), and the matching function had to
1914            back up over a mixture of the two cases, it incorrectly assumed they
1915            were both the same length.
1917        (b) When PCRE was configured to use the heap rather than the stack for
1918            recursion during matching, it was not correctly preserving the data for
1919            the other case of a UTF-8 character when checking ahead for a match
1920            while processing a minimizing repeat. If the check also involved
1921            matching a wide character, but failed, corruption could cause an
1922            erroneous result when trying to check for a repeat of the original
1923            character.
1925    13. Some tidying changes to the testing mechanism:
1927        (a) The RunTest script now detects the internal link size and whether there
1928            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1929            values substituted by "configure". (The RunGrepTest script already did
1930            this for UTF-8.) The configure.ac script no longer substitutes the
1931            relevant variables.
1933        (b) The debugging options /B and /D in pcretest show the compiled bytecode
1934            with length and offset values. This means that the output is different
1935            for different internal link sizes. Test 2 is skipped for link sizes
1936            other than 2 because of this, bypassing the problem. Unfortunately,
1937            there was also a test in test 3 (the locale tests) that used /B and
1938            failed for link sizes other than 2. Rather than cut the whole test out,
1939            I have added a new /Z option to pcretest that replaces the length and
1940            offset values with spaces. This is now used to make test 3 independent
1941            of link size. (Test 2 will be tidied up later.)
1943    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1944        segmentation fault instead of returning the appropriate error message.
1946    15. In multiline mode when the newline sequence was set to "any", the pattern
1947        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1948        This doesn't seem right; it now treats the CRLF combination as the line
1949        ending, and so does not match in that case. It's only a pattern such as ^$
1950        that would hit this one: something like ^ABC$ would have failed after \r
1951        and then tried again after \r\n.
1953    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1954        in an attempt to make files that differ only in their line terminators
1955        compare equal. This works on Linux.
1957    17. Under certain error circumstances pcregrep might try to free random memory
1958        as it exited. This is now fixed, thanks to valgrind.
1960    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1961        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1962        was because its rules for how to advance for /g after matching an empty
1963        string at the end of a line did not allow for this case. They now check for
1964        it specially.
1966    20. pcretest is supposed to handle patterns and data of any length, by
1967        extending its buffers when necessary. It was getting this wrong when the
1968        buffer for a data line had to be extended.
1970    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1971        CRLF as a newline sequence.
1973    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1974        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1975        I have nevertheless tidied it up.
1977    23. Added some casts to kill warnings from HP-UX ia64 compiler.
1979    24. Added a man page for pcre-config.
1982    Version 7.0 19-Dec-06
1983    ---------------------
1985     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1986        moving to gcc 4.1.1.
1988     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1989        sys/time.h, which is documented as needed for this function. It doesn't
1990        seem to matter on Linux, but it showed up on some releases of OS X.
1992     3. It seems that there are systems where bytes whose values are greater than
1993        127 match isprint() in the "C" locale. The "C" locale should be the
1994        default when a C program starts up. In most systems, only ASCII printing
1995        characters match isprint(). This difference caused the output from pcretest
1996        to vary, making some of the tests fail. I have changed pcretest so that:
1998        (a) When it is outputting text in the compiled version of a pattern, bytes
1999            other than 32-126 are always shown as hex escapes.
2001        (b) When it is outputting text that is a matched part of a subject string,
2002            it does the same, unless a different locale has been set for the match
2003            (using the /L modifier). In this case, it uses isprint() to decide.
2005     4. Fixed a major bug that caused incorrect computation of the amount of memory
2006        required for a compiled pattern when options that changed within the
2007        pattern affected the logic of the preliminary scan that determines the
2008        length. The relevant options are -x, and -i in UTF-8 mode. The result was
2009        that the computed length was too small. The symptoms of this bug were
2010        either the PCRE error "internal error: code overflow" from pcre_compile(),
2011        or a glibc crash with a message such as "pcretest: free(): invalid next
2012        size (fast)". Examples of patterns that provoked this bug (shown in
2013        pcretest format) are:
2015          /(?-x: )/x
2016          /(?x)(?-x: \s*#\s*)/
2017          /((?i)[\x{c0}])/8
2018          /(?i:[\x{c0}])/8
2020        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
2021        is now done differently.
2023     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
2024        wrapper classes; (b) implement a new function in the C++ scanner that is
2025        more efficient than the old way of doing things because it avoids levels of
2026        recursion in the regex matching; (c) add a paragraph to the documentation
2027        for the FullMatch() function.
2029     6. The escape sequence \n was being treated as whatever was defined as
2030        "newline". Not only was this contrary to the documentation, which states
2031        that \n is character 10 (hex 0A), but it also went horribly wrong when
2032        "newline" was defined as CRLF. This has been fixed.
2034     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
2035        was being set to -1 for the "end of line" case (supposedly a value that no
2036        character can have). Though this value is never used (the check for end of
2037        line is "zero bytes in current character"), it caused compiler complaints.
2038        I've changed it to 0xffffffff.
2040     8. In pcre_version.c, the version string was being built by a sequence of
2041        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
2042        string (as it is for production releases) called a macro with an empty
2043        argument. The C standard says the result of this is undefined. The gcc
2044        compiler treats it as an empty string (which was what was wanted) but it is
2045        reported that Visual C gives an error. The source has been hacked around to
2046        avoid this problem.
2048     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
2049        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
2050        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
2051        of them did).
2053    10. Originally, pcretest opened its input and output without "b"; then I was
2054        told that "b" was needed in some environments, so it was added for release
2055        5.0 to both the input and output. (It makes no difference on Unix-like
2056        systems.) Later I was told that it is wrong for the input on Windows. I've
2057        now abstracted the modes into two macros, to make it easier to fiddle with
2058        them, and removed "b" from the input mode under Windows.
2060    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
2062    12. Added -help and --help to pcretest as an official way of being reminded
2063        of the options.
2065    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
2066        and pcrecpp.cc because they annoy compilers at high warning levels.
2068    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
2070    15. Fixed an occurrence of == in configure.ac that should have been = (shell
2071        scripts are not C programs :-) and which was not noticed because it works
2072        on Linux.
2074    16. pcretest is supposed to handle any length of pattern and data line (as one
2075        line or as a continued sequence of lines) by extending its input buffer if
2076        necessary. This feature was broken for very long pattern lines, leading to
2077        a string of junk being passed to pcre_compile() if the pattern was longer
2078        than about 50K.
2080    17. I have done a major re-factoring of the way pcre_compile() computes the
2081        amount of memory needed for a compiled pattern. Previously, there was code
2082        that made a preliminary scan of the pattern in order to do this. That was
2083        OK when PCRE was new, but as the facilities have expanded, it has become
2084        harder and harder to keep it in step with the real compile phase, and there
2085        have been a number of bugs (see for example, 4 above). I have now found a
2086        cunning way of running the real compile function in a "fake" mode that
2087        enables it to compute how much memory it would need, while actually only
2088        ever using a few hundred bytes of working memory and without too many
2089        tests of the mode. This should make future maintenance and development
2090        easier. A side effect of this work is that the limit of 200 on the nesting
2091        depth of parentheses has been removed (though this was never a serious
2092        limitation, I suspect). However, there is a downside: pcre_compile() now
2093        runs more slowly than before (30% or more, depending on the pattern). I
2094        hope this isn't a big issue. There is no effect on runtime performance.
2096    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
2097        newline (only possible for the last line of a file) and it was a
2098        pattern that set a locale (followed by /Lsomething), pcretest crashed.
2100    19. Added additional timing features to pcretest. (1) The -tm option now times
2101        matching only, not compiling. (2) Both -t and -tm can be followed, as a
2102        separate command line item, by a number that specifies the number of
2103        repeats to use when timing. The default is 50000; this gives better
2104        precision, but takes uncomfortably long for very large patterns.
