/[pcre]/code/trunk/ChangeLog
ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 308 by ph10, Mon Jan 21 14:57:19 2008 UTC revision 787 by ph10, Tue Dec 6 15:37:24 2011 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4  Version 7.6 19-Jan-08  Version 8.21 05-Dec-2011
5    ------------------------
6    
7    1.  Updating the JIT compiler.
8    
9    2.  JIT compiler now supports OP_NCREF, OP_RREF and OP_NRREF. New test cases
10        are added as well.
11    
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.
15    
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.
19    
20    5.  Supporting of \x, \U and \u in JavaScript compatibility mode based on the
21        ECMA-262 standard.
22    
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.
26    
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.
36    
37    8.  A possessively repeated conditional subpattern such as (?(?=c)c|d)++ was
38        being incorrectly compiled and would have given unpredicatble results.
39    
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.
44    
45    10. Add a cast and remove a redundant test from the code.
46    
47    11. JIT should use pcre_malloc/pcre_free for allocation.
48    
49    12. Updated pcre-config so that it no longer shows -L/usr/lib, which seems
50        best practice nowadays, and helps with cross-compiling. (If the exec_prefix
51        is anything other than /usr, -L is still shown).
52    
53    13. In non-UTF-8 mode, \C is now supported in lookbehinds and DFA matching.
54    
55    14. Perl does not support \N without a following name in a [] class; PCRE now
56        also gives an error.
57    
58    15. If a forward reference was repeated with an upper limit of around 2000,
59        it caused the error "internal error: overran compiling workspace". The
60        maximum number of forward references (including repeats) was limited by the
61        internal workspace, and dependent on the LINK_SIZE. The code has been
62        rewritten so that the workspace expands (via pcre_malloc) if necessary, and
63        the default depends on LINK_SIZE. There is a new upper limit (for safety)
64        of around 200,000 forward references. While doing this, I also speeded up
65        the filling in of repeated forward references.
66    
67    16. A repeated forward reference in a pattern such as (a)(?2){2}(.) was
68        incorrectly expecting the subject to contain another "a" after the start.
69    
70    17. When (*SKIP:name) is activated without a corresponding (*MARK:name) earlier
71        in the match, the SKIP should be ignored. This was not happening; instead
72        the SKIP was being treated as NOMATCH. For patterns such as
73        /A(*MARK:A)A+(*SKIP:B)Z|AAC/ this meant that the AAC branch was never
74        tested.
75    
76    18. The behaviour of (*MARK), (*PRUNE), and (*THEN) has been reworked and is
77        now much more compatible with Perl, in particular in cases where the result
78        is a non-match for a non-anchored pattern. For example, if
79        /b(*:m)f|a(*:n)w/ is matched against "abc", the non-match returns the name
80        "m", where previously it did not return a name. A side effect of this
81        change is that for partial matches, the last encountered mark name is
82        returned, as for non matches. A number of tests that were previously not
83        Perl-compatible have been moved into the Perl-compatible test files. The
84        refactoring has had the pleasing side effect of removing one argument from
85        the match() function, thus reducing its stack requirements.
86    
87    19. If the /S+ option was used in pcretest to study a pattern using JIT,
88        subsequent uses of /S (without +) incorrectly behaved like /S+.
89    
90    21. Retrieve executable code size support for the JIT compiler and fixing
91        some warnings.
92    
93    22. A caseless match of a UTF-8 character whose other case uses fewer bytes did
94        not work when the shorter character appeared right at the end of the
95        subject string.
96    
97    23. Added some (int) casts to non-JIT modules to reduce warnings on 64-bit
98        systems.
99    
100    24. Added PCRE_INFO_JITSIZE to pass on the value from (21) above, and also
101        output it when the /M option is used in pcretest.
102    
103    25. The CheckMan script was not being included in the distribution. Also, added
104        an explicit "perl" to run Perl scripts from the PrepareRelease script
105        because this is reportedly needed in Windows.
106    
107    26. If study data was being save in a file and studying had not found a set of
108        "starts with" bytes for the pattern, the data written to the file (though
109        never used) was taken from uninitialized memory and so caused valgrind to
110        complain.
111    
112    
113    Version 8.20 21-Oct-2011
114    ------------------------
115    
116    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
117        a POSIX class. After further experiments with Perl, which convinced me that
118        Perl has bugs and confusions, a closing square bracket is no longer allowed
119        in a POSIX name. This bug also affected patterns with classes that started
120        with full stops.
121    
122    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
123        captured substring, but while checking the failing first alternative,
124        substring 1 is temporarily captured. If the output vector supplied to
125        pcre_exec() was not big enough for this capture, the yield of the function
126        was still zero ("insufficient space for captured substrings"). This cannot
127        be totally fixed without adding another stack variable, which seems a lot
128        of expense for a edge case. However, I have improved the situation in cases
129        such as /(a)(b)x|abc/ matched against "abc", where the return code
130        indicates that fewer than the maximum number of slots in the ovector have
131        been set.
132    
133    3.  Related to (2) above: when there are more back references in a pattern than
134        slots in the output vector, pcre_exec() uses temporary memory during
135        matching, and copies in the captures as far as possible afterwards. It was
136        using the entire output vector, but this conflicts with the specification
137        that only 2/3 is used for passing back captured substrings. Now it uses
138        only the first 2/3, for compatibility. This is, of course, another edge
139        case.
140    
141    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
142        main code base, and can be used by building with --enable-jit. When this is
143        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
144        runtime --no-jit option is given.
145    
146    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
147        ovector, the return from the function was zero, implying that there were
148        other matches that did not fit. The correct "exactly full" value is now
149        returned.
150    
151    6.  If a subpattern that was called recursively or as a subroutine contained
152        (*PRUNE) or any other control that caused it to give a non-standard return,
153        invalid errors such as "Error -26 (nested recursion at the same subject
154        position)" or even infinite loops could occur.
155    
156    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
157        computing the minimum length on reaching *ACCEPT, and so ended up with the
158        wrong value of 1 rather than 0. Further investigation indicates that
159        computing a minimum subject length in the presence of *ACCEPT is difficult
160        (think back references, subroutine calls), and so I have changed the code
161        so that no minimum is registered for a pattern that contains *ACCEPT.
162    
163    8.  If (*THEN) was present in the first (true) branch of a conditional group,
164        it was not handled as intended. [But see 16 below.]
165    
166    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
167        Sheri Pierce.
168    
169    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
170        the first byte in a match must be "a".
171    
172    11. Change 17 for 8.13 increased the recursion depth for patterns like
173        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
174        pattern contains any instances of (*THEN). If it does not, the old
175        optimizations are restored. It would be nice to do this on a per-group
176        basis, but at the moment that is not feasible.
177    
178    12. In some environments, the output of pcretest -C is CRLF terminated. This
179        broke RunTest's code that checks for the link size. A single white space
180        character after the value is now allowed for.
