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

Diff of /code/trunk/ChangeLog

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

revision 77 by nigel, Sat Feb 24 21:40:45 2007 UTC revision 792 by ph10, Wed Dec 7 16:44:48 2011 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    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    27. Updated RunTest.bat as provided by Sheri Pierce.
113    
114    28. Fixed a possible uninitialized memory bug in pcre_jit_compile.c.
115    
116    
117    Version 8.20 21-Oct-2011
118    ------------------------
119    
120    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
121        a POSIX class. After further experiments with Perl, which convinced me that
122        Perl has bugs and confusions, a closing square bracket is no longer allowed
123        in a POSIX name. This bug also affected patterns with classes that started
124        with full stops.
125    
126    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
127        captured substring, but while checking the failing first alternative,
128        substring 1 is temporarily captured. If the output vector supplied to
129        pcre_exec() was not big enough for this capture, the yield of the function
130        was still zero ("insufficient space for captured substrings"). This cannot
131        be totally fixed without adding another stack variable, which seems a lot
132        of expense for a edge case. However, I have improved the situation in cases
133        such as /(a)(b)x|abc/ matched against "abc", where the return code
134        indicates that fewer than the maximum number of slots in the ovector have
135        been set.
136    
137    3.  Related to (2) above: when there are more back references in a pattern than
138        slots in the output vector, pcre_exec() uses temporary memory during
139        matching, and copies in the captures as far as possible afterwards. It was
140        using the entire output vector, but this conflicts with the specification
141        that only 2/3 is used for passing back captured substrings. Now it uses
142        only the first 2/3, for compatibility. This is, of course, another edge
143        case.
144    
145    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
146        main code base, and can be used by building with --enable-jit. When this is
147        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
148        runtime --no-jit option is given.
149    
150    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
151        ovector, the return from the function was zero, implying that there were
152        other matches that did not fit. The correct "exactly full" value is now
153        returned.
154    
155    6.  If a subpattern that was called recursively or as a subroutine contained
156        (*PRUNE) or any other control that caused it to give a non-standard return,
157        invalid errors such as "Error -26 (nested recursion at the same subject
158        position)" or even infinite loops could occur.
159    
160    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
161        computing the minimum length on reaching *ACCEPT, and so ended up with the
162        wrong value of 1 rather than 0. Further investigation indicates that
163        computing a minimum subject length in the presence of *ACCEPT is difficult
164        (think back references, subroutine calls), and so I have changed the code
165        so that no minimum is registered for a pattern that contains *ACCEPT.
166    
167    8.  If (*THEN) was present in the first (true) branch of a conditional group,
168        it was not handled as intended. [But see 16 below.]
169    
170    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
171        Sheri Pierce.
172    
173    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
174        the first byte in a match must be "a".
175    
176    11. Change 17 for 8.13 increased the recursion depth for patterns like
177        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
178        pattern contains any instances of (*THEN). If it does not, the old
179        optimizations are restored. It would be nice to do this on a per-group
180        basis, but at the moment that is not feasible.
181    
182    12. In some environments, the output of pcretest -C is CRLF terminated. This
183        broke RunTest's code that checks for the link size. A single white space
184        character after the value is now allowed for.
185    
186    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
187        For "fr", it uses the Windows-specific input and output files.
188    
189    14. If (*THEN) appeared in a group that was called recursively or as a
190        subroutine, it did not work as intended. [But see next item.]
191    
192    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
193        pattern fragments (but not containing any | characters). If A and B are
194        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
195        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
196        D. In other words, Perl considers parentheses that do not contain any |
197        characters to be part of a surrounding alternative, whereas PCRE was
198        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
199        differently. PCRE now behaves in the same way as Perl, except in the case
200        of subroutine/recursion calls such as (?1) which have in any case always
201        been different (but PCRE had them first :-).
202    
203    16. Related to 15 above: Perl does not treat the | in a conditional group as
204        creating alternatives. Such a group is treated in the same way as an
205        ordinary group without any | characters when processing (*THEN). PCRE has
206        been changed to match Perl's behaviour.
207    
208    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
209        RunGrepTest script failed.
210    
211    18. Change 22 for version 13 caused atomic groups to use more stack. This is
212        inevitable for groups that contain captures, but it can lead to a lot of
213        stack use in large patterns. The old behaviour has been restored for atomic
214        groups that do not contain any capturing parentheses.
215    
216    19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not
217        suppress the check for a minimum subject length at run time. (If it was
218        given to pcre_exec() or pcre_dfa_exec() it did work.)
219    
220    20. Fixed an ASCII-dependent infelicity in pcretest that would have made it
221        fail to work when decoding hex characters in data strings in EBCDIC
222        environments.
223    
224    21. It appears that in at least one Mac OS environment, the isxdigit() function
225        is implemented as a macro that evaluates to its argument more than once,
226        contravening the C 90 Standard (I haven't checked a later standard). There
227        was an instance in pcretest which caused it to go wrong when processing
228        \x{...} escapes in subject strings. The has been rewritten to avoid using
229        things like p++ in the argument of isxdigit().
230    
231    
232    Version 8.13 16-Aug-2011
233    ------------------------
234    
235    1.  The Unicode data tables have been updated to Unicode 6.0.0.
236    
237    2.  Two minor typos in pcre_internal.h have been fixed.
238    
239    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
240        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
241        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
242    
243    4.  There were a number of related bugs in the code for matching backrefences
244        caselessly in UTF-8 mode when codes for the characters concerned were
245        different numbers of bytes. For example, U+023A and U+2C65 are an upper
246        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
247        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
248        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
249        2-byte code at the end of the subject (it thought there wasn't enough data
250        left).
251    
252    5.  Comprehensive information about what went wrong is now returned by
253        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
254        as the output vector has at least 2 elements. The offset of the start of
255        the failing character and a reason code are placed in the vector.
256    
257    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
258        now returned is for the first byte of the failing character, instead of the
259        last byte inspected. This is an incompatible change, but I hope it is small
260        enough not to be a problem. It makes the returned offset consistent with
261        pcre_exec() and pcre_dfa_exec().
262    
263    7.  pcretest now gives a text phrase as well as the error number when
264        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
265        failure, the offset and reason code are output.
266    
267    8.  When \R was used with a maximizing quantifier it failed to skip backwards
268        over a \r\n pair if the subsequent match failed. Instead, it just skipped
269        back over a single character (\n). This seems wrong (because it treated the
270        two characters as a single entity when going forwards), conflicts with the
271        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
272        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
273        has been changed.
274    
275    9.  Some internal refactoring has changed the processing so that the handling
276        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
277        time (the PCRE_DOTALL option was changed this way some time ago: version
278        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
279        which was always a bit of a fudge. It also means that there is one less
280        argument for the match() function, which reduces its stack requirements
281        slightly. This change also fixes an incompatibility with Perl: the pattern
282        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
283    
284    10. More internal refactoring has drastically reduced the number of recursive
285        calls to match() for possessively repeated groups such as (abc)++ when
286        using pcre_exec().
287    
288    11. While implementing 10, a number of bugs in the handling of groups were
289        discovered and fixed:
290    
291        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
292        (a|)*(?1) gave a compile-time internal error.
293        ((a|)+)+  did not notice that the outer group could match an empty string.
294        (^a|^)+   was not marked as anchored.
295        (.*a|.*)+ was not marked as matching at start or after a newline.
296    
297    12. Yet more internal refactoring has removed another argument from the match()
298        function. Special calls to this function are now indicated by setting a
299        value in a variable in the "match data" data block.
300    
301    13. Be more explicit in pcre_study() instead of relying on "default" for
302        opcodes that mean there is no starting character; this means that when new
303        ones are added and accidentally left out of pcre_study(), testing should
304        pick them up.
305    
306    14. The -s option of pcretest has been documented for ages as being an old
307        synonym of -m (show memory usage). I have changed it to mean "force study
308        for every regex", that is, assume /S for every regex. This is similar to -i
309        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
310        using it. It makes it easier to run collections of tests with and without
311        study enabled, and thereby test pcre_study() more easily. All the standard
312        tests are now run with and without -s (but some patterns can be marked as
313        "never study" - see 20 below).
314    
315    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
316        restoration of the capturing data to the outer values was not happening
317        correctly.
318    
319    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
320        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
321        pattern had matched an empty string, and so incorrectly returned a no
322        match.
323    
324    17. There was optimizing code for the last branch of non-capturing parentheses,
325        and also for the obeyed branch of a conditional subexpression, which used
326        tail recursion to cut down on stack usage. Unfortunately, now that there is
327        the possibility of (*THEN) occurring in these branches, tail recursion is
328        no longer possible because the return has to be checked for (*THEN). These
329        two optimizations have therefore been removed. [But see 8.20/11 above.]
330    
331    18. If a pattern containing \R was studied, it was assumed that \R always
332        matched two bytes, thus causing the minimum subject length to be
333        incorrectly computed because \R can also match just one byte.
334    
335    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
336        was incorrectly computed.
337    
338    20. If /S is present twice on a test pattern in pcretest input, it now
339        *disables* studying, thereby overriding the use of -s on the command line
340        (see 14 above). This is necessary for one or two tests to keep the output
341        identical in both cases.
342    
343    21. When (*ACCEPT) was used in an assertion that matched an empty string and
344        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
345    
346    22. When an atomic group that contained a capturing parenthesis was
347        successfully matched, but the branch in which it appeared failed, the
348        capturing was not being forgotten if a higher numbered group was later
349        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
350        group 1 to "a", when in fact it should be unset. This applied to multi-
351        branched capturing and non-capturing groups, repeated or not, and also to
352        positive assertions (capturing in negative assertions does not happen
353        in PCRE) and also to nested atomic groups.
354    
355    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
356        subject after a captured substring, to make it easier to tell which of a
357        number of identical substrings has been captured.
358    
359    24. The way atomic groups are processed by pcre_exec() has been changed so that
360        if they are repeated, backtracking one repetition now resets captured
361        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
362        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
363        "aaa". Previously, it would have been "a". As part of this code
364        refactoring, the way recursive calls are handled has also been changed.
365    
366    25. If an assertion condition captured any substrings, they were not passed
367        back unless some other capturing happened later. For example, if
368        (?(?=(a))a) was matched against "a", no capturing was returned.
369    
370    26. When studying a pattern that contained subroutine calls or assertions,
371        the code for finding the minimum length of a possible match was handling
372        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
373        group 1 called group 2 while simultaneously a separate group 2 called group
374        1). A stack overflow occurred in this case. I have fixed this by limiting
375        the recursion depth to 10.
376    
377    27. Updated RunTest.bat in the distribution to the version supplied by Tom
378        Fortmann. This supports explicit test numbers on the command line, and has
379        argument validation and error reporting.
380    
381    28. An instance of \X with an unlimited repeat could fail if at any point the
382        first character it looked at was a mark character.
383    
384    29. Some minor code refactoring concerning Unicode properties and scripts
385        should reduce the stack requirement of match() slightly.
386    
387    30. Added the '=' option to pcretest to check the setting of unused capturing
388        slots at the end of the pattern, which are documented as being -1, but are
389        not included in the return count.
390    
391    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
392        compiled something random. Now it gives a compile-time error (as does
393        Perl).
394    
395    32. A *MARK encountered during the processing of a positive assertion is now
396        recorded and passed back (compatible with Perl).
397    
398    33. If --only-matching or --colour was set on a pcregrep call whose pattern
399        had alternative anchored branches, the search for a second match in a line
400        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
401        matched the line "0102" twice. The same bug affected patterns that started
402        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
403        twice.
404    
405    34. Previously, PCRE did not allow quantification of assertions. However, Perl
406        does, and because of capturing effects, quantifying parenthesized
407        assertions may at times be useful. Quantifiers are now allowed for
408        parenthesized assertions.
409    
410    35. A minor code tidy in pcre_compile() when checking options for \R usage.
411    
412    36. \g was being checked for fancy things in a character class, when it should
413        just be a literal "g".
414    
415    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
416        appearance of a nested POSIX class supersedes an apparent external class.
417        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
418        unescaped square brackets may also appear as part of class names. For
419        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
420        more like Perl. (But see 8.20/1 above.)
421    
422    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
423        was because it thought it was \N{name}, which is not supported).
424    
425    39. Add minix to OS list not supporting the -S option in pcretest.
426    
427    40. PCRE tries to detect cases of infinite recursion at compile time, but it
428        cannot analyze patterns in sufficient detail to catch mutual recursions
429        such as ((?1))((?2)). There is now a runtime test that gives an error if a
430        subgroup is called recursively as a subpattern for a second time at the
431        same position in the subject string. In previous releases this might have
432        been caught by the recursion limit, or it might have run out of stack.
433    
434    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
435        happen only once. PCRE was, however incorrectly giving a compile time error
436        "recursive call could loop indefinitely" because it cannot analyze the
437        pattern in sufficient detail. The compile time test no longer happens when
438        PCRE is compiling a conditional subpattern, but actual runaway loops are
439        now caught at runtime (see 40 above).