2106    20. Extended pcre_study() to be more clever in cases where a branch of a
2107        subpattern has no definite first character. For example, (a*|b*)[cd] would
2108        previously give no result from pcre_study(). Now it recognizes that the
2109        first character must be a, b, c, or d.
2111    21. There was an incorrect error "recursive call could loop indefinitely" if
2112        a subpattern (or the entire pattern) that was being tested for matching an
2113        empty string contained only one non-empty item after a nested subpattern.
2114        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
2115        incorrectly, because the \d was being skipped in the check.
2117    22. The pcretest program now has a new pattern option /B and a command line
2118        option -b, which is equivalent to adding /B to every pattern. This causes
2119        it to show the compiled bytecode, without the additional information that
2120        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
2121        is the same as /B/I).
2123    23. A new optimization is now able automatically to treat some sequences such
2124        as a*b as a*+b. More specifically, if something simple (such as a character
2125        or a simple class like \d) has an unlimited quantifier, and is followed by
2126        something that cannot possibly match the quantified thing, the quantifier
2127        is automatically "possessified".
2129    24. A recursive reference to a subpattern whose number was greater than 39
2130        went wrong under certain circumstances in UTF-8 mode. This bug could also
2131        have affected the operation of pcre_study().
2133    25. Realized that a little bit of performance could be had by replacing
2134        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
2136    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
2138    27. Possessive quantifiers such as a++ were previously implemented by turning
2139        them into atomic groups such as ($>a+). Now they have their own opcodes,
2140        which improves performance. This includes the automatically created ones
2141        from 23 above.
2143    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
2144        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
2145        the first matched character to be a colon. This applied both to named and
2146        numbered groups.
2148    29. The ucpinternal.h header file was missing its idempotency #ifdef.
2150    30. I was sent a "project" file called libpcre.a.dev which I understand makes
2151        building PCRE on Windows easier, so I have included it in the distribution.
2153    31. There is now a check in pcretest against a ridiculously large number being
2154        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
2155        loop, the loop is abandoned.
2157    32. Forward references to subpatterns in conditions such as (?(2)...) where
2158        subpattern 2 is defined later cause pcre_compile() to search forwards in
2159        the pattern for the relevant set of parentheses. This search went wrong
2160        when there were unescaped parentheses in a character class, parentheses
2161        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
2163    33. "Subroutine" calls and backreferences were previously restricted to
2164        referencing subpatterns earlier in the regex. This restriction has now
2165        been removed.
2167    34. Added a number of extra features that are going to be in Perl 5.10. On the
2168        whole, these are just syntactic alternatives for features that PCRE had
2169        previously implemented using the Python syntax or my own invention. The
2170        other formats are all retained for compatibility.
2172        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
2173            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
2174            also .NET compatible.
2176        (b) A recursion or subroutine call to a named group can now be defined as
2177            (?&name) as well as (?P>name).
2179        (c) A backreference to a named group can now be defined as \k<name> or
2180            \k'name' as well as (?P=name). The new forms, as well as being in Perl
2181            5.10, are also .NET compatible.
2183        (d) A conditional reference to a named group can now use the syntax
2184            (?(<name>) or (?('name') as well as (?(name).
2186        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
2187            groups (named and numbered) that are never evaluated inline, but can be
2188            called as "subroutines" from elsewhere. In effect, the DEFINE condition
2189            is always false. There may be only one alternative in such a group.
2191        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
2192            as the simple (?(R). The condition is true only if the most recent
2193            recursion is that of the given number or name. It does not search out
2194            through the entire recursion stack.
2196        (g) The escape \gN or \g{N} has been added, where N is a positive or
2197            negative number, specifying an absolute or relative reference.
2199    35. Tidied to get rid of some further signed/unsigned compiler warnings and
2200        some "unreachable code" warnings.
2202    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
2203        things, this adds five new scripts.
2205    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
2206        There were also incompatibilities regarding the handling of \Q..\E inside
2207        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
2208        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
2210    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
2211        matches an empty string, and forcibly breaks the loop. There were bugs in
2212        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
2213        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
2214        separate and independent bugs (that affected different cases) have been
2215        fixed.
2217    39. Refactored the code to abolish the use of different opcodes for small
2218        capturing bracket numbers. This is a tidy that I avoided doing when I
2219        removed the limit on the number of capturing brackets for 3.5 back in 2001.
2220        The new approach is not only tidier, it makes it possible to reduce the
2221        memory needed to fix the previous bug (38).
2223    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
2224        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
2225        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
2226        mode.
2228    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
2229        report.
2231    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
2232        copy construction and assignment in the C++ wrapper.
2234    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
2235        couple of bugs that could have given wrong results in the "--newline=crlf"
2236        case.
2238    44. Added a number of casts and did some reorganization of signed/unsigned int
2239        variables following suggestions from Dair Grant. Also renamed the variable
2240        "this" as "item" because it is a C++ keyword.
2242    45. Arranged for dftables to add
2244          #include "pcre_internal.h"
2246        to pcre_chartables.c because without it, gcc 4.x may remove the array
2247        definition from the final binary if PCRE is built into a static library and
2248        dead code stripping is activated.
2250    46. For an unanchored pattern, if a match attempt fails at the start of a
2251        newline sequence, and the newline setting is CRLF or ANY, and the next two
2252        characters are CRLF, advance by two characters instead of one.
2255    Version 6.7 04-Jul-06
2256    ---------------------
2258     1. In order to handle tests when input lines are enormously long, pcretest has
2259        been re-factored so that it automatically extends its buffers when
2260        necessary. The code is crude, but this _is_ just a test program. The
2261        default size has been increased from 32K to 50K.
2263     2. The code in pcre_study() was using the value of the re argument before
2264        testing it for NULL. (Of course, in any sensible call of the function, it
2265        won't be NULL.)
2267     3. The memmove() emulation function in pcre_internal.h, which is used on
2268        systems that lack both memmove() and bcopy() - that is, hardly ever -
2269        was missing a "static" storage class specifier.
2271     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
2272        containing an extended class (one that cannot be represented by a bitmap
2273        because it contains high-valued characters or Unicode property items, e.g.
2274        [\pZ]). Almost always one would set UTF-8 mode when processing such a
2275        pattern, but PCRE should not loop if you do not (it no longer does).
2276        [Detail: two cases were found: (a) a repeated subpattern containing an
2277        extended class; (b) a recursive reference to a subpattern that followed a
2278        previous extended class. It wasn't skipping over the extended class
2279        correctly when UTF-8 mode was not set.]
2281     5. A negated single-character class was not being recognized as fixed-length
2282        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
2283        compile error "lookbehind assertion is not fixed length".
2285     6. The RunPerlTest auxiliary script was showing an unexpected difference
2286        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
2287        write a Perl script that can interpret lines of an input file either as
2288        byte characters or as UTF-8, which is what "perltest" was being required to
2289        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
2290        can't do is switch easily at run time between having the "use utf8;" pragma
2291        or not. In the end, I fudged it by using the RunPerlTest script to insert
2292        "use utf8;" explicitly for the UTF-8 tests.
2294     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
2295        the end of the subject string, contrary to the documentation and to what
2296        Perl does. This was true of both matching functions. Now it matches only at
2297        the start of the subject and immediately after *internal* newlines.
2299     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
2300        a pointer to an int instead of a pointer to an unsigned long int. This
2301        caused problems on 64-bit systems.
2303     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
2304        instance of the 'standard' template library not being so standard".
2306    10. There was no check on the number of named subpatterns nor the maximum
2307        length of a subpattern name. The product of these values is used to compute
2308        the size of the memory block for a compiled pattern. By supplying a very
2309        long subpattern name and a large number of named subpatterns, the size
2310        computation could be caused to overflow. This is now prevented by limiting
2311        the length of names to 32 characters, and the number of named subpatterns
2312        to 10,000.