181    
182    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
183        For "fr", it uses the Windows-specific input and output files.
184    
185    14. If (*THEN) appeared in a group that was called recursively or as a
186        subroutine, it did not work as intended. [But see next item.]
187    
188    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
189        pattern fragments (but not containing any | characters). If A and B are
190        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
191        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
192        D. In other words, Perl considers parentheses that do not contain any |
193        characters to be part of a surrounding alternative, whereas PCRE was
194        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
195        differently. PCRE now behaves in the same way as Perl, except in the case
196        of subroutine/recursion calls such as (?1) which have in any case always
197        been different (but PCRE had them first :-).
198    
199    16. Related to 15 above: Perl does not treat the | in a conditional group as
200        creating alternatives. Such a group is treated in the same way as an
201        ordinary group without any | characters when processing (*THEN). PCRE has
202        been changed to match Perl's behaviour.
203    
204    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
205        RunGrepTest script failed.
206    
207    18. Change 22 for version 13 caused atomic groups to use more stack. This is
208        inevitable for groups that contain captures, but it can lead to a lot of
209        stack use in large patterns. The old behaviour has been restored for atomic
210        groups that do not contain any capturing parentheses.
211    
212    19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not
213        suppress the check for a minimum subject length at run time. (If it was
214        given to pcre_exec() or pcre_dfa_exec() it did work.)
215    
216    20. Fixed an ASCII-dependent infelicity in pcretest that would have made it
217        fail to work when decoding hex characters in data strings in EBCDIC
218        environments.
219    
220    21. It appears that in at least one Mac OS environment, the isxdigit() function
221        is implemented as a macro that evaluates to its argument more than once,
222        contravening the C 90 Standard (I haven't checked a later standard). There
223        was an instance in pcretest which caused it to go wrong when processing
224        \x{...} escapes in subject strings. The has been rewritten to avoid using
225        things like p++ in the argument of isxdigit().
226    
227    
228    Version 8.13 16-Aug-2011
229    ------------------------
230    
231    1.  The Unicode data tables have been updated to Unicode 6.0.0.
232    
233    2.  Two minor typos in pcre_internal.h have been fixed.
234    
235    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
236        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
237        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
238    
239    4.  There were a number of related bugs in the code for matching backrefences
240        caselessly in UTF-8 mode when codes for the characters concerned were
241        different numbers of bytes. For example, U+023A and U+2C65 are an upper
242        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
243        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
244        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
245        2-byte code at the end of the subject (it thought there wasn't enough data
246        left).
247    
248    5.  Comprehensive information about what went wrong is now returned by
249        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
250        as the output vector has at least 2 elements. The offset of the start of
251        the failing character and a reason code are placed in the vector.
252    
253    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
254        now returned is for the first byte of the failing character, instead of the
255        last byte inspected. This is an incompatible change, but I hope it is small
256        enough not to be a problem. It makes the returned offset consistent with
257        pcre_exec() and pcre_dfa_exec().
258    
259    7.  pcretest now gives a text phrase as well as the error number when
260        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
261        failure, the offset and reason code are output.
262    
263    8.  When \R was used with a maximizing quantifier it failed to skip backwards
264        over a \r\n pair if the subsequent match failed. Instead, it just skipped
265        back over a single character (\n). This seems wrong (because it treated the
266        two characters as a single entity when going forwards), conflicts with the
267        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
268        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
269        has been changed.
270    
271    9.  Some internal refactoring has changed the processing so that the handling
272        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
273        time (the PCRE_DOTALL option was changed this way some time ago: version
274        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
275        which was always a bit of a fudge. It also means that there is one less
276        argument for the match() function, which reduces its stack requirements
277        slightly. This change also fixes an incompatibility with Perl: the pattern
278        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
279    
280    10. More internal refactoring has drastically reduced the number of recursive
281        calls to match() for possessively repeated groups such as (abc)++ when
282        using pcre_exec().
283    
284    11. While implementing 10, a number of bugs in the handling of groups were
285        discovered and fixed:
286    
287        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
288        (a|)*(?1) gave a compile-time internal error.
289        ((a|)+)+  did not notice that the outer group could match an empty string.
290        (^a|^)+   was not marked as anchored.
291        (.*a|.*)+ was not marked as matching at start or after a newline.
292    
293    12. Yet more internal refactoring has removed another argument from the match()
294        function. Special calls to this function are now indicated by setting a
295        value in a variable in the "match data" data block.
296    
297    13. Be more explicit in pcre_study() instead of relying on "default" for
298        opcodes that mean there is no starting character; this means that when new
299        ones are added and accidentally left out of pcre_study(), testing should
300        pick them up.
301    
302    14. The -s option of pcretest has been documented for ages as being an old
303        synonym of -m (show memory usage). I have changed it to mean "force study
304        for every regex", that is, assume /S for every regex. This is similar to -i
305        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
306        using it. It makes it easier to run collections of tests with and without
307        study enabled, and thereby test pcre_study() more easily. All the standard
308        tests are now run with and without -s (but some patterns can be marked as
309        "never study" - see 20 below).
310    
311    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
312        restoration of the capturing data to the outer values was not happening
313        correctly.
314    
315    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
316        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
317        pattern had matched an empty string, and so incorrectly returned a no
318        match.
319    
320    17. There was optimizing code for the last branch of non-capturing parentheses,
321        and also for the obeyed branch of a conditional subexpression, which used
322        tail recursion to cut down on stack usage. Unfortunately, now that there is
323        the possibility of (*THEN) occurring in these branches, tail recursion is
324        no longer possible because the return has to be checked for (*THEN). These
325        two optimizations have therefore been removed. [But see 8.20/11 above.]
326    
327    18. If a pattern containing \R was studied, it was assumed that \R always
328        matched two bytes, thus causing the minimum subject length to be
329        incorrectly computed because \R can also match just one byte.
330    
331    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
332        was incorrectly computed.
333    
334    20. If /S is present twice on a test pattern in pcretest input, it now
335        *disables* studying, thereby overriding the use of -s on the command line
336        (see 14 above). This is necessary for one or two tests to keep the output
337        identical in both cases.
338    
339    21. When (*ACCEPT) was used in an assertion that matched an empty string and
340        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
341    
342    22. When an atomic group that contained a capturing parenthesis was
343        successfully matched, but the branch in which it appeared failed, the
344        capturing was not being forgotten if a higher numbered group was later
345        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
346        group 1 to "a", when in fact it should be unset. This applied to multi-
347        branched capturing and non-capturing groups, repeated or not, and also to
348        positive assertions (capturing in negative assertions does not happen
349        in PCRE) and also to nested atomic groups.
350    
351    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
352        subject after a captured substring, to make it easier to tell which of a
353        number of identical substrings has been captured.