440    
441    42. It seems that Perl allows any characters other than a closing parenthesis
442        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
443        has been changed to be the same.
444    
445    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
446        as not to get warnings when autogen.sh is called. Also changed
447        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
448    
449    44. To help people who use pcregrep to scan files containing exceedingly long
450        lines, the following changes have been made:
451    
452        (a) The default value of the buffer size parameter has been increased from
453            8K to 20K. (The actual buffer used is three times this size.)
454    
455        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
456            PCRE is built.
457    
458        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
459            to be set at run time.
460    
461        (d) Numerical values in pcregrep options can be followed by K or M, for
462            example --buffer-size=50K.
463    
464        (e) If a line being scanned overflows pcregrep's buffer, an error is now
465            given and the return code is set to 2.
466    
467    45. Add a pointer to the latest mark to the callout data block.
468    
469    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
470        partial match of an empty string instead of no match. This was specific to
471        the use of ".".
472    
473    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
474        complete match instead of a partial match. This bug was dependent on both
475        the PCRE_UTF8 and PCRE_DOTALL options being set.
476    
477    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
478        starting byte set, because \b was not being ignored.
479    
480    
481    Version 8.12 15-Jan-2011
482    ------------------------
483    
484    1.  Fixed some typos in the markup of the man pages, and wrote a script that
485        checks for such things as part of the documentation building process.
486    
487    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
488        --match-limit and --recursion-limit options (added for 8.11). In
489        particular, this made one of the standard tests fail. (The integer value
490        went into the wrong half of a long int.)
491    
492    3.  If the --colour option was given to pcregrep with -v (invert match), it
493        did strange things, either producing crazy output, or crashing. It should,
494        of course, ignore a request for colour when reporting lines that do not
495        match.
496    
497    4.  Another pcregrep bug caused similar problems if --colour was specified with
498        -M (multiline) and the pattern match finished with a line ending.
499    
500    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
501        matched in multiline mode, the following line was shown as part of the
502        match. This seems wrong, so I have changed it.
503    
504    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
505        the check for further matches in the same line (so they could be coloured)
506        to overrun the end of the current line. If another match was found, it was
507        incorrectly shown (and then shown again when found in the next line).
508    
509    7.  If pcregrep was compiled under Windows, there was a reference to the
510        function pcregrep_exit() before it was defined. I am assuming this was
511        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
512        reported by a user. I've moved the definition above the reference.
513    
514    
515    Version 8.11 10-Dec-2010
516    ------------------------
517    
518    1.  (*THEN) was not working properly if there were untried alternatives prior
519        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
520        backtracked to try for "b" instead of moving to the next alternative branch
521        at the same level (in this case, to look for "c"). The Perl documentation
522        is clear that when (*THEN) is backtracked onto, it goes to the "next
523        alternative in the innermost enclosing group".
524    
525    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
526        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
527        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
528        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
529        (*THEN).
530    
531    3.  If \s appeared in a character class, it removed the VT character from
532        the class, even if it had been included by some previous item, for example
533        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
534        of \s, but is part of the POSIX "space" class.)
535    
536    4.  A partial match never returns an empty string (because you can always
537        match an empty string at the end of the subject); however the checking for
538        an empty string was starting at the "start of match" point. This has been
539        changed to the "earliest inspected character" point, because the returned
540        data for a partial match starts at this character. This means that, for
541        example, /(?<=abc)def/ gives a partial match for the subject "abc"
542        (previously it gave "no match").
543    
544    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
545        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
546        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
547        has an implication that the given string is incomplete (because a partial
548        match is preferred over a full match). For this reason, these items now
549        give a partial match in this situation. [Aside: previously, the one case
550        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
551        match rather than a full match, which was wrong by the old rules, but is
552        now correct.]
553    
554    6.  There was a bug in the handling of #-introduced comments, recognized when
555        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
556        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
557        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
558        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
559        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
560        places in pcre_compile().
561    
562    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
563        comments when looking ahead for named forward references to subpatterns,
564        the only newline sequence it recognized was NL. It now handles newlines
565        according to the set newline convention.
566    
567    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
568        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
569        cater for a lack of strerror(). These oversights have been fixed.
570    
571    9.  Added --match-limit and --recursion-limit to pcregrep.
572    
573    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
574    
575    11. When the -o option was used, pcregrep was setting a return code of 1, even
576        when matches were found, and --line-buffered was not being honoured.
577    
578    12. Added an optional parentheses number to the -o and --only-matching options
579        of pcregrep.
580    
581    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
582        can match an empty string. The code to do it in pcretest and pcredemo
583        needed fixing:
584    
585        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
586            only one byte after an empty string match just before CRLF (this case
587            just got forgotten; "any" and "anycrlf" were OK).
588    
589        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
590            mode when an empty string match preceded an ASCII character followed by
591            a non-ASCII character. (The code for advancing by one character rather
592            than one byte was nonsense.)
593    
594        (c) The pcredemo.c sample program did not have any code at all to handle
595            the cases when CRLF is a valid newline sequence.
596    
597    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
598        as a starting offset was within the subject string. There is now a new
599        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
600        negative or greater than the length of the string. In order to test this,
601        pcretest is extended to allow the setting of negative starting offsets.
602    
603    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
604        starting offset points to the beginning of a UTF-8 character was
605        unnecessarily clumsy. I tidied it up.
606    
607    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
608        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
609    
610    17. Nobody had reported that the --include_dir option, which was added in
611        release 7.7 should have been called --include-dir (hyphen, not underscore)
612        for compatibility with GNU grep. I have changed it to --include-dir, but
613        left --include_dir as an undocumented synonym, and the same for
614        --exclude-dir, though that is not available in GNU grep, at least as of
615        release 2.5.4.
616    
617    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
618        characters from a string of bytes) have been redefined so as not to use
619        loops, in order to improve performance in some environments. At the same
620        time, I abstracted some of the common code into auxiliary macros to save
621        repetition (this should not affect the compiled code).
622    
623    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
624        compile-time error is now given if \c is not followed by an ASCII
625        character, that is, a byte less than 128. (In EBCDIC mode, the code is
626        different, and any byte value is allowed.)
627    
628    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
629        START_OPTIMIZE option, which is now allowed at compile time - but just
630        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
631        to pcregrep and other applications that have no direct access to PCRE
632        options. The new /Y option in pcretest sets this option when calling
633        pcre_compile().
634    
635    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
636        back references. Groups containing recursive back references were forced to
637        be atomic by that change, but in the case of named groups, the amount of
638        memory required was incorrectly computed, leading to "Failed: internal
639        error: code overflow". This has been fixed.
640    
641    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
642        pcretest.c, to avoid build problems in some Borland environments.
643    
644    
645    Version 8.10 25-Jun-2010
646    ------------------------
647    
648    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
649        THEN.
650    
651    2.  (*ACCEPT) was not working when inside an atomic group.
652    
653    3.  Inside a character class, \B is treated as a literal by default, but
654        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
655        causes the error). The code is unchanged, but I tidied the documentation.
656    
657    4.  Inside a character class, PCRE always treated \R and \X as literals,
658        whereas Perl faults them if its -w option is set. I have changed PCRE so
659        that it faults them when PCRE_EXTRA is set.
660    
661    5.  Added support for \N, which always matches any character other than
662        newline. (It is the same as "." when PCRE_DOTALL is not set.)
663    
664    6.  When compiling pcregrep with newer versions of gcc which may have
665        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
666        declared with attribute warn_unused_result" were given. Just casting the
667        result to (void) does not stop the warnings; a more elaborate fudge is
668        needed. I've used a macro to implement this.
669    
670    7.  Minor change to pcretest.c to avoid a compiler warning.
671    
672    8.  Added four artifical Unicode properties to help with an option to make
673        \s etc use properties (see next item). The new properties are: Xan
674        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
675    
676    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
677        use Unicode properties. (*UCP) at the start of a pattern can be used to set
678        this option. Modified pcretest to add /W to test this facility. Added
679        REG_UCP to make it available via the POSIX interface.
680    
681    10. Added --line-buffered to pcregrep.
682    
683    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
684        studied, and the match started with a letter with a code point greater than
685        127 whose first byte was different to the first byte of the other case of
686        the letter, the other case of this starting letter was not recognized
687        (#976).
688    
689    12. If a pattern that was studied started with a repeated Unicode property
690        test, for example, \p{Nd}+, there was the theoretical possibility of
691        setting up an incorrect bitmap of starting bytes, but fortunately it could
692        not have actually happened in practice until change 8 above was made (it
693        added property types that matched character-matching opcodes).
694    
695    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
696        possible starting bytes for non-anchored patterns.
697    
698    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
699        \R, and also a number of cases that involve Unicode properties, both
700        explicit and implicit when PCRE_UCP is set.
701    
702    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
703        input, it could crash or give wrong results if characters with values
704        greater than 0xc0 were present in the subject string. (Detail: it assumed
705        UTF-8 input when processing these items.)
706    
707    16. Added a lot of (int) casts to avoid compiler warnings in systems where
708        size_t is 64-bit (#991).
709    
710    17. Added a check for running out of memory when PCRE is compiled with
711        --disable-stack-for-recursion (#990).
712    
713    18. If the last data line in a file for pcretest does not have a newline on
714        the end, a newline was missing in the output.
715    
716    19. The default pcre_chartables.c file recognizes only ASCII characters (values
717        less than 128) in its various bitmaps. However, there is a facility for
718        generating tables according to the current locale when PCRE is compiled. It
719        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
720        characters, are recognized by isspace() and therefore were getting set in
721        these tables, and indeed these tables seem to approximate to ISO 8859. This
722        caused a problem in UTF-8 mode when pcre_study() was used to create a list
723        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
724        which of course cannot start UTF-8 characters. I have changed the code so
725        that only real ASCII characters (less than 128) and the correct starting
726        bytes for UTF-8 encodings are set for characters greater than 127 when in
727        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
728        altogether.)
729    
730    20. Added the /T option to pcretest so as to be able to run tests with non-
731        standard character tables, thus making it possible to include the tests
732        used for 19 above in the standard set of tests.
733    
734    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
735        reference to a subpattern the other side of a comment that contains an
736        opening parenthesis caused either an internal compiling error, or a
737        reference to the wrong subpattern.
738    
739    
740    Version 8.02 19-Mar-2010
741    ------------------------
742    
743    1.  The Unicode data tables have been updated to Unicode 5.2.0.
744    
745    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
746        configured.
747    
748    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
749        original author of that file, following a query about its status.
750    
751    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
752        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
753    
754    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
755        quantifier applied to a forward-referencing subroutine call, could compile
756        incorrect code or give the error "internal error: previously-checked
757        referenced subpattern not found".
758    
759    6.  Both MS Visual Studio and Symbian OS have problems with initializing
760        variables to point to external functions. For these systems, therefore,
761        pcre_malloc etc. are now initialized to local functions that call the
762        relevant global functions.
763    
764    7.  There were two entries missing in the vectors called coptable and poptable
765        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
766        I've fixed the data, and added a kludgy way of testing at compile time that
767        the lengths are correct (equal to the number of opcodes).
768    
769    8.  Following on from 7, I added a similar kludge to check the length of the
770        eint vector in pcreposix.c.
771    
772    9.  Error texts for pcre_compile() are held as one long string to avoid too
773        much relocation at load time. To find a text, the string is searched,
774        counting zeros. There was no check for running off the end of the string,
775        which could happen if a new error number was added without updating the
776        string.
777    
778    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
779    
780    11. \K was not working if it appeared in an atomic group or in a group that
781        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
782        \K is "not well defined" if used in an assertion. PCRE now accepts it if
783        the assertion is positive, but not if it is negative.
784    
785    12. Change 11 fortuitously reduced the size of the stack frame used in the
786        "match()" function of pcre_exec.c by one pointer. Forthcoming
787        implementation of support for (*MARK) will need an extra pointer on the
788        stack; I have reserved it now, so that the stack frame size does not
789        decrease.
790    
791    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
792        item in branch that calls a recursion is a subroutine call - as in the
793        second branch in the above example - was incorrectly given the compile-
794        time error "recursive call could loop indefinitely" because pcre_compile()
795        was not correctly checking the subroutine for matching a non-empty string.
796    
797    14. The checks for overrunning compiling workspace could trigger after an
798        overrun had occurred. This is a "should never occur" error, but it can be
799        triggered by pathological patterns such as hundreds of nested parentheses.
800        The checks now trigger 100 bytes before the end of the workspace.
801    
802    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
803    
804    
805    Version 8.01 19-Jan-2010
806    ------------------------
807    
808    1.  If a pattern contained a conditional subpattern with only one branch (in
809        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
810        computed the wrong minimum data length (which is of course zero for such
811        subpatterns). This could cause incorrect "no match" results.