2314    11. Subpatterns that are repeated with specific counts have to be replicated in
2315        the compiled pattern. The size of memory for this was computed from the
2316        length of the subpattern and the repeat count. The latter is limited to
2317        65535, but there was no limit on the former, meaning that integer overflow
2318        could in principle occur. The compiled length of a repeated subpattern is
2319        now limited to 30,000 bytes in order to prevent this.
2321    12. Added the optional facility to have named substrings with the same name.
2323    13. Added the ability to use a named substring as a condition, using the
2324        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
2325        are numbers (not recommended). Forward references are permitted.
2327    14. Added forward references in named backreferences (if you see what I mean).
2329    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
2330        pattern could run off the end of the subject. For example, the pattern
2331        "(?s)(.{1,5})"8 did this with the subject "ab".
2333    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
2334        PCRE_CASELESS was set when matching characters that were quantified with ?
2335        or *.
2337    17. A character class other than a single negated character that had a minimum
2338        but no maximum quantifier - for example [ab]{6,} - was not handled
2339        correctly by pce_dfa_exec(). It would match only one character.
2341    18. A valid (though odd) pattern that looked like a POSIX character
2342        class but used an invalid character after [ (for example [[,abc,]]) caused
2343        pcre_compile() to give the error "Failed: internal error: code overflow" or
2344        in some cases to crash with a glibc free() error. This could even happen if
2345        the pattern terminated after [[ but there just happened to be a sequence of
2346        letters, a binary zero, and a closing ] in the memory that followed.
2348    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
2349        over the years. Originally (before any Unicode support), just the bottom 8
2350        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
2351        output from "man perlunicode" includes this:
2353          The regular expression compiler produces polymorphic opcodes.  That
2354          is, the pattern adapts to the data and automatically switches to
2355          the Unicode character scheme when presented with Unicode data--or
2356          instead uses a traditional byte scheme when presented with byte
2357          data.
2359        Sadly, a wide octal escape does not cause a switch, and in a string with
2360        no other multibyte characters, these octal escapes are treated as before.
2361        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
2362        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
2363        Unicode string.
2365        I have not perpetrated such confusion in PCRE. Up till now, it took just
2366        the bottom 8 bits, as in old Perl. I have now made octal escapes with
2367        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
2368        translate to the appropriate multibyte character.
2370    29. Applied some refactoring to reduce the number of warnings from Microsoft
2371        and Borland compilers. This has included removing the fudge introduced
2372        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
2373        a warning about an unused variable.
2375    21. PCRE has not included VT (character 0x0b) in the set of whitespace
2376        characters since release 4.0, because Perl (from release 5.004) does not.
2377        [Or at least, is documented not to: some releases seem to be in conflict
2378        with the documentation.] However, when a pattern was studied with
2379        pcre_study() and all its branches started with \s, PCRE still included VT
2380        as a possible starting character. Of course, this did no harm; it just
2381        caused an unnecessary match attempt.
2383    22. Removed a now-redundant internal flag bit that recorded the fact that case
2384        dependency changed within the pattern. This was once needed for "required
2385        byte" processing, but is no longer used. This recovers a now-scarce options
2386        bit. Also moved the least significant internal flag bit to the most-
2387        significant bit of the word, which was not previously used (hangover from
2388        the days when it was an int rather than a uint) to free up another bit for
2389        the future.
2391    23. Added support for CRLF line endings as well as CR and LF. As well as the
2392        default being selectable at build time, it can now be changed at runtime
2393        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
2394        specify that it is scanning data with non-default line endings.
2396    24. Changed the definition of CXXLINK to make it agree with the definition of
2397        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
2399    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
2400        recursions. This makes a big different to stack usage for some patterns.
2402    26. If a subpattern containing a named recursion or subroutine reference such
2403        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
2404        the space required for the compiled pattern went wrong and gave too small a
2405        value. Depending on the environment, this could lead to "Failed: internal
2406        error: code overflow at offset 49" or "glibc detected double free or
2407        corruption" errors.
2409    27. Applied patches from Google (a) to support the new newline modes and (b) to
2410        advance over multibyte UTF-8 characters in GlobalReplace.
2412    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
2413        difference for some implementation of PCRE in some Windows version.
2415    29. Added some extra testing facilities to pcretest:
2417        \q<number>   in a data line sets the "match limit" value
2418        \Q<number>   in a data line sets the "match recursion limt" value
2419        -S <number>  sets the stack size, where <number> is in megabytes
2421        The -S option isn't available for Windows.
2424    Version 6.6 06-Feb-06
2425    ---------------------
2427     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
2428        in pcreposix.h. I have copied the definition from pcre.h.
2430     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
2431        because pcre.h is no longer a built file.
2433     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
2434        not normally included in the compiled code.
2437    Version 6.5 01-Feb-06
2438    ---------------------
2440     1. When using the partial match feature with pcre_dfa_exec(), it was not
2441        anchoring the second and subsequent partial matches at the new starting
2442        point. This could lead to incorrect results. For example, with the pattern
2443        /1234/, partially matching against "123" and then "a4" gave a match.
2445     2. Changes to pcregrep:
2447        (a) All non-match returns from pcre_exec() were being treated as failures
2448            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
2449            error message is output. Some extra information is given for the
2450            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
2451            probably the only errors that are likely to be caused by users (by
2452            specifying a regex that has nested indefinite repeats, for instance).
2453            If there are more than 20 of these errors, pcregrep is abandoned.
2455        (b) A binary zero was treated as data while matching, but terminated the
2456            output line if it was written out. This has been fixed: binary zeroes
2457            are now no different to any other data bytes.
2459        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
2460            used to set a locale for matching. The --locale=xxxx long option has
2461            been added (no short equivalent) to specify a locale explicitly on the
2462            pcregrep command, overriding the environment variables.
2464        (d) When -B was used with -n, some line numbers in the output were one less
2465            than they should have been.
2467        (e) Added the -o (--only-matching) option.
2469        (f) If -A or -C was used with -c (count only), some lines of context were
2470            accidentally printed for the final match.
2472        (g) Added the -H (--with-filename) option.
2474        (h) The combination of options -rh failed to suppress file names for files
2475            that were found from directory arguments.
2477        (i) Added the -D (--devices) and -d (--directories) options.
2479        (j) Added the -F (--fixed-strings) option.
2481        (k) Allow "-" to be used as a file name for -f as well as for a data file.
2483        (l) Added the --colo(u)r option.
2485        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
2486            is not present by default.
2488     3. A nasty bug was discovered in the handling of recursive patterns, that is,
2489        items such as (?R) or (?1), when the recursion could match a number of
2490        alternatives. If it matched one of the alternatives, but subsequently,
2491        outside the recursion, there was a failure, the code tried to back up into
2492        the recursion. However, because of the way PCRE is implemented, this is not
2493        possible, and the result was an incorrect result from the match.
2495        In order to prevent this happening, the specification of recursion has
2496        been changed so that all such subpatterns are automatically treated as
2497        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
2499     4. I had overlooked the fact that, in some locales, there are characters for
2500        which isalpha() is true but neither isupper() nor islower() are true. In
2501        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
2502        and ordfeminine) are like this. This affected the treatment of \w and \W
2503        when they appeared in character classes, but not when they appeared outside
2504        a character class. The bit map for "word" characters is now created
2505        separately from the results of isalnum() instead of just taking it from the
2506        upper, lower, and digit maps. (Plus the underscore character, of course.)
2508     5. The above bug also affected the handling of POSIX character classes such as
2509        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
2510        permanent tables. Instead, the bit maps for such a class were previously
2511        created as the appropriate unions of the upper, lower, and digit bitmaps.
2512        Now they are created by subtraction from the [[:word:]] class, which has
2513        its own bitmap.
