ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 203 by ph10, Fri Aug 3 10:09:57 2007 UTC revision 752 by zherczeg, Sat Nov 19 15:28:29 2011 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
4  Version 7.3 05-Jul-07  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.
45    10. Add a cast and remove a redundant test from the code.
47    11. JIT should use pcre_malloc/pcre_free for allocation.
50    Version 8.20 21-Oct-2011
51    ------------------------
53    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
54        a POSIX class. After further experiments with Perl, which convinced me that
55        Perl has bugs and confusions, a closing square bracket is no longer allowed
56        in a POSIX name. This bug also affected patterns with classes that started
57        with full stops.
59    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
60        captured substring, but while checking the failing first alternative,
61        substring 1 is temporarily captured. If the output vector supplied to
62        pcre_exec() was not big enough for this capture, the yield of the function
63        was still zero ("insufficient space for captured substrings"). This cannot
64        be totally fixed without adding another stack variable, which seems a lot
65        of expense for a edge case. However, I have improved the situation in cases
66        such as /(a)(b)x|abc/ matched against "abc", where the return code
67        indicates that fewer than the maximum number of slots in the ovector have
68        been set.
70    3.  Related to (2) above: when there are more back references in a pattern than
71        slots in the output vector, pcre_exec() uses temporary memory during
72        matching, and copies in the captures as far as possible afterwards. It was
73        using the entire output vector, but this conflicts with the specification
74        that only 2/3 is used for passing back captured substrings. Now it uses
75        only the first 2/3, for compatibility. This is, of course, another edge
76        case.
78    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
79        main code base, and can be used by building with --enable-jit. When this is
80        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
81        runtime --no-jit option is given.
83    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
84        ovector, the return from the function was zero, implying that there were
85        other matches that did not fit. The correct "exactly full" value is now
86        returned.
88    6.  If a subpattern that was called recursively or as a subroutine contained
89        (*PRUNE) or any other control that caused it to give a non-standard return,
90        invalid errors such as "Error -26 (nested recursion at the same subject
91        position)" or even infinite loops could occur.
93    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
94        computing the minimum length on reaching *ACCEPT, and so ended up with the
95        wrong value of 1 rather than 0. Further investigation indicates that
96        computing a minimum subject length in the presence of *ACCEPT is difficult
97        (think back references, subroutine calls), and so I have changed the code
98        so that no minimum is registered for a pattern that contains *ACCEPT.
100    8.  If (*THEN) was present in the first (true) branch of a conditional group,
101        it was not handled as intended. [But see 16 below.]
103    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
104        Sheri Pierce.
106    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
107        the first byte in a match must be "a".
109    11. Change 17 for 8.13 increased the recursion depth for patterns like
110        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
111        pattern contains any instances of (*THEN). If it does not, the old
112        optimizations are restored. It would be nice to do this on a per-group
113        basis, but at the moment that is not feasible.
115    12. In some environments, the output of pcretest -C is CRLF terminated. This
116        broke RunTest's code that checks for the link size. A single white space
117        character after the value is now allowed for.
119    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
120        For "fr", it uses the Windows-specific input and output files.
122    14. If (*THEN) appeared in a group that was called recursively or as a
123        subroutine, it did not work as intended. [But see next item.]
125    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
126        pattern fragments (but not containing any | characters). If A and B are
127        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
128        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
129        D. In other words, Perl considers parentheses that do not contain any |
130        characters to be part of a surrounding alternative, whereas PCRE was
131        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
132        differently. PCRE now behaves in the same way as Perl, except in the case
133        of subroutine/recursion calls such as (?1) which have in any case always
134        been different (but PCRE had them first :-).
136    16. Related to 15 above: Perl does not treat the | in a conditional group as
137        creating alternatives. Such a group is treated in the same way as an
138        ordinary group without any | characters when processing (*THEN). PCRE has
139        been changed to match Perl's behaviour.
141    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
142        RunGrepTest script failed.
144    18. Change 22 for version 13 caused atomic groups to use more stack. This is
145        inevitable for groups that contain captures, but it can lead to a lot of
146        stack use in large patterns. The old behaviour has been restored for atomic
147        groups that do not contain any capturing parentheses.
149    19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not
150        suppress the check for a minimum subject length at run time. (If it was
151        given to pcre_exec() or pcre_dfa_exec() it did work.)
153    20. Fixed an ASCII-dependent infelicity in pcretest that would have made it
154        fail to work when decoding hex characters in data strings in EBCDIC
155        environments.
157    21. It appears that in at least one Mac OS environment, the isxdigit() function
158        is implemented as a macro that evaluates to its argument more than once,
159        contravening the C 90 Standard (I haven't checked a later standard). There
160        was an instance in pcretest which caused it to go wrong when processing
161        \x{...} escapes in subject strings. The has been rewritten to avoid using
162        things like p++ in the argument of isxdigit().
165    Version 8.13 16-Aug-2011
166    ------------------------
168    1.  The Unicode data tables have been updated to Unicode 6.0.0.
170    2.  Two minor typos in pcre_internal.h have been fixed.
172    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
173        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
174        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
176    4.  There were a number of related bugs in the code for matching backrefences
177        caselessly in UTF-8 mode when codes for the characters concerned were
178        different numbers of bytes. For example, U+023A and U+2C65 are an upper
179        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
180        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
181        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
182        2-byte code at the end of the subject (it thought there wasn't enough data
183        left).
185    5.  Comprehensive information about what went wrong is now returned by
186        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
187        as the output vector has at least 2 elements. The offset of the start of
188        the failing character and a reason code are placed in the vector.
190    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
191        now returned is for the first byte of the failing character, instead of the
192        last byte inspected. This is an incompatible change, but I hope it is small
193        enough not to be a problem. It makes the returned offset consistent with
194        pcre_exec() and pcre_dfa_exec().
196    7.  pcretest now gives a text phrase as well as the error number when
197        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
198        failure, the offset and reason code are output.
200    8.  When \R was used with a maximizing quantifier it failed to skip backwards
201        over a \r\n pair if the subsequent match failed. Instead, it just skipped
202        back over a single character (\n). This seems wrong (because it treated the
203        two characters as a single entity when going forwards), conflicts with the
204        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
205        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
206        has been changed.
208    9.  Some internal refactoring has changed the processing so that the handling
209        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
210        time (the PCRE_DOTALL option was changed this way some time ago: version
211        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
212        which was always a bit of a fudge. It also means that there is one less
213        argument for the match() function, which reduces its stack requirements
214        slightly. This change also fixes an incompatibility with Perl: the pattern
215        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
217    10. More internal refactoring has drastically reduced the number of recursive
218        calls to match() for possessively repeated groups such as (abc)++ when
219        using pcre_exec().
221    11. While implementing 10, a number of bugs in the handling of groups were
222        discovered and fixed:
224        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
225        (a|)*(?1) gave a compile-time internal error.
226        ((a|)+)+  did not notice that the outer group could match an empty string.
227        (^a|^)+   was not marked as anchored.
228        (.*a|.*)+ was not marked as matching at start or after a newline.
230    12. Yet more internal refactoring has removed another argument from the match()
231        function. Special calls to this function are now indicated by setting a
232        value in a variable in the "match data" data block.