354    
355    24. The way atomic groups are processed by pcre_exec() has been changed so that
356        if they are repeated, backtracking one repetition now resets captured
357        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
358        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
359        "aaa". Previously, it would have been "a". As part of this code
360        refactoring, the way recursive calls are handled has also been changed.
361    
362    25. If an assertion condition captured any substrings, they were not passed
363        back unless some other capturing happened later. For example, if
364        (?(?=(a))a) was matched against "a", no capturing was returned.
365    
366    26. When studying a pattern that contained subroutine calls or assertions,
367        the code for finding the minimum length of a possible match was handling
368        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
369        group 1 called group 2 while simultaneously a separate group 2 called group
370        1). A stack overflow occurred in this case. I have fixed this by limiting
371        the recursion depth to 10.
372    
373    27. Updated RunTest.bat in the distribution to the version supplied by Tom
374        Fortmann. This supports explicit test numbers on the command line, and has
375        argument validation and error reporting.
376    
377    28. An instance of \X with an unlimited repeat could fail if at any point the
378        first character it looked at was a mark character.
379    
380    29. Some minor code refactoring concerning Unicode properties and scripts
381        should reduce the stack requirement of match() slightly.
382    
383    30. Added the '=' option to pcretest to check the setting of unused capturing
384        slots at the end of the pattern, which are documented as being -1, but are
385        not included in the return count.
386    
387    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
388        compiled something random. Now it gives a compile-time error (as does
389        Perl).
390    
391    32. A *MARK encountered during the processing of a positive assertion is now
392        recorded and passed back (compatible with Perl).
393    
394    33. If --only-matching or --colour was set on a pcregrep call whose pattern
395        had alternative anchored branches, the search for a second match in a line
396        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
397        matched the line "0102" twice. The same bug affected patterns that started
398        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
399        twice.
400    
401    34. Previously, PCRE did not allow quantification of assertions. However, Perl
402        does, and because of capturing effects, quantifying parenthesized
403        assertions may at times be useful. Quantifiers are now allowed for
404        parenthesized assertions.
405    
406    35. A minor code tidy in pcre_compile() when checking options for \R usage.
407    
408    36. \g was being checked for fancy things in a character class, when it should
409        just be a literal "g".
410    
411    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
412        appearance of a nested POSIX class supersedes an apparent external class.
413        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
414        unescaped square brackets may also appear as part of class names. For
415        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
416        more like Perl. (But see 8.20/1 above.)
417    
418    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
419        was because it thought it was \N{name}, which is not supported).
420    
421    39. Add minix to OS list not supporting the -S option in pcretest.
422    
423    40. PCRE tries to detect cases of infinite recursion at compile time, but it
424        cannot analyze patterns in sufficient detail to catch mutual recursions
425        such as ((?1))((?2)). There is now a runtime test that gives an error if a
426        subgroup is called recursively as a subpattern for a second time at the
427        same position in the subject string. In previous releases this might have
428        been caught by the recursion limit, or it might have run out of stack.
429    
430    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
431        happen only once. PCRE was, however incorrectly giving a compile time error
432        "recursive call could loop indefinitely" because it cannot analyze the
433        pattern in sufficient detail. The compile time test no longer happens when
434        PCRE is compiling a conditional subpattern, but actual runaway loops are
435        now caught at runtime (see 40 above).
436    
437    42. It seems that Perl allows any characters other than a closing parenthesis
438        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
439        has been changed to be the same.
440    
441    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
442        as not to get warnings when autogen.sh is called. Also changed
443        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
444    
445    44. To help people who use pcregrep to scan files containing exceedingly long
446        lines, the following changes have been made:
447    
448        (a) The default value of the buffer size parameter has been increased from
449            8K to 20K. (The actual buffer used is three times this size.)
450    
451        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
452            PCRE is built.
453    
454        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
455            to be set at run time.
456    
457        (d) Numerical values in pcregrep options can be followed by K or M, for
458            example --buffer-size=50K.
459    
460        (e) If a line being scanned overflows pcregrep's buffer, an error is now
461            given and the return code is set to 2.
462    
463    45. Add a pointer to the latest mark to the callout data block.
464    
465    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
466        partial match of an empty string instead of no match. This was specific to
467        the use of ".".
468    
469    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
470        complete match instead of a partial match. This bug was dependent on both
471        the PCRE_UTF8 and PCRE_DOTALL options being set.
472    
473    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
474        starting byte set, because \b was not being ignored.
475    
476    
477    Version 8.12 15-Jan-2011
478    ------------------------
479    
480    1.  Fixed some typos in the markup of the man pages, and wrote a script that
481        checks for such things as part of the documentation building process.
482    
483    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
484        --match-limit and --recursion-limit options (added for 8.11). In
485        particular, this made one of the standard tests fail. (The integer value
486        went into the wrong half of a long int.)
487    
488    3.  If the --colour option was given to pcregrep with -v (invert match), it
489        did strange things, either producing crazy output, or crashing. It should,
490        of course, ignore a request for colour when reporting lines that do not
491        match.
492    
493    4.  Another pcregrep bug caused similar problems if --colour was specified with
494        -M (multiline) and the pattern match finished with a line ending.
495    
496    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
497        matched in multiline mode, the following line was shown as part of the
498        match. This seems wrong, so I have changed it.
499    
500    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
501        the check for further matches in the same line (so they could be coloured)
502        to overrun the end of the current line. If another match was found, it was
503        incorrectly shown (and then shown again when found in the next line).
504    
505    7.  If pcregrep was compiled under Windows, there was a reference to the
506        function pcregrep_exit() before it was defined. I am assuming this was
507        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
508        reported by a user. I've moved the definition above the reference.
509    
510    
511    Version 8.11 10-Dec-2010
512    ------------------------
513    
514    1.  (*THEN) was not working properly if there were untried alternatives prior
515        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
516        backtracked to try for "b" instead of moving to the next alternative branch
517        at the same level (in this case, to look for "c"). The Perl documentation
518        is clear that when (*THEN) is backtracked onto, it goes to the "next
519        alternative in the innermost enclosing group".
520    
521    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
522        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
523        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
524        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
525        (*THEN).
526    
527    3.  If \s appeared in a character class, it removed the VT character from
528        the class, even if it had been included by some previous item, for example
529        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
530        of \s, but is part of the POSIX "space" class.)
531    
532    4.  A partial match never returns an empty string (because you can always
533        match an empty string at the end of the subject); however the checking for
534        an empty string was starting at the "start of match" point. This has been
535        changed to the "earliest inspected character" point, because the returned
536        data for a partial match starts at this character. This means that, for
537        example, /(?<=abc)def/ gives a partial match for the subject "abc"
538        (previously it gave "no match").
539    
540    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
541        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
542        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
543        has an implication that the given string is incomplete (because a partial
544        match is preferred over a full match). For this reason, these items now
545        give a partial match in this situation. [Aside: previously, the one case
546        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
547        match rather than a full match, which was wrong by the old rules, but is
548        now correct.]