2515     6. The [[:blank:]] character class matches horizontal, but not vertical space.
2516        It is created by subtracting the vertical space characters (\x09, \x0a,
2517        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
2518        subtraction was done in the overall bitmap for a character class, meaning
2519        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
2520        be recognized. This bug has been fixed.
2522     7. Patches from the folks at Google:
2524          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
2525          real life, but is still worth protecting against".
2527          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
2528          regular expressions".
2530          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
2531          have it.
2533          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
2534          "configure" and the latter not, in order to fix a problem somebody had
2535          with compiling the Arg class on HP-UX.
2537          (e) Improve the error-handling of the C++ wrapper a little bit.
2539          (f) New tests for checking recursion limiting.
2541     8. The pcre_memmove() function, which is used only if the environment does not
2542        have a standard memmove() function (and is therefore rarely compiled),
2543        contained two bugs: (a) use of int instead of size_t, and (b) it was not
2544        returning a result (though PCRE never actually uses the result).
2546     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
2547        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
2548        returned instead of calling malloc() with an overflowing number that would
2549        most likely cause subsequent chaos.
2551    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
2553    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
2554        with this option is matched, the nmatch and pmatch options of regexec() are
2555        ignored.
2557    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
2558        provided in case anyone wants to the the POSIX interface with UTF-8
2559        strings.
2561    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
2562        C++ linking (needed for some HP-UX environments).
2564    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
2565        (unused parameter) and in the pcre_printint() function (omitted "default"
2566        switch label when the default is to do nothing).
2568    15. Added some code to make it possible, when PCRE is compiled as a C++
2569        library, to replace subject pointers for pcre_exec() with a smart pointer
2570        class, thus making it possible to process discontinuous strings.
2572    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
2573        much the same function. They were added by different people who were trying
2574        to make PCRE easy to compile on non-Unix systems. It has been suggested
2575        that PCRE_EXPORT be abolished now that there is more automatic apparatus
2576        for compiling on Windows systems. I have therefore replaced it with
2577        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
2578        defaults to "extern" for C or "extern C" for C++, which works fine on
2579        Unix-like systems. It is now possible to override the value of PCRE_DATA_
2580        SCOPE with something explicit in config.h. In addition:
2582        (a) pcreposix.h still had just "extern" instead of either of these macros;
2583            I have replaced it with PCRE_DATA_SCOPE.
2585        (b) Functions such as _pcre_xclass(), which are internal to the library,
2586            but external in the C sense, all had PCRE_EXPORT in their definitions.
2587            This is apparently wrong for the Windows case, so I have removed it.
2588            (It makes no difference on Unix-like systems.)
2590    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
2591        of recursive calls to match(). This is different to MATCH_LIMIT because
2592        that limits the total number of calls to match(), not all of which increase
2593        the depth of recursion. Limiting the recursion depth limits the amount of
2594        stack (or heap if NO_RECURSE is set) that is used. The default can be set
2595        when PCRE is compiled, and changed at run time. A patch from Google adds
2596        this functionality to the C++ interface.
2598    18. Changes to the handling of Unicode character properties:
2600        (a) Updated the table to Unicode 4.1.0.
2602        (b) Recognize characters that are not in the table as "Cn" (undefined).
2604        (c) I revised the way the table is implemented to a much improved format
2605            which includes recognition of ranges. It now supports the ranges that
2606            are defined in UnicodeData.txt, and it also amalgamates other
2607            characters into ranges. This has reduced the number of entries in the
2608            table from around 16,000 to around 3,000, thus reducing its size
2609            considerably. I realized I did not need to use a tree structure after
2610            all - a binary chop search is just as efficient. Having reduced the
2611            number of entries, I extended their size from 6 bytes to 8 bytes to
2612            allow for more data.
2614        (d) Added support for Unicode script names via properties such as \p{Han}.
2616    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
2617        matching that character.
2619    20. When matching a repeated Unicode property with a minimum greater than zero,
2620        (for example \pL{2,}), PCRE could look past the end of the subject if it
2621        reached it while seeking the minimum number of characters. This could
2622        happen only if some of the characters were more than one byte long, because
2623        there is a check for at least the minimum number of bytes.
2625    21. Refactored the implementation of \p and \P so as to be more general, to
2626        allow for more different types of property in future. This has changed the
2627        compiled form incompatibly. Anybody with saved compiled patterns that use
2628        \p or \P will have to recompile them.
2630    22. Added "Any" and "L&" to the supported property types.
2632    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
2633        but give a compile time error if the value is greater than 0xff.
2635    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
2636        accidentally not being installed or uninstalled.
2638    25. The pcre.h file was built from pcre.h.in, but the only changes that were
2639        made were to insert the current release number. This seemed silly, because
2640        it made things harder for people building PCRE on systems that don't run
2641        "configure". I have turned pcre.h into a distributed file, no longer built
2642        by "configure", with the version identification directly included. There is
2643        no longer a pcre.h.in file.
2645        However, this change necessitated a change to the pcre-config script as
2646        well. It is built from pcre-config.in, and one of the substitutions was the
2647        release number. I have updated configure.ac so that ./configure now finds
2648        the release number by grepping pcre.h.
2650    26. Added the ability to run the tests under valgrind.
2653    Version 6.4 05-Sep-05
2654    ---------------------
2656     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
2657        "--" to be printed when multiple files were scanned, even when none of the
2658        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
2659        consider it to be a bug, and have restored the previous behaviour.
2661     2. A couple of code tidies to get rid of compiler warnings.
2663     3. The pcretest program used to cheat by referring to symbols in the library
2664        whose names begin with _pcre_. These are internal symbols that are not
2665        really supposed to be visible externally, and in some environments it is
2666        possible to suppress them. The cheating is now confined to including
2667        certain files from the library's source, which is a bit cleaner.
2669     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
2670        file's purpose clearer.
2672     5. Reorganized pcre_ucp_findchar().
2675    Version 6.3 15-Aug-05
2676    ---------------------
2678     1. The file libpcre.pc.in did not have general read permission in the tarball.
2680     2. There were some problems when building without C++ support:
2682        (a) If C++ support was not built, "make install" and "make test" still
2683            tried to test it.
2685        (b) There were problems when the value of CXX was explicitly set. Some
2686            changes have been made to try to fix these, and ...
2688        (c) --disable-cpp can now be used to explicitly disable C++ support.
2690        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
2691            backslash in a target when C++ was disabled. This confuses some
2692            versions of "make", apparently. Using an intermediate variable solves
2693            this. (Same for CPP_LOBJ.)
2695     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
2696        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
2697        necessary on certain architectures.
2699     4. Added a setting of -export-symbols-regex to the link command to remove
2700        those symbols that are exported in the C sense, but actually are local
2701        within the library, and not documented. Their names all begin with
2702        "_pcre_". This is not a perfect job, because (a) we have to except some
2703        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
2704        available (and never for static libraries). I have made a note to try to
2705        find a way round (a) in the future.
2708    Version 6.2 01-Aug-05
2709    ---------------------
2711     1. There was no test for integer overflow of quantifier values. A construction
2712        such as {1111111111111111} would give undefined results. What is worse, if
2713        a minimum quantifier for a parenthesized subpattern overflowed and became
2714        negative, the calculation of the memory size went wrong. This could have
2715        led to memory overwriting.
2717     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
2719     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
2720        operating environments where this matters.
2722     4. Applied Giuseppe Maxia's patch to add additional features for controlling
2723        PCRE options from within the C++ wrapper.
2725     5. Named capturing subpatterns were not being correctly counted when a pattern
2726        was compiled. This caused two problems: (a) If there were more than 100
2727        such subpatterns, the calculation of the memory needed for the whole
2728        compiled pattern went wrong, leading to an overflow error. (b) Numerical
2729        back references of the form \12, where the number was greater than 9, were
2730        not recognized as back references, even though there were sufficient
2731        previous subpatterns.