234    13. Be more explicit in pcre_study() instead of relying on "default" for
235        opcodes that mean there is no starting character; this means that when new
236        ones are added and accidentally left out of pcre_study(), testing should
237        pick them up.
239    14. The -s option of pcretest has been documented for ages as being an old
240        synonym of -m (show memory usage). I have changed it to mean "force study
241        for every regex", that is, assume /S for every regex. This is similar to -i
242        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
243        using it. It makes it easier to run collections of tests with and without
244        study enabled, and thereby test pcre_study() more easily. All the standard
245        tests are now run with and without -s (but some patterns can be marked as
246        "never study" - see 20 below).
248    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
249        restoration of the capturing data to the outer values was not happening
250        correctly.
252    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
253        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
254        pattern had matched an empty string, and so incorrectly returned a no
255        match.
257    17. There was optimizing code for the last branch of non-capturing parentheses,
258        and also for the obeyed branch of a conditional subexpression, which used
259        tail recursion to cut down on stack usage. Unfortunately, now that there is
260        the possibility of (*THEN) occurring in these branches, tail recursion is
261        no longer possible because the return has to be checked for (*THEN). These
262        two optimizations have therefore been removed. [But see 8.20/11 above.]
264    18. If a pattern containing \R was studied, it was assumed that \R always
265        matched two bytes, thus causing the minimum subject length to be
266        incorrectly computed because \R can also match just one byte.
268    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
269        was incorrectly computed.
271    20. If /S is present twice on a test pattern in pcretest input, it now
272        *disables* studying, thereby overriding the use of -s on the command line
273        (see 14 above). This is necessary for one or two tests to keep the output
274        identical in both cases.
276    21. When (*ACCEPT) was used in an assertion that matched an empty string and
277        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
279    22. When an atomic group that contained a capturing parenthesis was
280        successfully matched, but the branch in which it appeared failed, the
281        capturing was not being forgotten if a higher numbered group was later
282        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
283        group 1 to "a", when in fact it should be unset. This applied to multi-
284        branched capturing and non-capturing groups, repeated or not, and also to
285        positive assertions (capturing in negative assertions does not happen
286        in PCRE) and also to nested atomic groups.
288    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
289        subject after a captured substring, to make it easier to tell which of a
290        number of identical substrings has been captured.
292    24. The way atomic groups are processed by pcre_exec() has been changed so that
293        if they are repeated, backtracking one repetition now resets captured
294        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
295        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
296        "aaa". Previously, it would have been "a". As part of this code
297        refactoring, the way recursive calls are handled has also been changed.
299    25. If an assertion condition captured any substrings, they were not passed
300        back unless some other capturing happened later. For example, if
301        (?(?=(a))a) was matched against "a", no capturing was returned.
303    26. When studying a pattern that contained subroutine calls or assertions,
304        the code for finding the minimum length of a possible match was handling
305        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
306        group 1 called group 2 while simultaneously a separate group 2 called group
307        1). A stack overflow occurred in this case. I have fixed this by limiting
308        the recursion depth to 10.
310    27. Updated RunTest.bat in the distribution to the version supplied by Tom
311        Fortmann. This supports explicit test numbers on the command line, and has
312        argument validation and error reporting.
314    28. An instance of \X with an unlimited repeat could fail if at any point the
315        first character it looked at was a mark character.
317    29. Some minor code refactoring concerning Unicode properties and scripts
318        should reduce the stack requirement of match() slightly.
320    30. Added the '=' option to pcretest to check the setting of unused capturing
321        slots at the end of the pattern, which are documented as being -1, but are
322        not included in the return count.
324    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
325        compiled something random. Now it gives a compile-time error (as does
326        Perl).
328    32. A *MARK encountered during the processing of a positive assertion is now
329        recorded and passed back (compatible with Perl).
331    33. If --only-matching or --colour was set on a pcregrep call whose pattern
332        had alternative anchored branches, the search for a second match in a line
333        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
334        matched the line "0102" twice. The same bug affected patterns that started
335        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
336        twice.
338    34. Previously, PCRE did not allow quantification of assertions. However, Perl
339        does, and because of capturing effects, quantifying parenthesized
340        assertions may at times be useful. Quantifiers are now allowed for
341        parenthesized assertions.
343    35. A minor code tidy in pcre_compile() when checking options for \R usage.
345    36. \g was being checked for fancy things in a character class, when it should
346        just be a literal "g".
348    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
349        appearance of a nested POSIX class supersedes an apparent external class.
350        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
351        unescaped square brackets may also appear as part of class names. For
352        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
353        more like Perl. (But see 8.20/1 above.)
355    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
356        was because it thought it was \N{name}, which is not supported).
358    39. Add minix to OS list not supporting the -S option in pcretest.
360    40. PCRE tries to detect cases of infinite recursion at compile time, but it
361        cannot analyze patterns in sufficient detail to catch mutual recursions
362        such as ((?1))((?2)). There is now a runtime test that gives an error if a
363        subgroup is called recursively as a subpattern for a second time at the
364        same position in the subject string. In previous releases this might have
365        been caught by the recursion limit, or it might have run out of stack.
367    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
368        happen only once. PCRE was, however incorrectly giving a compile time error
369        "recursive call could loop indefinitely" because it cannot analyze the
370        pattern in sufficient detail. The compile time test no longer happens when
371        PCRE is compiling a conditional subpattern, but actual runaway loops are
372        now caught at runtime (see 40 above).
374    42. It seems that Perl allows any characters other than a closing parenthesis
375        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
376        has been changed to be the same.
378    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
379        as not to get warnings when autogen.sh is called. Also changed
380        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
382    44. To help people who use pcregrep to scan files containing exceedingly long
383        lines, the following changes have been made:
385        (a) The default value of the buffer size parameter has been increased from
386            8K to 20K. (The actual buffer used is three times this size.)
388        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
389            PCRE is built.
391        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
392            to be set at run time.
394        (d) Numerical values in pcregrep options can be followed by K or M, for
395            example --buffer-size=50K.
397        (e) If a line being scanned overflows pcregrep's buffer, an error is now
398            given and the return code is set to 2.
400    45. Add a pointer to the latest mark to the callout data block.
402    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
403        partial match of an empty string instead of no match. This was specific to
404        the use of ".".
406    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
407        complete match instead of a partial match. This bug was dependent on both
408        the PCRE_UTF8 and PCRE_DOTALL options being set.
410    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
411        starting byte set, because \b was not being ignored.
414    Version 8.12 15-Jan-2011
415    ------------------------
417    1.  Fixed some typos in the markup of the man pages, and wrote a script that
418        checks for such things as part of the documentation building process.
420    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
421        --match-limit and --recursion-limit options (added for 8.11). In
422        particular, this made one of the standard tests fail. (The integer value
423        went into the wrong half of a long int.)
425    3.  If the --colour option was given to pcregrep with -v (invert match), it
426        did strange things, either producing crazy output, or crashing. It should,
427        of course, ignore a request for colour when reporting lines that do not
428        match.
430    4.  Another pcregrep bug caused similar problems if --colour was specified with
431        -M (multiline) and the pattern match finished with a line ending.
433    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
434        matched in multiline mode, the following line was shown as part of the
435        match. This seems wrong, so I have changed it.
437    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
438        the check for further matches in the same line (so they could be coloured)
439        to overrun the end of the current line. If another match was found, it was
440        incorrectly shown (and then shown again when found in the next line).