812    
813    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
814        the pattern is reset in the first branch, pcre_compile() failed with
815        "internal error: code overflow at offset...". This happened only when
816        the reset was to the original external option setting. (An optimization
817        abstracts leading options settings into an external setting, which was the
818        cause of this.)
819    
820    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
821        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
822        assertion pattern did not match (meaning that the assertion was true), it
823        was incorrectly treated as false if the SKIP had been reached during the
824        matching. This also applied to assertions used as conditions.
825    
826    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
827        assertion subpattern, including such a pattern used as a condition,
828        unpredictable results occurred, instead of the error return
829        PCRE_ERROR_DFA_UITEM.
830    
831    5.  The C++ GlobalReplace function was not working like Perl for the special
832        situation when an empty string is matched. It now does the fancy magic
833        stuff that is necessary.
834    
835    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
836        removed. (These were left over from very, very early versions of PCRE.)
837    
838    7.  Some cosmetic changes to the code to make life easier when compiling it
839        as part of something else:
840    
841        (a) Change DEBUG to PCRE_DEBUG.
842    
843        (b) In pcre_compile(), rename the member of the "branch_chain" structure
844            called "current" as "current_branch", to prevent a collision with the
845            Linux macro when compiled as a kernel module.
846    
847        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
848            prevent a collision with the Linux macro when compiled as a kernel
849            module.
850    
851    8.  In pcre_compile() there are some checks for integer overflows that used to
852        cast potentially large values to (double). This has been changed to that
853        when building, a check for int64_t is made, and if it is found, it is used
854        instead, thus avoiding the use of floating point arithmetic. (There is no
855        other use of FP in PCRE.) If int64_t is not found, the fallback is to
856        double.
857    
858    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
859        2005 (difference between two addresses compared to an unsigned value).
860    
861    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
862        custom one, because of the following reported problem in Windows:
863    
864          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
865              under Win32.
866          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
867              therefore missing the function definition.
868          - The compiler thus generates a "C" signature for the test function.
869          - The linker fails to find the "C" function.
870          - PCRE fails to configure if asked to do so against libbz2.
871    
872    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
873        messages were output:
874    
875          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
876          rerunning libtoolize, to keep the correct libtool macros in-tree.
877          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
878    
879        I have done both of these things.
880    
881    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
882        most of the time, it *can* run out if it is given a pattern that contains a
883        runaway infinite recursion. I updated the discussion in the pcrestack man
884        page.
885    
886    13. Now that we have gone to the x.xx style of version numbers, the minor
887        version may start with zero. Using 08 or 09 is a bad idea because users
888        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
889        interpreted as invalid octal numbers. I've updated the previous comment in
890        configure.ac, and also added a check that gives an error if 08 or 09 are
891        used.
892    
893    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
894        causing partial matching to fail when the end of the subject matched \W
895        in a UTF-8 pattern where \W was quantified with a minimum of 3.
896    
897    15. There were some discrepancies between the declarations in pcre_internal.h
898        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
899        their definitions. The declarations used "const uschar *" and the
900        definitions used USPTR. Even though USPTR is normally defined as "const
901        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
902        reported that: "This difference in casting confuses some C++ compilers, for
903        example, SunCC recognizes above declarations as different functions and
904        generates broken code for hbpcre." I have changed the declarations to use
905        USPTR.
906    
907    16. GNU libtool is named differently on some systems. The autogen.sh script now
908        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
909        (FreeBSD).
910    
911    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
912        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
913        comment: "Figure out how to create a longlong from a string: strtoll and
914        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
915        instance, but it only takes 2 args instead of 3!"
916    
917    18. A subtle bug concerned with back references has been fixed by a change of
918        specification, with a corresponding code fix. A pattern such as
919        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
920        refers, was giving matches when it shouldn't. For example, xa=xaaa would
921        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
922        same bug. Such groups have to be quantified to be useful, or contained
923        inside another quantified group. (If there's no repetition, the reference
924        can never match.) The problem arises because, having left the group and
925        moved on to the rest of the pattern, a later failure that backtracks into
926        the group uses the captured value from the final iteration of the group
927        rather than the correct earlier one. I have fixed this in PCRE by forcing
928        any group that contains a reference to itself to be an atomic group; that
929        is, there cannot be any backtracking into it once it has completed. This is
930        similar to recursive and subroutine calls.
931    
932    
933    Version 8.00 19-Oct-09
934    ----------------------
935    
936    1.  The table for translating pcre_compile() error codes into POSIX error codes
937        was out-of-date, and there was no check on the pcre_compile() error code
938        being within the table. This could lead to an OK return being given in
939        error.
940    
941    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
942        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
943        in a Windows environment.
944    
945    3.  The pcregrep --count option prints the count for each file even when it is
946        zero, as does GNU grep. However, pcregrep was also printing all files when
947        --files-with-matches was added. Now, when both options are given, it prints
948        counts only for those files that have at least one match. (GNU grep just
949        prints the file name in this circumstance, but including the count seems
950        more useful - otherwise, why use --count?) Also ensured that the
951        combination -clh just lists non-zero counts, with no names.
952    
953    4.  The long form of the pcregrep -F option was incorrectly implemented as
954        --fixed_strings instead of --fixed-strings. This is an incompatible change,
955        but it seems right to fix it, and I didn't think it was worth preserving
956        the old behaviour.
957    
958    5.  The command line items --regex=pattern and --regexp=pattern were not
959        recognized by pcregrep, which required --regex pattern or --regexp pattern
960        (with a space rather than an '='). The man page documented the '=' forms,
961        which are compatible with GNU grep; these now work.
962    
963    6.  No libpcreposix.pc file was created for pkg-config; there was just
964        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
965    
966    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
967        when UCP support is not needed, by modifying the Python script that
968        generates it from Unicode data files. This should not matter if the module
969        is correctly used as a library, but I received one complaint about 50K of
970        unwanted data. My guess is that the person linked everything into his
971        program rather than using a library. Anyway, it does no harm.
972    
973    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
974        was a minimum greater than 1 for a wide character in a possessive
975        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
976        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
977        character. Chaos in the form of incorrect output or a compiling loop could
978        result.
979    
980    9.  The restrictions on what a pattern can contain when partial matching is
981        requested for pcre_exec() have been removed. All patterns can now be
982        partially matched by this function. In addition, if there are at least two
983        slots in the offset vector, the offset of the earliest inspected character
984        for the match and the offset of the end of the subject are set in them when
985        PCRE_ERROR_PARTIAL is returned.
986    
987    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
988        synonymous with PCRE_PARTIAL, for backwards compatibility, and
989        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
990        and may be more useful for multi-segment matching.
991    
992    11. Partial matching with pcre_exec() is now more intuitive. A partial match
993        used to be given if ever the end of the subject was reached; now it is
994        given only if matching could not proceed because another character was
995        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
996        string "Z", which now yields "no match" instead of "partial match". In the
997        case of pcre_dfa_exec(), "no match" is given if every matching path for the
998        final character ended with (*FAIL).
999    
1000    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
1001        if the pattern had a "must contain" character that was already found in the
1002        earlier partial match, unless partial matching was again requested. For
1003        example, with the pattern /dog.(body)?/, the "must contain" character is
1004        "g". If the first part-match was for the string "dog", restarting with
1005        "sbody" failed. This bug has been fixed.
1006    
1007    13. The string returned by pcre_dfa_exec() after a partial match has been
1008        changed so that it starts at the first inspected character rather than the
1009        first character of the match. This makes a difference only if the pattern
1010        starts with a lookbehind assertion or \b or \B (\K is not supported by
1011        pcre_dfa_exec()). It's an incompatible change, but it makes the two
1012        matching functions compatible, and I think it's the right thing to do.
1013    
1014    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
1015        so that the demonstration program is easily available in environments where
1016        PCRE has not been installed from source.
1017    
1018    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
1019        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
1020        library.
1021    
1022    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
1023        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
1024        is not the first non-POSIX option to be added. Clearly some people find
1025        these options useful.
1026    
1027    17. If a caller to the POSIX matching function regexec() passes a non-zero
1028        value for nmatch with a NULL value for pmatch, the value of
1029        nmatch is forced to zero.
1030    
1031    18. RunGrepTest did not have a test for the availability of the -u option of
1032        the diff command, as RunTest does. It now checks in the same way as
1033        RunTest, and also checks for the -b option.
1034    
1035    19. If an odd number of negated classes containing just a single character
1036        interposed, within parentheses, between a forward reference to a named
1037        subpattern and the definition of the subpattern, compilation crashed with
1038        an internal error, complaining that it could not find the referenced
1039        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
1040        [The bug was that it was starting one character too far in when skipping
1041        over the character class, thus treating the ] as data rather than
1042        terminating the class. This meant it could skip too much.]
1043    
1044    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
1045        /g option in pcretest when the pattern contains \K, which makes it possible
1046        to have an empty string match not at the start, even when the pattern is
1047        anchored. Updated pcretest and pcredemo to use this option.
1048    
1049    21. If the maximum number of capturing subpatterns in a recursion was greater
1050        than the maximum at the outer level, the higher number was returned, but
1051        with unset values at the outer level. The correct (outer level) value is
1052        now given.
1053    
1054    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
1055        PCRE did not set those parentheses (unlike Perl). I have now found a way to
1056        make it do so. The string so far is captured, making this feature
1057        compatible with Perl.
1058    
1059    23. The tests have been re-organized, adding tests 11 and 12, to make it
1060        possible to check the Perl 5.10 features against Perl 5.10.
1061    
1062    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
1063        pattern matches a fixed length string. PCRE did not allow this; now it
1064        does. Neither allows recursion.
1065    
1066    25. I finally figured out how to implement a request to provide the minimum
1067        length of subject string that was needed in order to match a given pattern.
1068        (It was back references and recursion that I had previously got hung up
1069        on.) This code has now been added to pcre_study(); it finds a lower bound
1070        to the length of subject needed. It is not necessarily the greatest lower
1071        bound, but using it to avoid searching strings that are too short does give
1072        some useful speed-ups. The value is available to calling programs via
1073        pcre_fullinfo().
1074    
1075    26. While implementing 25, I discovered to my embarrassment that pcretest had
1076        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
1077        study optimizations had never been tested with that matching function.
1078        Oops. What is worse, even when it was passed study data, there was a bug in
1079        pcre_dfa_exec() that meant it never actually used it. Double oops. There
1080        were also very few tests of studied patterns with pcre_dfa_exec().
1081    
1082    27. If (?| is used to create subpatterns with duplicate numbers, they are now
1083        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
1084        on the other side of the coin, they are no longer allowed to have different
1085        names, because these cannot be distinguished in PCRE, and this has caused
1086        confusion. (This is a difference from Perl.)
1087    
1088    28. When duplicate subpattern names are present (necessarily with different
1089        numbers, as required by 27 above), and a test is made by name in a
1090        conditional pattern, either for a subpattern having been matched, or for
1091        recursion in such a pattern, all the associated numbered subpatterns are
1092        tested, and the overall condition is true if the condition is true for any
1093        one of them. This is the way Perl works, and is also more like the way
1094        testing by number works.
1095    
1096    
1097    Version 7.9 11-Apr-09
1098    ---------------------
1099    
1100    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
1101        (pcretest), all targets were linked against these libraries. This included
1102        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
1103        libraries. This caused unwanted dependencies to be created. This problem
1104        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
1105        pcretest is linked with readline.
1106    
1107    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
1108        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
1109        moved outside it again, because FALSE and TRUE are already defined in AIX,
1110        but BOOL is not.
1111    
1112    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
1113        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
1114    
1115    4.  The pcregrep documentation said spaces were inserted as well as colons (or
1116        hyphens) following file names and line numbers when outputting matching
1117        lines. This is not true; no spaces are inserted. I have also clarified the
1118        wording for the --colour (or --color) option.
1119    
1120    5.  In pcregrep, when --colour was used with -o, the list of matching strings
1121        was not coloured; this is different to GNU grep, so I have changed it to be
1122        the same.
1123    
1124    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
1125        each matching line was coloured. Now it goes on to look for further matches
1126        of any of the test patterns, which is the same behaviour as GNU grep.
1127    
1128    7.  A pattern that could match an empty string could cause pcregrep to loop; it
1129        doesn't make sense to accept an empty string match in pcregrep, so I have
1130        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
1131        seems to be how GNU grep behaves.
1132    
1133    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
1134        start or after a newline", because the conditional assertion was not being
1135        correctly handled. The rule now is that both the assertion and what follows
1136        in the first alternative must satisfy the test.
1137    
1138    9.  If auto-callout was enabled in a pattern with a conditional group whose
1139        condition was an assertion, PCRE could crash during matching, both with
1140        pcre_exec() and pcre_dfa_exec().
1141    
1142    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
1143        used for matching.
1144    
1145    11. Unicode property support in character classes was not working for
1146        characters (bytes) greater than 127 when not in UTF-8 mode.
1147    
1148    12. Added the -M command line option to pcretest.
1149    
1150    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
1151    
1152    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
1153    
1154    16. Added comments and documentation about mis-use of no_arg in the C++
1155        wrapper.
1156    
1157    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
1158        from Martin Jerabek that uses macro names for all relevant character and
1159        string constants.
1160    
1161    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
1162        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
1163        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
1164        these, but not everybody uses configure.
1165    
1166    19. A conditional group that had only one branch was not being correctly
1167        recognized as an item that could match an empty string. This meant that an
1168        enclosing group might also not be so recognized, causing infinite looping
1169        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
1170        with the subject "ab", where knowledge that the repeated group can match
1171        nothing is needed in order to break the loop.
1172    
1173    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
1174        exec(), but without supplying a callout function, matching went wrong.
1175    
1176    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
1177        leak if the size of the offset vector was greater than 30. When the vector
1178        is smaller, the saved offsets during recursion go onto a local stack
1179        vector, but for larger vectors malloc() is used. It was failing to free
1180        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
1181        error, in fact).
1182    
1183    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
1184        heapframe that is used only when UTF-8 support is enabled. This caused no
1185        problem, but was untidy.
1186    
1187    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
1188        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
1189        included within another project.
1190    
1191    24. Steven Van Ingelgem's patches to add more options to the CMake support,
1192        slightly modified by me:
1193    
1194          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
1195              not building pcregrep.
1196    
1197          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
1198              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
1199    
1200    25. Forward references, both numeric and by name, in patterns that made use of
1201        duplicate group numbers, could behave incorrectly or give incorrect errors,
1202        because when scanning forward to find the reference group, PCRE was not
1203        taking into account the duplicate group numbers. A pattern such as
1204        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
1205    
1206    26. Changed a few more instances of "const unsigned char *" to USPTR, making
1207        the feature of a custom pointer more persuasive (as requested by a user).
1208    
1209    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
1210        pcretest.c, inside #ifndefs, because it seems they are sometimes already
1211        pre-defined.
1212    
1213    28. Added support for (*UTF8) at the start of a pattern.
1214    
1215    29. Arrange for flags added by the "release type" setting in CMake to be shown
1216        in the configuration summary.
1217    
1218    
1219    Version 7.8 05-Sep-08
1220    ---------------------
1221    
1222    1.  Replaced UCP searching code with optimized version as implemented for Ad
1223        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
1224        stage table and inline lookup instead of a function, giving speed ups of 2
1225        to 5 times on some simple patterns that I tested. Permission was given to
1226        distribute the MultiStage2.py script that generates the tables (it's not in
1227        the tarball, but is in the Subversion repository).
1228    
1229    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
1230        scripts.
1231    
1232    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
1233        a group with a zero qualifier. The result of the study could be incorrect,
1234        or the function might crash, depending on the pattern.
1235    
1236    4.  Caseless matching was not working for non-ASCII characters in back
1237        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
1238        It now works when Unicode Property Support is available.
1239    
1240    5.  In pcretest, an escape such as \x{de} in the data was always generating
1241        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
1242        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
1243        truncation.
1244    
1245    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
1246    
1247    7.  Added two (int) casts to pcregrep when printing the difference of two
1248        pointers, in case they are 64-bit values.
1249    
1250    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
1251        test 2 if it fails.
1252    
1253    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
1254        and a #define of that name to empty if it is not externally set. This is to
1255        allow users of MSVC to set it if necessary.
1256    
1257    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
1258        the convenience functions in the pcre_get.c source file.
1259    
1260    11. An option change at the start of a pattern that had top-level alternatives
1261        could cause overwriting and/or a crash. This command provoked a crash in
1262        some environments:
1263    
1264          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
1265    
1266        This potential security problem was recorded as CVE-2008-2371.
1267    
1268    12. For a pattern where the match had to start at the beginning or immediately
1269        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
1270        pcre_dfa_exec() could read past the end of the passed subject if there was
1271        no match. To help with detecting such bugs (e.g. with valgrind), I modified
1272        pcretest so that it places the subject at the end of its malloc-ed buffer.
1273    
1274    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
1275        exec() might read past the end of the data buffer in UTF-8 mode.
1276    
1277    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
1278        the data contained the byte 0x85 as part of a UTF-8 character within its
1279        first line. This applied both to normal and DFA matching.
1280    
1281    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
1282        /^[^d]*?$/8 failed to match "abc".
1283    
1284    16. Added a missing copyright notice to pcrecpp_internal.h.
1285    
1286    17. Make it more clear in the documentation that values returned from
1287        pcre_exec() in ovector are byte offsets, not character counts.
1288    
1289    18. Tidied a few places to stop certain compilers from issuing warnings.
1290    
1291    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
1292        supplied by Stefan Weber. I made a further small update for 7.8 because
1293        there is a change of source arrangements: the pcre_searchfuncs.c module is
1294        replaced by pcre_ucd.c.
1295    
1296    
1297    Version 7.7 07-May-08
1298    ---------------------
1299    
1300    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
1301        a string to a long long, pretend we don't even have a long long." This is
1302        done by checking for the strtoq, strtoll, and _strtoi64 functions.
1303    
1304    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
1305        pre-7.6 versions, which defined a global no_arg variable instead of putting
1306        it in the RE class. (See also #8 below.)
1307    
1308    3.  Remove a line of dead code, identified by coverity and reported by Nuno
1309        Lopes.
1310    
1311    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
1312    
1313        (1) The include/exclude patterns were being applied to the whole pathnames
1314            of files, instead of just to the final components.
1315    
1316        (2) If there was more than one level of directory, the subdirectories were
1317            skipped unless they satisfied the include/exclude conditions. This is
1318            inconsistent with GNU grep (and could even be seen as contrary to the
1319            pcregrep specification - which I improved to make it absolutely clear).
1320            The action now is always to scan all levels of directory, and just
1321            apply the include/exclude patterns to regular files.
1322    
1323    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
1324        --exclude_dir in the tests to avoid scanning .svn directories.
1325    
1326    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
1327        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
1328        doesn't support NULs in patterns.
1329    
1330    7.  Added some missing "const"s to declarations of static tables in
1331        pcre_compile.c and pcre_dfa_exec.c.
1332    
1333    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
1334        caused by fix #2  above. (Subsequently also a second patch to fix the
1335        first patch. And a third patch - this was a messy problem.)
1336    
1337    9.  Applied Craig's patch to remove the use of push_back().
1338    
1339    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
1340        matching function regexec().
1341    
1342    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
1343        which, however, unlike Perl's \g{...}, are subroutine calls, not back
1344        references. PCRE supports relative numbers with this syntax (I don't think
1345        Oniguruma does).
1346    
1347    12. Previously, a group with a zero repeat such as (...){0} was completely
1348        omitted from the compiled regex. However, this means that if the group
1349        was called as a subroutine from elsewhere in the pattern, things went wrong
1350        (an internal error was given). Such groups are now left in the compiled
1351        pattern, with a new opcode that causes them to be skipped at execution
1352        time.
1353    
1354    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
1355        to the way PCRE behaves:
1356    
1357        (a) A lone ] character is dis-allowed (Perl treats it as data).
1358    
1359        (b) A back reference to an unmatched subpattern matches an empty string
1360            (Perl fails the current match path).
1361    
1362        (c) A data ] in a character class must be notated as \] because if the
1363            first data character in a class is ], it defines an empty class. (In
1364            Perl it is not possible to have an empty class.) The empty class []
1365            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
1366            The negative empty class [^] matches any one character, independently
1367            of the DOTALL setting.
1368    
1369    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
1370        non-existent subpattern following a character class starting with ']' and
1371        containing () gave an internal compiling error instead of "reference to
1372        non-existent subpattern". Fortunately, when the pattern did exist, the
1373        compiled code was correct. (When scanning forwards to check for the
1374        existencd of the subpattern, it was treating the data ']' as terminating
1375        the class, so got the count wrong. When actually compiling, the reference
1376        was subsequently set up correctly.)
1377    
1378    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
1379        it was being rejected as not supported by pcre_dfa_exec(), even though
1380        other assertions are supported. I have made pcre_dfa_exec() support
1381        (*FAIL).
1382    
1383    16. The implementation of 13c above involved the invention of a new opcode,
1384        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
1385        cannot be changed at match time, I realized I could make a small
1386        improvement to matching performance by compiling OP_ALLANY instead of
1387        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
1388        on the OP_ANY path.
1389    
1390    17. Compiling pcretest on Windows with readline support failed without the
1391        following two fixes: (1) Make the unistd.h include conditional on
1392        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
1393    
1394    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
1395        ncurses library to be included for pcretest when ReadLine support is
1396        requested, but also to allow for it to be overridden. This patch came from
1397        Daniel Bergström.
1398    
1399    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
1400        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
1401        any errors with the current Unicode tables. Thanks to Peter Kankowski for
1402        spotting this.
1403    
1404    
1405    Version 7.6 28-Jan-08
1406    ---------------------
1407    
1408    1.  A character class containing a very large number of characters with
1409        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
1410        overflow.
1411    
1412    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
1413        HAVE_LONG_LONG is not defined.
1414    
1415    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
1416        bring them up to date and include new features. This patch includes:
1417    
1418        - Fixed PH's badly added libz and libbz2 support.
1419        - Fixed a problem with static linking.
1420        - Added pcredemo. [But later removed - see 7 below.]
1421        - Fixed dftables problem and added an option.
1422        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
1423            HAVE_LONG_LONG.
1424        - Added readline support for pcretest.
1425        - Added an listing of the option settings after cmake has run.
1426    
1427    4.  A user submitted a patch to Makefile that makes it easy to create
1428        "pcre.dll" under mingw when using Configure/Make. I added stuff to
1429        Makefile.am that cause it to include this special target, without
1430        affecting anything else. Note that the same mingw target plus all
1431        the other distribution libraries and programs are now supported
1432        when configuring with CMake (see 6 below) instead of with
1433        Configure/Make.
1434    
1435    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
1436        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
1437        exported in the Windows port". It has not yet been confirmed that the patch
1438        solves the problem, but it does no harm.
1439    
1440    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
1441        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
1442        with CMake, and also correct the comment about stack recursion.
1443    
1444    7.  Remove the automatic building of pcredemo from the ./configure system and
1445        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
1446        of a program that users should build themselves after PCRE is installed, so
1447        building it automatically is not really right. What is more, it gave
1448        trouble in some build environments.
1449    
1450    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
1451    
1452    
1453    Version 7.5 10-Jan-08
1454    ---------------------
1455    
1456    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
1457        values in parens when parsing an RE using the C++ wrapper."
1458    
1459    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
1460        Characters greater than 255 were excluded from the class instead of being
1461        included.
1462    
1463    3.  The same bug as (2) above applied to negated POSIX classes such as
1464        [:^space:].
1465    
1466    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
1467        defined or documented. It seems to have been a typo for PCRE_STATIC, so
1468        I have changed it.
1469    
1470    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
1471        first named subpattern) and a construct such as (?&a) would reference the
1472        first named subpattern whose name started with "a" (in other words, the
1473        length check was missing). Both these problems are fixed. "Subpattern name
1474        expected" is now given for (?&) (a zero-length name), and this patch also
1475        makes it give the same error for \k'' (previously it complained that that
1476        was a reference to a non-existent subpattern).
1477    
1478    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
1479        this is right because (?- can be followed by option settings as well as by
1480        digits. I have, however, made the messages clearer.
1481    
1482    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
1483        than the number used in the conditional) now cause a compile-time error.
1484        This is actually not compatible with Perl, which accepts such patterns, but
1485        treats the conditional as always being FALSE (as PCRE used to), but it
1486        seems to me that giving a diagnostic is better.
1487    
1488    8.  Change "alphameric" to the more common word "alphanumeric" in comments
1489        and messages.
1490    
1491    9.  Fix two occurrences of "backslash" in comments that should have been
1492        "backspace".
1493    
1494    10. Remove two redundant lines of code that can never be obeyed (their function
1495        was moved elsewhere).
1496    
1497    11. The program that makes PCRE's Unicode character property table had a bug
1498        which caused it to generate incorrect table entries for sequences of
1499        characters that have the same character type, but are in different scripts.
1500        It amalgamated them into a single range, with the script of the first of
1501        them. In other words, some characters were in the wrong script. There were
1502        thirteen such cases, affecting characters in the following ranges:
1503    
1504          U+002b0 - U+002c1
1505          U+0060c - U+0060d
1506          U+0061e - U+00612
1507          U+0064b - U+0065e
1508          U+0074d - U+0076d
1509          U+01800 - U+01805
1510          U+01d00 - U+01d77
1511          U+01d9b - U+01dbf
1512          U+0200b - U+0200f
1513          U+030fc - U+030fe
1514          U+03260 - U+0327f
1515          U+0fb46 - U+0fbb1
1516          U+10450 - U+1049d
1517    
1518    12. The -o option (show only the matching part of a line) for pcregrep was not
1519        compatible with GNU grep in that, if there was more than one match in a
1520        line, it showed only the first of them. It now behaves in the same way as
1521        GNU grep.
1522    
1523    13. If the -o and -v options were combined for pcregrep, it printed a blank
1524        line for every non-matching line. GNU grep prints nothing, and pcregrep now
1525        does the same. The return code can be used to tell if there were any
1526        non-matching lines.
1527    
1528    14. Added --file-offsets and --line-offsets to pcregrep.
1529    
1530    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
1531        infinitely looping recursion. The bug was that positive lookaheads were not
1532        being skipped when checking for a possible empty match (negative lookaheads
1533        and both kinds of lookbehind were skipped).
1534    
1535    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
1536        inclusion of <windows.h> to before rather than after the definition of
1537        INVALID_FILE_ATTRIBUTES (patch from David Byron).
1538    
1539    17. Specifying a possessive quantifier with a specific limit for a Unicode
1540        character property caused pcre_compile() to compile bad code, which led at
1541        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
1542        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
1543        caused the error; without that there was no problem.
1544    
1545    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
1546    
1547    19. Added --enable-pcretest-libreadline.
1548    
1549    20. In pcrecpp.cc, the variable 'count' was incremented twice in
1550        RE::GlobalReplace(). As a result, the number of replacements returned was
1551        double what it should be. I removed one of the increments, but Craig sent a
1552        later patch that removed the other one (the right fix) and added unit tests
1553        that check the return values (which was not done before).
1554    
1555    21. Several CMake things:
1556    
1557        (1) Arranged that, when cmake is used on Unix, the libraries end up with
1558            the names libpcre and libpcreposix, not just pcre and pcreposix.
1559    
1560        (2) The above change means that pcretest and pcregrep are now correctly
1561            linked with the newly-built libraries, not previously installed ones.
1562    
1563        (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
1564    
1565    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
1566        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
1567        UTF-8 newline character). The key issue is that the pattern starts .*;
1568        this means that the match must be either at the beginning, or after a
1569        newline. The bug was in the code for advancing after a failed match and
1570        checking that the new position followed a newline. It was not taking
1571        account of UTF-8 characters correctly.
1572    
1573    23. PCRE was behaving differently from Perl in the way it recognized POSIX
1574        character classes. PCRE was not treating the sequence [:...:] as a
1575        character class unless the ... were all letters. Perl, however, seems to
1576        allow any characters between [: and :], though of course it rejects as
1577        unknown any "names" that contain non-letters, because all the known class
1578        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
1579        for example, whereas PCRE did not - it did not recognize a POSIX character
1580        class. This seemed a bit dangerous, so the code has been changed to be
1581        closer to Perl. The behaviour is not identical to Perl, because PCRE will
1582        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
1583        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
1584        Perl does, and where it didn't before.
1585    
1586    24. Rewrite so as to remove the single use of %n from pcregrep because in some
1587        Windows environments %n is disabled by default.
1588    
1589    
1590    Version 7.4 21-Sep-07
1591    ---------------------
1592    
1593    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
1594        means that a class such as [\s] counted as "explicit reference to CR or
1595        LF". That isn't really right - the whole point of the change was to try to
1596        help when there was an actual mention of one of the two characters. So now
1597        the change happens only if \r or \n (or a literal CR or LF) character is
1598        encountered.
1599    
1600    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
1601        of both had grown to the point where there were only 3 bits left.
1602        Fortunately, there was spare space in the data structure, and so I have
1603        moved the internal flags into a new 16-bit field to free up more option
1604        bits.
1605    
1606    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
1607        but did not set the internal JCHANGED flag - either of these is enough to
1608        control the way the "get" function works - but the PCRE_INFO_JCHANGED
1609        facility is supposed to tell if (?J) was ever used, so now (?J) at the
1610        start sets both bits.
1611    
1612    4.  Added options (at build time, compile time, exec time) to change \R from
1613        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
1614    
1615    5.  doc/pcresyntax.html was missing from the distribution.
1616    
1617    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
1618        compatibility, even though it is no longer used.
1619    
1620    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
1621        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
1622        windows.h file is present (where different names are used). [This was
1623        reversed later after testing - see 16 below.]
1624    
1625    8.  Changed all #include <config.h> to #include "config.h". There were also
1626        some further <pcre.h> cases that I changed to "pcre.h".
1627    
1628    9.  When pcregrep was used with the --colour option, it missed the line ending
1629        sequence off the lines that it output.
1630    
1631    10. It was pointed out to me that arrays of string pointers cause lots of
1632        relocations when a shared library is dynamically loaded. A technique of
1633        using a single long string with a table of offsets can drastically reduce
1634        these. I have refactored PCRE in four places to do this. The result is
1635        dramatic:
1636    
1637          Originally:                          290
1638          After changing UCP table:            187
1639          After changing error message table:   43
1640          After changing table of "verbs"       36
1641          After changing table of Posix names   22
1642    
1643        Thanks to the folks working on Gregex for glib for this insight.
1644    
1645    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
1646        unicode-properties was also set.
1647    
1648    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
1649    
1650    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
1651        checked only for CRLF.
1652    
1653    14. Added casts to pcretest.c to avoid compiler warnings.
1654    
1655    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
1656    
1657    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
1658        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
1659        entirely. This removes changes made in 7 above.
1660    
1661    17. The CMake files have been updated, and there is now more information about
1662        building with CMake in the NON-UNIX-USE document.
1663    
1664    
1665    Version 7.3 28-Aug-07
1666    ---------------------
1667    
1668     1. In the rejigging of the build system that eventually resulted in 7.1, the
1669        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
1670        brackets there is not right, since it causes compilers to look for an
1671        installed pcre.h, not the version that is in the source that is being
1672        compiled (which of course may be different). I have changed it back to:
1673    
1674          #include "pcre.h"
1675    
1676        I have a vague recollection that the change was concerned with compiling in
1677        different directories, but in the new build system, that is taken care of
1678        by the VPATH setting the Makefile.
1679    
1680     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
1681        when the subject happened to end in the byte 0x85 (e.g. if the last
1682        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
1683        characters but of course it shouldn't be taken as a newline when it is part
1684        of another character. The bug was that, for an unlimited repeat of . in
1685        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
1686        characters when looking for a newline.
1687    
1688     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
1689    
1690     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
1691        in debug output.
1692    
1693     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
1694        long printing in the pcrecpp unittest when running under MinGW.
1695    
1696     6. ESC_K was left out of the EBCDIC table.
1697    
1698     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
1699        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
1700        limit also applies to "virtual nesting" when a pattern is recursive, and in
1701        this case 1000 isn't so big. I have been able to remove this limit at the
1702        expense of backing off one optimization in certain circumstances. Normally,
1703        when pcre_exec() would call its internal match() function recursively and
1704        immediately return the result unconditionally, it uses a "tail recursion"
1705        feature to save stack. However, when a subpattern that can match an empty
1706        string has an unlimited repetition quantifier, it no longer makes this
1707        optimization. That gives it a stack frame in which to save the data for
1708        checking that an empty string has been matched. Previously this was taken
1709        from the 1000-entry workspace that had been reserved. So now there is no
1710        explicit limit, but more stack is used.
1711    
1712     8. Applied Daniel's patches to solve problems with the import/export magic
1713        syntax that is required for Windows, and which was going wrong for the
1714        pcreposix and pcrecpp parts of the library. These were overlooked when this
1715        problem was solved for the main library.
1716    
1717     9. There were some crude static tests to avoid integer overflow when computing
1718        the size of patterns that contain repeated groups with explicit upper
1719        limits. As the maximum quantifier is 65535, the maximum group length was
1720        set at 30,000 so that the product of these two numbers did not overflow a
1721        32-bit integer. However, it turns out that people want to use groups that
1722        are longer than 30,000 bytes (though not repeat them that many times).
1723        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1724        made it possible to implement the integer overflow checks in a much more
1725        dynamic way, which I have now done. The artificial limitation on group
1726        length has been removed - we now have only the limit on the total length of
1727        the compiled pattern, which depends on the LINK_SIZE setting.
1728    
1729    10. Fixed a bug in the documentation for get/copy named substring when
1730        duplicate names are permitted. If none of the named substrings are set, the
1731        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1732        empty string.
1733    
1734    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1735        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1736        because the ] is interpreted as the first data character and the
1737        terminating ] is not found. PCRE has been made compatible with Perl in this
1738        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1739        cause memory overwriting.
1740    
1741    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1742        string has been matched (to stop an infinite loop). It was not recognizing
1743        a conditional subpattern that could match an empty string if that
1744        subpattern was within another subpattern. For example, it looped when
1745        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
1746        condition was not nested. This bug has been fixed.
1747    
1748    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1749        past the start of the subject in the presence of bytes with the top bit
1750        set, for example "\x8aBCD".
1751    
1752    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1753        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1754    
1755    14. Optimized (?!) to (*FAIL).
1756    
1757    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1758        This restricts code points to be within the range 0 to 0x10FFFF, excluding
1759        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1760        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1761        does: it's just the validity check that is more restrictive.
1762    
1763    16. Inserted checks for integer overflows during escape sequence (backslash)
1764        processing, and also fixed erroneous offset values for syntax errors during
1765        backslash processing.
1766    
1767    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1768        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1769    
1770    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1771        caused an overrun.
1772    
1773    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1774        something other than just ASCII characters) inside a group that had an
1775        unlimited repeat caused a loop at compile time (while checking to see
1776        whether the group could match an empty string).
1777    
1778    20. Debugging a pattern containing \p or \P could cause a crash. For example,
1779        [\P{Any}] did so. (Error in the code for printing property names.)
1780    
1781    21. An orphan \E inside a character class could cause a crash.
1782    
1783    22. A repeated capturing bracket such as (A)? could cause a wild memory
1784        reference during compilation.
1785    
1786    23. There are several functions in pcre_compile() that scan along a compiled
1787        expression for various reasons (e.g. to see if it's fixed length for look
1788        behind). There were bugs in these functions when a repeated \p or \P was
1789        present in the pattern. These operators have additional parameters compared
1790        with \d, etc, and these were not being taken into account when moving along
1791        the compiled data. Specifically:
1792    
1793        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1794            length.
1795    
1796        (b) An item such as \pL+ within a repeated group could cause crashes or
1797            loops.
1798    
1799        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1800            "reference to non-existent subpattern" error.
1801    
1802        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1803    
1804    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1805        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1806    
1807    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1808    
1809    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1810        character were causing crashes (broken optimization).
1811    
1812    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1813        \p or \P) caused a compile-time loop.
1814    
1815    28. More problems have arisen in unanchored patterns when CRLF is a valid line
1816        break. For example, the unstudied pattern [\r\n]A does not match the string
1817        "\r\nA" because change 7.0/46 below moves the current point on by two
1818        characters after failing to match at the start. However, the pattern \nA
1819        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
1820        the same is true. There doesn't seem any very clean way out of this, but
1821        what I have chosen to do makes the common cases work: PCRE now takes note
1822        of whether there can be an explicit match for \r or \n anywhere in the
1823        pattern, and if so, 7.0/46 no longer applies. As part of this change,
1824        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1825        pattern has explicit CR or LF references.
1826    
1827    29. Added (*CR) etc for changing newline setting at start of pattern.
1828    
1829    
1830    Version 7.2 19-Jun-07
1831    ---------------------
1832    
1833     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
1834        which is apparently normally available under Windows.
1835    
1836     2. Re-jig the pcregrep tests with different newline settings in an attempt
1837        to make them independent of the local environment's newline setting.
1838    
1839     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
1840    
1841     4. Some of the "internals" tests were previously cut out when the link size
1842        was not 2, because the output contained actual offsets. The recent new
1843        "Z" feature of pcretest means that these can be cut out, making the tests
1844        usable with all link sizes.
1845    
1846     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
1847        stack recursion. This gives a massive performance boost under BSD, but just
1848        a small improvement under Linux. However, it saves one field in the frame
1849        in all cases.
1850    
1851     6. Added more features from the forthcoming Perl 5.10:
1852    
1853        (a) (?-n) (where n is a string of digits) is a relative subroutine or
1854            recursion call. It refers to the nth most recently opened parentheses.
1855    
1856        (b) (?+n) is also a relative subroutine call; it refers to the nth next
1857            to be opened parentheses.
1858    
1859        (c) Conditions that refer to capturing parentheses can be specified
1860            relatively, for example, (?(-2)... or (?(+3)...
1861    
1862        (d) \K resets the start of the current match so that everything before
1863            is not part of it.
1864    
1865        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1866    
1867        (f) \g{name} is another synonym - part of Perl 5.10's unification of
1868            reference syntax.
1869    
1870        (g) (?| introduces a group in which the numbering of parentheses in each
1871            alternative starts with the same number.
1872    
1873        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
1874    
1875     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
1876        PCRE_INFO_JCHANGED.
1877    
1878     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
1879        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
1880        for detecting groups that can match an empty string.
1881    
1882     9. A pattern with a very large number of alternatives (more than several
1883        hundred) was running out of internal workspace during the pre-compile
1884        phase, where pcre_compile() figures out how much memory will be needed. A
1885        bit of new cunning has reduced the workspace needed for groups with
1886        alternatives. The 1000-alternative test pattern now uses 12 bytes of
1887        workspace instead of running out of the 4096 that are available.
1888    
1889    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
1890    
1891    11. Applied patch from Google to remove an optimization that didn't quite work.
1892        The report of the bug said:
1893    
1894          pcrecpp::RE("a*").FullMatch("aaa") matches, while
1895          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
1896          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
1897    
1898    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
1899        it matched the wrong number of bytes.
1900    
1901    
1902    Version 7.1 24-Apr-07
1903    ---------------------
1904    
1905     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
1906        that is more "standard", making use of automake and other Autotools. There
1907        is some re-arrangement of the files and adjustment of comments consequent
1908        on this.
1909    
1910     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
1911        for recursive directory scanning broke on some systems because the files
1912        are not scanned in any specific order and on different systems the order
1913        was different. A call to "sort" has been inserted into RunGrepTest for the
1914        approprate test as a short-term fix. In the longer term there may be an
1915        alternative.
1916    
1917     3. I had an email from Eric Raymond about problems translating some of PCRE's
1918        man pages to HTML (despite the fact that I distribute HTML pages, some
1919        people do their own conversions for various reasons). The problems
1920        concerned the use of low-level troff macros .br and .in. I have therefore
1921        removed all such uses from the man pages (some were redundant, some could
1922        be replaced by .nf/.fi pairs). The 132html script that I use to generate
1923        HTML has been updated to handle .nf/.fi and to complain if it encounters
1924        .br or .in.
1925    
1926     4. Updated comments in configure.ac that get placed in config.h.in and also
1927        arranged for config.h to be included in the distribution, with the name
1928        config.h.generic, for the benefit of those who have to compile without
1929        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
1930    
1931     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
1932        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
1933        makevp.bat for the current PCRE, using the additional files
1934        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
1935    
1936     6. A Windows user reported a minor discrepancy with test 2, which turned out
1937        to be caused by a trailing space on an input line that had got lost in his
1938        copy. The trailing space was an accident, so I've just removed it.
1939    
1940     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
1941        that is needed.
1942    
1943     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
1944        as "const" (a) because they are and (b) because it helps the PHP
1945        maintainers who have recently made a script to detect big data structures
1946        in the php code that should be moved to the .rodata section. I remembered
1947        to update Builducptable as well, so it won't revert if ucptable.h is ever
1948        re-created.
1949    
1950     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
1951        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
1952        order to be able to cut out the UTF-8 tables in the latter when UTF-8
1953        support is not required. This saves 1.5-2K of code, which is important in
1954        some applications.
1955    
1956        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
1957        so as not to refer to the tables, even though these functions will never be
1958        called when UTF-8 support is disabled. Otherwise there are problems with a
1959        shared library.
1960    
1961    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
1962    
1963        (a) It was defining its arguments as char * instead of void *.
1964    
1965        (b) It was assuming that all moves were upwards in memory; this was true
1966            a long time ago when I wrote it, but is no longer the case.
1967    
1968        The emulated memove() is provided for those environments that have neither
1969        memmove() nor bcopy(). I didn't think anyone used it these days, but that
1970        is clearly not the case, as these two bugs were recently reported.
1971    
1972    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
1973        and Detrail to create the HTML documentation, the .txt form of the man
1974        pages, and it removes trailing spaces from listed files. It also creates
1975        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
1976        case, it wraps all the #defines with #ifndefs. This script should be run
1977        before "make dist".
1978    
1979    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1980        with Unicode property support.
1981    
1982        (a) For a maximizing quantifier, if the two different cases of the
1983            character were of different lengths in their UTF-8 codings (there are
1984            some cases like this - I found 11), and the matching function had to
1985            back up over a mixture of the two cases, it incorrectly assumed they
1986            were both the same length.
1987    
1988        (b) When PCRE was configured to use the heap rather than the stack for
1989            recursion during matching, it was not correctly preserving the data for
1990            the other case of a UTF-8 character when checking ahead for a match
1991            while processing a minimizing repeat. If the check also involved
1992            matching a wide character, but failed, corruption could cause an
1993            erroneous result when trying to check for a repeat of the original
1994            character.
1995    
1996    13. Some tidying changes to the testing mechanism:
1997    
1998        (a) The RunTest script now detects the internal link size and whether there
1999            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
2000            values substituted by "configure". (The RunGrepTest script already did
2001            this for UTF-8.) The configure.ac script no longer substitutes the
2002            relevant variables.
2003    
2004        (b) The debugging options /B and /D in pcretest show the compiled bytecode
2005            with length and offset values. This means that the output is different
2006            for different internal link sizes. Test 2 is skipped for link sizes
2007            other than 2 because of this, bypassing the problem. Unfortunately,
2008            there was also a test in test 3 (the locale tests) that used /B and
2009            failed for link sizes other than 2. Rather than cut the whole test out,
2010            I have added a new /Z option to pcretest that replaces the length and
2011            offset values with spaces. This is now used to make test 3 independent
2012            of link size. (Test 2 will be tidied up later.)
2013    
2014    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
2015        segmentation fault instead of returning the appropriate error message.
2016    
2017    15. In multiline mode when the newline sequence was set to "any", the pattern
2018        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
2019        This doesn't seem right; it now treats the CRLF combination as the line
2020        ending, and so does not match in that case. It's only a pattern such as ^$
2021        that would hit this one: something like ^ABC$ would have failed after \r
2022        and then tried again after \r\n.
2023    
2024    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
2025        in an attempt to make files that differ only in their line terminators
2026        compare equal. This works on Linux.
2027    
2028    17. Under certain error circumstances pcregrep might try to free random memory
2029        as it exited. This is now fixed, thanks to valgrind.
2030    
2031    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
2032        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
2033        was because its rules for how to advance for /g after matching an empty
2034        string at the end of a line did not allow for this case. They now check for
2035        it specially.
2036    
2037    20. pcretest is supposed to handle patterns and data of any length, by
2038        extending its buffers when necessary. It was getting this wrong when the
2039        buffer for a data line had to be extended.
2040    
2041    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
2042        CRLF as a newline sequence.
2043    
2044    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
2045        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
2046        I have nevertheless tidied it up.
2047    
2048    23. Added some casts to kill warnings from HP-UX ia64 compiler.
2049    
2050    24. Added a man page for pcre-config.
2051    
2052    
2053    Version 7.0 19-Dec-06
2054    ---------------------
2055    
2056     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
2057        moving to gcc 4.1.1.
2058    
2059     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
2060        sys/time.h, which is documented as needed for this function. It doesn't
2061        seem to matter on Linux, but it showed up on some releases of OS X.
2062    
2063     3. It seems that there are systems where bytes whose values are greater than
2064        127 match isprint() in the "C" locale. The "C" locale should be the
2065        default when a C program starts up. In most systems, only ASCII printing
2066        characters match isprint(). This difference caused the output from pcretest
2067        to vary, making some of the tests fail. I have changed pcretest so that:
2068    
2069        (a) When it is outputting text in the compiled version of a pattern, bytes
2070            other than 32-126 are always shown as hex escapes.
2071    
2072        (b) When it is outputting text that is a matched part of a subject string,
2073            it does the same, unless a different locale has been set for the match
2074            (using the /L modifier). In this case, it uses isprint() to decide.
2075    
2076     4. Fixed a major bug that caused incorrect computation of the amount of memory
2077        required for a compiled pattern when options that changed within the
2078        pattern affected the logic of the preliminary scan that determines the
2079        length. The relevant options are -x, and -i in UTF-8 mode. The result was
2080        that the computed length was too small. The symptoms of this bug were
2081        either the PCRE error "internal error: code overflow" from pcre_compile(),
2082        or a glibc crash with a message such as "pcretest: free(): invalid next
2083        size (fast)". Examples of patterns that provoked this bug (shown in
2084        pcretest format) are:
2085    
2086          /(?-x: )/x
2087          /(?x)(?-x: \s*#\s*)/
2088          /((?i)[\x{c0}])/8
2089          /(?i:[\x{c0}])/8
2090    
2091        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
2092        is now done differently.
2093    
2094     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
2095        wrapper classes; (b) implement a new function in the C++ scanner that is
2096        more efficient than the old way of doing things because it avoids levels of
2097        recursion in the regex matching; (c) add a paragraph to the documentation
2098        for the FullMatch() function.
2099    
2100     6. The escape sequence \n was being treated as whatever was defined as
2101        "newline". Not only was this contrary to the documentation, which states
2102        that \n is character 10 (hex 0A), but it also went horribly wrong when
2103        "newline" was defined as CRLF. This has been fixed.
2104    
2105     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
2106        was being set to -1 for the "end of line" case (supposedly a value that no
2107        character can have). Though this value is never used (the check for end of
2108        line is "zero bytes in current character"), it caused compiler complaints.
2109        I've changed it to 0xffffffff.
2110    
2111     8. In pcre_version.c, the version string was being built by a sequence of
2112        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
2113        string (as it is for production releases) called a macro with an empty
2114        argument. The C standard says the result of this is undefined. The gcc
2115        compiler treats it as an empty string (which was what was wanted) but it is
2116        reported that Visual C gives an error. The source has been hacked around to
2117        avoid this problem.
2118    
2119     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
2120        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
2121        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
2122        of them did).
2123    
2124    10. Originally, pcretest opened its input and output without "b"; then I was
2125        told that "b" was needed in some environments, so it was added for release
2126        5.0 to both the input and output. (It makes no difference on Unix-like
2127        systems.) Later I was told that it is wrong for the input on Windows. I've
2128        now abstracted the modes into two macros, to make it easier to fiddle with
2129        them, and removed "b" from the input mode under Windows.
2130    
2131    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
2132    
2133    12. Added -help and --help to pcretest as an official way of being reminded
2134        of the options.
2135    
2136    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
2137        and pcrecpp.cc because they annoy compilers at high warning levels.
2138    
2139    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
2140    
2141    15. Fixed an occurrence of == in configure.ac that should have been = (shell
2142        scripts are not C programs :-) and which was not noticed because it works
2143        on Linux.
2144    
2145    16. pcretest is supposed to handle any length of pattern and data line (as one
2146        line or as a continued sequence of lines) by extending its input buffer if
2147        necessary. This feature was broken for very long pattern lines, leading to
2148        a string of junk being passed to pcre_compile() if the pattern was longer
2149        than about 50K.
2150    
2151    17. I have done a major re-factoring of the way pcre_compile() computes the
2152        amount of memory needed for a compiled pattern. Previously, there was code
2153        that made a preliminary scan of the pattern in order to do this. That was
2154        OK when PCRE was new, but as the facilities have expanded, it has become
2155        harder and harder to keep it in step with the real compile phase, and there
2156        have been a number of bugs (see for example, 4 above). I have now found a
2157        cunning way of running the real compile function in a "fake" mode that
2158        enables it to compute how much memory it would need, while actually only
2159        ever using a few hundred bytes of working memory and without too many
2160        tests of the mode. This should make future maintenance and development
2161        easier. A side effect of this work is that the limit of 200 on the nesting
2162        depth of parentheses has been removed (though this was never a serious
2163        limitation, I suspect). However, there is a downside: pcre_compile() now
2164        runs more slowly than before (30% or more, depending on the pattern). I
2165        hope this isn't a big issue. There is no effect on runtime performance.
2166    
2167    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
2168        newline (only possible for the last line of a file) and it was a
2169        pattern that set a locale (followed by /Lsomething), pcretest crashed.
2170    
2171    19. Added additional timing features to pcretest. (1) The -tm option now times
2172        matching only, not compiling. (2) Both -t and -tm can be followed, as a
2173        separate command line item, by a number that specifies the number of
2174        repeats to use when timing. The default is 50000; this gives better
2175        precision, but takes uncomfortably long for very large patterns.
2176    
2177    20. Extended pcre_study() to be more clever in cases where a branch of a
2178        subpattern has no definite first character. For example, (a*|b*)[cd] would
2179        previously give no result from pcre_study(). Now it recognizes that the
2180        first character must be a, b, c, or d.
2181    
2182    21. There was an incorrect error "recursive call could loop indefinitely" if
2183        a subpattern (or the entire pattern) that was being tested for matching an
2184        empty string contained only one non-empty item after a nested subpattern.
2185        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
2186        incorrectly, because the \d was being skipped in the check.
2187    
2188    22. The pcretest program now has a new pattern option /B and a command line
2189        option -b, which is equivalent to adding /B to every pattern. This causes
2190        it to show the compiled bytecode, without the additional information that
2191        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
2192        is the same as /B/I).
2193    
2194    23. A new optimization is now able automatically to treat some sequences such
2195        as a*b as a*+b. More specifically, if something simple (such as a character
2196        or a simple class like \d) has an unlimited quantifier, and is followed by
2197        something that cannot possibly match the quantified thing, the quantifier
2198        is automatically "possessified".
2199    
2200    24. A recursive reference to a subpattern whose number was greater than 39
2201        went wrong under certain circumstances in UTF-8 mode. This bug could also
2202        have affected the operation of pcre_study().
2203    
2204    25. Realized that a little bit of performance could be had by replacing
2205        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
2206    
2207    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
2208    
2209    27. Possessive quantifiers such as a++ were previously implemented by turning
2210        them into atomic groups such as ($>a+). Now they have their own opcodes,
2211        which improves performance. This includes the automatically created ones
2212        from 23 above.
2213    
2214    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
2215        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
2216        the first matched character to be a colon. This applied both to named and
2217        numbered groups.
2218    
2219    29. The ucpinternal.h header file was missing its idempotency #ifdef.
2220    
2221    30. I was sent a "project" file called libpcre.a.dev which I understand makes
2222        building PCRE on Windows easier, so I have included it in the distribution.
2223    
2224    31. There is now a check in pcretest against a ridiculously large number being
2225        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
2226        loop, the loop is abandoned.
2227    
2228    32. Forward references to subpatterns in conditions such as (?(2)...) where
2229        subpattern 2 is defined later cause pcre_compile() to search forwards in
2230        the pattern for the relevant set of parentheses. This search went wrong
2231        when there were unescaped parentheses in a character class, parentheses
2232        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
2233    
2234    33. "Subroutine" calls and backreferences were previously restricted to
2235        referencing subpatterns earlier in the regex. This restriction has now
2236        been removed.
2237    
2238    34. Added a number of extra features that are going to be in Perl 5.10. On the
2239        whole, these are just syntactic alternatives for features that PCRE had
2240        previously implemented using the Python syntax or my own invention. The
2241        other formats are all retained for compatibility.
2242    
2243        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
2244            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
2245            also .NET compatible.
2246    
2247        (b) A recursion or subroutine call to a named group can now be defined as
2248            (?&name) as well as (?P>name).
2249    
2250        (c) A backreference to a named group can now be defined as \k<name> or
2251            \k'name' as well as (?P=name). The new forms, as well as being in Perl
2252            5.10, are also .NET compatible.
2253    
2254        (d) A conditional reference to a named group can now use the syntax
2255            (?(<name>) or (?('name') as well as (?(name).
2256    
2257        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
2258            groups (named and numbered) that are never evaluated inline, but can be
2259            called as "subroutines" from elsewhere. In effect, the DEFINE condition
2260            is always false. There may be only one alternative in such a group.
2261    
2262        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
2263            as the simple (?(R). The condition is true only if the most recent
2264            recursion is that of the given number or name. It does not search out
2265            through the entire recursion stack.
2266    
2267        (g) The escape \gN or \g{N} has been added, where N is a positive or
2268            negative number, specifying an absolute or relative reference.
2269    
2270    35. Tidied to get rid of some further signed/unsigned compiler warnings and
2271        some "unreachable code" warnings.
2272    
2273    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
2274        things, this adds five new scripts.
2275    
2276    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
2277        There were also incompatibilities regarding the handling of \Q..\E inside
2278        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
2279        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
2280    
2281    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
2282        matches an empty string, and forcibly breaks the loop. There were bugs in
2283        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
2284        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
2285        separate and independent bugs (that affected different cases) have been
2286        fixed.
2287    
2288    39. Refactored the code to abolish the use of different opcodes for small
2289        capturing bracket numbers. This is a tidy that I avoided doing when I
2290        removed the limit on the number of capturing brackets for 3.5 back in 2001.
2291        The new approach is not only tidier, it makes it possible to reduce the
2292        memory needed to fix the previous bug (38).
2293    
2294    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
2295        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
2296        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
2297        mode.
2298    
2299    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
2300        report.
2301    
2302    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
2303        copy construction and assignment in the C++ wrapper.
2304    
2305    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
2306        couple of bugs that could have given wrong results in the "--newline=crlf"
2307        case.
2308    
2309    44. Added a number of casts and did some reorganization of signed/unsigned int
2310        variables following suggestions from Dair Grant. Also renamed the variable
2311        "this" as "item" because it is a C++ keyword.
2312    
2313    45. Arranged for dftables to add
2314    
2315          #include "pcre_internal.h"
2316    
2317        to pcre_chartables.c because without it, gcc 4.x may remove the array
2318        definition from the final binary if PCRE is built into a static library and
2319        dead code stripping is activated.
2320    
2321    46. For an unanchored pattern, if a match attempt fails at the start of a
2322        newline sequence, and the newline setting is CRLF or ANY, and the next two
2323        characters are CRLF, advance by two characters instead of one.
2324    
2325    
2326    Version 6.7 04-Jul-06
2327    ---------------------
2328    
2329     1. In order to handle tests when input lines are enormously long, pcretest has
2330        been re-factored so that it automatically extends its buffers when
2331        necessary. The code is crude, but this _is_ just a test program. The
2332        default size has been increased from 32K to 50K.
2333    
2334     2. The code in pcre_study() was using the value of the re argument before
2335        testing it for NULL. (Of course, in any sensible call of the function, it
2336        won't be NULL.)
2337    
2338     3. The memmove() emulation function in pcre_internal.h, which is used on
2339        systems that lack both memmove() and bcopy() - that is, hardly ever -
2340        was missing a "static" storage class specifier.
2341    
2342     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
2343        containing an extended class (one that cannot be represented by a bitmap
2344        because it contains high-valued characters or Unicode property items, e.g.
2345        [\pZ]). Almost always one would set UTF-8 mode when processing such a
2346        pattern, but PCRE should not loop if you do not (it no longer does).
2347        [Detail: two cases were found: (a) a repeated subpattern containing an
2348        extended class; (b) a recursive reference to a subpattern that followed a
2349        previous extended class. It wasn't skipping over the extended class
2350        correctly when UTF-8 mode was not set.]
2351    
2352     5. A negated single-character class was not being recognized as fixed-length
2353        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
2354        compile error "lookbehind assertion is not fixed length".
2355    
2356     6. The RunPerlTest auxiliary script was showing an unexpected difference
2357        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
2358        write a Perl script that can interpret lines of an input file either as
2359        byte characters or as UTF-8, which is what "perltest" was being required to
2360        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
2361        can't do is switch easily at run time between having the "use utf8;" pragma
2362        or not. In the end, I fudged it by using the RunPerlTest script to insert
2363        "use utf8;" explicitly for the UTF-8 tests.
2364    
2365     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
2366        the end of the subject string, contrary to the documentation and to what
2367        Perl does. This was true of both matching functions. Now it matches only at
2368        the start of the subject and immediately after *internal* newlines.
2369    
2370     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
2371        a pointer to an int instead of a pointer to an unsigned long int. This
2372        caused problems on 64-bit systems.
2373    
2374     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
2375        instance of the 'standard' template library not being so standard".
2376    
2377    10. There was no check on the number of named subpatterns nor the maximum
2378        length of a subpattern name. The product of these values is used to compute
2379        the size of the memory block for a compiled pattern. By supplying a very
2380        long subpattern name and a large number of named subpatterns, the size
2381        computation could be caused to overflow. This is now prevented by limiting
2382        the length of names to 32 characters, and the number of named subpatterns
2383        to 10,000.
2384    
2385    11. Subpatterns that are repeated with specific counts have to be replicated in
2386        the compiled pattern. The size of memory for this was computed from the
2387        length of the subpattern and the repeat count. The latter is limited to
2388        65535, but there was no limit on the former, meaning that integer overflow
2389        could in principle occur. The compiled length of a repeated subpattern is
2390        now limited to 30,000 bytes in order to prevent this.
2391    
2392    12. Added the optional facility to have named substrings with the same name.
2393    
2394    13. Added the ability to use a named substring as a condition, using the
2395        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
2396        are numbers (not recommended). Forward references are permitted.
2397    
2398    14. Added forward references in named backreferences (if you see what I mean).
2399    
2400    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
2401        pattern could run off the end of the subject. For example, the pattern
2402        "(?s)(.{1,5})"8 did this with the subject "ab".
2403    
2404    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
2405        PCRE_CASELESS was set when matching characters that were quantified with ?
2406        or *.
2407    
2408    17. A character class other than a single negated character that had a minimum
2409        but no maximum quantifier - for example [ab]{6,} - was not handled
2410        correctly by pce_dfa_exec(). It would match only one character.
2411    
2412    18. A valid (though odd) pattern that looked like a POSIX character
2413        class but used an invalid character after [ (for example [[,abc,]]) caused
2414        pcre_compile() to give the error "Failed: internal error: code overflow" or
2415        in some cases to crash with a glibc free() error. This could even happen if
2416        the pattern terminated after [[ but there just happened to be a sequence of
2417        letters, a binary zero, and a closing ] in the memory that followed.
2418    
2419    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
2420        over the years. Originally (before any Unicode support), just the bottom 8
2421        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
2422        output from "man perlunicode" includes this:
2423    
2424          The regular expression compiler produces polymorphic opcodes.  That
2425          is, the pattern adapts to the data and automatically switches to
2426          the Unicode character scheme when presented with Unicode data--or
2427          instead uses a traditional byte scheme when presented with byte
2428          data.
2429    
2430        Sadly, a wide octal escape does not cause a switch, and in a string with
2431        no other multibyte characters, these octal escapes are treated as before.
2432        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
2433        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
2434        Unicode string.
2435    
2436        I have not perpetrated such confusion in PCRE. Up till now, it took just
2437        the bottom 8 bits, as in old Perl. I have now made octal escapes with
2438        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
2439        translate to the appropriate multibyte character.
2440    
2441    29. Applied some refactoring to reduce the number of warnings from Microsoft
2442        and Borland compilers. This has included removing the fudge introduced
2443        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
2444        a warning about an unused variable.
2445    
2446    21. PCRE has not included VT (character 0x0b) in the set of whitespace
2447        characters since release 4.0, because Perl (from release 5.004) does not.
2448        [Or at least, is documented not to: some releases seem to be in conflict
2449        with the documentation.] However, when a pattern was studied with
2450        pcre_study() and all its branches started with \s, PCRE still included VT
2451        as a possible starting character. Of course, this did no harm; it just
2452        caused an unnecessary match attempt.
2453    
2454    22. Removed a now-redundant internal flag bit that recorded the fact that case
2455        dependency changed within the pattern. This was once needed for "required
2456        byte" processing, but is no longer used. This recovers a now-scarce options
2457        bit. Also moved the least significant internal flag bit to the most-
2458        significant bit of the word, which was not previously used (hangover from
2459        the days when it was an int rather than a uint) to free up another bit for
2460        the future.
2461    
2462    23. Added support for CRLF line endings as well as CR and LF. As well as the
2463        default being selectable at build time, it can now be changed at runtime
2464        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
2465        specify that it is scanning data with non-default line endings.
2466    
2467    24. Changed the definition of CXXLINK to make it agree with the definition of
2468        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
2469    
2470    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
2471        recursions. This makes a big different to stack usage for some patterns.
2472    
2473    26. If a subpattern containing a named recursion or subroutine reference such
2474        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
2475        the space required for the compiled pattern went wrong and gave too small a
2476        value. Depending on the environment, this could lead to "Failed: internal
2477        error: code overflow at offset 49" or "glibc detected double free or
2478        corruption" errors.
2479    
2480    27. Applied patches from Google (a) to support the new newline modes and (b) to
2481        advance over multibyte UTF-8 characters in GlobalReplace.
2482    
2483    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
2484        difference for some implementation of PCRE in some Windows version.
2485    
2486    29. Added some extra testing facilities to pcretest:
2487    
2488        \q<number>   in a data line sets the "match limit" value
2489        \Q<number>   in a data line sets the "match recursion limt" value
2490        -S <number>  sets the stack size, where <number> is in megabytes
2491    
2492        The -S option isn't available for Windows.
2493    
2494    
2495    Version 6.6 06-Feb-06
2496    ---------------------
2497    
2498     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
2499        in pcreposix.h. I have copied the definition from pcre.h.
2500    
2501     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
2502        because pcre.h is no longer a built file.
2503    
2504     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
2505        not normally included in the compiled code.
2506    
2507    
2508    Version 6.5 01-Feb-06
2509    ---------------------
2510    
2511     1. When using the partial match feature with pcre_dfa_exec(), it was not
2512        anchoring the second and subsequent partial matches at the new starting
2513        point. This could lead to incorrect results. For example, with the pattern
2514        /1234/, partially matching against "123" and then "a4" gave a match.
2515    
2516     2. Changes to pcregrep:
2517    
2518        (a) All non-match returns from pcre_exec() were being treated as failures
2519            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
2520            error message is output. Some extra information is given for the
2521            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
2522            probably the only errors that are likely to be caused by users (by
2523            specifying a regex that has nested indefinite repeats, for instance).
2524            If there are more than 20 of these errors, pcregrep is abandoned.
2525    
2526        (b) A binary zero was treated as data while matching, but terminated the
2527            output line if it was written out. This has been fixed: binary zeroes
2528            are now no different to any other data bytes.
2529    
2530        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
2531            used to set a locale for matching. The --locale=xxxx long option has
2532            been added (no short equivalent) to specify a locale explicitly on the
2533            pcregrep command, overriding the environment variables.
2534    
2535        (d) When -B was used with -n, some line numbers in the output were one less
2536            than they should have been.
2537    
2538        (e) Added the -o (--only-matching) option.
2539    
2540        (f) If -A or -C was used with -c (count only), some lines of context were
2541            accidentally printed for the final match.
2542    
2543        (g) Added the -H (--with-filename) option.
2544    
2545        (h) The combination of options -rh failed to suppress file names for files
2546            that were found from directory arguments.
2547    
2548        (i) Added the -D (--devices) and -d (--directories) options.
2549    
2550        (j) Added the -F (--fixed-strings) option.
2551    
2552        (k) Allow "-" to be used as a file name for -f as well as for a data file.
2553    
2554        (l) Added the --colo(u)r option.
2555    
2556        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
2557            is not present by default.
2558    
2559     3. A nasty bug was discovered in the handling of recursive patterns, that is,
2560        items such as (?R) or (?1), when the recursion could match a number of
2561        alternatives. If it matched one of the alternatives, but subsequently,
2562        outside the recursion, there was a failure, the code tried to back up into
2563        the recursion. However, because of the way PCRE is implemented, this is not
2564        possible, and the result was an incorrect result from the match.
2565    
2566        In order to prevent this happening, the specification of recursion has
2567        been changed so that all such subpatterns are automatically treated as
2568        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
2569    
2570     4. I had overlooked the fact that, in some locales, there are characters for
2571        which isalpha() is true but neither isupper() nor islower() are true. In
2572        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
2573        and ordfeminine) are like this. This affected the treatment of \w and \W
2574        when they appeared in character classes, but not when they appeared outside
2575        a character class. The bit map for "word" characters is now created
2576        separately from the results of isalnum() instead of just taking it from the
2577        upper, lower, and digit maps. (Plus the underscore character, of course.)
2578    
2579     5. The above bug also affected the handling of POSIX character classes such as
2580        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
2581        permanent tables. Instead, the bit maps for such a class were previously
2582        created as the appropriate unions of the upper, lower, and digit bitmaps.
2583        Now they are created by subtraction from the [[:word:]] class, which has
2584        its own bitmap.
2585    
2586     6. The [[:blank:]] character class matches horizontal, but not vertical space.
2587        It is created by subtracting the vertical space characters (\x09, \x0a,
2588        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
2589        subtraction was done in the overall bitmap for a character class, meaning
2590        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
2591        be recognized. This bug has been fixed.
2592    
2593     7. Patches from the folks at Google:
2594    
2595          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
2596          real life, but is still worth protecting against".
2597    
2598          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
2599          regular expressions".
2600    
2601          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
2602          have it.
2603    
2604          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
2605          "configure" and the latter not, in order to fix a problem somebody had
2606          with compiling the Arg class on HP-UX.
2607    
2608          (e) Improve the error-handling of the C++ wrapper a little bit.
2609    
2610          (f) New tests for checking recursion limiting.
2611    
2612     8. The pcre_memmove() function, which is used only if the environment does not
2613        have a standard memmove() function (and is therefore rarely compiled),
2614        contained two bugs: (a) use of int instead of size_t, and (b) it was not
2615        returning a result (though PCRE never actually uses the result).
2616    
2617     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
2618        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
2619        returned instead of calling malloc() with an overflowing number that would
2620        most likely cause subsequent chaos.
2621    
2622    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
2623    
2624    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
2625        with this option is matched, the nmatch and pmatch options of regexec() are
2626        ignored.
2627    
2628    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
2629        provided in case anyone wants to the the POSIX interface with UTF-8
2630        strings.
2631    
2632    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
2633        C++ linking (needed for some HP-UX environments).
2634    
2635    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
2636        (unused parameter) and in the pcre_printint() function (omitted "default"
2637        switch label when the default is to do nothing).
2638    
2639    15. Added some code to make it possible, when PCRE is compiled as a C++
2640        library, to replace subject pointers for pcre_exec() with a smart pointer
2641        class, thus making it possible to process discontinuous strings.
2642    
2643    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
2644        much the same function. They were added by different people who were trying
2645        to make PCRE easy to compile on non-Unix systems. It has been suggested
2646        that PCRE_EXPORT be abolished now that there is more automatic apparatus
2647        for compiling on Windows systems. I have therefore replaced it with
2648        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
2649        defaults to "extern" for C or "extern C" for C++, which works fine on
2650        Unix-like systems. It is now possible to override the value of PCRE_DATA_
2651        SCOPE with something explicit in config.h. In addition:
2652    
2653        (a) pcreposix.h still had just "extern" instead of either of these macros;
2654            I have replaced it with PCRE_DATA_SCOPE.
2655    
2656        (b) Functions such as _pcre_xclass(), which are internal to the library,
2657            but external in the C sense, all had PCRE_EXPORT in their definitions.
2658            This is apparently wrong for the Windows case, so I have removed it.
2659            (It makes no difference on Unix-like systems.)
2660    
2661    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
2662        of recursive calls to match(). This is different to MATCH_LIMIT because
2663        that limits the total number of calls to match(), not all of which increase
2664        the depth of recursion. Limiting the recursion depth limits the amount of
2665        stack (or heap if NO_RECURSE is set) that is used. The default can be set
2666        when PCRE is compiled, and changed at run time. A patch from Google adds
2667        this functionality to the C++ interface.
2668    
2669    18. Changes to the handling of Unicode character properties:
2670    
2671        (a) Updated the table to Unicode 4.1.0.
2672    
2673        (b) Recognize characters that are not in the table as "Cn" (undefined).
2674    
2675        (c) I revised the way the table is implemented to a much improved format
2676            which includes recognition of ranges. It now supports the ranges that
2677            are defined in UnicodeData.txt, and it also amalgamates other
2678            characters into ranges. This has reduced the number of entries in the
2679            table from around 16,000 to around 3,000, thus reducing its size
2680            considerably. I realized I did not need to use a tree structure after
2681            all - a binary chop search is just as efficient. Having reduced the
2682            number of entries, I extended their size from 6 bytes to 8 bytes to
2683            allow for more data.
2684    
2685        (d) Added support for Unicode script names via properties such as \p{Han}.
2686    
2687    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
2688        matching that character.
2689    
2690    20. When matching a repeated Unicode property with a minimum greater than zero,
2691        (for example \pL{2,}), PCRE could look past the end of the subject if it
2692        reached it while seeking the minimum number of characters. This could
2693        happen only if some of the characters were more than one byte long, because
2694        there is a check for at least the minimum number of bytes.
2695    
2696    21. Refactored the implementation of \p and \P so as to be more general, to
2697        allow for more different types of property in future. This has changed the
2698        compiled form incompatibly. Anybody with saved compiled patterns that use
2699        \p or \P will have to recompile them.
2700    
2701    22. Added "Any" and "L&" to the supported property types.
2702    
2703    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
2704        but give a compile time error if the value is greater than 0xff.
2705    
2706    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
2707        accidentally not being installed or uninstalled.
2708    
2709    25. The pcre.h file was built from pcre.h.in, but the only changes that were
2710        made were to insert the current release number. This seemed silly, because
2711        it made things harder for people building PCRE on systems that don't run
2712        "configure". I have turned pcre.h into a distributed file, no longer built
2713        by "configure", with the version identification directly included. There is
2714        no longer a pcre.h.in file.
2715    
2716        However, this change necessitated a change to the pcre-config script as
2717        well. It is built from pcre-config.in, and one of the substitutions was the
2718        release number. I have updated configure.ac so that ./configure now finds
2719        the release number by grepping pcre.h.
2720    
2721    26. Added the ability to run the tests under valgrind.
2722    
2723    
2724    Version 6.4 05-Sep-05
2725    ---------------------
2726    
2727     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
2728        "--" to be printed when multiple files were scanned, even when none of the
2729        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
2730        consider it to be a bug, and have restored the previous behaviour.
2731    
2732     2. A couple of code tidies to get rid of compiler warnings.
2733    
2734     3. The pcretest program used to cheat by referring to symbols in the library
2735        whose names begin with _pcre_. These are internal symbols that are not
2736        really supposed to be visible externally, and in some environments it is
2737        possible to suppress them. The cheating is now confined to including
2738        certain files from the library's source, which is a bit cleaner.
2739    
2740     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
2741        file's purpose clearer.
2742    
2743     5. Reorganized pcre_ucp_findchar().
2744    
2745    
2746    Version 6.3 15-Aug-05
2747    ---------------------
2748    
2749     1. The file libpcre.pc.in did not have general read permission in the tarball.
2750    
2751     2. There were some problems when building without C++ support:
2752    
2753        (a) If C++ support was not built, "make install" and "make test" still
2754            tried to test it.
2755    
2756        (b) There were problems when the value of CXX was explicitly set. Some
2757            changes have been made to try to fix these, and ...
2758    
2759        (c) --disable-cpp can now be used to explicitly disable C++ support.
2760    
2761        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
2762            backslash in a target when C++ was disabled. This confuses some
2763            versions of "make", apparently. Using an intermediate variable solves
2764            this. (Same for CPP_LOBJ.)
2765    
2766     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
2767        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
2768        necessary on certain architectures.
2769    
2770     4. Added a setting of -export-symbols-regex to the link command to remove
2771        those symbols that are exported in the C sense, but actually are local
2772        within the library, and not documented. Their names all begin with
2773        "_pcre_". This is not a perfect job, because (a) we have to except some
2774        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
2775        available (and never for static libraries). I have made a note to try to
2776        find a way round (a) in the future.
2777    
2778    
2779    Version 6.2 01-Aug-05
2780    ---------------------
2781    
2782     1. There was no test for integer overflow of quantifier values. A construction
2783        such as {1111111111111111} would give undefined results. What is worse, if
2784        a minimum quantifier for a parenthesized subpattern overflowed and became
2785        negative, the calculation of the memory size went wrong. This could have
2786        led to memory overwriting.
2787    
2788     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
2789    
2790     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
2791        operating environments where this matters.
2792    
2793     4. Applied Giuseppe Maxia's patch to add additional features for controlling
2794        PCRE options from within the C++ wrapper.
2795    
2796     5. Named capturing subpatterns were not being correctly counted when a pattern
2797        was compiled. This caused two problems: (a) If there were more than 100
2798        such subpatterns, the calculation of the memory needed for the whole
2799        compiled pattern went wrong, leading to an overflow error. (b) Numerical
2800        back references of the form \12, where the number was greater than 9, were
2801        not recognized as back references, even though there were sufficient
2802        previous subpatterns.
2803    
2804     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
2805        versions of gcc, e.g. 2.95.4.
2806    
2807    
2808    Version 6.1 21-Jun-05
2809    ---------------------
2810    
2811     1. There was one reference to the variable "posix" in pcretest.c that was not
2812        surrounded by "#if !defined NOPOSIX".
2813    
2814     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
2815        the cross-check on the old pcre_info() function, for the benefit of the
2816        cut-down version of PCRE that is currently imported into Exim.
2817    
2818     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
2819        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
2820        patterns that start (?i) but I don't think that matters. The use of (?i) is
2821        just an example; this all applies to the other options as well.
2822    
2823     4. Since libtool seems to echo the compile commands it is issuing, the output
2824        from "make" can be reduced a bit by putting "@" in front of each libtool
2825        compile command.
2826    
2827     5. Patch from the folks at Google for configure.in to be a bit more thorough
2828        in checking for a suitable C++ installation before trying to compile the
2829        C++ stuff. This should fix a reported problem when a compiler was present,
2830        but no suitable headers.
2831    
2832     6. The man pages all had just "PCRE" as their title. I have changed them to
2833        be the relevant file name. I have also arranged that these names are
2834        retained in the file doc/pcre.txt, which is a concatenation in text format
2835        of all the man pages except the little individual ones for each function.
2836    
2837     7. The NON-UNIX-USE file had not been updated for the different set of source
2838        files that come with release 6. I also added a few comments about the C++
2839        wrapper.
2840    
2841    
2842  Version 6.0 07-Jun-05  Version 6.0 07-Jun-05
2843  ---------------------  ---------------------
2844    

Legend:
Removed from v.77  
changed lines
  Added in v.792

  ViewVC Help
Powered by ViewVC 1.1.5