2733     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
2734        versions of gcc, e.g. 2.95.4.
2737    Version 6.1 21-Jun-05
2738    ---------------------
2740     1. There was one reference to the variable "posix" in pcretest.c that was not
2741        surrounded by "#if !defined NOPOSIX".
2743     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
2744        the cross-check on the old pcre_info() function, for the benefit of the
2745        cut-down version of PCRE that is currently imported into Exim.
2747     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
2748        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
2749        patterns that start (?i) but I don't think that matters. The use of (?i) is
2750        just an example; this all applies to the other options as well.
2752     4. Since libtool seems to echo the compile commands it is issuing, the output
2753        from "make" can be reduced a bit by putting "@" in front of each libtool
2754        compile command.
2756     5. Patch from the folks at Google for configure.in to be a bit more thorough
2757        in checking for a suitable C++ installation before trying to compile the
2758        C++ stuff. This should fix a reported problem when a compiler was present,
2759        but no suitable headers.
2761     6. The man pages all had just "PCRE" as their title. I have changed them to
2762        be the relevant file name. I have also arranged that these names are
2763        retained in the file doc/pcre.txt, which is a concatenation in text format
2764        of all the man pages except the little individual ones for each function.
2766     7. The NON-UNIX-USE file had not been updated for the different set of source
2767        files that come with release 6. I also added a few comments about the C++
2768        wrapper.
2771    Version 6.0 07-Jun-05
2772    ---------------------
2774     1. Some minor internal re-organization to help with my DFA experiments.
2776     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
2777        didn't matter for the library itself when fully configured, but did matter
2778        when compiling without UCP support, or within Exim, where the ucp files are
2779        not imported.
2781     3. Refactoring of the library code to split up the various functions into
2782        different source modules. The addition of the new DFA matching code (see
2783        below) to a single monolithic source would have made it really too
2784        unwieldy, quite apart from causing all the code to be include in a
2785        statically linked application, when only some functions are used. This is
2786        relevant even without the DFA addition now that patterns can be compiled in
2787        one application and matched in another.
2789        The downside of splitting up is that there have to be some external
2790        functions and data tables that are used internally in different modules of
2791        the library but which are not part of the API. These have all had their
2792        names changed to start with "_pcre_" so that they are unlikely to clash
2793        with other external names.
2795     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
2796        a different (DFA) algorithm. Although it is slower than the original
2797        function, it does have some advantages for certain types of matching
2798        problem.
2800     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
2801        including restarting after a partial match.
2803     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
2804        defined when compiling for Windows was sent to me. I have put it into the
2805        code, though I have no means of testing or verifying it.
2807     7. Added the pcre_refcount() auxiliary function.
2809     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
2810        match before or at the first newline in the subject string. In pcretest,
2811        the /f option on a pattern can be used to set this.
2813     9. A repeated \w when used in UTF-8 mode with characters greater than 256
2814        would behave wrongly. This has been present in PCRE since release 4.0.
2816    10. A number of changes to the pcregrep command:
2818        (a) Refactored how -x works; insert ^(...)$ instead of setting
2819            PCRE_ANCHORED and checking the length, in preparation for adding
2820            something similar for -w.
2822        (b) Added the -w (match as a word) option.
2824        (c) Refactored the way lines are read and buffered so as to have more
2825            than one at a time available.
2827        (d) Implemented a pcregrep test script.
2829        (e) Added the -M (multiline match) option. This allows patterns to match
2830            over several lines of the subject. The buffering ensures that at least
2831            8K, or the rest of the document (whichever is the shorter) is available
2832            for matching (and similarly the previous 8K for lookbehind assertions).
2834        (f) Changed the --help output so that it now says
2836              -w, --word-regex(p)
2838            instead of two lines, one with "regex" and the other with "regexp"
2839            because that confused at least one person since the short forms are the
2840            same. (This required a bit of code, as the output is generated
2841            automatically from a table. It wasn't just a text change.)
2843        (g) -- can be used to terminate pcregrep options if the next thing isn't an
2844            option but starts with a hyphen. Could be a pattern or a path name
2845            starting with a hyphen, for instance.
2847        (h) "-" can be given as a file name to represent stdin.
2849        (i) When file names are being printed, "(standard input)" is used for
2850            the standard input, for compatibility with GNU grep. Previously
2851            "<stdin>" was used.
2853        (j) The option --label=xxx can be used to supply a name to be used for
2854            stdin when file names are being printed. There is no short form.
2856        (k) Re-factored the options decoding logic because we are going to add
2857            two more options that take data. Such options can now be given in four
2858            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
2860        (l) Added the -A, -B, and -C options for requesting that lines of context
2861            around matches be printed.
2863        (m) Added the -L option to print the names of files that do not contain
2864            any matching lines, that is, the complement of -l.
2866        (n) The return code is 2 if any file cannot be opened, but pcregrep does
2867            continue to scan other files.
2869        (o) The -s option was incorrectly implemented. For compatibility with other
2870            greps, it now suppresses the error message for a non-existent or non-
2871            accessible file (but not the return code). There is a new option called
2872            -q that suppresses the output of matching lines, which was what -s was
2873            previously doing.
2875        (p) Added --include and --exclude options to specify files for inclusion
2876            and exclusion when recursing.
2878    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
2879        Hopefully, it now does.
2881    12. Missing cast in pcre_study().
2883    13. Added an "uninstall" target to the makefile.
2885    14. Replaced "extern" in the function prototypes in Makefile.in with
2886        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
2887        world, but is set differently for Windows.
2889    15. Added a second compiling function called pcre_compile2(). The only
2890        difference is that it has an extra argument, which is a pointer to an
2891        integer error code. When there is a compile-time failure, this is set
2892        non-zero, in addition to the error test pointer being set to point to an
2893        error message. The new argument may be NULL if no error number is required
2894        (but then you may as well call pcre_compile(), which is now just a
2895        wrapper). This facility is provided because some applications need a
2896        numeric error indication, but it has also enabled me to tidy up the way
2897        compile-time errors are handled in the POSIX wrapper.
2899    16. Added VPATH=.libs to the makefile; this should help when building with one
2900        prefix path and installing with another. (Or so I'm told by someone who
2901        knows more about this stuff than I do.)
2903    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
2904        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
2905        match everything, including newlines. This is not POSIX-compatible, but
2906        somebody wanted the feature. From pcretest it can be activated by using
2907        both the P and the s flags.
2909    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
2911    19. libpcre.pc was being incorrectly installed as executable.
2913    20. A couple of places in pcretest check for end-of-line by looking for '\n';
2914        it now also looks for '\r' so that it will work unmodified on Windows.
2916    21. Added Google's contributed C++ wrapper to the distribution.
2918    22. Added some untidy missing memory free() calls in pcretest, to keep
2919        Electric Fence happy when testing.
2923    Version 5.0 13-Sep-04
2924    ---------------------
2926     1. Internal change: literal characters are no longer packed up into items
2927        containing multiple characters in a single byte-string. Each character
2928        is now matched using a separate opcode. However, there may be more than one
2929        byte in the character in UTF-8 mode.
2931     2. The pcre_callout_block structure has two new fields: pattern_position and
2932        next_item_length. These contain the offset in the pattern to the next match
2933        item, and its length, respectively.
2935     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
2936        insertion of callouts before each pattern item. Added the /C option to
2937        pcretest to make use of this.
2939     4. On the advice of a Windows user, the lines
2941          #if defined(_WIN32) || defined(WIN32)
2942          _setmode( _fileno( stdout ), 0x8000 );
2943          #endif  /* defined(_WIN32) || defined(WIN32) */
2945        have been added to the source of pcretest. This apparently does useful
2946        magic in relation to line terminators.