442    7.  If pcregrep was compiled under Windows, there was a reference to the
443        function pcregrep_exit() before it was defined. I am assuming this was
444        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
445        reported by a user. I've moved the definition above the reference.
448    Version 8.11 10-Dec-2010
449    ------------------------
451    1.  (*THEN) was not working properly if there were untried alternatives prior
452        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
453        backtracked to try for "b" instead of moving to the next alternative branch
454        at the same level (in this case, to look for "c"). The Perl documentation
455        is clear that when (*THEN) is backtracked onto, it goes to the "next
456        alternative in the innermost enclosing group".
458    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
459        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
460        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
461        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
462        (*THEN).
464    3.  If \s appeared in a character class, it removed the VT character from
465        the class, even if it had been included by some previous item, for example
466        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
467        of \s, but is part of the POSIX "space" class.)
469    4.  A partial match never returns an empty string (because you can always
470        match an empty string at the end of the subject); however the checking for
471        an empty string was starting at the "start of match" point. This has been
472        changed to the "earliest inspected character" point, because the returned
473        data for a partial match starts at this character. This means that, for
474        example, /(?<=abc)def/ gives a partial match for the subject "abc"
475        (previously it gave "no match").
477    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
478        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
479        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
480        has an implication that the given string is incomplete (because a partial
481        match is preferred over a full match). For this reason, these items now
482        give a partial match in this situation. [Aside: previously, the one case
483        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
484        match rather than a full match, which was wrong by the old rules, but is
485        now correct.]
487    6.  There was a bug in the handling of #-introduced comments, recognized when
488        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
489        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
490        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
491        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
492        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
493        places in pcre_compile().
495    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
496        comments when looking ahead for named forward references to subpatterns,
497        the only newline sequence it recognized was NL. It now handles newlines
498        according to the set newline convention.
500    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
501        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
502        cater for a lack of strerror(). These oversights have been fixed.
504    9.  Added --match-limit and --recursion-limit to pcregrep.
506    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
508    11. When the -o option was used, pcregrep was setting a return code of 1, even
509        when matches were found, and --line-buffered was not being honoured.
511    12. Added an optional parentheses number to the -o and --only-matching options
512        of pcregrep.
514    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
515        can match an empty string. The code to do it in pcretest and pcredemo
516        needed fixing:
518        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
519            only one byte after an empty string match just before CRLF (this case
520            just got forgotten; "any" and "anycrlf" were OK).
522        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
523            mode when an empty string match preceded an ASCII character followed by
524            a non-ASCII character. (The code for advancing by one character rather
525            than one byte was nonsense.)
527        (c) The pcredemo.c sample program did not have any code at all to handle
528            the cases when CRLF is a valid newline sequence.
530    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
531        as a starting offset was within the subject string. There is now a new
532        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
533        negative or greater than the length of the string. In order to test this,
534        pcretest is extended to allow the setting of negative starting offsets.
536    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
537        starting offset points to the beginning of a UTF-8 character was
538        unnecessarily clumsy. I tidied it up.
540    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
541        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
543    17. Nobody had reported that the --include_dir option, which was added in
544        release 7.7 should have been called --include-dir (hyphen, not underscore)
545        for compatibility with GNU grep. I have changed it to --include-dir, but
546        left --include_dir as an undocumented synonym, and the same for
547        --exclude-dir, though that is not available in GNU grep, at least as of
548        release 2.5.4.
550    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
551        characters from a string of bytes) have been redefined so as not to use
552        loops, in order to improve performance in some environments. At the same
553        time, I abstracted some of the common code into auxiliary macros to save
554        repetition (this should not affect the compiled code).
556    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
557        compile-time error is now given if \c is not followed by an ASCII
558        character, that is, a byte less than 128. (In EBCDIC mode, the code is
559        different, and any byte value is allowed.)
561    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
562        START_OPTIMIZE option, which is now allowed at compile time - but just
563        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
564        to pcregrep and other applications that have no direct access to PCRE
565        options. The new /Y option in pcretest sets this option when calling
566        pcre_compile().
568    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
569        back references. Groups containing recursive back references were forced to
570        be atomic by that change, but in the case of named groups, the amount of
571        memory required was incorrectly computed, leading to "Failed: internal
572        error: code overflow". This has been fixed.
574    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
575        pcretest.c, to avoid build problems in some Borland environments.
578    Version 8.10 25-Jun-2010
579    ------------------------
581    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
582        THEN.
584    2.  (*ACCEPT) was not working when inside an atomic group.
586    3.  Inside a character class, \B is treated as a literal by default, but
587        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
588        causes the error). The code is unchanged, but I tidied the documentation.
590    4.  Inside a character class, PCRE always treated \R and \X as literals,
591        whereas Perl faults them if its -w option is set. I have changed PCRE so
592        that it faults them when PCRE_EXTRA is set.
594    5.  Added support for \N, which always matches any character other than
595        newline. (It is the same as "." when PCRE_DOTALL is not set.)
597    6.  When compiling pcregrep with newer versions of gcc which may have
598        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
599        declared with attribute warn_unused_result" were given. Just casting the
600        result to (void) does not stop the warnings; a more elaborate fudge is
601        needed. I've used a macro to implement this.
603    7.  Minor change to pcretest.c to avoid a compiler warning.
605    8.  Added four artifical Unicode properties to help with an option to make
606        \s etc use properties (see next item). The new properties are: Xan
607        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
609    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
610        use Unicode properties. (*UCP) at the start of a pattern can be used to set
611        this option. Modified pcretest to add /W to test this facility. Added
612        REG_UCP to make it available via the POSIX interface.
614    10. Added --line-buffered to pcregrep.
616    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
617        studied, and the match started with a letter with a code point greater than
618        127 whose first byte was different to the first byte of the other case of
619        the letter, the other case of this starting letter was not recognized
620        (#976).
622    12. If a pattern that was studied started with a repeated Unicode property
623        test, for example, \p{Nd}+, there was the theoretical possibility of
624        setting up an incorrect bitmap of starting bytes, but fortunately it could
625        not have actually happened in practice until change 8 above was made (it
626        added property types that matched character-matching opcodes).
628    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
629        possible starting bytes for non-anchored patterns.
631    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
632        \R, and also a number of cases that involve Unicode properties, both
633        explicit and implicit when PCRE_UCP is set.
635    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
636        input, it could crash or give wrong results if characters with values
637        greater than 0xc0 were present in the subject string. (Detail: it assumed
638        UTF-8 input when processing these items.)
640    16. Added a lot of (int) casts to avoid compiler warnings in systems where
641        size_t is 64-bit (#991).
643    17. Added a check for running out of memory when PCRE is compiled with
644        --disable-stack-for-recursion (#990).
646    18. If the last data line in a file for pcretest does not have a newline on
647        the end, a newline was missing in the output.
649    19. The default pcre_chartables.c file recognizes only ASCII characters (values
650        less than 128) in its various bitmaps. However, there is a facility for
651        generating tables according to the current locale when PCRE is compiled. It
652        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
653        characters, are recognized by isspace() and therefore were getting set in
654        these tables, and indeed these tables seem to approximate to ISO 8859. This
655        caused a problem in UTF-8 mode when pcre_study() was used to create a list
656        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
657        which of course cannot start UTF-8 characters. I have changed the code so
658        that only real ASCII characters (less than 128) and the correct starting
659        bytes for UTF-8 encodings are set for characters greater than 127 when in
660        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
661        altogether.)
663    20. Added the /T option to pcretest so as to be able to run tests with non-
664        standard character tables, thus making it possible to include the tests
665        used for 19 above in the standard set of tests.
667    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
668        reference to a subpattern the other side of a comment that contains an
669        opening parenthesis caused either an internal compiling error, or a
670        reference to the wrong subpattern.
673    Version 8.02 19-Mar-2010
674    ------------------------
676    1.  The Unicode data tables have been updated to Unicode 5.2.0.
678    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
679        configured.
681    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
682        original author of that file, following a query about its status.
684    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
685        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
687    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
688        quantifier applied to a forward-referencing subroutine call, could compile
689        incorrect code or give the error "internal error: previously-checked
690        referenced subpattern not found".
692    6.  Both MS Visual Studio and Symbian OS have problems with initializing
693        variables to point to external functions. For these systems, therefore,
694        pcre_malloc etc. are now initialized to local functions that call the
695        relevant global functions.
697    7.  There were two entries missing in the vectors called coptable and poptable
698        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
699        I've fixed the data, and added a kludgy way of testing at compile time that
700        the lengths are correct (equal to the number of opcodes).
702    8.  Following on from 7, I added a similar kludge to check the length of the
703        eint vector in pcreposix.c.
705    9.  Error texts for pcre_compile() are held as one long string to avoid too
706        much relocation at load time. To find a text, the string is searched,
707        counting zeros. There was no check for running off the end of the string,
708        which could happen if a new error number was added without updating the
709        string.
711    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
713    11. \K was not working if it appeared in an atomic group or in a group that
714        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
715        \K is "not well defined" if used in an assertion. PCRE now accepts it if
716        the assertion is positive, but not if it is negative.
718    12. Change 11 fortuitously reduced the size of the stack frame used in the
719        "match()" function of pcre_exec.c by one pointer. Forthcoming
720        implementation of support for (*MARK) will need an extra pointer on the
721        stack; I have reserved it now, so that the stack frame size does not
722        decrease.
724    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
725        item in branch that calls a recursion is a subroutine call - as in the
726        second branch in the above example - was incorrectly given the compile-
727        time error "recursive call could loop indefinitely" because pcre_compile()
728        was not correctly checking the subroutine for matching a non-empty string.
730    14. The checks for overrunning compiling workspace could trigger after an
731        overrun had occurred. This is a "should never occur" error, but it can be
732        triggered by pathological patterns such as hundreds of nested parentheses.
733        The checks now trigger 100 bytes before the end of the workspace.
735    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
738    Version 8.01 19-Jan-2010
739    ------------------------
741    1.  If a pattern contained a conditional subpattern with only one branch (in
742        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
743        computed the wrong minimum data length (which is of course zero for such
744        subpatterns). This could cause incorrect "no match" results.
746    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
747        the pattern is reset in the first branch, pcre_compile() failed with
748        "internal error: code overflow at offset...". This happened only when
749        the reset was to the original external option setting. (An optimization
750        abstracts leading options settings into an external setting, which was the
751        cause of this.)
753    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
754        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
755        assertion pattern did not match (meaning that the assertion was true), it
756        was incorrectly treated as false if the SKIP had been reached during the
757        matching. This also applied to assertions used as conditions.
759    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
760        assertion subpattern, including such a pattern used as a condition,
761        unpredictable results occurred, instead of the error return
764    5.  The C++ GlobalReplace function was not working like Perl for the special
765        situation when an empty string is matched. It now does the fancy magic
766        stuff that is necessary.
768    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
769        removed. (These were left over from very, very early versions of PCRE.)
771    7.  Some cosmetic changes to the code to make life easier when compiling it
772        as part of something else:
774        (a) Change DEBUG to PCRE_DEBUG.
776        (b) In pcre_compile(), rename the member of the "branch_chain" structure
777            called "current" as "current_branch", to prevent a collision with the
778            Linux macro when compiled as a kernel module.
780        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
781            prevent a collision with the Linux macro when compiled as a kernel
782            module.
784    8.  In pcre_compile() there are some checks for integer overflows that used to
785        cast potentially large values to (double). This has been changed to that
786        when building, a check for int64_t is made, and if it is found, it is used
787        instead, thus avoiding the use of floating point arithmetic. (There is no
788        other use of FP in PCRE.) If int64_t is not found, the fallback is to
789        double.
791    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
792        2005 (difference between two addresses compared to an unsigned value).
794    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
795        custom one, because of the following reported problem in Windows:
797          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
798              under Win32.
799          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
800              therefore missing the function definition.
801          - The compiler thus generates a "C" signature for the test function.
802          - The linker fails to find the "C" function.
803          - PCRE fails to configure if asked to do so against libbz2.
805    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
806        messages were output:
808          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
809          rerunning libtoolize, to keep the correct libtool macros in-tree.
810          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
812        I have done both of these things.
814    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
815        most of the time, it *can* run out if it is given a pattern that contains a
816        runaway infinite recursion. I updated the discussion in the pcrestack man
817        page.
819    13. Now that we have gone to the x.xx style of version numbers, the minor
820        version may start with zero. Using 08 or 09 is a bad idea because users
821        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
822        interpreted as invalid octal numbers. I've updated the previous comment in
823        configure.ac, and also added a check that gives an error if 08 or 09 are
824        used.
826    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
827        causing partial matching to fail when the end of the subject matched \W
828        in a UTF-8 pattern where \W was quantified with a minimum of 3.
830    15. There were some discrepancies between the declarations in pcre_internal.h
831        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
832        their definitions. The declarations used "const uschar *" and the
833        definitions used USPTR. Even though USPTR is normally defined as "const
834        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
835        reported that: "This difference in casting confuses some C++ compilers, for
836        example, SunCC recognizes above declarations as different functions and
837        generates broken code for hbpcre." I have changed the declarations to use
838        USPTR.
840    16. GNU libtool is named differently on some systems. The autogen.sh script now
841        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
842        (FreeBSD).
844    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
845        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
846        comment: "Figure out how to create a longlong from a string: strtoll and
847        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
848        instance, but it only takes 2 args instead of 3!"
850    18. A subtle bug concerned with back references has been fixed by a change of
851        specification, with a corresponding code fix. A pattern such as
852        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
853        refers, was giving matches when it shouldn't. For example, xa=xaaa would
854        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
855        same bug. Such groups have to be quantified to be useful, or contained
856        inside another quantified group. (If there's no repetition, the reference
857        can never match.) The problem arises because, having left the group and
858        moved on to the rest of the pattern, a later failure that backtracks into
859        the group uses the captured value from the final iteration of the group
860        rather than the correct earlier one. I have fixed this in PCRE by forcing
861        any group that contains a reference to itself to be an atomic group; that
862        is, there cannot be any backtracking into it once it has completed. This is
863        similar to recursive and subroutine calls.
866    Version 8.00 19-Oct-09
867    ----------------------
869    1.  The table for translating pcre_compile() error codes into POSIX error codes
870        was out-of-date, and there was no check on the pcre_compile() error code
871        being within the table. This could lead to an OK return being given in
872        error.
874    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
875        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
876        in a Windows environment.
878    3.  The pcregrep --count option prints the count for each file even when it is
879        zero, as does GNU grep. However, pcregrep was also printing all files when
880        --files-with-matches was added. Now, when both options are given, it prints
881        counts only for those files that have at least one match. (GNU grep just
882        prints the file name in this circumstance, but including the count seems
883        more useful - otherwise, why use --count?) Also ensured that the
884        combination -clh just lists non-zero counts, with no names.
886    4.  The long form of the pcregrep -F option was incorrectly implemented as
887        --fixed_strings instead of --fixed-strings. This is an incompatible change,
888        but it seems right to fix it, and I didn't think it was worth preserving
889        the old behaviour.
891    5.  The command line items --regex=pattern and --regexp=pattern were not
892        recognized by pcregrep, which required --regex pattern or --regexp pattern
893        (with a space rather than an '='). The man page documented the '=' forms,
894        which are compatible with GNU grep; these now work.
896    6.  No libpcreposix.pc file was created for pkg-config; there was just
897        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
899    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
900        when UCP support is not needed, by modifying the Python script that
901        generates it from Unicode data files. This should not matter if the module
902        is correctly used as a library, but I received one complaint about 50K of
903        unwanted data. My guess is that the person linked everything into his
904        program rather than using a library. Anyway, it does no harm.
906    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
907        was a minimum greater than 1 for a wide character in a possessive
908        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
909        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
910        character. Chaos in the form of incorrect output or a compiling loop could
911        result.
913    9.  The restrictions on what a pattern can contain when partial matching is
914        requested for pcre_exec() have been removed. All patterns can now be
915        partially matched by this function. In addition, if there are at least two
916        slots in the offset vector, the offset of the earliest inspected character
917        for the match and the offset of the end of the subject are set in them when
918        PCRE_ERROR_PARTIAL is returned.
920    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
921        synonymous with PCRE_PARTIAL, for backwards compatibility, and
922        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
923        and may be more useful for multi-segment matching.
925    11. Partial matching with pcre_exec() is now more intuitive. A partial match
926        used to be given if ever the end of the subject was reached; now it is
927        given only if matching could not proceed because another character was
928        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
929        string "Z", which now yields "no match" instead of "partial match". In the
930        case of pcre_dfa_exec(), "no match" is given if every matching path for the
931        final character ended with (*FAIL).
933    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
934        if the pattern had a "must contain" character that was already found in the
935        earlier partial match, unless partial matching was again requested. For
936        example, with the pattern /dog.(body)?/, the "must contain" character is
937        "g". If the first part-match was for the string "dog", restarting with
938        "sbody" failed. This bug has been fixed.
940    13. The string returned by pcre_dfa_exec() after a partial match has been
941        changed so that it starts at the first inspected character rather than the
942        first character of the match. This makes a difference only if the pattern
943        starts with a lookbehind assertion or \b or \B (\K is not supported by
944        pcre_dfa_exec()). It's an incompatible change, but it makes the two
945        matching functions compatible, and I think it's the right thing to do.
947    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
948        so that the demonstration program is easily available in environments where
949        PCRE has not been installed from source.
951    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
952        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
953        library.
955    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
956        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
957        is not the first non-POSIX option to be added. Clearly some people find
958        these options useful.
960    17. If a caller to the POSIX matching function regexec() passes a non-zero
961        value for nmatch with a NULL value for pmatch, the value of
962        nmatch is forced to zero.
964    18. RunGrepTest did not have a test for the availability of the -u option of
965        the diff command, as RunTest does. It now checks in the same way as
966        RunTest, and also checks for the -b option.
968    19. If an odd number of negated classes containing just a single character
969        interposed, within parentheses, between a forward reference to a named
970        subpattern and the definition of the subpattern, compilation crashed with
971        an internal error, complaining that it could not find the referenced
972        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
973        [The bug was that it was starting one character too far in when skipping
974        over the character class, thus treating the ] as data rather than
975        terminating the class. This meant it could skip too much.]
977    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
978        /g option in pcretest when the pattern contains \K, which makes it possible
979        to have an empty string match not at the start, even when the pattern is
980        anchored. Updated pcretest and pcredemo to use this option.
982    21. If the maximum number of capturing subpatterns in a recursion was greater
983        than the maximum at the outer level, the higher number was returned, but
984        with unset values at the outer level. The correct (outer level) value is
985        now given.
987    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
988        PCRE did not set those parentheses (unlike Perl). I have now found a way to
989        make it do so. The string so far is captured, making this feature
990        compatible with Perl.
992    23. The tests have been re-organized, adding tests 11 and 12, to make it
993        possible to check the Perl 5.10 features against Perl 5.10.
995    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
996        pattern matches a fixed length string. PCRE did not allow this; now it
997        does. Neither allows recursion.
999    25. I finally figured out how to implement a request to provide the minimum
1000        length of subject string that was needed in order to match a given pattern.
1001        (It was back references and recursion that I had previously got hung up
1002        on.) This code has now been added to pcre_study(); it finds a lower bound
1003        to the length of subject needed. It is not necessarily the greatest lower
1004        bound, but using it to avoid searching strings that are too short does give
1005        some useful speed-ups. The value is available to calling programs via
1006        pcre_fullinfo().
1008    26. While implementing 25, I discovered to my embarrassment that pcretest had
1009        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
1010        study optimizations had never been tested with that matching function.
1011        Oops. What is worse, even when it was passed study data, there was a bug in
1012        pcre_dfa_exec() that meant it never actually used it. Double oops. There
1013        were also very few tests of studied patterns with pcre_dfa_exec().
1015    27. If (?| is used to create subpatterns with duplicate numbers, they are now
1016        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
1017        on the other side of the coin, they are no longer allowed to have different
1018        names, because these cannot be distinguished in PCRE, and this has caused
1019        confusion. (This is a difference from Perl.)
1021    28. When duplicate subpattern names are present (necessarily with different
1022        numbers, as required by 27 above), and a test is made by name in a
1023        conditional pattern, either for a subpattern having been matched, or for
1024        recursion in such a pattern, all the associated numbered subpatterns are
1025        tested, and the overall condition is true if the condition is true for any
1026        one of them. This is the way Perl works, and is also more like the way
1027        testing by number works.
1030    Version 7.9 11-Apr-09
1031    ---------------------
1033    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
1034        (pcretest), all targets were linked against these libraries. This included
1035        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
1036        libraries. This caused unwanted dependencies to be created. This problem
1037        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
1038        pcretest is linked with readline.
1040    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
1041        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
1042        moved outside it again, because FALSE and TRUE are already defined in AIX,
1043        but BOOL is not.
1045    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
1046        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
1048    4.  The pcregrep documentation said spaces were inserted as well as colons (or
1049        hyphens) following file names and line numbers when outputting matching
1050        lines. This is not true; no spaces are inserted. I have also clarified the
1051        wording for the --colour (or --color) option.
1053    5.  In pcregrep, when --colour was used with -o, the list of matching strings
1054        was not coloured; this is different to GNU grep, so I have changed it to be
1055        the same.
1057    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
1058        each matching line was coloured. Now it goes on to look for further matches
1059        of any of the test patterns, which is the same behaviour as GNU grep.
1061    7.  A pattern that could match an empty string could cause pcregrep to loop; it
1062        doesn't make sense to accept an empty string match in pcregrep, so I have
1063        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
1064        seems to be how GNU grep behaves.
1066    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
1067        start or after a newline", because the conditional assertion was not being
1068        correctly handled. The rule now is that both the assertion and what follows
1069        in the first alternative must satisfy the test.
1071    9.  If auto-callout was enabled in a pattern with a conditional group whose
1072        condition was an assertion, PCRE could crash during matching, both with
1073        pcre_exec() and pcre_dfa_exec().
1075    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
1076        used for matching.
1078    11. Unicode property support in character classes was not working for
1079        characters (bytes) greater than 127 when not in UTF-8 mode.
1081    12. Added the -M command line option to pcretest.
1083    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
1085    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
1087    16. Added comments and documentation about mis-use of no_arg in the C++
1088        wrapper.
1090    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
1091        from Martin Jerabek that uses macro names for all relevant character and
1092        string constants.
1094    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
1095        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
1096        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
1097        these, but not everybody uses configure.
1099    19. A conditional group that had only one branch was not being correctly
1100        recognized as an item that could match an empty string. This meant that an
1101        enclosing group might also not be so recognized, causing infinite looping
1102        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
1103        with the subject "ab", where knowledge that the repeated group can match
1104        nothing is needed in order to break the loop.
1106    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
1107        exec(), but without supplying a callout function, matching went wrong.
1109    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
1110        leak if the size of the offset vector was greater than 30. When the vector
1111        is smaller, the saved offsets during recursion go onto a local stack
1112        vector, but for larger vectors malloc() is used. It was failing to free
1113        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
1114        error, in fact).
1116    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
1117        heapframe that is used only when UTF-8 support is enabled. This caused no
1118        problem, but was untidy.
1120    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
1121        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
1122        included within another project.
1124    24. Steven Van Ingelgem's patches to add more options to the CMake support,
1125        slightly modified by me:
1127          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
1128              not building pcregrep.
1130          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
1131              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
1133    25. Forward references, both numeric and by name, in patterns that made use of
1134        duplicate group numbers, could behave incorrectly or give incorrect errors,
1135        because when scanning forward to find the reference group, PCRE was not
1136        taking into account the duplicate group numbers. A pattern such as
1137        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
1139    26. Changed a few more instances of "const unsigned char *" to USPTR, making
1140        the feature of a custom pointer more persuasive (as requested by a user).
1142    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
1143        pcretest.c, inside #ifndefs, because it seems they are sometimes already
1144        pre-defined.
1146    28. Added support for (*UTF8) at the start of a pattern.
1148    29. Arrange for flags added by the "release type" setting in CMake to be shown
1149        in the configuration summary.
1152    Version 7.8 05-Sep-08
1153    ---------------------
1155    1.  Replaced UCP searching code with optimized version as implemented for Ad
1156        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
1157        stage table and inline lookup instead of a function, giving speed ups of 2
1158        to 5 times on some simple patterns that I tested. Permission was given to
1159        distribute the MultiStage2.py script that generates the tables (it's not in
1160        the tarball, but is in the Subversion repository).
1162    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
1163        scripts.
1165    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
1166        a group with a zero qualifier. The result of the study could be incorrect,
1167        or the function might crash, depending on the pattern.
1169    4.  Caseless matching was not working for non-ASCII characters in back
1170        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
1171        It now works when Unicode Property Support is available.
1173    5.  In pcretest, an escape such as \x{de} in the data was always generating
1174        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
1175        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
1176        truncation.
1178    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
1180    7.  Added two (int) casts to pcregrep when printing the difference of two
1181        pointers, in case they are 64-bit values.
1183    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
1184        test 2 if it fails.
1186    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
1187        and a #define of that name to empty if it is not externally set. This is to
1188        allow users of MSVC to set it if necessary.
1190    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
1191        the convenience functions in the pcre_get.c source file.
1193    11. An option change at the start of a pattern that had top-level alternatives
1194        could cause overwriting and/or a crash. This command provoked a crash in
1195        some environments:
1197          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
1199        This potential security problem was recorded as CVE-2008-2371.
1201    12. For a pattern where the match had to start at the beginning or immediately
1202        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
1203        pcre_dfa_exec() could read past the end of the passed subject if there was
1204        no match. To help with detecting such bugs (e.g. with valgrind), I modified
1205        pcretest so that it places the subject at the end of its malloc-ed buffer.
1207    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
1208        exec() might read past the end of the data buffer in UTF-8 mode.
1210    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
1211        the data contained the byte 0x85 as part of a UTF-8 character within its
1212        first line. This applied both to normal and DFA matching.
1214    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
1215        /^[^d]*?$/8 failed to match "abc".
1217    16. Added a missing copyright notice to pcrecpp_internal.h.
1219    17. Make it more clear in the documentation that values returned from
1220        pcre_exec() in ovector are byte offsets, not character counts.
1222    18. Tidied a few places to stop certain compilers from issuing warnings.
1224    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
1225        supplied by Stefan Weber. I made a further small update for 7.8 because
1226        there is a change of source arrangements: the pcre_searchfuncs.c module is
1227        replaced by pcre_ucd.c.
1230    Version 7.7 07-May-08
1231    ---------------------
1233    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
1234        a string to a long long, pretend we don't even have a long long." This is
1235        done by checking for the strtoq, strtoll, and _strtoi64 functions.
1237    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
1238        pre-7.6 versions, which defined a global no_arg variable instead of putting
1239        it in the RE class. (See also #8 below.)
1241    3.  Remove a line of dead code, identified by coverity and reported by Nuno
1242        Lopes.
1244    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
1246        (1) The include/exclude patterns were being applied to the whole pathnames
1247            of files, instead of just to the final components.
1249        (2) If there was more than one level of directory, the subdirectories were
1250            skipped unless they satisfied the include/exclude conditions. This is
1251            inconsistent with GNU grep (and could even be seen as contrary to the
1252            pcregrep specification - which I improved to make it absolutely clear).
1253            The action now is always to scan all levels of directory, and just
1254            apply the include/exclude patterns to regular files.
1256    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
1257        --exclude_dir in the tests to avoid scanning .svn directories.
1259    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
1260        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
1261        doesn't support NULs in patterns.
1263    7.  Added some missing "const"s to declarations of static tables in
1264        pcre_compile.c and pcre_dfa_exec.c.
1266    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
1267        caused by fix #2  above. (Subsequently also a second patch to fix the
1268        first patch. And a third patch - this was a messy problem.)
1270    9.  Applied Craig's patch to remove the use of push_back().
1272    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
1273        matching function regexec().
1275    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
1276        which, however, unlike Perl's \g{...}, are subroutine calls, not back
1277        references. PCRE supports relative numbers with this syntax (I don't think
1278        Oniguruma does).
1280    12. Previously, a group with a zero repeat such as (...){0} was completely
1281        omitted from the compiled regex. However, this means that if the group
1282        was called as a subroutine from elsewhere in the pattern, things went wrong
1283        (an internal error was given). Such groups are now left in the compiled
1284        pattern, with a new opcode that causes them to be skipped at execution
1285        time.
1287    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
1288        to the way PCRE behaves:
1290        (a) A lone ] character is dis-allowed (Perl treats it as data).
1292        (b) A back reference to an unmatched subpattern matches an empty string
1293            (Perl fails the current match path).
1295        (c) A data ] in a character class must be notated as \] because if the
1296            first data character in a class is ], it defines an empty class. (In
1297            Perl it is not possible to have an empty class.) The empty class []
1298            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
1299            The negative empty class [^] matches any one character, independently
1300            of the DOTALL setting.
1302    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
1303        non-existent subpattern following a character class starting with ']' and
1304        containing () gave an internal compiling error instead of "reference to
1305        non-existent subpattern". Fortunately, when the pattern did exist, the
1306        compiled code was correct. (When scanning forwards to check for the
1307        existencd of the subpattern, it was treating the data ']' as terminating
1308        the class, so got the count wrong. When actually compiling, the reference
1309        was subsequently set up correctly.)
1311    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
1312        it was being rejected as not supported by pcre_dfa_exec(), even though
1313        other assertions are supported. I have made pcre_dfa_exec() support
1314        (*FAIL).
1316    16. The implementation of 13c above involved the invention of a new opcode,
1317        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
1318        cannot be changed at match time, I realized I could make a small
1319        improvement to matching performance by compiling OP_ALLANY instead of
1320        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
1321        on the OP_ANY path.
1323    17. Compiling pcretest on Windows with readline support failed without the
1324        following two fixes: (1) Make the unistd.h include conditional on
1325        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
1327    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
1328        ncurses library to be included for pcretest when ReadLine support is
1329        requested, but also to allow for it to be overridden. This patch came from
1330        Daniel Bergström.
1332    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
1333        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
1334        any errors with the current Unicode tables. Thanks to Peter Kankowski for
1335        spotting this.
1338    Version 7.6 28-Jan-08
1339    ---------------------
1341    1.  A character class containing a very large number of characters with
1342        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
1343        overflow.
1345    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
1346        HAVE_LONG_LONG is not defined.
1348    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
1349        bring them up to date and include new features. This patch includes:
1351        - Fixed PH's badly added libz and libbz2 support.
1352        - Fixed a problem with static linking.
1353        - Added pcredemo. [But later removed - see 7 below.]
1354        - Fixed dftables problem and added an option.
1355        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
1356            HAVE_LONG_LONG.
1357        - Added readline support for pcretest.
1358        - Added an listing of the option settings after cmake has run.
1360    4.  A user submitted a patch to Makefile that makes it easy to create
1361        "pcre.dll" under mingw when using Configure/Make. I added stuff to
1362        Makefile.am that cause it to include this special target, without
1363        affecting anything else. Note that the same mingw target plus all
1364        the other distribution libraries and programs are now supported
1365        when configuring with CMake (see 6 below) instead of with
1366        Configure/Make.
1368    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
1369        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
1370        exported in the Windows port". It has not yet been confirmed that the patch
1371        solves the problem, but it does no harm.
1373    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
1374        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
1375        with CMake, and also correct the comment about stack recursion.
1377    7.  Remove the automatic building of pcredemo from the ./configure system and
1378        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
1379        of a program that users should build themselves after PCRE is installed, so
1380        building it automatically is not really right. What is more, it gave
1381        trouble in some build environments.
1383    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
1386    Version 7.5 10-Jan-08
1387    ---------------------
1389    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
1390        values in parens when parsing an RE using the C++ wrapper."
1392    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
1393        Characters greater than 255 were excluded from the class instead of being
1394        included.
1396    3.  The same bug as (2) above applied to negated POSIX classes such as
1397        [:^space:].
1399    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
1400        defined or documented. It seems to have been a typo for PCRE_STATIC, so
1401        I have changed it.
1403    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
1404        first named subpattern) and a construct such as (?&a) would reference the
1405        first named subpattern whose name started with "a" (in other words, the
1406        length check was missing). Both these problems are fixed. "Subpattern name
1407        expected" is now given for (?&) (a zero-length name), and this patch also
1408        makes it give the same error for \k'' (previously it complained that that
1409        was a reference to a non-existent subpattern).
1411    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
1412        this is right because (?- can be followed by option settings as well as by
1413        digits. I have, however, made the messages clearer.
1415    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
1416        than the number used in the conditional) now cause a compile-time error.
1417        This is actually not compatible with Perl, which accepts such patterns, but
1418        treats the conditional as always being FALSE (as PCRE used to), but it
1419        seems to me that giving a diagnostic is better.
1421    8.  Change "alphameric" to the more common word "alphanumeric" in comments
1422        and messages.
1424    9.  Fix two occurrences of "backslash" in comments that should have been
1425        "backspace".
1427    10. Remove two redundant lines of code that can never be obeyed (their function
1428        was moved elsewhere).
1430    11. The program that makes PCRE's Unicode character property table had a bug
1431        which caused it to generate incorrect table entries for sequences of
1432        characters that have the same character type, but are in different scripts.
1433        It amalgamated them into a single range, with the script of the first of
1434        them. In other words, some characters were in the wrong script. There were
1435        thirteen such cases, affecting characters in the following ranges:
1437          U+002b0 - U+002c1
1438          U+0060c - U+0060d
1439          U+0061e - U+00612
1440          U+0064b - U+0065e
1441          U+0074d - U+0076d
1442          U+01800 - U+01805
1443          U+01d00 - U+01d77
1444          U+01d9b - U+01dbf
1445          U+0200b - U+0200f
1446          U+030fc - U+030fe
1447          U+03260 - U+0327f
1448          U+0fb46 - U+0fbb1
1449          U+10450 - U+1049d
1451    12. The -o option (show only the matching part of a line) for pcregrep was not
1452        compatible with GNU grep in that, if there was more than one match in a
1453        line, it showed only the first of them. It now behaves in the same way as
1454        GNU grep.
1456    13. If the -o and -v options were combined for pcregrep, it printed a blank
1457        line for every non-matching line. GNU grep prints nothing, and pcregrep now
1458        does the same. The return code can be used to tell if there were any
1459        non-matching lines.
1461    14. Added --file-offsets and --line-offsets to pcregrep.
1463    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
1464        infinitely looping recursion. The bug was that positive lookaheads were not
1465        being skipped when checking for a possible empty match (negative lookaheads
1466        and both kinds of lookbehind were skipped).
1468    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
1469        inclusion of <windows.h> to before rather than after the definition of
1470        INVALID_FILE_ATTRIBUTES (patch from David Byron).
1472    17. Specifying a possessive quantifier with a specific limit for a Unicode
1473        character property caused pcre_compile() to compile bad code, which led at
1474        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
1475        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
1476        caused the error; without that there was no problem.
1478    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
1480    19. Added --enable-pcretest-libreadline.
1482    20. In pcrecpp.cc, the variable 'count' was incremented twice in
1483        RE::GlobalReplace(). As a result, the number of replacements returned was
1484        double what it should be. I removed one of the increments, but Craig sent a
1485        later patch that removed the other one (the right fix) and added unit tests
1486        that check the return values (which was not done before).
1488    21. Several CMake things:
1490        (1) Arranged that, when cmake is used on Unix, the libraries end up with
1491            the names libpcre and libpcreposix, not just pcre and pcreposix.
1493        (2) The above change means that pcretest and pcregrep are now correctly
1494            linked with the newly-built libraries, not previously installed ones.
1498    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
1499        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
1500        UTF-8 newline character). The key issue is that the pattern starts .*;
1501        this means that the match must be either at the beginning, or after a
1502        newline. The bug was in the code for advancing after a failed match and
1503        checking that the new position followed a newline. It was not taking
1504        account of UTF-8 characters correctly.
1506    23. PCRE was behaving differently from Perl in the way it recognized POSIX
1507        character classes. PCRE was not treating the sequence [:...:] as a
1508        character class unless the ... were all letters. Perl, however, seems to
1509        allow any characters between [: and :], though of course it rejects as
1510        unknown any "names" that contain non-letters, because all the known class
1511        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
1512        for example, whereas PCRE did not - it did not recognize a POSIX character
1513        class. This seemed a bit dangerous, so the code has been changed to be
1514        closer to Perl. The behaviour is not identical to Perl, because PCRE will
1515        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
1516        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
1517        Perl does, and where it didn't before.
1519    24. Rewrite so as to remove the single use of %n from pcregrep because in some
1520        Windows environments %n is disabled by default.
1523    Version 7.4 21-Sep-07
1524    ---------------------
1526    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
1527        means that a class such as [\s] counted as "explicit reference to CR or
1528        LF". That isn't really right - the whole point of the change was to try to
1529        help when there was an actual mention of one of the two characters. So now
1530        the change happens only if \r or \n (or a literal CR or LF) character is
1531        encountered.
1533    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
1534        of both had grown to the point where there were only 3 bits left.
1535        Fortunately, there was spare space in the data structure, and so I have
1536        moved the internal flags into a new 16-bit field to free up more option
1537        bits.
1539    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
1540        but did not set the internal JCHANGED flag - either of these is enough to
1541        control the way the "get" function works - but the PCRE_INFO_JCHANGED
1542        facility is supposed to tell if (?J) was ever used, so now (?J) at the
1543        start sets both bits.
1545    4.  Added options (at build time, compile time, exec time) to change \R from
1546        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
1548    5.  doc/pcresyntax.html was missing from the distribution.
1550    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
1551        compatibility, even though it is no longer used.
1553    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
1554        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
1555        windows.h file is present (where different names are used). [This was
1556        reversed later after testing - see 16 below.]
1558    8.  Changed all #include <config.h> to #include "config.h". There were also
1559        some further <pcre.h> cases that I changed to "pcre.h".
1561    9.  When pcregrep was used with the --colour option, it missed the line ending
1562        sequence off the lines that it output.
1564    10. It was pointed out to me that arrays of string pointers cause lots of
1565        relocations when a shared library is dynamically loaded. A technique of
1566        using a single long string with a table of offsets can drastically reduce
1567        these. I have refactored PCRE in four places to do this. The result is
1568        dramatic:
1570          Originally:                          290
1571          After changing UCP table:            187
1572          After changing error message table:   43
1573          After changing table of "verbs"       36
1574          After changing table of Posix names   22
1576        Thanks to the folks working on Gregex for glib for this insight.
1578    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
1579        unicode-properties was also set.
1581    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
1583    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
1584        checked only for CRLF.
1586    14. Added casts to pcretest.c to avoid compiler warnings.
1588    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
1590    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
1591        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
1592        entirely. This removes changes made in 7 above.
1594    17. The CMake files have been updated, and there is now more information about
1595        building with CMake in the NON-UNIX-USE document.
1598    Version 7.3 28-Aug-07
1599  ---------------------  ---------------------
1601   1. In the rejigging of the build system that eventually resulted in 7.1, the   1. In the rejigging of the build system that eventually resulted in 7.1, the
# Line 64  Version 7.3 05-Jul-07 Line 1658  Version 7.3 05-Jul-07
1658      dynamic way, which I have now done. The artificial limitation on group      dynamic way, which I have now done. The artificial limitation on group
1659      length has been removed - we now have only the limit on the total length of      length has been removed - we now have only the limit on the total length of
1660      the compiled pattern, which depends on the LINK_SIZE setting.      the compiled pattern, which depends on the LINK_SIZE setting.
1662  10. Fixed a bug in the documentation for get/copy named substring when  10. Fixed a bug in the documentation for get/copy named substring when
1663      duplicate names are permitted. If none of the named substrings are set, the      duplicate names are permitted. If none of the named substrings are set, the
1664      functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an      functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1665      empty string.      empty string.
1667    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1668        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1669        because the ] is interpreted as the first data character and the
1670        terminating ] is not found. PCRE has been made compatible with Perl in this
1671        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1672        cause memory overwriting.
1674    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1675        string has been matched (to stop an infinite loop). It was not recognizing
1676        a conditional subpattern that could match an empty string if that
1677        subpattern was within another subpattern. For example, it looped when
1678        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
1679        condition was not nested. This bug has been fixed.
1681    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1682        past the start of the subject in the presence of bytes with the top bit
1683        set, for example "\x8aBCD".
1685    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1686        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1688    14. Optimized (?!) to (*FAIL).
1690    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1691        This restricts code points to be within the range 0 to 0x10FFFF, excluding
1692        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1693        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1694        does: it's just the validity check that is more restrictive.
1696    16. Inserted checks for integer overflows during escape sequence (backslash)
1697        processing, and also fixed erroneous offset values for syntax errors during
1698        backslash processing.
1700    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1701        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1703    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1704        caused an overrun.
1706    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1707        something other than just ASCII characters) inside a group that had an
1708        unlimited repeat caused a loop at compile time (while checking to see
1709        whether the group could match an empty string).
1711    20. Debugging a pattern containing \p or \P could cause a crash. For example,
1712        [\P{Any}] did so. (Error in the code for printing property names.)
1714    21. An orphan \E inside a character class could cause a crash.
1716    22. A repeated capturing bracket such as (A)? could cause a wild memory
1717        reference during compilation.
1719    23. There are several functions in pcre_compile() that scan along a compiled
1720        expression for various reasons (e.g. to see if it's fixed length for look
1721        behind). There were bugs in these functions when a repeated \p or \P was
1722        present in the pattern. These operators have additional parameters compared
1723        with \d, etc, and these were not being taken into account when moving along
1724        the compiled data. Specifically:
1726        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1727            length.
1729        (b) An item such as \pL+ within a repeated group could cause crashes or
1730            loops.
1732        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1733            "reference to non-existent subpattern" error.
1735        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1737    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1738        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1740    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1742    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1743        character were causing crashes (broken optimization).
1745    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1746        \p or \P) caused a compile-time loop.
1748    28. More problems have arisen in unanchored patterns when CRLF is a valid line
1749        break. For example, the unstudied pattern [\r\n]A does not match the string
1750        "\r\nA" because change 7.0/46 below moves the current point on by two
1751        characters after failing to match at the start. However, the pattern \nA
1752        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
1753        the same is true. There doesn't seem any very clean way out of this, but
1754        what I have chosen to do makes the common cases work: PCRE now takes note
1755        of whether there can be an explicit match for \r or \n anywhere in the
1756        pattern, and if so, 7.0/46 no longer applies. As part of this change,
1757        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1758        pattern has explicit CR or LF references.
1760    29. Added (*CR) etc for changing newline setting at start of pattern.
1763  Version 7.2 19-Jun-07  Version 7.2 19-Jun-07

Removed from v.203  
changed lines
  Added in v.752

  ViewVC Help
Powered by ViewVC 1.1.5