549    
550    6.  There was a bug in the handling of #-introduced comments, recognized when
551        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
552        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
553        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
554        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
555        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
556        places in pcre_compile().
557    
558    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
559        comments when looking ahead for named forward references to subpatterns,
560        the only newline sequence it recognized was NL. It now handles newlines
561        according to the set newline convention.
562    
563    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
564        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
565        cater for a lack of strerror(). These oversights have been fixed.
566    
567    9.  Added --match-limit and --recursion-limit to pcregrep.
568    
569    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
570    
571    11. When the -o option was used, pcregrep was setting a return code of 1, even
572        when matches were found, and --line-buffered was not being honoured.
573    
574    12. Added an optional parentheses number to the -o and --only-matching options
575        of pcregrep.
576    
577    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
578        can match an empty string. The code to do it in pcretest and pcredemo
579        needed fixing:
580    
581        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
582            only one byte after an empty string match just before CRLF (this case
583            just got forgotten; "any" and "anycrlf" were OK).
584    
585        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
586            mode when an empty string match preceded an ASCII character followed by
587            a non-ASCII character. (The code for advancing by one character rather
588            than one byte was nonsense.)
589    
590        (c) The pcredemo.c sample program did not have any code at all to handle
591            the cases when CRLF is a valid newline sequence.
592    
593    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
594        as a starting offset was within the subject string. There is now a new
595        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
596        negative or greater than the length of the string. In order to test this,
597        pcretest is extended to allow the setting of negative starting offsets.
598    
599    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
600        starting offset points to the beginning of a UTF-8 character was
601        unnecessarily clumsy. I tidied it up.
602    
603    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
604        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
605    
606    17. Nobody had reported that the --include_dir option, which was added in
607        release 7.7 should have been called --include-dir (hyphen, not underscore)
608        for compatibility with GNU grep. I have changed it to --include-dir, but
609        left --include_dir as an undocumented synonym, and the same for
610        --exclude-dir, though that is not available in GNU grep, at least as of
611        release 2.5.4.
612    
613    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
614        characters from a string of bytes) have been redefined so as not to use
615        loops, in order to improve performance in some environments. At the same
616        time, I abstracted some of the common code into auxiliary macros to save
617        repetition (this should not affect the compiled code).
618    
619    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
620        compile-time error is now given if \c is not followed by an ASCII
621        character, that is, a byte less than 128. (In EBCDIC mode, the code is
622        different, and any byte value is allowed.)
623    
624    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
625        START_OPTIMIZE option, which is now allowed at compile time - but just
626        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
627        to pcregrep and other applications that have no direct access to PCRE
628        options. The new /Y option in pcretest sets this option when calling
629        pcre_compile().
630    
631    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
632        back references. Groups containing recursive back references were forced to
633        be atomic by that change, but in the case of named groups, the amount of
634        memory required was incorrectly computed, leading to "Failed: internal
635        error: code overflow". This has been fixed.
636    
637    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
638        pcretest.c, to avoid build problems in some Borland environments.
639    
640    
641    Version 8.10 25-Jun-2010
642    ------------------------
643    
644    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
645        THEN.
646    
647    2.  (*ACCEPT) was not working when inside an atomic group.
648    
649    3.  Inside a character class, \B is treated as a literal by default, but
650        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
651        causes the error). The code is unchanged, but I tidied the documentation.
652    
653    4.  Inside a character class, PCRE always treated \R and \X as literals,
654        whereas Perl faults them if its -w option is set. I have changed PCRE so
655        that it faults them when PCRE_EXTRA is set.
656    
657    5.  Added support for \N, which always matches any character other than
658        newline. (It is the same as "." when PCRE_DOTALL is not set.)
659    
660    6.  When compiling pcregrep with newer versions of gcc which may have
661        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
662        declared with attribute warn_unused_result" were given. Just casting the
663        result to (void) does not stop the warnings; a more elaborate fudge is
664        needed. I've used a macro to implement this.
665    
666    7.  Minor change to pcretest.c to avoid a compiler warning.
667    
668    8.  Added four artifical Unicode properties to help with an option to make
669        \s etc use properties (see next item). The new properties are: Xan
670        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
671    
672    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
673        use Unicode properties. (*UCP) at the start of a pattern can be used to set
674        this option. Modified pcretest to add /W to test this facility. Added
675        REG_UCP to make it available via the POSIX interface.
676    
677    10. Added --line-buffered to pcregrep.
678    
679    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
680        studied, and the match started with a letter with a code point greater than
681        127 whose first byte was different to the first byte of the other case of
682        the letter, the other case of this starting letter was not recognized
683        (#976).
684    
685    12. If a pattern that was studied started with a repeated Unicode property
686        test, for example, \p{Nd}+, there was the theoretical possibility of
687        setting up an incorrect bitmap of starting bytes, but fortunately it could
688        not have actually happened in practice until change 8 above was made (it
689        added property types that matched character-matching opcodes).
690    
691    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
692        possible starting bytes for non-anchored patterns.
693    
694    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
695        \R, and also a number of cases that involve Unicode properties, both
696        explicit and implicit when PCRE_UCP is set.
697    
698    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
699        input, it could crash or give wrong results if characters with values
700        greater than 0xc0 were present in the subject string. (Detail: it assumed
701        UTF-8 input when processing these items.)
702    
703    16. Added a lot of (int) casts to avoid compiler warnings in systems where
704        size_t is 64-bit (#991).
705    
706    17. Added a check for running out of memory when PCRE is compiled with
707        --disable-stack-for-recursion (#990).
708    
709    18. If the last data line in a file for pcretest does not have a newline on
710        the end, a newline was missing in the output.
711    
712    19. The default pcre_chartables.c file recognizes only ASCII characters (values
713        less than 128) in its various bitmaps. However, there is a facility for
714        generating tables according to the current locale when PCRE is compiled. It
715        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
716        characters, are recognized by isspace() and therefore were getting set in
717        these tables, and indeed these tables seem to approximate to ISO 8859. This
718        caused a problem in UTF-8 mode when pcre_study() was used to create a list
719        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
720        which of course cannot start UTF-8 characters. I have changed the code so
721        that only real ASCII characters (less than 128) and the correct starting
722        bytes for UTF-8 encodings are set for characters greater than 127 when in
723        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
724        altogether.)
725    
726    20. Added the /T option to pcretest so as to be able to run tests with non-
727        standard character tables, thus making it possible to include the tests
728        used for 19 above in the standard set of tests.
729    
730    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
731        reference to a subpattern the other side of a comment that contains an
732        opening parenthesis caused either an internal compiling error, or a
733        reference to the wrong subpattern.
734    
735    
736    Version 8.02 19-Mar-2010
737    ------------------------
738    
739    1.  The Unicode data tables have been updated to Unicode 5.2.0.
740    
741    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
742        configured.
743    
744    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
745        original author of that file, following a query about its status.
746    
747    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
748        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
749    
750    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
751        quantifier applied to a forward-referencing subroutine call, could compile
752        incorrect code or give the error "internal error: previously-checked
753        referenced subpattern not found".
754    
755    6.  Both MS Visual Studio and Symbian OS have problems with initializing
756        variables to point to external functions. For these systems, therefore,
757        pcre_malloc etc. are now initialized to local functions that call the
758        relevant global functions.
759    
760    7.  There were two entries missing in the vectors called coptable and poptable
761        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
762        I've fixed the data, and added a kludgy way of testing at compile time that
763        the lengths are correct (equal to the number of opcodes).
764    
765    8.  Following on from 7, I added a similar kludge to check the length of the
766        eint vector in pcreposix.c.
767    
768    9.  Error texts for pcre_compile() are held as one long string to avoid too
769        much relocation at load time. To find a text, the string is searched,
770        counting zeros. There was no check for running off the end of the string,
771        which could happen if a new error number was added without updating the
772        string.
773    
774    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
775    
776    11. \K was not working if it appeared in an atomic group or in a group that
777        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
778        \K is "not well defined" if used in an assertion. PCRE now accepts it if
779        the assertion is positive, but not if it is negative.
780    
781    12. Change 11 fortuitously reduced the size of the stack frame used in the
782        "match()" function of pcre_exec.c by one pointer. Forthcoming
783        implementation of support for (*MARK) will need an extra pointer on the
784        stack; I have reserved it now, so that the stack frame size does not
785        decrease.
786    
787    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
788        item in branch that calls a recursion is a subroutine call - as in the
789        second branch in the above example - was incorrectly given the compile-
790        time error "recursive call could loop indefinitely" because pcre_compile()
791        was not correctly checking the subroutine for matching a non-empty string.
792    
793    14. The checks for overrunning compiling workspace could trigger after an
794        overrun had occurred. This is a "should never occur" error, but it can be
795        triggered by pathological patterns such as hundreds of nested parentheses.
796        The checks now trigger 100 bytes before the end of the workspace.
797    
798    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
799    
800    
801    Version 8.01 19-Jan-2010
802    ------------------------
803    
804    1.  If a pattern contained a conditional subpattern with only one branch (in
805        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
806        computed the wrong minimum data length (which is of course zero for such
807        subpatterns). This could cause incorrect "no match" results.
808    
809    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
810        the pattern is reset in the first branch, pcre_compile() failed with
811        "internal error: code overflow at offset...". This happened only when
812        the reset was to the original external option setting. (An optimization
813        abstracts leading options settings into an external setting, which was the
814        cause of this.)
815    
816    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
817        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
818        assertion pattern did not match (meaning that the assertion was true), it
819        was incorrectly treated as false if the SKIP had been reached during the
820        matching. This also applied to assertions used as conditions.
821    
822    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
823        assertion subpattern, including such a pattern used as a condition,
824        unpredictable results occurred, instead of the error return
825        PCRE_ERROR_DFA_UITEM.
826    
827    5.  The C++ GlobalReplace function was not working like Perl for the special
828        situation when an empty string is matched. It now does the fancy magic
829        stuff that is necessary.
830    
831    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
832        removed. (These were left over from very, very early versions of PCRE.)
833    
834    7.  Some cosmetic changes to the code to make life easier when compiling it
835        as part of something else:
836    
837        (a) Change DEBUG to PCRE_DEBUG.
838    
839        (b) In pcre_compile(), rename the member of the "branch_chain" structure
840            called "current" as "current_branch", to prevent a collision with the
841            Linux macro when compiled as a kernel module.
842    
843        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
844            prevent a collision with the Linux macro when compiled as a kernel
845            module.
846    
847    8.  In pcre_compile() there are some checks for integer overflows that used to
848        cast potentially large values to (double). This has been changed to that
849        when building, a check for int64_t is made, and if it is found, it is used
850        instead, thus avoiding the use of floating point arithmetic. (There is no
851        other use of FP in PCRE.) If int64_t is not found, the fallback is to
852        double.
853    
854    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
855        2005 (difference between two addresses compared to an unsigned value).
856    
857    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
858        custom one, because of the following reported problem in Windows:
859    
860          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
861              under Win32.
862          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
863              therefore missing the function definition.
864          - The compiler thus generates a "C" signature for the test function.
865          - The linker fails to find the "C" function.
866          - PCRE fails to configure if asked to do so against libbz2.
867    
868    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
869        messages were output:
870    
871          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
872          rerunning libtoolize, to keep the correct libtool macros in-tree.
873          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
874    
875        I have done both of these things.
876    
877    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
878        most of the time, it *can* run out if it is given a pattern that contains a
879        runaway infinite recursion. I updated the discussion in the pcrestack man
880        page.
881    
882    13. Now that we have gone to the x.xx style of version numbers, the minor
883        version may start with zero. Using 08 or 09 is a bad idea because users
884        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
885        interpreted as invalid octal numbers. I've updated the previous comment in
886        configure.ac, and also added a check that gives an error if 08 or 09 are
887        used.
888    
889    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
890        causing partial matching to fail when the end of the subject matched \W
891        in a UTF-8 pattern where \W was quantified with a minimum of 3.
892    
893    15. There were some discrepancies between the declarations in pcre_internal.h
894        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
895        their definitions. The declarations used "const uschar *" and the
896        definitions used USPTR. Even though USPTR is normally defined as "const
897        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
898        reported that: "This difference in casting confuses some C++ compilers, for
899        example, SunCC recognizes above declarations as different functions and
900        generates broken code for hbpcre." I have changed the declarations to use
901        USPTR.
902    
903    16. GNU libtool is named differently on some systems. The autogen.sh script now
904        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
905        (FreeBSD).
906    
907    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
908        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
909        comment: "Figure out how to create a longlong from a string: strtoll and
910        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
911        instance, but it only takes 2 args instead of 3!"
912    
913    18. A subtle bug concerned with back references has been fixed by a change of
914        specification, with a corresponding code fix. A pattern such as
915        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
916        refers, was giving matches when it shouldn't. For example, xa=xaaa would
917        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
918        same bug. Such groups have to be quantified to be useful, or contained
919        inside another quantified group. (If there's no repetition, the reference
920        can never match.) The problem arises because, having left the group and
921        moved on to the rest of the pattern, a later failure that backtracks into
922        the group uses the captured value from the final iteration of the group
923        rather than the correct earlier one. I have fixed this in PCRE by forcing
924        any group that contains a reference to itself to be an atomic group; that
925        is, there cannot be any backtracking into it once it has completed. This is
926        similar to recursive and subroutine calls.
927    
928    
929    Version 8.00 19-Oct-09
930    ----------------------
931    
932    1.  The table for translating pcre_compile() error codes into POSIX error codes
933        was out-of-date, and there was no check on the pcre_compile() error code
934        being within the table. This could lead to an OK return being given in
935        error.
936    
937    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
938        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
939        in a Windows environment.
940    
941    3.  The pcregrep --count option prints the count for each file even when it is
942        zero, as does GNU grep. However, pcregrep was also printing all files when
943        --files-with-matches was added. Now, when both options are given, it prints
944        counts only for those files that have at least one match. (GNU grep just
945        prints the file name in this circumstance, but including the count seems
946        more useful - otherwise, why use --count?) Also ensured that the
947        combination -clh just lists non-zero counts, with no names.
948    
949    4.  The long form of the pcregrep -F option was incorrectly implemented as
950        --fixed_strings instead of --fixed-strings. This is an incompatible change,
951        but it seems right to fix it, and I didn't think it was worth preserving
952        the old behaviour.
953    
954    5.  The command line items --regex=pattern and --regexp=pattern were not
955        recognized by pcregrep, which required --regex pattern or --regexp pattern
956        (with a space rather than an '='). The man page documented the '=' forms,
957        which are compatible with GNU grep; these now work.
958    
959    6.  No libpcreposix.pc file was created for pkg-config; there was just
960        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
961    
962    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
963        when UCP support is not needed, by modifying the Python script that
964        generates it from Unicode data files. This should not matter if the module
965        is correctly used as a library, but I received one complaint about 50K of
966        unwanted data. My guess is that the person linked everything into his
967        program rather than using a library. Anyway, it does no harm.
968    
969    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
970        was a minimum greater than 1 for a wide character in a possessive
971        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
972        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
973        character. Chaos in the form of incorrect output or a compiling loop could
974        result.
975    
976    9.  The restrictions on what a pattern can contain when partial matching is
977        requested for pcre_exec() have been removed. All patterns can now be
978        partially matched by this function. In addition, if there are at least two
979        slots in the offset vector, the offset of the earliest inspected character
980        for the match and the offset of the end of the subject are set in them when
981        PCRE_ERROR_PARTIAL is returned.
982    
983    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
984        synonymous with PCRE_PARTIAL, for backwards compatibility, and
985        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
986        and may be more useful for multi-segment matching.
987    
988    11. Partial matching with pcre_exec() is now more intuitive. A partial match
989        used to be given if ever the end of the subject was reached; now it is
990        given only if matching could not proceed because another character was
991        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
992        string "Z", which now yields "no match" instead of "partial match". In the
993        case of pcre_dfa_exec(), "no match" is given if every matching path for the
994        final character ended with (*FAIL).
995    
996    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
997        if the pattern had a "must contain" character that was already found in the
998        earlier partial match, unless partial matching was again requested. For
999        example, with the pattern /dog.(body)?/, the "must contain" character is
1000        "g". If the first part-match was for the string "dog", restarting with
1001        "sbody" failed. This bug has been fixed.
1002    
1003    13. The string returned by pcre_dfa_exec() after a partial match has been
1004        changed so that it starts at the first inspected character rather than the
1005        first character of the match. This makes a difference only if the pattern
1006        starts with a lookbehind assertion or \b or \B (\K is not supported by
1007        pcre_dfa_exec()). It's an incompatible change, but it makes the two
1008        matching functions compatible, and I think it's the right thing to do.
1009    
1010    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
1011        so that the demonstration program is easily available in environments where
1012        PCRE has not been installed from source.
1013    
1014    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
1015        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
1016        library.
1017    
1018    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
1019        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
1020        is not the first non-POSIX option to be added. Clearly some people find
1021        these options useful.
1022    
1023    17. If a caller to the POSIX matching function regexec() passes a non-zero
1024        value for nmatch with a NULL value for pmatch, the value of
1025        nmatch is forced to zero.
1026    
1027    18. RunGrepTest did not have a test for the availability of the -u option of
1028        the diff command, as RunTest does. It now checks in the same way as
1029        RunTest, and also checks for the -b option.
1030    
1031    19. If an odd number of negated classes containing just a single character
1032        interposed, within parentheses, between a forward reference to a named
1033        subpattern and the definition of the subpattern, compilation crashed with
1034        an internal error, complaining that it could not find the referenced
1035        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
1036        [The bug was that it was starting one character too far in when skipping
1037        over the character class, thus treating the ] as data rather than
1038        terminating the class. This meant it could skip too much.]
1039    
1040    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
1041        /g option in pcretest when the pattern contains \K, which makes it possible
1042        to have an empty string match not at the start, even when the pattern is
1043        anchored. Updated pcretest and pcredemo to use this option.
1044    
1045    21. If the maximum number of capturing subpatterns in a recursion was greater
1046        than the maximum at the outer level, the higher number was returned, but
1047        with unset values at the outer level. The correct (outer level) value is
1048        now given.
1049    
1050    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
1051        PCRE did not set those parentheses (unlike Perl). I have now found a way to
1052        make it do so. The string so far is captured, making this feature
1053        compatible with Perl.
1054    
1055    23. The tests have been re-organized, adding tests 11 and 12, to make it
1056        possible to check the Perl 5.10 features against Perl 5.10.
1057    
1058    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
1059        pattern matches a fixed length string. PCRE did not allow this; now it
1060        does. Neither allows recursion.
1061    
1062    25. I finally figured out how to implement a request to provide the minimum
1063        length of subject string that was needed in order to match a given pattern.
1064        (It was back references and recursion that I had previously got hung up
1065        on.) This code has now been added to pcre_study(); it finds a lower bound
1066        to the length of subject needed. It is not necessarily the greatest lower
1067        bound, but using it to avoid searching strings that are too short does give
1068        some useful speed-ups. The value is available to calling programs via
1069        pcre_fullinfo().
1070    
1071    26. While implementing 25, I discovered to my embarrassment that pcretest had
1072        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
1073        study optimizations had never been tested with that matching function.
1074        Oops. What is worse, even when it was passed study data, there was a bug in
1075        pcre_dfa_exec() that meant it never actually used it. Double oops. There
1076        were also very few tests of studied patterns with pcre_dfa_exec().
1077    
1078    27. If (?| is used to create subpatterns with duplicate numbers, they are now
1079        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
1080        on the other side of the coin, they are no longer allowed to have different
1081        names, because these cannot be distinguished in PCRE, and this has caused
1082        confusion. (This is a difference from Perl.)
1083    
1084    28. When duplicate subpattern names are present (necessarily with different
1085        numbers, as required by 27 above), and a test is made by name in a
1086        conditional pattern, either for a subpattern having been matched, or for
1087        recursion in such a pattern, all the associated numbered subpatterns are
1088        tested, and the overall condition is true if the condition is true for any
1089        one of them. This is the way Perl works, and is also more like the way
1090        testing by number works.
1091    
1092    
1093    Version 7.9 11-Apr-09
1094    ---------------------
1095    
1096    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
1097        (pcretest), all targets were linked against these libraries. This included
1098        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
1099        libraries. This caused unwanted dependencies to be created. This problem
1100        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
1101        pcretest is linked with readline.
1102    
1103    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
1104        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
1105        moved outside it again, because FALSE and TRUE are already defined in AIX,
1106        but BOOL is not.
1107    
1108    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
1109        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
1110    
1111    4.  The pcregrep documentation said spaces were inserted as well as colons (or
1112        hyphens) following file names and line numbers when outputting matching
1113        lines. This is not true; no spaces are inserted. I have also clarified the
1114        wording for the --colour (or --color) option.
1115    
1116    5.  In pcregrep, when --colour was used with -o, the list of matching strings
1117        was not coloured; this is different to GNU grep, so I have changed it to be
1118        the same.
1119    
1120    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
1121        each matching line was coloured. Now it goes on to look for further matches
1122        of any of the test patterns, which is the same behaviour as GNU grep.
1123    
1124    7.  A pattern that could match an empty string could cause pcregrep to loop; it
1125        doesn't make sense to accept an empty string match in pcregrep, so I have
1126        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
1127        seems to be how GNU grep behaves.
1128    
1129    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
1130        start or after a newline", because the conditional assertion was not being
1131        correctly handled. The rule now is that both the assertion and what follows
1132        in the first alternative must satisfy the test.
1133    
1134    9.  If auto-callout was enabled in a pattern with a conditional group whose
1135        condition was an assertion, PCRE could crash during matching, both with
1136        pcre_exec() and pcre_dfa_exec().
1137    
1138    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
1139        used for matching.
1140    
1141    11. Unicode property support in character classes was not working for
1142        characters (bytes) greater than 127 when not in UTF-8 mode.
1143    
1144    12. Added the -M command line option to pcretest.
1145    
1146    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
1147    
1148    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
1149    
1150    16. Added comments and documentation about mis-use of no_arg in the C++
1151        wrapper.
1152    
1153    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
1154        from Martin Jerabek that uses macro names for all relevant character and
1155        string constants.
1156    
1157    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
1158        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
1159        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
1160        these, but not everybody uses configure.
1161    
1162    19. A conditional group that had only one branch was not being correctly
1163        recognized as an item that could match an empty string. This meant that an
1164        enclosing group might also not be so recognized, causing infinite looping
1165        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
1166        with the subject "ab", where knowledge that the repeated group can match
1167        nothing is needed in order to break the loop.
1168    
1169    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
1170        exec(), but without supplying a callout function, matching went wrong.
1171    
1172    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
1173        leak if the size of the offset vector was greater than 30. When the vector
1174        is smaller, the saved offsets during recursion go onto a local stack
1175        vector, but for larger vectors malloc() is used. It was failing to free
1176        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
1177        error, in fact).
1178    
1179    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
1180        heapframe that is used only when UTF-8 support is enabled. This caused no
1181        problem, but was untidy.
1182    
1183    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
1184        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
1185        included within another project.
1186    
1187    24. Steven Van Ingelgem's patches to add more options to the CMake support,
1188        slightly modified by me:
1189    
1190          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
1191              not building pcregrep.
1192    
1193          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
1194              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
1195    
1196    25. Forward references, both numeric and by name, in patterns that made use of
1197        duplicate group numbers, could behave incorrectly or give incorrect errors,
1198        because when scanning forward to find the reference group, PCRE was not
1199        taking into account the duplicate group numbers. A pattern such as
1200        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
1201    
1202    26. Changed a few more instances of "const unsigned char *" to USPTR, making
1203        the feature of a custom pointer more persuasive (as requested by a user).
1204    
1205    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
1206        pcretest.c, inside #ifndefs, because it seems they are sometimes already
1207        pre-defined.
1208    
1209    28. Added support for (*UTF8) at the start of a pattern.
1210    
1211    29. Arrange for flags added by the "release type" setting in CMake to be shown
1212        in the configuration summary.
1213    
1214    
1215    Version 7.8 05-Sep-08
1216    ---------------------
1217    
1218    1.  Replaced UCP searching code with optimized version as implemented for Ad
1219        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
1220        stage table and inline lookup instead of a function, giving speed ups of 2
1221        to 5 times on some simple patterns that I tested. Permission was given to
1222        distribute the MultiStage2.py script that generates the tables (it's not in
1223        the tarball, but is in the Subversion repository).
1224    
1225    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
1226        scripts.
1227    
1228    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
1229        a group with a zero qualifier. The result of the study could be incorrect,
1230        or the function might crash, depending on the pattern.
1231    
1232    4.  Caseless matching was not working for non-ASCII characters in back
1233        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
1234        It now works when Unicode Property Support is available.
1235    
1236    5.  In pcretest, an escape such as \x{de} in the data was always generating
1237        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
1238        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
1239        truncation.
1240    
1241    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
1242    
1243    7.  Added two (int) casts to pcregrep when printing the difference of two
1244        pointers, in case they are 64-bit values.
1245    
1246    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
1247        test 2 if it fails.
1248    
1249    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
1250        and a #define of that name to empty if it is not externally set. This is to
1251        allow users of MSVC to set it if necessary.
1252    
1253    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
1254        the convenience functions in the pcre_get.c source file.
1255    
1256    11. An option change at the start of a pattern that had top-level alternatives
1257        could cause overwriting and/or a crash. This command provoked a crash in
1258        some environments:
1259    
1260          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
1261    
1262        This potential security problem was recorded as CVE-2008-2371.
1263    
1264    12. For a pattern where the match had to start at the beginning or immediately
1265        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
1266        pcre_dfa_exec() could read past the end of the passed subject if there was
1267        no match. To help with detecting such bugs (e.g. with valgrind), I modified
1268        pcretest so that it places the subject at the end of its malloc-ed buffer.
1269    
1270    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
1271        exec() might read past the end of the data buffer in UTF-8 mode.
1272    
1273    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
1274        the data contained the byte 0x85 as part of a UTF-8 character within its
1275        first line. This applied both to normal and DFA matching.
1276    
1277    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
1278        /^[^d]*?$/8 failed to match "abc".
1279    
1280    16. Added a missing copyright notice to pcrecpp_internal.h.
1281    
1282    17. Make it more clear in the documentation that values returned from
1283        pcre_exec() in ovector are byte offsets, not character counts.
1284    
1285    18. Tidied a few places to stop certain compilers from issuing warnings.
1286    
1287    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
1288        supplied by Stefan Weber. I made a further small update for 7.8 because
1289        there is a change of source arrangements: the pcre_searchfuncs.c module is
1290        replaced by pcre_ucd.c.
1291    
1292    
1293    Version 7.7 07-May-08
1294    ---------------------
1295    
1296    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
1297        a string to a long long, pretend we don't even have a long long." This is
1298        done by checking for the strtoq, strtoll, and _strtoi64 functions.
1299    
1300    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
1301        pre-7.6 versions, which defined a global no_arg variable instead of putting
1302        it in the RE class. (See also #8 below.)
1303    
1304    3.  Remove a line of dead code, identified by coverity and reported by Nuno
1305        Lopes.
1306    
1307    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
1308    
1309        (1) The include/exclude patterns were being applied to the whole pathnames
1310            of files, instead of just to the final components.
1311    
1312        (2) If there was more than one level of directory, the subdirectories were
1313            skipped unless they satisfied the include/exclude conditions. This is
1314            inconsistent with GNU grep (and could even be seen as contrary to the
1315            pcregrep specification - which I improved to make it absolutely clear).
1316            The action now is always to scan all levels of directory, and just
1317            apply the include/exclude patterns to regular files.
1318    
1319    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
1320        --exclude_dir in the tests to avoid scanning .svn directories.
1321    
1322    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
1323        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
1324        doesn't support NULs in patterns.
1325    
1326    7.  Added some missing "const"s to declarations of static tables in
1327        pcre_compile.c and pcre_dfa_exec.c.
1328    
1329    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
1330        caused by fix #2  above. (Subsequently also a second patch to fix the
1331        first patch. And a third patch - this was a messy problem.)
1332    
1333    9.  Applied Craig's patch to remove the use of push_back().
1334    
1335    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
1336        matching function regexec().
1337    
1338    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
1339        which, however, unlike Perl's \g{...}, are subroutine calls, not back
1340        references. PCRE supports relative numbers with this syntax (I don't think
1341        Oniguruma does).
1342    
1343    12. Previously, a group with a zero repeat such as (...){0} was completely
1344        omitted from the compiled regex. However, this means that if the group
1345        was called as a subroutine from elsewhere in the pattern, things went wrong
1346        (an internal error was given). Such groups are now left in the compiled
1347        pattern, with a new opcode that causes them to be skipped at execution
1348        time.
1349    
1350    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
1351        to the way PCRE behaves:
1352    
1353        (a) A lone ] character is dis-allowed (Perl treats it as data).
1354    
1355        (b) A back reference to an unmatched subpattern matches an empty string
1356            (Perl fails the current match path).
1357    
1358        (c) A data ] in a character class must be notated as \] because if the
1359            first data character in a class is ], it defines an empty class. (In
1360            Perl it is not possible to have an empty class.) The empty class []
1361            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
1362            The negative empty class [^] matches any one character, independently
1363            of the DOTALL setting.
1364    
1365    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
1366        non-existent subpattern following a character class starting with ']' and
1367        containing () gave an internal compiling error instead of "reference to
1368        non-existent subpattern". Fortunately, when the pattern did exist, the
1369        compiled code was correct. (When scanning forwards to check for the
1370        existencd of the subpattern, it was treating the data ']' as terminating
1371        the class, so got the count wrong. When actually compiling, the reference
1372        was subsequently set up correctly.)
1373    
1374    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
1375        it was being rejected as not supported by pcre_dfa_exec(), even though
1376        other assertions are supported. I have made pcre_dfa_exec() support
1377        (*FAIL).
1378    
1379    16. The implementation of 13c above involved the invention of a new opcode,
1380        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
1381        cannot be changed at match time, I realized I could make a small
1382        improvement to matching performance by compiling OP_ALLANY instead of
1383        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
1384        on the OP_ANY path.
1385    
1386    17. Compiling pcretest on Windows with readline support failed without the
1387        following two fixes: (1) Make the unistd.h include conditional on
1388        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
1389    
1390    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
1391        ncurses library to be included for pcretest when ReadLine support is
1392        requested, but also to allow for it to be overridden. This patch came from
1393        Daniel Bergström.
1394    
1395    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
1396        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
1397        any errors with the current Unicode tables. Thanks to Peter Kankowski for
1398        spotting this.
1399    
1400    
1401    Version 7.6 28-Jan-08
1402  ---------------------  ---------------------
1403    
1404  1.  A character class containing a very large number of characters with  1.  A character class containing a very large number of characters with
1405      codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer      codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
1406      overflow.      overflow.
1407    
1408  2.  Patch to cut out the "long long" test in pcrecpp_unittest when  2.  Patch to cut out the "long long" test in pcrecpp_unittest when
1409      HAVE_LONG_LONG is not defined.      HAVE_LONG_LONG is not defined.
1410    
1411  3.  Applied Christian Ehrlicher's patch to update the CMake build files to  3.  Applied Christian Ehrlicher's patch to update the CMake build files to
1412      bring them up to date and include new features. This patch includes:      bring them up to date and include new features. This patch includes:
1413    
1414      - Fixed PH's badly added libz and libbz2 support.      - Fixed PH's badly added libz and libbz2 support.
1415      - Fixed a problem with static linking.      - Fixed a problem with static linking.
1416      - Added pcredemo.      - Added pcredemo. [But later removed - see 7 below.]
1417      - Fixed dftables problem and added an option.      - Fixed dftables problem and added an option.
1418      - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and      - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
1419          HAVE_LONG_LONG.          HAVE_LONG_LONG.
1420      - Added readline support for pcretest.      - Added readline support for pcretest.
1421      - Added an listing of the option settings after cmake has run.      - Added an listing of the option settings after cmake has run.
1422    
1423  4.  A user submitted a patch to Makefile that makes it easy to created a dll  4.  A user submitted a patch to Makefile that makes it easy to create
1424      under mingw. I added stuff to Makefile.am that cause it to include this      "pcre.dll" under mingw when using Configure/Make. I added stuff to
1425      special target, without affecting anything else.      Makefile.am that cause it to include this special target, without
1426        affecting anything else. Note that the same mingw target plus all
1427        the other distribution libraries and programs are now supported
1428        when configuring with CMake (see 6 below) instead of with
1429        Configure/Make.
1430    
1431  5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.  5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
1432      This is an attempt to solve the reported problem "pcrecpp::no_arg is not      This is an attempt to solve the reported problem "pcrecpp::no_arg is not
1433      exported in the Windows port". It has not yet been confirmed that the patch      exported in the Windows port". It has not yet been confirmed that the patch
1434      solves the problem, but it does no harm.      solves the problem, but it does no harm.
1435    
1436    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
1437        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
1438        with CMake, and also correct the comment about stack recursion.
1439    
1440    7.  Remove the automatic building of pcredemo from the ./configure system and
1441        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
1442        of a program that users should build themselves after PCRE is installed, so
1443        building it automatically is not really right. What is more, it gave
1444        trouble in some build environments.
1445    
1446    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
1447    
1448    
1449  Version 7.5 10-Jan-08  Version 7.5 10-Jan-08

Legend:
Removed from v.308  
changed lines
  Added in v.787

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12