2948     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
2949        for the benefit of those environments where the "b" makes a difference.
2951     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
2952        to know about it. I have put a hack into configure.in that adds in code
2953        to set GCC=yes if CC=icc. This seems to end up at a point in the
2954        generated configure script that is early enough to affect the setting of
2955        compiler options, which is what is needed, but I have no means of testing
2956        whether it really works. (The user who reported this had patched the
2957        generated configure script, which of course I cannot do.)
2959        LATER: After change 22 below (new libtool files), the configure script
2960        seems to know about icc (and also ecc). Therefore, I have commented out
2961        this hack in configure.in.
2963     7. Added support for pkg-config (2 patches were sent in).
2965     8. Negated POSIX character classes that used a combination of internal tables
2966        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2967        [[:^ascii]]. Typically, they would match almost any characters. The other
2968        POSIX classes were not broken in this way.
2970     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2971        to find the match, as PCRE was deluded into thinking that the match had to
2972        start at the start point or following a newline. The same bug applied to
2973        patterns with negative forward assertions or any backward assertions
2974        preceding ".*" at the start, unless the pattern required a fixed first
2975        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2977    10. In UTF-8 mode, when moving forwards in the subject after a failed match
2978        starting at the last subject character, bytes beyond the end of the subject
2979        string were read.
2981    11. Renamed the variable "class" as "classbits" to make life easier for C++
2982        users. (Previously there was a macro definition, but it apparently wasn't
2983        enough.)
2985    12. Added the new field "tables" to the extra data so that tables can be passed
2986        in at exec time, or the internal tables can be re-selected. This allows
2987        a compiled regex to be saved and re-used at a later time by a different
2988        program that might have everything at different addresses.
2990    13. Modified the pcre-config script so that, when run on Solaris, it shows a
2991        -R library as well as a -L library.
2993    14. The debugging options of pcretest (-d on the command line or D on a
2994        pattern) showed incorrect output for anything following an extended class
2995        that contained multibyte characters and which was followed by a quantifier.
2997    15. Added optional support for general category Unicode character properties
2998        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2999        support. It adds about 90K to the size of the library. The meanings of the
3000        inbuilt class escapes such as \d and \s have NOT been changed.
3002    16. Updated pcredemo.c to include calls to free() to release the memory for the
3003        compiled pattern.
3005    17. The generated file chartables.c was being created in the source directory
3006        instead of in the building directory. This caused the build to fail if the
3007        source directory was different from the building directory, and was
3008        read-only.
3010    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
3011        file. No doubt somebody will tell me if they don't make sense... Also added
3012        Dan Mooney's comments about building on OpenVMS.
3014    19. Added support for partial matching via the PCRE_PARTIAL option for
3015        pcre_exec() and the \P data escape in pcretest.
3017    20. Extended pcretest with 3 new pattern features:
3019        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
3020              write the compiled pattern to the file whose name is "rest-of-line".
3021              This is a straight binary dump of the data, with the saved pointer to
3022              the character tables forced to be NULL. The study data, if any, is
3023              written too. After writing, pcretest reads a new pattern.
3025        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
3026              compiled pattern from the given file. There must not be any
3027              occurrences of "<" in the file name (pretty unlikely); if there are,
3028              pcretest will instead treat the initial "<" as a pattern delimiter.
3029              After reading in the pattern, pcretest goes on to read data lines as
3030              usual.
3032        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
3033              and 16-bit fields in a compiled pattern, to simulate a pattern that
3034              was compiled on a host of opposite endianness.
3036    21. The pcre-exec() function can now cope with patterns that were compiled on
3037        hosts of opposite endianness, with this restriction:
3039          As for any compiled expression that is saved and used later, the tables
3040          pointer field cannot be preserved; the extra_data field in the arguments
3041          to pcre_exec() should be used to pass in a tables address if a value
3042          other than the default internal tables were used at compile time.
3044    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
3045        now diagnosed as an error. Previously, most of the time, a negative number
3046        would have been treated as zero, but if in addition "ovector" was passed as
3047        NULL, a crash could occur.
3049    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
3050        new versions from the libtool 1.5 distribution (the last one is a copy of
3051        a file called libtool.m4). This seems to have fixed the need to patch
3052        "configure" to support Darwin 1.3 (which I used to do). However, I still
3053        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
3054        workstation).
3056    24. Changed the PCRE licence to be the more standard "BSD" licence.
3059    Version 4.5 01-Dec-03
3060    ---------------------
3062     1. There has been some re-arrangement of the code for the match() function so
3063        that it can be compiled in a version that does not call itself recursively.
3064        Instead, it keeps those local variables that need separate instances for
3065        each "recursion" in a frame on the heap, and gets/frees frames whenever it
3066        needs to "recurse". Keeping track of where control must go is done by means
3067        of setjmp/longjmp. The whole thing is implemented by a set of macros that
3068        hide most of the details from the main code, and operates only if
3069        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
3070        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
3071        operating.
3073        To make it easier for callers to provide specially tailored get/free
3074        functions for this usage, two new functions, pcre_stack_malloc, and
3075        pcre_stack_free, are used. They are always called in strict stacking order,
3076        and the size of block requested is always the same.
3078        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
3079        PCRE has been compiled to use the stack or the heap for recursion. The
3080        -C option of pcretest uses this to show which version is compiled.
3082        A new data escape \S, is added to pcretest; it causes the amounts of store
3083        obtained and freed by both kinds of malloc/free at match time to be added
3084        to the output.
3086     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
3087        what's available on my current Linux desktop machine.
3089     3. When matching a UTF-8 string, the test for a valid string at the start has
3090        been extended. If start_offset is not zero, PCRE now checks that it points
3091        to a byte that is the start of a UTF-8 character. If not, it returns
3092        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
3093        this is necessary because there may be backward assertions in the pattern.
3094        When matching the same subject several times, it may save resources to use
3095        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
3097     4. The code for checking the validity of UTF-8 strings has been tightened so
3098        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
3099        containing "overlong sequences".
3101     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
3102        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
3103        should have been "&", but it just so happened that all the cases this let
3104        through by mistake were picked up later in the function.
3106     6. I had used a variable called "isblank" - this is a C99 function, causing
3107        some compilers to warn. To avoid this, I renamed it (as "blankclass").
3109     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
3110        prompting; (b) run "./pcretest /dev/null" at the start of the test script
3111        so the version is shown; (c) stop "make test" echoing "./RunTest".
3113     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
3115     9. The prototype for memmove() for systems that don't have it was using
3116        size_t, but the inclusion of the header that defines size_t was later. I've
3117        moved the #includes for the C headers earlier to avoid this.
3119    10. Added some adjustments to the code to make it easier to compiler on certain
3120        special systems:
3122          (a) Some "const" qualifiers were missing.
3123          (b) Added the macro EXPORT before all exported functions; by default this
3124              is defined to be empty.
3125          (c) Changed the dftables auxiliary program (that builds chartables.c) so
3126              that it reads its output file name as an argument instead of writing
3127              to the standard output and assuming this can be redirected.
3129    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
3130        class containing characters with values greater than 255, PCRE compilation
3131        went into a loop.
3133    12. A recursive reference to a subpattern that was within another subpattern
3134        that had a minimum quantifier of zero caused PCRE to crash. For example,
3135        (x(y(?2))z)? provoked this bug with a subject that got as far as the
3136        recursion. If the recursively-called subpattern itself had a zero repeat,
3137        that was OK.
3139    13. In pcretest, the buffer for reading a data line was set at 30K, but the
3140        buffer into which it was copied (for escape processing) was still set at
3141        1024, so long lines caused crashes.
3143    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
3144        "internal error: code overflow...". This applied to any character class
3145        that was followed by a possessive quantifier.
3147    15. Modified the Makefile to add libpcre.la as a prerequisite for
3148        libpcreposix.la because I was told this is needed for a parallel build to
3149        work.
3151    16. If a pattern that contained .* following optional items at the start was
3152        studied, the wrong optimizing data was generated, leading to matching
3153        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
3154        matching string must start with a or b or c. The correct conclusion for
3155        this pattern is that a match can start with any character.
3158    Version 4.4 13-Aug-03
3159    ---------------------
3161     1. In UTF-8 mode, a character class containing characters with values between
3162        127 and 255 was not handled correctly if the compiled pattern was studied.
3163        In fixing this, I have also improved the studying algorithm for such
3164        classes (slightly).
3166     2. Three internal functions had redundant arguments passed to them. Removal
3167        might give a very teeny performance improvement.
3169     3. Documentation bug: the value of the capture_top field in a callout is *one
3170        more than* the number of the hightest numbered captured substring.
3172     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
3173        in incorrectly linking with a previously installed version. They now link
3174        explicitly with libpcre.la.
3176     5. configure.in no longer needs to recognize Cygwin specially.
3178     6. A problem in pcre.in for Windows platforms is fixed.
3180     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
3181        pcretest, it used to include the size of the study block as part of its
3182        output. Unfortunately, the structure contains a field that has a different
3183        size on different hardware architectures. This meant that the tests that
3184        showed this size failed. As the block is currently always of a fixed size,
3185        this information isn't actually particularly useful in pcretest output, so
3186        I have just removed it.
3188     8. Three pre-processor statements accidentally did not start in column 1.
3189        Sadly, there are *still* compilers around that complain, even though
3190        standard C has not required this for well over a decade. Sigh.
3192     9. In pcretest, the code for checking callouts passed small integers in the
3193        callout_data field, which is a void * field. However, some picky compilers
3194        complained about the casts involved for this on 64-bit systems. Now
3195        pcretest passes the address of the small integer instead, which should get
3196        rid of the warnings.
3198    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
3199        both compile and run time, and gives an error if an invalid UTF-8 sequence
3200        is found. There is a option for disabling this check in cases where the
3201        string is known to be correct and/or the maximum performance is wanted.
3203    11. In response to a bug report, I changed one line in Makefile.in from
3205            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
3206        to
3207            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
3209        to look similar to other lines, but I have no way of telling whether this
3210        is the right thing to do, as I do not use Windows. No doubt I'll get told
3211        if it's wrong...
3214    Version 4.3 21-May-03
3215    ---------------------
3217    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
3218       Makefile.
3220    2. Some refactoring to improve the quality of the code:
3222       (i)   The utf8_table... variables are now declared "const".
3224       (ii)  The code for \cx, which used the "case flipping" table to upper case
3225             lower case letters, now just substracts 32. This is ASCII-specific,
3226             but the whole concept of \cx is ASCII-specific, so it seems
3227             reasonable.
3229       (iii) PCRE was using its character types table to recognize decimal and
3230             hexadecimal digits in the pattern. This is silly, because it handles
3231             only 0-9, a-f, and A-F, but the character types table is locale-
3232             specific, which means strange things might happen. A private
3233             table is now used for this - though it costs 256 bytes, a table is
3234             much faster than multiple explicit tests. Of course, the standard
3235             character types table is still used for matching digits in subject
3236             strings against \d.
3238       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
3239             ending in _t are). So I've renamed it as ESC_tee.
3241    3. The first argument for regexec() in the POSIX wrapper should have been
3242       defined as "const".
3244    4. Changed pcretest to use malloc() for its buffers so that they can be
3245       Electric Fenced for debugging.
3247    5. There were several places in the code where, in UTF-8 mode, PCRE would try
3248       to read one or more bytes before the start of the subject string. Often this
3249       had no effect on PCRE's behaviour, but in some circumstances it could
3250       provoke a segmentation fault.
3252    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
3253       to try to read one or more bytes before the start of the subject string.
3255    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
3256       UTF-8 support could misbehave in various ways if the subject string
3257       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
3258       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
3259       back over UTF-8 characters.)
3262    Version 4.2 14-Apr-03
3263    ---------------------
3265    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
3267    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
3268         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
3269         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
3270         [WIN_PREFIX]: new variable, "cyg" for Cygwin
3271         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
3272           and BUILD_EXEEXT
3273         Note: automatic setting of the BUILD variables is not yet working
3274         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
3275           compile-time but not at link-time
3276         [LINK]: use for linking executables only
3277         make different versions for Windows and non-Windows
3278         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
3279           libraries
3280         [LINK_FOR_BUILD]: new variable
3281         [OBJEXT]: use throughout
3282         [EXEEXT]: use throughout
3283         <winshared>: new target
3284         <wininstall>: new target
3285         <dftables.o>: use native compiler
3286         <dftables>: use native linker
3287         <install>: handle Windows platform correctly
3288         <clean>: ditto
3289         <check>: ditto
3290         copy DLL to top builddir before testing
3292       As part of these changes, -no-undefined was removed again. This was reported
3293       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
3294       in any case.
3296    3. Some tidies to get rid of compiler warnings:
3298       . In the match_data structure, match_limit was an unsigned long int, whereas
3299         match_call_count was an int. I've made them both unsigned long ints.
3301       . In pcretest the fact that a const uschar * doesn't automatically cast to
3302         a void * provoked a warning.
3304       . Turning on some more compiler warnings threw up some "shadow" variables
3305         and a few more missing casts.
3307    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
3308       option, a class that contained a single character with a value between 128
3309       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
3311    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
3312       option, a class that contained several characters, but with at least one
3313       whose value was between 128 and 255 caused PCRE to crash.
3316    Version 4.1 12-Mar-03
3317    ---------------------
3319    1. Compiling with gcc -pedantic found a couple of places where casts were
3320    needed, and a string in dftables.c that was longer than standard compilers are
3321    required to support.
3323    2. Compiling with Sun's compiler found a few more places where the code could
3324    be tidied up in order to avoid warnings.
3326    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
3327    first of these names is deprecated in the latest Autoconf in favour of the name
3328    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
3329    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
3330    analogy I have changed it to CFLAGS_FOR_BUILD.
3332    4. Added -no-undefined to the linking command in the Makefile, because this is
3333    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
3334    linking step for the pcreposix library.
3336    5. PCRE was failing to diagnose the case of two named groups with the same
3337    name.
3339    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
3340    literal character that is needed in the subject for a match, and scans along to
3341    ensure that it is present before embarking on the full matching process. This
3342    saves time in cases of nested unlimited repeats that are never going to match.
3343    Problem: the scan can take a lot of time if the subject is very long (e.g.
3344    megabytes), thus penalizing straightforward matches. It is now done only if the
3345    amount of subject to be scanned is less than 1000 bytes.
3347    7. A lesser problem with the same optimization is that it was recording the
3348    first character of an anchored pattern as "needed", thus provoking a search
3349    right along the subject, even when the first match of the pattern was going to
3350    fail. The "needed" character is now not set for anchored patterns, unless it
3351    follows something in the pattern that is of non-fixed length. Thus, it still
3352    fulfils its original purpose of finding quick non-matches in cases of nested
3353    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
3356    Version 4.0 17-Feb-03
3357    ---------------------
3359    1. If a comment in an extended regex that started immediately after a meta-item
3360    extended to the end of string, PCRE compiled incorrect data. This could lead to
3361    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
3363    2. Moved to autoconf 2.53 and libtool 1.4.2.
3365    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
3366    the special perltest8 script is no longer needed - all the tests can be run
3367    from a single perltest script.
3369    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
3370    by \s. It has now been removed in PCRE. This means it isn't recognized as
3371    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
3372    class [:space:] *does* include VT, thereby creating a mess.
3374    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
3375    space and tab.
3377    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
3378    its new features into the main test script, reducing the number of scripts.
3380    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
3381    were backward compatible, and made the (?i) apply to the whole pattern, as if
3382    /i were given. Now it behaves more logically, and applies the option setting
3383    only to what follows. PCRE has been changed to follow suit. However, if it
3384    finds options settings right at the start of the pattern, it extracts them into
3385    the global options, as before. Thus, they show up in the info data.
3387    8. Added support for the \Q...\E escape sequence. Characters in between are
3388    treated as literals. This is slightly different from Perl in that $ and @ are
3389    also handled as literals inside the quotes. In Perl, they will cause variable
3390    interpolation. Note the following examples:
3392        Pattern            PCRE matches      Perl matches
3394        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
3395        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
3396        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
3398    For compatibility with Perl, \Q...\E sequences are recognized inside character
3399    classes as well as outside them.
3401    9. Re-organized 3 code statements in pcretest to avoid "overflow in
3402    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
3403    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
3404    signed/unsigned warnings.
3406    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
3407    option for pcretest, so I've replaced it by a simple function that does just
3408    that job.
3410    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
3411    "pcregrep -".
3413    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
3414    Java package. This provides some syntactic sugar for simple cases of what my
3415    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
3416    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
3417    item, you can use this simplified notation. Note that only makes sense with
3418    greedy quantifiers. Consequently, the use of the possessive quantifier forces
3419    greediness, whatever the setting of the PCRE_UNGREEDY option.
3421    13. A change of greediness default within a pattern was not taking effect at
3422    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
3423    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
3424    was abstracted outside.
3426    14. PCRE now supports the \G assertion. It is true when the current matching
3427    position is at the start point of the match. This differs from \A when the
3428    starting offset is non-zero. Used with the /g option of pcretest (or similar
3429    code), it works in the same way as it does for Perl's /g option. If all
3430    alternatives of a regex begin with \G, the expression is anchored to the start
3431    match position, and the "anchored" flag is set in the compiled expression.
3433    15. Some bugs concerning the handling of certain option changes within patterns
3434    have been fixed. These applied to options other than (?ims). For example,
3435    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
3436    been the other way round. Some of this was related to change 7 above.
3438    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
3439    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
3440    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
3441    POSIX classes only within a class (e.g. /[[:alpha:]]/).
3443    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
3444    mode. Unlike ".", it always matches newline, whatever the setting of
3445    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
3446    assertions. Perl allows it, but it doesn't (in general) work because it can't
3447    calculate the length of the lookbehind. At least, that's the case for Perl
3448    5.8.0 - I've been told they are going to document that it doesn't work in
3449    future.
3451    18. Added an error diagnosis for escapes that PCRE does not support: these are
3452    \L, \l, \N, \P, \p, \U, \u, and \X.
3454    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
3455    reading past the end of the pattern in cases such as /[abcd/.
3457    20. PCRE was getting more memory than necessary for patterns with classes that
3458    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
3460    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
3461    compiling PCRE for use with Virtual Pascal.
3463    22. Small fix to the Makefile to make it work properly if the build is done
3464    outside the source tree.
3466    23. Added a new extension: a condition to go with recursion. If a conditional
3467    subpattern starts with (?(R) the "true" branch is used if recursion has
3468    happened, whereas the "false" branch is used only at the top level.
3470    24. When there was a very long string of literal characters (over 255 bytes
3471    without UTF support, over 250 bytes with UTF support), the computation of how
3472    much memory was required could be incorrect, leading to segfaults or other
3473    strange effects.
3475    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
3476    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
3477    there was a subsequent back reference to those brackets. This meant that, for
3478    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
3479    possible to check for precisely this case. All we can do is abandon the
3480    optimization if .* occurs inside capturing brackets when there are any back
3481    references whatsoever. (See below for a better fix that came later.)
3483    26. The handling of the optimization for finding the first character of a
3484    non-anchored pattern, and for finding a character that is required later in the
3485    match were failing in some cases. This didn't break the matching; it just
3486    failed to optimize when it could. The way this is done has been re-implemented.
3488    27. Fixed typo in error message for invalid (?R item (it said "(?p").
3490    28. Added a new feature that provides some of the functionality that Perl
3491    provides with (?{...}). The facility is termed a "callout". The way it is done
3492    in PCRE is for the caller to provide an optional function, by setting
3493    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
3494    global variable. By default it is unset, which disables all calling out. To get
3495    the function called, the regex must include (?C) at appropriate points. This
3496    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
3497    This provides a means of identifying different callout points. When PCRE
3498    reaches such a point in the regex, if pcre_callout has been set, the external
3499    function is called. It is provided with data in a structure called
3500    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
3501    matching continues; if it returns a non-zero value, the match at the current
3502    point fails. However, backtracking will occur if possible. [This was changed
3503    later and other features added - see item 49 below.]
3505    29. pcretest is upgraded to test the callout functionality. It provides a
3506    callout function that displays information. By default, it shows the start of
3507    the match and the current position in the text. There are some new data escapes
3508    to vary what happens:
3510        \C+         in addition, show current contents of captured substrings
3511        \C-         do not supply a callout function
3512        \C!n        return 1 when callout number n is reached
3513        \C!n!m      return 1 when callout number n is reached for the mth time
3515    30. If pcregrep was called with the -l option and just a single file name, it
3516    output "<stdin>" if a match was found, instead of the file name.
3518    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
3519    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
3520    pcre_exec(). This saves a malloc/free per call. The default value of
3521    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
3522    when configuring.
3524    32. The default maximum size of a compiled pattern is 64K. There have been a
3525    few cases of people hitting this limit. The code now uses macros to handle the
3526    storing of links as offsets within the compiled pattern. It defaults to 2-byte
3527    links, but this can be changed to 3 or 4 bytes by --with-link-size when
3528    configuring. Tests 2 and 5 work only with 2-byte links because they output
3529    debugging information about compiled patterns.
3531    33. Internal code re-arrangements:
3533    (a) Moved the debugging function for printing out a compiled regex into
3534        its own source file (printint.c) and used #include to pull it into
3535        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
3536        separate copies.
3538    (b) Defined the list of op-code names for debugging as a macro in
3539        internal.h so that it is next to the definition of the opcodes.
3541    (c) Defined a table of op-code lengths for simpler skipping along compiled
3542        code. This is again a macro in internal.h so that it is next to the
3543        definition of the opcodes.
3545    34. Added support for recursive calls to individual subpatterns, along the
3546    lines of Robin Houston's patch (but implemented somewhat differently).
3548    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
3549    allow it to read and process whole directories in Win32. This code was
3550    contributed by Lionel Fourquaux; it has not been tested by me.
3552    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
3553    used to name a group. Names consist of alphanumerics and underscores, and must
3554    be unique. Back references use the syntax (?P=name) and recursive calls use
3555    (?P>name) which is a PCRE extension to the Python extension. Groups still have
3556    numbers. The function pcre_fullinfo() can be used after compilation to extract
3557    a name/number map. There are three relevant calls:
3559      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
3560      PCRE_INFO_NAMECOUNT            yields the number of entries
3561      PCRE_INFO_NAMETABLE            yields a pointer to the map.
3563    The map is a vector of fixed-size entries. The size of each entry depends on
3564    the length of the longest name used. The first two bytes of each entry are the
3565    group number, most significant byte first. There follows the corresponding
3566    name, zero terminated. The names are in alphabetical order.
3568    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
3569    case instead of 255. Making it the same both with and without UTF-8 support
3570    means that the same test output works with both.
3572    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
3573    calling malloc() with a zero argument.
3575    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
3576    optimization. I've improved things by keeping a bitmap of backreferences with
3577    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
3578    fact referenced, the optimization can be applied. It is unlikely that a
3579    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
3580    the match to follow \n) will appear inside brackets with a number greater than
3581    31, but if it does, any back reference > 31 suppresses the optimization.
3583    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect