ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 11 by nigel, Sat Feb 24 21:38:17 2007 UTC revision 944 by ph10, Tue Feb 28 16:49:21 2012 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
4    Version 8.31
5    -----------------------------
7    1.  Fixing a wrong JIT test case and some compiler warnings.
9    2.  Removed a bashism from the RunTest script.
11    3.  Add a cast to pcre_exec.c to fix the warning "unary minus operator applied
12        to unsigned type, result still unsigned" that was given by an MS compiler
13        on encountering the code "-sizeof(xxx)".
15    4.  Partial matching support is added to the JIT compiler.
17    5.  Fixed several bugs concerned with partial matching of items that consist
18        of more than one character:
20        (a) /^(..)\1/ did not partially match "aba" because checking references was
21            done on an "all or nothing" basis. This also applied to repeated
22            references.
24        (b) \R did not give a hard partial match if \r was found at the end of the
25            subject.
27        (c) \X did not give a hard partial match after matching one or more
28            characters at the end of the subject.
30        (d) When newline was set to CRLF, a pattern such as /a$/ did not recognize
31            a partial match for the string "\r".
33        (e) When newline was set to CRLF, the metacharacter "." did not recognize
34            a partial match for a CR character at the end of the subject string.
36    6.  If JIT is requested using /S++ or -s++ (instead of just /S+ or -s+) when
37        running pcretest, the text "(JIT)" added to the output whenever JIT is
38        actually used to run the match.
40    7.  Individual JIT compile options can be set in pcretest by following -s+[+]
41        or /S+[+] with a digit between 1 and 7.
43    8.  OP_NOT now supports any UTF character not just single-byte ones.
45    9.  (*MARK) control verb is now supported by the JIT compiler.
47    10. The command "./RunTest list" lists the available tests without actually
48        running any of them. (Because I keep forgetting what they all are.)
52    12. Applied a (slightly modified) user-supplied patch that improves performance
53        when the heap is used for recursion (compiled with --disable-stack-for-
54        recursion). Instead of malloc and free for each heap frame each time a
55        logical recursion happens, frames are retained on a chain and re-used where
56        possible. This sometimes gives as much as 30% improvement.
58    13. As documented, (*COMMIT) is now confined to within a recursive subpattern
59        call.
61    14. As documented, (*COMMIT) is now confined to within a positive assertion.
63    15. It is now possible to link pcretest with libedit as an alternative to
64        libreadline.
66    16. (*COMMIT) control verb is now supported by the JIT compiler.
68    17. The Unicode data tables have been updated to Unicode 6.1.0.
70    18. Added --file-list option to pcregrep.
73    Version 8.30 04-February-2012
74    -----------------------------
76    1.  Renamed "isnumber" as "is_a_number" because in some Mac environments this
77        name is defined in ctype.h.
79    2.  Fixed a bug in fixed-length calculation for lookbehinds that would show up
80        only in quite long subpatterns.
82    3.  Removed the function pcre_info(), which has been obsolete and deprecated
83        since it was replaced by pcre_fullinfo() in February 2000.
85    4.  For a non-anchored pattern, if (*SKIP) was given with a name that did not
86        match a (*MARK), and the match failed at the start of the subject, a
87        reference to memory before the start of the subject could occur. This bug
88        was introduced by fix 17 of release 8.21.
90    5.  A reference to an unset group with zero minimum repetition was giving
91        totally wrong answers (in non-JavaScript-compatibility mode). For example,
92        /(another)?(\1?)test/ matched against "hello world test". This bug was
93        introduced in release 8.13.
95    6.  Add support for 16-bit character strings (a large amount of work involving
96        many changes and refactorings).
98    7.  RunGrepTest failed on msys because \r\n was replaced by whitespace when the
99        command "pattern=`printf 'xxx\r\njkl'`" was run. The pattern is now taken
100        from a file.
102    8.  Ovector size of 2 is also supported by JIT based pcre_exec (the ovector size
103        rounding is not applied in this particular case).
105    9.  The invalid Unicode surrogate codepoints U+D800 to U+DFFF are now rejected
106        if they appear, or are escaped, in patterns.
108    10. Get rid of a number of -Wunused-but-set-variable warnings.
110    11. The pattern /(?=(*:x))(q|)/ matches an empty string, and returns the mark
111        "x". The similar pattern /(?=(*:x))((*:y)q|)/ did not return a mark at all.
112        Oddly, Perl behaves the same way. PCRE has been fixed so that this pattern
113        also returns the mark "x". This bug applied to capturing parentheses,
114        non-capturing parentheses, and atomic parentheses. It also applied to some
115        assertions.
117    12. Stephen Kelly's patch to CMakeLists.txt allows it to parse the version
118        information out of configure.ac instead of relying on pcre.h.generic, which
119        is not stored in the repository.
121    13. Applied Dmitry V. Levin's patch for a more portable method for linking with
122        -lreadline.
124    14. ZH added PCRE_CONFIG_JITTARGET; added its output to pcretest -C.
126    15. Applied Graycode's patch to put the top-level frame on the stack rather
127        than the heap when not using the stack for recursion. This gives a
128        performance improvement in many cases when recursion is not deep.
130    16. Experimental code added to "pcretest -C" to output the stack frame size.
133    Version 8.21 12-Dec-2011
134    ------------------------
136    1.  Updating the JIT compiler.
138    2.  JIT compiler now supports OP_NCREF, OP_RREF and OP_NRREF. New test cases
139        are added as well.
141    3.  Fix cache-flush issue on PowerPC (It is still an experimental JIT port).
142        PCRE_EXTRA_TABLES is not suported by JIT, and should be checked before
143        calling _pcre_jit_exec. Some extra comments are added.
145    4.  (*MARK) settings inside atomic groups that do not contain any capturing
146        parentheses, for example, (?>a(*:m)), were not being passed out. This bug
147        was introduced by change 18 for 8.20.
149    5.  Supporting of \x, \U and \u in JavaScript compatibility mode based on the
150        ECMA-262 standard.
152    6.  Lookbehinds such as (?<=a{2}b) that contained a fixed repetition were
153        erroneously being rejected as "not fixed length" if PCRE_CASELESS was set.
154        This bug was probably introduced by change 9 of 8.13.
156    7.  While fixing 6 above, I noticed that a number of other items were being
157        incorrectly rejected as "not fixed length". This arose partly because newer
158        opcodes had not been added to the fixed-length checking code. I have (a)
159        corrected the bug and added tests for these items, and (b) arranged for an
160        error to occur if an unknown opcode is encountered while checking for fixed
161        length instead of just assuming "not fixed length". The items that were
162        rejected were: (*ACCEPT), (*COMMIT), (*FAIL), (*MARK), (*PRUNE), (*SKIP),
163        (*THEN), \h, \H, \v, \V, and single character negative classes with fixed
164        repetitions, e.g. [^a]{3}, with and without PCRE_CASELESS.
166    8.  A possessively repeated conditional subpattern such as (?(?=c)c|d)++ was
167        being incorrectly compiled and would have given unpredicatble results.
169    9.  A possessively repeated subpattern with minimum repeat count greater than
170        one behaved incorrectly. For example, (A){2,}+ behaved as if it was
171        (A)(A)++ which meant that, after a subsequent mismatch, backtracking into
172        the first (A) could occur when it should not.
174    10. Add a cast and remove a redundant test from the code.
176    11. JIT should use pcre_malloc/pcre_free for allocation.
178    12. Updated pcre-config so that it no longer shows -L/usr/lib, which seems
179        best practice nowadays, and helps with cross-compiling. (If the exec_prefix
180        is anything other than /usr, -L is still shown).
182    13. In non-UTF-8 mode, \C is now supported in lookbehinds and DFA matching.
184    14. Perl does not support \N without a following name in a [] class; PCRE now
185        also gives an error.
187    15. If a forward reference was repeated with an upper limit of around 2000,
188        it caused the error "internal error: overran compiling workspace". The
189        maximum number of forward references (including repeats) was limited by the
190        internal workspace, and dependent on the LINK_SIZE. The code has been
191        rewritten so that the workspace expands (via pcre_malloc) if necessary, and
192        the default depends on LINK_SIZE. There is a new upper limit (for safety)
193        of around 200,000 forward references. While doing this, I also speeded up
194        the filling in of repeated forward references.
196    16. A repeated forward reference in a pattern such as (a)(?2){2}(.) was
197        incorrectly expecting the subject to contain another "a" after the start.
199    17. When (*SKIP:name) is activated without a corresponding (*MARK:name) earlier
200        in the match, the SKIP should be ignored. This was not happening; instead
201        the SKIP was being treated as NOMATCH. For patterns such as
202        /A(*MARK:A)A+(*SKIP:B)Z|AAC/ this meant that the AAC branch was never
203        tested.
205    18. The behaviour of (*MARK), (*PRUNE), and (*THEN) has been reworked and is
206        now much more compatible with Perl, in particular in cases where the result
207        is a non-match for a non-anchored pattern. For example, if
208        /b(*:m)f|a(*:n)w/ is matched against "abc", the non-match returns the name
209        "m", where previously it did not return a name. A side effect of this
210        change is that for partial matches, the last encountered mark name is
211        returned, as for non matches. A number of tests that were previously not
212        Perl-compatible have been moved into the Perl-compatible test files. The
213        refactoring has had the pleasing side effect of removing one argument from
214        the match() function, thus reducing its stack requirements.
216    19. If the /S+ option was used in pcretest to study a pattern using JIT,
217        subsequent uses of /S (without +) incorrectly behaved like /S+.
219    21. Retrieve executable code size support for the JIT compiler and fixing
220        some warnings.
222    22. A caseless match of a UTF-8 character whose other case uses fewer bytes did
223        not work when the shorter character appeared right at the end of the
224        subject string.
226    23. Added some (int) casts to non-JIT modules to reduce warnings on 64-bit
227        systems.
229    24. Added PCRE_INFO_JITSIZE to pass on the value from (21) above, and also
230        output it when the /M option is used in pcretest.
232    25. The CheckMan script was not being included in the distribution. Also, added
233        an explicit "perl" to run Perl scripts from the PrepareRelease script
234        because this is reportedly needed in Windows.
236    26. If study data was being save in a file and studying had not found a set of
237        "starts with" bytes for the pattern, the data written to the file (though
238        never used) was taken from uninitialized memory and so caused valgrind to
239        complain.
241    27. Updated RunTest.bat as provided by Sheri Pierce.
243    28. Fixed a possible uninitialized memory bug in pcre_jit_compile.c.
245    29. Computation of memory usage for the table of capturing group names was
246        giving an unnecessarily large value.
249    Version 8.20 21-Oct-2011
250    ------------------------
252    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
253        a POSIX class. After further experiments with Perl, which convinced me that
254        Perl has bugs and confusions, a closing square bracket is no longer allowed
255        in a POSIX name. This bug also affected patterns with classes that started
256        with full stops.
258    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
259        captured substring, but while checking the failing first alternative,
260        substring 1 is temporarily captured. If the output vector supplied to
261        pcre_exec() was not big enough for this capture, the yield of the function
262        was still zero ("insufficient space for captured substrings"). This cannot
263        be totally fixed without adding another stack variable, which seems a lot
264        of expense for a edge case. However, I have improved the situation in cases
265        such as /(a)(b)x|abc/ matched against "abc", where the return code
266        indicates that fewer than the maximum number of slots in the ovector have
267        been set.
269    3.  Related to (2) above: when there are more back references in a pattern than
270        slots in the output vector, pcre_exec() uses temporary memory during
271        matching, and copies in the captures as far as possible afterwards. It was
272        using the entire output vector, but this conflicts with the specification
273        that only 2/3 is used for passing back captured substrings. Now it uses
274        only the first 2/3, for compatibility. This is, of course, another edge
275        case.
277    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
278        main code base, and can be used by building with --enable-jit. When this is
279        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
280        runtime --no-jit option is given.
282    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
283        ovector, the return from the function was zero, implying that there were
284        other matches that did not fit. The correct "exactly full" value is now
285        returned.
287    6.  If a subpattern that was called recursively or as a subroutine contained
288        (*PRUNE) or any other control that caused it to give a non-standard return,
289        invalid errors such as "Error -26 (nested recursion at the same subject
290        position)" or even infinite loops could occur.
292    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
293        computing the minimum length on reaching *ACCEPT, and so ended up with the
294        wrong value of 1 rather than 0. Further investigation indicates that
295        computing a minimum subject length in the presence of *ACCEPT is difficult
296        (think back references, subroutine calls), and so I have changed the code
297        so that no minimum is registered for a pattern that contains *ACCEPT.
299    8.  If (*THEN) was present in the first (true) branch of a conditional group,
300        it was not handled as intended. [But see 16 below.]
302    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
303        Sheri Pierce.
305    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
306        the first byte in a match must be "a".
308    11. Change 17 for 8.13 increased the recursion depth for patterns like
309        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
310        pattern contains any instances of (*THEN). If it does not, the old
311        optimizations are restored. It would be nice to do this on a per-group
312        basis, but at the moment that is not feasible.
314    12. In some environments, the output of pcretest -C is CRLF terminated. This
315        broke RunTest's code that checks for the link size. A single white space
316        character after the value is now allowed for.
318    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
319        For "fr", it uses the Windows-specific input and output files.
321    14. If (*THEN) appeared in a group that was called recursively or as a
322        subroutine, it did not work as intended. [But see next item.]
324    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
325        pattern fragments (but not containing any | characters). If A and B are
326        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
327        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
328        D. In other words, Perl considers parentheses that do not contain any |
329        characters to be part of a surrounding alternative, whereas PCRE was
330        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
331        differently. PCRE now behaves in the same way as Perl, except in the case
332        of subroutine/recursion calls such as (?1) which have in any case always
333        been different (but PCRE had them first :-).
335    16. Related to 15 above: Perl does not treat the | in a conditional group as
336        creating alternatives. Such a group is treated in the same way as an
337        ordinary group without any | characters when processing (*THEN). PCRE has
338        been changed to match Perl's behaviour.
340    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
341        RunGrepTest script failed.
343    18. Change 22 for version 13 caused atomic groups to use more stack. This is
344        inevitable for groups that contain captures, but it can lead to a lot of
345        stack use in large patterns. The old behaviour has been restored for atomic
346        groups that do not contain any capturing parentheses.
348    19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not
349        suppress the check for a minimum subject length at run time. (If it was
350        given to pcre_exec() or pcre_dfa_exec() it did work.)
352    20. Fixed an ASCII-dependent infelicity in pcretest that would have made it
353        fail to work when decoding hex characters in data strings in EBCDIC
354        environments.
356    21. It appears that in at least one Mac OS environment, the isxdigit() function
357        is implemented as a macro that evaluates to its argument more than once,
358        contravening the C 90 Standard (I haven't checked a later standard). There
359        was an instance in pcretest which caused it to go wrong when processing
360        \x{...} escapes in subject strings. The has been rewritten to avoid using
361        things like p++ in the argument of isxdigit().
364    Version 8.13 16-Aug-2011
365    ------------------------
367    1.  The Unicode data tables have been updated to Unicode 6.0.0.
369    2.  Two minor typos in pcre_internal.h have been fixed.
371    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
372        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
373        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
375    4.  There were a number of related bugs in the code for matching backrefences
376        caselessly in UTF-8 mode when codes for the characters concerned were
377        different numbers of bytes. For example, U+023A and U+2C65 are an upper
378        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
379        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
380        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
381        2-byte code at the end of the subject (it thought there wasn't enough data
382        left).
384    5.  Comprehensive information about what went wrong is now returned by
385        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
386        as the output vector has at least 2 elements. The offset of the start of
387        the failing character and a reason code are placed in the vector.
389    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
390        now returned is for the first byte of the failing character, instead of the
391        last byte inspected. This is an incompatible change, but I hope it is small
392        enough not to be a problem. It makes the returned offset consistent with
393        pcre_exec() and pcre_dfa_exec().
395    7.  pcretest now gives a text phrase as well as the error number when
396        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
397        failure, the offset and reason code are output.
399    8.  When \R was used with a maximizing quantifier it failed to skip backwards
400        over a \r\n pair if the subsequent match failed. Instead, it just skipped
401        back over a single character (\n). This seems wrong (because it treated the
402        two characters as a single entity when going forwards), conflicts with the
403        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
404        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
405        has been changed.
407    9.  Some internal refactoring has changed the processing so that the handling
408        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
409        time (the PCRE_DOTALL option was changed this way some time ago: version
410        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
411        which was always a bit of a fudge. It also means that there is one less
412        argument for the match() function, which reduces its stack requirements
413        slightly. This change also fixes an incompatibility with Perl: the pattern
414        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
416    10. More internal refactoring has drastically reduced the number of recursive
417        calls to match() for possessively repeated groups such as (abc)++ when
418        using pcre_exec().
420    11. While implementing 10, a number of bugs in the handling of groups were
421        discovered and fixed:
423        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
424        (a|)*(?1) gave a compile-time internal error.
425        ((a|)+)+  did not notice that the outer group could match an empty string.
426        (^a|^)+   was not marked as anchored.
427        (.*a|.*)+ was not marked as matching at start or after a newline.
429    12. Yet more internal refactoring has removed another argument from the match()
430        function. Special calls to this function are now indicated by setting a
431        value in a variable in the "match data" data block.
433    13. Be more explicit in pcre_study() instead of relying on "default" for
434        opcodes that mean there is no starting character; this means that when new
435        ones are added and accidentally left out of pcre_study(), testing should
436        pick them up.
438    14. The -s option of pcretest has been documented for ages as being an old
439        synonym of -m (show memory usage). I have changed it to mean "force study
440        for every regex", that is, assume /S for every regex. This is similar to -i
441        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
442        using it. It makes it easier to run collections of tests with and without
443        study enabled, and thereby test pcre_study() more easily. All the standard
444        tests are now run with and without -s (but some patterns can be marked as
445        "never study" - see 20 below).
447    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
448        restoration of the capturing data to the outer values was not happening
449        correctly.
451    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
452        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
453        pattern had matched an empty string, and so incorrectly returned a no
454        match.
456    17. There was optimizing code for the last branch of non-capturing parentheses,
457        and also for the obeyed branch of a conditional subexpression, which used
458        tail recursion to cut down on stack usage. Unfortunately, now that there is
459        the possibility of (*THEN) occurring in these branches, tail recursion is
460        no longer possible because the return has to be checked for (*THEN). These
461        two optimizations have therefore been removed. [But see 8.20/11 above.]
463    18. If a pattern containing \R was studied, it was assumed that \R always
464        matched two bytes, thus causing the minimum subject length to be
465        incorrectly computed because \R can also match just one byte.
467    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
468        was incorrectly computed.
470    20. If /S is present twice on a test pattern in pcretest input, it now
471        *disables* studying, thereby overriding the use of -s on the command line
472        (see 14 above). This is necessary for one or two tests to keep the output
473        identical in both cases.
475    21. When (*ACCEPT) was used in an assertion that matched an empty string and
476        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
478    22. When an atomic group that contained a capturing parenthesis was
479        successfully matched, but the branch in which it appeared failed, the
480        capturing was not being forgotten if a higher numbered group was later
481        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
482        group 1 to "a", when in fact it should be unset. This applied to multi-
483        branched capturing and non-capturing groups, repeated or not, and also to
484        positive assertions (capturing in negative assertions does not happen
485        in PCRE) and also to nested atomic groups.
487    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
488        subject after a captured substring, to make it easier to tell which of a
489        number of identical substrings has been captured.
491    24. The way atomic groups are processed by pcre_exec() has been changed so that
492        if they are repeated, backtracking one repetition now resets captured
493        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
494        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
495        "aaa". Previously, it would have been "a". As part of this code
496        refactoring, the way recursive calls are handled has also been changed.
498    25. If an assertion condition captured any substrings, they were not passed
499        back unless some other capturing happened later. For example, if
500        (?(?=(a))a) was matched against "a", no capturing was returned.
502    26. When studying a pattern that contained subroutine calls or assertions,
503        the code for finding the minimum length of a possible match was handling
504        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
505        group 1 called group 2 while simultaneously a separate group 2 called group
506        1). A stack overflow occurred in this case. I have fixed this by limiting
507        the recursion depth to 10.
509    27. Updated RunTest.bat in the distribution to the version supplied by Tom
510        Fortmann. This supports explicit test numbers on the command line, and has
511        argument validation and error reporting.
513    28. An instance of \X with an unlimited repeat could fail if at any point the
514        first character it looked at was a mark character.
516    29. Some minor code refactoring concerning Unicode properties and scripts
517        should reduce the stack requirement of match() slightly.
519    30. Added the '=' option to pcretest to check the setting of unused capturing
520        slots at the end of the pattern, which are documented as being -1, but are
521        not included in the return count.
523    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
524        compiled something random. Now it gives a compile-time error (as does
525        Perl).
527    32. A *MARK encountered during the processing of a positive assertion is now
528        recorded and passed back (compatible with Perl).
530    33. If --only-matching or --colour was set on a pcregrep call whose pattern
531        had alternative anchored branches, the search for a second match in a line
532        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
533        matched the line "0102" twice. The same bug affected patterns that started
534        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
535        twice.
537    34. Previously, PCRE did not allow quantification of assertions. However, Perl
538        does, and because of capturing effects, quantifying parenthesized
539        assertions may at times be useful. Quantifiers are now allowed for
540        parenthesized assertions.
542    35. A minor code tidy in pcre_compile() when checking options for \R usage.
544    36. \g was being checked for fancy things in a character class, when it should
545        just be a literal "g".
547    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
548        appearance of a nested POSIX class supersedes an apparent external class.
549        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
550        unescaped square brackets may also appear as part of class names. For
551        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
552        more like Perl. (But see 8.20/1 above.)
554    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
555        was because it thought it was \N{name}, which is not supported).
557    39. Add minix to OS list not supporting the -S option in pcretest.
559    40. PCRE tries to detect cases of infinite recursion at compile time, but it
560        cannot analyze patterns in sufficient detail to catch mutual recursions
561        such as ((?1))((?2)). There is now a runtime test that gives an error if a
562        subgroup is called recursively as a subpattern for a second time at the
563        same position in the subject string. In previous releases this might have
564        been caught by the recursion limit, or it might have run out of stack.
566    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
567        happen only once. PCRE was, however incorrectly giving a compile time error
568        "recursive call could loop indefinitely" because it cannot analyze the
569        pattern in sufficient detail. The compile time test no longer happens when
570        PCRE is compiling a conditional subpattern, but actual runaway loops are
571        now caught at runtime (see 40 above).
573    42. It seems that Perl allows any characters other than a closing parenthesis
574        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
575        has been changed to be the same.
577    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
578        as not to get warnings when autogen.sh is called. Also changed
579        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
581    44. To help people who use pcregrep to scan files containing exceedingly long
582        lines, the following changes have been made:
584        (a) The default value of the buffer size parameter has been increased from
585            8K to 20K. (The actual buffer used is three times this size.)
587        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
588            PCRE is built.
590        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
591            to be set at run time.
593        (d) Numerical values in pcregrep options can be followed by K or M, for
594            example --buffer-size=50K.
596        (e) If a line being scanned overflows pcregrep's buffer, an error is now
597            given and the return code is set to 2.
599    45. Add a pointer to the latest mark to the callout data block.
601    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
602        partial match of an empty string instead of no match. This was specific to
603        the use of ".".
605    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
606        complete match instead of a partial match. This bug was dependent on both
607        the PCRE_UTF8 and PCRE_DOTALL options being set.
609    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
610        starting byte set, because \b was not being ignored.
613    Version 8.12 15-Jan-2011
614    ------------------------
616    1.  Fixed some typos in the markup of the man pages, and wrote a script that
617        checks for such things as part of the documentation building process.
619    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
620        --match-limit and --recursion-limit options (added for 8.11). In
621        particular, this made one of the standard tests fail. (The integer value
622        went into the wrong half of a long int.)
624    3.  If the --colour option was given to pcregrep with -v (invert match), it
625        did strange things, either producing crazy output, or crashing. It should,
626        of course, ignore a request for colour when reporting lines that do not
627        match.
629    4.  Another pcregrep bug caused similar problems if --colour was specified with
630        -M (multiline) and the pattern match finished with a line ending.
632    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
633        matched in multiline mode, the following line was shown as part of the
634        match. This seems wrong, so I have changed it.
636    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
637        the check for further matches in the same line (so they could be coloured)
638        to overrun the end of the current line. If another match was found, it was
639        incorrectly shown (and then shown again when found in the next line).
641    7.  If pcregrep was compiled under Windows, there was a reference to the
642        function pcregrep_exit() before it was defined. I am assuming this was
643        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
644        reported by a user. I've moved the definition above the reference.
647    Version 8.11 10-Dec-2010
648    ------------------------
650    1.  (*THEN) was not working properly if there were untried alternatives prior
651        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
652        backtracked to try for "b" instead of moving to the next alternative branch
653        at the same level (in this case, to look for "c"). The Perl documentation
654        is clear that when (*THEN) is backtracked onto, it goes to the "next
655        alternative in the innermost enclosing group".
657    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
658        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
659        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
660        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
661        (*THEN).
663    3.  If \s appeared in a character class, it removed the VT character from
664        the class, even if it had been included by some previous item, for example
665        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
666        of \s, but is part of the POSIX "space" class.)
668    4.  A partial match never returns an empty string (because you can always
669        match an empty string at the end of the subject); however the checking for
670        an empty string was starting at the "start of match" point. This has been
671        changed to the "earliest inspected character" point, because the returned
672        data for a partial match starts at this character. This means that, for
673        example, /(?<=abc)def/ gives a partial match for the subject "abc"
674        (previously it gave "no match").
676    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
677        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
678        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
679        has an implication that the given string is incomplete (because a partial
680        match is preferred over a full match). For this reason, these items now
681        give a partial match in this situation. [Aside: previously, the one case
682        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
683        match rather than a full match, which was wrong by the old rules, but is
684        now correct.]
686    6.  There was a bug in the handling of #-introduced comments, recognized when
687        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
688        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
689        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
690        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
691        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
692        places in pcre_compile().
694    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
695        comments when looking ahead for named forward references to subpatterns,
696        the only newline sequence it recognized was NL. It now handles newlines
697        according to the set newline convention.
699    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
700        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
701        cater for a lack of strerror(). These oversights have been fixed.
703    9.  Added --match-limit and --recursion-limit to pcregrep.
705    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
707    11. When the -o option was used, pcregrep was setting a return code of 1, even
708        when matches were found, and --line-buffered was not being honoured.
710    12. Added an optional parentheses number to the -o and --only-matching options
711        of pcregrep.
713    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
714        can match an empty string. The code to do it in pcretest and pcredemo
715        needed fixing:
717        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
718            only one byte after an empty string match just before CRLF (this case
719            just got forgotten; "any" and "anycrlf" were OK).
721        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
722            mode when an empty string match preceded an ASCII character followed by
723            a non-ASCII character. (The code for advancing by one character rather
724            than one byte was nonsense.)
726        (c) The pcredemo.c sample program did not have any code at all to handle
727            the cases when CRLF is a valid newline sequence.
729    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
730        as a starting offset was within the subject string. There is now a new
731        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
732        negative or greater than the length of the string. In order to test this,
733        pcretest is extended to allow the setting of negative starting offsets.
735    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
736        starting offset points to the beginning of a UTF-8 character was
737        unnecessarily clumsy. I tidied it up.
739    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
740        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
742    17. Nobody had reported that the --include_dir option, which was added in
743        release 7.7 should have been called --include-dir (hyphen, not underscore)
744        for compatibility with GNU grep. I have changed it to --include-dir, but
745        left --include_dir as an undocumented synonym, and the same for
746        --exclude-dir, though that is not available in GNU grep, at least as of
747        release 2.5.4.
749    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
750        characters from a string of bytes) have been redefined so as not to use
751        loops, in order to improve performance in some environments. At the same
752        time, I abstracted some of the common code into auxiliary macros to save
753        repetition (this should not affect the compiled code).
755    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
756        compile-time error is now given if \c is not followed by an ASCII
757        character, that is, a byte less than 128. (In EBCDIC mode, the code is
758        different, and any byte value is allowed.)
760    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
761        START_OPTIMIZE option, which is now allowed at compile time - but just
762        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
763        to pcregrep and other applications that have no direct access to PCRE
764        options. The new /Y option in pcretest sets this option when calling
765        pcre_compile().
767    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
768        back references. Groups containing recursive back references were forced to
769        be atomic by that change, but in the case of named groups, the amount of
770        memory required was incorrectly computed, leading to "Failed: internal
771        error: code overflow". This has been fixed.
773    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
774        pcretest.c, to avoid build problems in some Borland environments.
777    Version 8.10 25-Jun-2010
778    ------------------------
780    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
781        THEN.
783    2.  (*ACCEPT) was not working when inside an atomic group.
785    3.  Inside a character class, \B is treated as a literal by default, but
786        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
787        causes the error). The code is unchanged, but I tidied the documentation.
789    4.  Inside a character class, PCRE always treated \R and \X as literals,
790        whereas Perl faults them if its -w option is set. I have changed PCRE so
791        that it faults them when PCRE_EXTRA is set.
793    5.  Added support for \N, which always matches any character other than
794        newline. (It is the same as "." when PCRE_DOTALL is not set.)
796    6.  When compiling pcregrep with newer versions of gcc which may have
797        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
798        declared with attribute warn_unused_result" were given. Just casting the
799        result to (void) does not stop the warnings; a more elaborate fudge is
800        needed. I've used a macro to implement this.
802    7.  Minor change to pcretest.c to avoid a compiler warning.
804    8.  Added four artifical Unicode properties to help with an option to make
805        \s etc use properties (see next item). The new properties are: Xan
806        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
808    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
809        use Unicode properties. (*UCP) at the start of a pattern can be used to set
810        this option. Modified pcretest to add /W to test this facility. Added
811        REG_UCP to make it available via the POSIX interface.
813    10. Added --line-buffered to pcregrep.
815    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
816        studied, and the match started with a letter with a code point greater than
817        127 whose first byte was different to the first byte of the other case of
818        the letter, the other case of this starting letter was not recognized
819        (#976).
821    12. If a pattern that was studied started with a repeated Unicode property
822        test, for example, \p{Nd}+, there was the theoretical possibility of
823        setting up an incorrect bitmap of starting bytes, but fortunately it could
824        not have actually happened in practice until change 8 above was made (it
825        added property types that matched character-matching opcodes).
827    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
828        possible starting bytes for non-anchored patterns.
830    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
831        \R, and also a number of cases that involve Unicode properties, both
832        explicit and implicit when PCRE_UCP is set.
834    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
835        input, it could crash or give wrong results if characters with values
836        greater than 0xc0 were present in the subject string. (Detail: it assumed
837        UTF-8 input when processing these items.)
839    16. Added a lot of (int) casts to avoid compiler warnings in systems where
840        size_t is 64-bit (#991).
842    17. Added a check for running out of memory when PCRE is compiled with
843        --disable-stack-for-recursion (#990).
845    18. If the last data line in a file for pcretest does not have a newline on
846        the end, a newline was missing in the output.
848    19. The default pcre_chartables.c file recognizes only ASCII characters (values
849        less than 128) in its various bitmaps. However, there is a facility for
850        generating tables according to the current locale when PCRE is compiled. It
851        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
852        characters, are recognized by isspace() and therefore were getting set in
853        these tables, and indeed these tables seem to approximate to ISO 8859. This
854        caused a problem in UTF-8 mode when pcre_study() was used to create a list
855        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
856        which of course cannot start UTF-8 characters. I have changed the code so
857        that only real ASCII characters (less than 128) and the correct starting
858        bytes for UTF-8 encodings are set for characters greater than 127 when in
859        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
860        altogether.)
862    20. Added the /T option to pcretest so as to be able to run tests with non-
863        standard character tables, thus making it possible to include the tests
864        used for 19 above in the standard set of tests.
866    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
867        reference to a subpattern the other side of a comment that contains an
868        opening parenthesis caused either an internal compiling error, or a
869        reference to the wrong subpattern.
872    Version 8.02 19-Mar-2010
873    ------------------------
875    1.  The Unicode data tables have been updated to Unicode 5.2.0.
877    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
878        configured.
880    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
881        original author of that file, following a query about its status.
883    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
884        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
886    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
887        quantifier applied to a forward-referencing subroutine call, could compile
888        incorrect code or give the error "internal error: previously-checked
889        referenced subpattern not found".
891    6.  Both MS Visual Studio and Symbian OS have problems with initializing
892        variables to point to external functions. For these systems, therefore,
893        pcre_malloc etc. are now initialized to local functions that call the
894        relevant global functions.
896    7.  There were two entries missing in the vectors called coptable and poptable
897        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
898        I've fixed the data, and added a kludgy way of testing at compile time that
899        the lengths are correct (equal to the number of opcodes).
901    8.  Following on from 7, I added a similar kludge to check the length of the
902        eint vector in pcreposix.c.
904    9.  Error texts for pcre_compile() are held as one long string to avoid too
905        much relocation at load time. To find a text, the string is searched,
906        counting zeros. There was no check for running off the end of the string,
907        which could happen if a new error number was added without updating the
908        string.
910    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
912    11. \K was not working if it appeared in an atomic group or in a group that
913        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
914        \K is "not well defined" if used in an assertion. PCRE now accepts it if
915        the assertion is positive, but not if it is negative.
917    12. Change 11 fortuitously reduced the size of the stack frame used in the
918        "match()" function of pcre_exec.c by one pointer. Forthcoming
919        implementation of support for (*MARK) will need an extra pointer on the
920        stack; I have reserved it now, so that the stack frame size does not
921        decrease.
923    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
924        item in branch that calls a recursion is a subroutine call - as in the
925        second branch in the above example - was incorrectly given the compile-
926        time error "recursive call could loop indefinitely" because pcre_compile()
927        was not correctly checking the subroutine for matching a non-empty string.
929    14. The checks for overrunning compiling workspace could trigger after an
930        overrun had occurred. This is a "should never occur" error, but it can be
931        triggered by pathological patterns such as hundreds of nested parentheses.
932        The checks now trigger 100 bytes before the end of the workspace.
934    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
937    Version 8.01 19-Jan-2010
938    ------------------------
940    1.  If a pattern contained a conditional subpattern with only one branch (in
941        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
942        computed the wrong minimum data length (which is of course zero for such
943        subpatterns). This could cause incorrect "no match" results.
945    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
946        the pattern is reset in the first branch, pcre_compile() failed with
947        "internal error: code overflow at offset...". This happened only when
948        the reset was to the original external option setting. (An optimization
949        abstracts leading options settings into an external setting, which was the
950        cause of this.)
952    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
953        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
954        assertion pattern did not match (meaning that the assertion was true), it
955        was incorrectly treated as false if the SKIP had been reached during the
956        matching. This also applied to assertions used as conditions.
958    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
959        assertion subpattern, including such a pattern used as a condition,
960        unpredictable results occurred, instead of the error return
963    5.  The C++ GlobalReplace function was not working like Perl for the special
964        situation when an empty string is matched. It now does the fancy magic
965        stuff that is necessary.
967    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
968        removed. (These were left over from very, very early versions of PCRE.)
970    7.  Some cosmetic changes to the code to make life easier when compiling it
971        as part of something else:
973        (a) Change DEBUG to PCRE_DEBUG.
975        (b) In pcre_compile(), rename the member of the "branch_chain" structure
976            called "current" as "current_branch", to prevent a collision with the
977            Linux macro when compiled as a kernel module.
979        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
980            prevent a collision with the Linux macro when compiled as a kernel
981            module.
983    8.  In pcre_compile() there are some checks for integer overflows that used to
984        cast potentially large values to (double). This has been changed to that
985        when building, a check for int64_t is made, and if it is found, it is used
986        instead, thus avoiding the use of floating point arithmetic. (There is no
987        other use of FP in PCRE.) If int64_t is not found, the fallback is to
988        double.
990    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
991        2005 (difference between two addresses compared to an unsigned value).
993    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
994        custom one, because of the following reported problem in Windows:
996          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
997              under Win32.
998          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
999              therefore missing the function definition.
1000          - The compiler thus generates a "C" signature for the test function.
1001          - The linker fails to find the "C" function.
1002          - PCRE fails to configure if asked to do so against libbz2.
1004    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
1005        messages were output:
1007          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
1008          rerunning libtoolize, to keep the correct libtool macros in-tree.
1009          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
1011        I have done both of these things.
1013    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
1014        most of the time, it *can* run out if it is given a pattern that contains a
1015        runaway infinite recursion. I updated the discussion in the pcrestack man
1016        page.
1018    13. Now that we have gone to the x.xx style of version numbers, the minor
1019        version may start with zero. Using 08 or 09 is a bad idea because users
1020        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
1021        interpreted as invalid octal numbers. I've updated the previous comment in
1022        configure.ac, and also added a check that gives an error if 08 or 09 are
1023        used.
1025    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
1026        causing partial matching to fail when the end of the subject matched \W
1027        in a UTF-8 pattern where \W was quantified with a minimum of 3.
1029    15. There were some discrepancies between the declarations in pcre_internal.h
1030        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
1031        their definitions. The declarations used "const uschar *" and the
1032        definitions used USPTR. Even though USPTR is normally defined as "const
1033        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
1034        reported that: "This difference in casting confuses some C++ compilers, for
1035        example, SunCC recognizes above declarations as different functions and
1036        generates broken code for hbpcre." I have changed the declarations to use
1037        USPTR.
1039    16. GNU libtool is named differently on some systems. The autogen.sh script now
1040        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
1041        (FreeBSD).
1043    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
1044        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
1045        comment: "Figure out how to create a longlong from a string: strtoll and
1046        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
1047        instance, but it only takes 2 args instead of 3!"
1049    18. A subtle bug concerned with back references has been fixed by a change of
1050        specification, with a corresponding code fix. A pattern such as
1051        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
1052        refers, was giving matches when it shouldn't. For example, xa=xaaa would
1053        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
1054        same bug. Such groups have to be quantified to be useful, or contained
1055        inside another quantified group. (If there's no repetition, the reference
1056        can never match.) The problem arises because, having left the group and
1057        moved on to the rest of the pattern, a later failure that backtracks into
1058        the group uses the captured value from the final iteration of the group
1059        rather than the correct earlier one. I have fixed this in PCRE by forcing
1060        any group that contains a reference to itself to be an atomic group; that
1061        is, there cannot be any backtracking into it once it has completed. This is
1062        similar to recursive and subroutine calls.
1065    Version 8.00 19-Oct-09
1066    ----------------------
1068    1.  The table for translating pcre_compile() error codes into POSIX error codes
1069        was out-of-date, and there was no check on the pcre_compile() error code
1070        being within the table. This could lead to an OK return being given in
1071        error.
1073    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
1074        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
1075        in a Windows environment.
1077    3.  The pcregrep --count option prints the count for each file even when it is
1078        zero, as does GNU grep. However, pcregrep was also printing all files when
1079        --files-with-matches was added. Now, when both options are given, it prints
1080        counts only for those files that have at least one match. (GNU grep just
1081        prints the file name in this circumstance, but including the count seems
1082        more useful - otherwise, why use --count?) Also ensured that the
1083        combination -clh just lists non-zero counts, with no names.
1085    4.  The long form of the pcregrep -F option was incorrectly implemented as
1086        --fixed_strings instead of --fixed-strings. This is an incompatible change,
1087        but it seems right to fix it, and I didn't think it was worth preserving
1088        the old behaviour.
1090    5.  The command line items --regex=pattern and --regexp=pattern were not
1091        recognized by pcregrep, which required --regex pattern or --regexp pattern
1092        (with a space rather than an '='). The man page documented the '=' forms,
1093        which are compatible with GNU grep; these now work.
1095    6.  No libpcreposix.pc file was created for pkg-config; there was just
1096        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
1098    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
1099        when UCP support is not needed, by modifying the Python script that
1100        generates it from Unicode data files. This should not matter if the module
1101        is correctly used as a library, but I received one complaint about 50K of
1102        unwanted data. My guess is that the person linked everything into his
1103        program rather than using a library. Anyway, it does no harm.
1105    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
1106        was a minimum greater than 1 for a wide character in a possessive
1107        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
1108        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
1109        character. Chaos in the form of incorrect output or a compiling loop could
1110        result.
1112    9.  The restrictions on what a pattern can contain when partial matching is
1113        requested for pcre_exec() have been removed. All patterns can now be
1114        partially matched by this function. In addition, if there are at least two
1115        slots in the offset vector, the offset of the earliest inspected character
1116        for the match and the offset of the end of the subject are set in them when
1117        PCRE_ERROR_PARTIAL is returned.
1119    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
1120        synonymous with PCRE_PARTIAL, for backwards compatibility, and
1121        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
1122        and may be more useful for multi-segment matching.
1124    11. Partial matching with pcre_exec() is now more intuitive. A partial match
1125        used to be given if ever the end of the subject was reached; now it is
1126        given only if matching could not proceed because another character was
1127        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
1128        string "Z", which now yields "no match" instead of "partial match". In the
1129        case of pcre_dfa_exec(), "no match" is given if every matching path for the
1130        final character ended with (*FAIL).
1132    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
1133        if the pattern had a "must contain" character that was already found in the
1134        earlier partial match, unless partial matching was again requested. For
1135        example, with the pattern /dog.(body)?/, the "must contain" character is
1136        "g". If the first part-match was for the string "dog", restarting with
1137        "sbody" failed. This bug has been fixed.
1139    13. The string returned by pcre_dfa_exec() after a partial match has been
1140        changed so that it starts at the first inspected character rather than the
1141        first character of the match. This makes a difference only if the pattern
1142        starts with a lookbehind assertion or \b or \B (\K is not supported by
1143        pcre_dfa_exec()). It's an incompatible change, but it makes the two
1144        matching functions compatible, and I think it's the right thing to do.
1146    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
1147        so that the demonstration program is easily available in environments where
1148        PCRE has not been installed from source.
1150    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
1151        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
1152        library.
1154    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
1155        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
1156        is not the first non-POSIX option to be added. Clearly some people find
1157        these options useful.
1159    17. If a caller to the POSIX matching function regexec() passes a non-zero
1160        value for nmatch with a NULL value for pmatch, the value of
1161        nmatch is forced to zero.
1163    18. RunGrepTest did not have a test for the availability of the -u option of
1164        the diff command, as RunTest does. It now checks in the same way as
1165        RunTest, and also checks for the -b option.
1167    19. If an odd number of negated classes containing just a single character
1168        interposed, within parentheses, between a forward reference to a named
1169        subpattern and the definition of the subpattern, compilation crashed with
1170        an internal error, complaining that it could not find the referenced
1171        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
1172        [The bug was that it was starting one character too far in when skipping
1173        over the character class, thus treating the ] as data rather than
1174        terminating the class. This meant it could skip too much.]
1176    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
1177        /g option in pcretest when the pattern contains \K, which makes it possible
1178        to have an empty string match not at the start, even when the pattern is
1179        anchored. Updated pcretest and pcredemo to use this option.
1181    21. If the maximum number of capturing subpatterns in a recursion was greater
1182        than the maximum at the outer level, the higher number was returned, but
1183        with unset values at the outer level. The correct (outer level) value is
1184        now given.
1186    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
1187        PCRE did not set those parentheses (unlike Perl). I have now found a way to
1188        make it do so. The string so far is captured, making this feature
1189        compatible with Perl.
1191    23. The tests have been re-organized, adding tests 11 and 12, to make it
1192        possible to check the Perl 5.10 features against Perl 5.10.
1194    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
1195        pattern matches a fixed length string. PCRE did not allow this; now it
1196        does. Neither allows recursion.
1198    25. I finally figured out how to implement a request to provide the minimum
1199        length of subject string that was needed in order to match a given pattern.
1200        (It was back references and recursion that I had previously got hung up
1201        on.) This code has now been added to pcre_study(); it finds a lower bound
1202        to the length of subject needed. It is not necessarily the greatest lower
1203        bound, but using it to avoid searching strings that are too short does give
1204        some useful speed-ups. The value is available to calling programs via
1205        pcre_fullinfo().
1207    26. While implementing 25, I discovered to my embarrassment that pcretest had
1208        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
1209        study optimizations had never been tested with that matching function.
1210        Oops. What is worse, even when it was passed study data, there was a bug in
1211        pcre_dfa_exec() that meant it never actually used it. Double oops. There
1212        were also very few tests of studied patterns with pcre_dfa_exec().
1214    27. If (?| is used to create subpatterns with duplicate numbers, they are now
1215        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
1216        on the other side of the coin, they are no longer allowed to have different
1217        names, because these cannot be distinguished in PCRE, and this has caused
1218        confusion. (This is a difference from Perl.)
1220    28. When duplicate subpattern names are present (necessarily with different
1221        numbers, as required by 27 above), and a test is made by name in a
1222        conditional pattern, either for a subpattern having been matched, or for
1223        recursion in such a pattern, all the associated numbered subpatterns are
1224        tested, and the overall condition is true if the condition is true for any
1225        one of them. This is the way Perl works, and is also more like the way
1226        testing by number works.
1229    Version 7.9 11-Apr-09
1230    ---------------------
1232    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
1233        (pcretest), all targets were linked against these libraries. This included
1234        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
1235        libraries. This caused unwanted dependencies to be created. This problem
1236        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
1237        pcretest is linked with readline.
1239    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
1240        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
1241        moved outside it again, because FALSE and TRUE are already defined in AIX,
1242        but BOOL is not.
1244    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
1245        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
1247    4.  The pcregrep documentation said spaces were inserted as well as colons (or
1248        hyphens) following file names and line numbers when outputting matching
1249        lines. This is not true; no spaces are inserted. I have also clarified the
1250        wording for the --colour (or --color) option.
1252    5.  In pcregrep, when --colour was used with -o, the list of matching strings
1253        was not coloured; this is different to GNU grep, so I have changed it to be
1254        the same.
1256    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
1257        each matching line was coloured. Now it goes on to look for further matches
1258        of any of the test patterns, which is the same behaviour as GNU grep.
1260    7.  A pattern that could match an empty string could cause pcregrep to loop; it
1261        doesn't make sense to accept an empty string match in pcregrep, so I have
1262        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
1263        seems to be how GNU grep behaves.
1265    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
1266        start or after a newline", because the conditional assertion was not being
1267        correctly handled. The rule now is that both the assertion and what follows
1268        in the first alternative must satisfy the test.
1270    9.  If auto-callout was enabled in a pattern with a conditional group whose
1271        condition was an assertion, PCRE could crash during matching, both with
1272        pcre_exec() and pcre_dfa_exec().
1274    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
1275        used for matching.
1277    11. Unicode property support in character classes was not working for
1278        characters (bytes) greater than 127 when not in UTF-8 mode.
1280    12. Added the -M command line option to pcretest.
1282    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
1284    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
1286    16. Added comments and documentation about mis-use of no_arg in the C++
1287        wrapper.
1289    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
1290        from Martin Jerabek that uses macro names for all relevant character and
1291        string constants.
1293    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
1294        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
1295        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
1296        these, but not everybody uses configure.
1298    19. A conditional group that had only one branch was not being correctly
1299        recognized as an item that could match an empty string. This meant that an
1300        enclosing group might also not be so recognized, causing infinite looping
1301        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
1302        with the subject "ab", where knowledge that the repeated group can match
1303        nothing is needed in order to break the loop.
1305    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
1306        exec(), but without supplying a callout function, matching went wrong.
1308    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
1309        leak if the size of the offset vector was greater than 30. When the vector
1310        is smaller, the saved offsets during recursion go onto a local stack
1311        vector, but for larger vectors malloc() is used. It was failing to free
1312        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
1313        error, in fact).
1315    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
1316        heapframe that is used only when UTF-8 support is enabled. This caused no
1317        problem, but was untidy.
1319    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
1320        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
1321        included within another project.
1323    24. Steven Van Ingelgem's patches to add more options to the CMake support,
1324        slightly modified by me:
1326          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
1327              not building pcregrep.
1329          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
1330              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
1332    25. Forward references, both numeric and by name, in patterns that made use of
1333        duplicate group numbers, could behave incorrectly or give incorrect errors,
1334        because when scanning forward to find the reference group, PCRE was not
1335        taking into account the duplicate group numbers. A pattern such as
1336        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
1338    26. Changed a few more instances of "const unsigned char *" to USPTR, making
1339        the feature of a custom pointer more persuasive (as requested by a user).
1341    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
1342        pcretest.c, inside #ifndefs, because it seems they are sometimes already
1343        pre-defined.
1345    28. Added support for (*UTF8) at the start of a pattern.
1347    29. Arrange for flags added by the "release type" setting in CMake to be shown
1348        in the configuration summary.
1351    Version 7.8 05-Sep-08
1352    ---------------------
1354    1.  Replaced UCP searching code with optimized version as implemented for Ad
1355        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
1356        stage table and inline lookup instead of a function, giving speed ups of 2
1357        to 5 times on some simple patterns that I tested. Permission was given to
1358        distribute the MultiStage2.py script that generates the tables (it's not in
1359        the tarball, but is in the Subversion repository).
1361    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
1362        scripts.
1364    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
1365        a group with a zero qualifier. The result of the study could be incorrect,
1366        or the function might crash, depending on the pattern.
1368    4.  Caseless matching was not working for non-ASCII characters in back
1369        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
1370        It now works when Unicode Property Support is available.
1372    5.  In pcretest, an escape such as \x{de} in the data was always generating
1373        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
1374        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
1375        truncation.
1377    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
1379    7.  Added two (int) casts to pcregrep when printing the difference of two
1380        pointers, in case they are 64-bit values.
1382    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
1383        test 2 if it fails.
1385    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
1386        and a #define of that name to empty if it is not externally set. This is to
1387        allow users of MSVC to set it if necessary.
1389    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
1390        the convenience functions in the pcre_get.c source file.
1392    11. An option change at the start of a pattern that had top-level alternatives
1393        could cause overwriting and/or a crash. This command provoked a crash in
1394        some environments:
1396          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
1398        This potential security problem was recorded as CVE-2008-2371.
1400    12. For a pattern where the match had to start at the beginning or immediately
1401        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
1402        pcre_dfa_exec() could read past the end of the passed subject if there was
1403        no match. To help with detecting such bugs (e.g. with valgrind), I modified
1404        pcretest so that it places the subject at the end of its malloc-ed buffer.
1406    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
1407        exec() might read past the end of the data buffer in UTF-8 mode.
1409    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
1410        the data contained the byte 0x85 as part of a UTF-8 character within its
1411        first line. This applied both to normal and DFA matching.
1413    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
1414        /^[^d]*?$/8 failed to match "abc".
1416    16. Added a missing copyright notice to pcrecpp_internal.h.
1418    17. Make it more clear in the documentation that values returned from
1419        pcre_exec() in ovector are byte offsets, not character counts.
1421    18. Tidied a few places to stop certain compilers from issuing warnings.
1423    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
1424        supplied by Stefan Weber. I made a further small update for 7.8 because
1425        there is a change of source arrangements: the pcre_searchfuncs.c module is
1426        replaced by pcre_ucd.c.
1429    Version 7.7 07-May-08
1430    ---------------------
1432    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
1433        a string to a long long, pretend we don't even have a long long." This is
1434        done by checking for the strtoq, strtoll, and _strtoi64 functions.
1436    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
1437        pre-7.6 versions, which defined a global no_arg variable instead of putting
1438        it in the RE class. (See also #8 below.)
1440    3.  Remove a line of dead code, identified by coverity and reported by Nuno
1441        Lopes.
1443    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
1445        (1) The include/exclude patterns were being applied to the whole pathnames
1446            of files, instead of just to the final components.
1448        (2) If there was more than one level of directory, the subdirectories were
1449            skipped unless they satisfied the include/exclude conditions. This is
1450            inconsistent with GNU grep (and could even be seen as contrary to the
1451            pcregrep specification - which I improved to make it absolutely clear).
1452            The action now is always to scan all levels of directory, and just
1453            apply the include/exclude patterns to regular files.
1455    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
1456        --exclude_dir in the tests to avoid scanning .svn directories.
1458    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
1459        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
1460        doesn't support NULs in patterns.
1462    7.  Added some missing "const"s to declarations of static tables in
1463        pcre_compile.c and pcre_dfa_exec.c.
1465    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
1466        caused by fix #2  above. (Subsequently also a second patch to fix the
1467        first patch. And a third patch - this was a messy problem.)
1469    9.  Applied Craig's patch to remove the use of push_back().
1471    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
1472        matching function regexec().
1474    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
1475        which, however, unlike Perl's \g{...}, are subroutine calls, not back
1476        references. PCRE supports relative numbers with this syntax (I don't think
1477        Oniguruma does).
1479    12. Previously, a group with a zero repeat such as (...){0} was completely
1480        omitted from the compiled regex. However, this means that if the group
1481        was called as a subroutine from elsewhere in the pattern, things went wrong
1482        (an internal error was given). Such groups are now left in the compiled
1483        pattern, with a new opcode that causes them to be skipped at execution
1484        time.
1486    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
1487        to the way PCRE behaves:
1489        (a) A lone ] character is dis-allowed (Perl treats it as data).
1491        (b) A back reference to an unmatched subpattern matches an empty string
1492            (Perl fails the current match path).
1494        (c) A data ] in a character class must be notated as \] because if the
1495            first data character in a class is ], it defines an empty class. (In
1496            Perl it is not possible to have an empty class.) The empty class []
1497            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
1498            The negative empty class [^] matches any one character, independently
1499            of the DOTALL setting.
1501    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
1502        non-existent subpattern following a character class starting with ']' and
1503        containing () gave an internal compiling error instead of "reference to
1504        non-existent subpattern". Fortunately, when the pattern did exist, the
1505        compiled code was correct. (When scanning forwards to check for the
1506        existencd of the subpattern, it was treating the data ']' as terminating
1507        the class, so got the count wrong. When actually compiling, the reference
1508        was subsequently set up correctly.)
1510    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
1511        it was being rejected as not supported by pcre_dfa_exec(), even though
1512        other assertions are supported. I have made pcre_dfa_exec() support
1513        (*FAIL).
1515    16. The implementation of 13c above involved the invention of a new opcode,
1516        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
1517        cannot be changed at match time, I realized I could make a small
1518        improvement to matching performance by compiling OP_ALLANY instead of
1519        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
1520        on the OP_ANY path.
1522    17. Compiling pcretest on Windows with readline support failed without the
1523        following two fixes: (1) Make the unistd.h include conditional on
1524        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
1526    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
1527        ncurses library to be included for pcretest when ReadLine support is
1528        requested, but also to allow for it to be overridden. This patch came from
1529        Daniel Bergström.
1531    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
1532        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
1533        any errors with the current Unicode tables. Thanks to Peter Kankowski for
1534        spotting this.
1537    Version 7.6 28-Jan-08
1538    ---------------------
1540    1.  A character class containing a very large number of characters with
1541        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
1542        overflow.
1544    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
1545        HAVE_LONG_LONG is not defined.
1547    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
1548        bring them up to date and include new features. This patch includes:
1550        - Fixed PH's badly added libz and libbz2 support.
1551        - Fixed a problem with static linking.
1552        - Added pcredemo. [But later removed - see 7 below.]
1553        - Fixed dftables problem and added an option.
1554        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
1555            HAVE_LONG_LONG.
1556        - Added readline support for pcretest.
1557        - Added an listing of the option settings after cmake has run.
1559    4.  A user submitted a patch to Makefile that makes it easy to create
1560        "pcre.dll" under mingw when using Configure/Make. I added stuff to
1561        Makefile.am that cause it to include this special target, without
1562        affecting anything else. Note that the same mingw target plus all
1563        the other distribution libraries and programs are now supported
1564        when configuring with CMake (see 6 below) instead of with
1565        Configure/Make.
1567    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
1568        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
1569        exported in the Windows port". It has not yet been confirmed that the patch
1570        solves the problem, but it does no harm.
1572    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
1573        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
1574        with CMake, and also correct the comment about stack recursion.
1576    7.  Remove the automatic building of pcredemo from the ./configure system and
1577        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
1578        of a program that users should build themselves after PCRE is installed, so
1579        building it automatically is not really right. What is more, it gave
1580        trouble in some build environments.
1582    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
1585    Version 7.5 10-Jan-08
1586    ---------------------
1588    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
1589        values in parens when parsing an RE using the C++ wrapper."
1591    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
1592        Characters greater than 255 were excluded from the class instead of being
1593        included.
1595    3.  The same bug as (2) above applied to negated POSIX classes such as
1596        [:^space:].
1598    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
1599        defined or documented. It seems to have been a typo for PCRE_STATIC, so
1600        I have changed it.
1602    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
1603        first named subpattern) and a construct such as (?&a) would reference the
1604        first named subpattern whose name started with "a" (in other words, the
1605        length check was missing). Both these problems are fixed. "Subpattern name
1606        expected" is now given for (?&) (a zero-length name), and this patch also
1607        makes it give the same error for \k'' (previously it complained that that
1608        was a reference to a non-existent subpattern).
1610    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
1611        this is right because (?- can be followed by option settings as well as by
1612        digits. I have, however, made the messages clearer.
1614    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
1615        than the number used in the conditional) now cause a compile-time error.
1616        This is actually not compatible with Perl, which accepts such patterns, but
1617        treats the conditional as always being FALSE (as PCRE used to), but it
1618        seems to me that giving a diagnostic is better.
1620    8.  Change "alphameric" to the more common word "alphanumeric" in comments
1621        and messages.
1623    9.  Fix two occurrences of "backslash" in comments that should have been
1624        "backspace".
1626    10. Remove two redundant lines of code that can never be obeyed (their function
1627        was moved elsewhere).
1629    11. The program that makes PCRE's Unicode character property table had a bug
1630        which caused it to generate incorrect table entries for sequences of
1631        characters that have the same character type, but are in different scripts.
1632        It amalgamated them into a single range, with the script of the first of
1633        them. In other words, some characters were in the wrong script. There were
1634        thirteen such cases, affecting characters in the following ranges:
1636          U+002b0 - U+002c1
1637          U+0060c - U+0060d
1638          U+0061e - U+00612
1639          U+0064b - U+0065e
1640          U+0074d - U+0076d
1641          U+01800 - U+01805
1642          U+01d00 - U+01d77
1643          U+01d9b - U+01dbf
1644          U+0200b - U+0200f
1645          U+030fc - U+030fe
1646          U+03260 - U+0327f
1647          U+0fb46 - U+0fbb1
1648          U+10450 - U+1049d
1650    12. The -o option (show only the matching part of a line) for pcregrep was not
1651        compatible with GNU grep in that, if there was more than one match in a
1652        line, it showed only the first of them. It now behaves in the same way as
1653        GNU grep.
1655    13. If the -o and -v options were combined for pcregrep, it printed a blank
1656        line for every non-matching line. GNU grep prints nothing, and pcregrep now
1657        does the same. The return code can be used to tell if there were any
1658        non-matching lines.
1660    14. Added --file-offsets and --line-offsets to pcregrep.
1662    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
1663        infinitely looping recursion. The bug was that positive lookaheads were not
1664        being skipped when checking for a possible empty match (negative lookaheads
1665        and both kinds of lookbehind were skipped).
1667    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
1668        inclusion of <windows.h> to before rather than after the definition of
1669        INVALID_FILE_ATTRIBUTES (patch from David Byron).
1671    17. Specifying a possessive quantifier with a specific limit for a Unicode
1672        character property caused pcre_compile() to compile bad code, which led at
1673        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
1674        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
1675        caused the error; without that there was no problem.
1677    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
1679    19. Added --enable-pcretest-libreadline.
1681    20. In pcrecpp.cc, the variable 'count' was incremented twice in
1682        RE::GlobalReplace(). As a result, the number of replacements returned was
1683        double what it should be. I removed one of the increments, but Craig sent a
1684        later patch that removed the other one (the right fix) and added unit tests
1685        that check the return values (which was not done before).
1687    21. Several CMake things:
1689        (1) Arranged that, when cmake is used on Unix, the libraries end up with
1690            the names libpcre and libpcreposix, not just pcre and pcreposix.
1692        (2) The above change means that pcretest and pcregrep are now correctly
1693            linked with the newly-built libraries, not previously installed ones.
1697    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
1698        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
1699        UTF-8 newline character). The key issue is that the pattern starts .*;
1700        this means that the match must be either at the beginning, or after a
1701        newline. The bug was in the code for advancing after a failed match and
1702        checking that the new position followed a newline. It was not taking
1703        account of UTF-8 characters correctly.
1705    23. PCRE was behaving differently from Perl in the way it recognized POSIX
1706        character classes. PCRE was not treating the sequence [:...:] as a
1707        character class unless the ... were all letters. Perl, however, seems to
1708        allow any characters between [: and :], though of course it rejects as
1709        unknown any "names" that contain non-letters, because all the known class
1710        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
1711        for example, whereas PCRE did not - it did not recognize a POSIX character
1712        class. This seemed a bit dangerous, so the code has been changed to be
1713        closer to Perl. The behaviour is not identical to Perl, because PCRE will
1714        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
1715        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
1716        Perl does, and where it didn't before.
1718    24. Rewrite so as to remove the single use of %n from pcregrep because in some
1719        Windows environments %n is disabled by default.
1722    Version 7.4 21-Sep-07
1723    ---------------------
1725    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
1726        means that a class such as [\s] counted as "explicit reference to CR or
1727        LF". That isn't really right - the whole point of the change was to try to
1728        help when there was an actual mention of one of the two characters. So now
1729        the change happens only if \r or \n (or a literal CR or LF) character is
1730        encountered.
1732    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
1733        of both had grown to the point where there were only 3 bits left.
1734        Fortunately, there was spare space in the data structure, and so I have
1735        moved the internal flags into a new 16-bit field to free up more option
1736        bits.
1738    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
1739        but did not set the internal JCHANGED flag - either of these is enough to
1740        control the way the "get" function works - but the PCRE_INFO_JCHANGED
1741        facility is supposed to tell if (?J) was ever used, so now (?J) at the
1742        start sets both bits.
1744    4.  Added options (at build time, compile time, exec time) to change \R from
1745        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
1747    5.  doc/pcresyntax.html was missing from the distribution.
1749    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
1750        compatibility, even though it is no longer used.
1752    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
1753        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
1754        windows.h file is present (where different names are used). [This was
1755        reversed later after testing - see 16 below.]
1757    8.  Changed all #include <config.h> to #include "config.h". There were also
1758        some further <pcre.h> cases that I changed to "pcre.h".
1760    9.  When pcregrep was used with the --colour option, it missed the line ending
1761        sequence off the lines that it output.
1763    10. It was pointed out to me that arrays of string pointers cause lots of
1764        relocations when a shared library is dynamically loaded. A technique of
1765        using a single long string with a table of offsets can drastically reduce
1766        these. I have refactored PCRE in four places to do this. The result is
1767        dramatic:
1769          Originally:                          290
1770          After changing UCP table:            187
1771          After changing error message table:   43
1772          After changing table of "verbs"       36
1773          After changing table of Posix names   22
1775        Thanks to the folks working on Gregex for glib for this insight.
1777    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
1778        unicode-properties was also set.
1780    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
1782    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
1783        checked only for CRLF.
1785    14. Added casts to pcretest.c to avoid compiler warnings.
1787    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
1789    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
1790        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
1791        entirely. This removes changes made in 7 above.
1793    17. The CMake files have been updated, and there is now more information about
1794        building with CMake in the NON-UNIX-USE document.
1797    Version 7.3 28-Aug-07
1798    ---------------------
1800     1. In the rejigging of the build system that eventually resulted in 7.1, the
1801        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
1802        brackets there is not right, since it causes compilers to look for an
1803        installed pcre.h, not the version that is in the source that is being
1804        compiled (which of course may be different). I have changed it back to:
1806          #include "pcre.h"
1808        I have a vague recollection that the change was concerned with compiling in
1809        different directories, but in the new build system, that is taken care of
1810        by the VPATH setting the Makefile.
1812     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
1813        when the subject happened to end in the byte 0x85 (e.g. if the last
1814        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
1815        characters but of course it shouldn't be taken as a newline when it is part
1816        of another character. The bug was that, for an unlimited repeat of . in
1817        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
1818        characters when looking for a newline.
1820     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
1822     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
1823        in debug output.
1825     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
1826        long printing in the pcrecpp unittest when running under MinGW.
1828     6. ESC_K was left out of the EBCDIC table.
1830     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
1831        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
1832        limit also applies to "virtual nesting" when a pattern is recursive, and in
1833        this case 1000 isn't so big. I have been able to remove this limit at the
1834        expense of backing off one optimization in certain circumstances. Normally,
1835        when pcre_exec() would call its internal match() function recursively and
1836        immediately return the result unconditionally, it uses a "tail recursion"
1837        feature to save stack. However, when a subpattern that can match an empty
1838        string has an unlimited repetition quantifier, it no longer makes this
1839        optimization. That gives it a stack frame in which to save the data for
1840        checking that an empty string has been matched. Previously this was taken
1841        from the 1000-entry workspace that had been reserved. So now there is no
1842        explicit limit, but more stack is used.
1844     8. Applied Daniel's patches to solve problems with the import/export magic
1845        syntax that is required for Windows, and which was going wrong for the
1846        pcreposix and pcrecpp parts of the library. These were overlooked when this
1847        problem was solved for the main library.
1849     9. There were some crude static tests to avoid integer overflow when computing
1850        the size of patterns that contain repeated groups with explicit upper
1851        limits. As the maximum quantifier is 65535, the maximum group length was
1852        set at 30,000 so that the product of these two numbers did not overflow a
1853        32-bit integer. However, it turns out that people want to use groups that
1854        are longer than 30,000 bytes (though not repeat them that many times).
1855        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1856        made it possible to implement the integer overflow checks in a much more
1857        dynamic way, which I have now done. The artificial limitation on group
1858        length has been removed - we now have only the limit on the total length of
1859        the compiled pattern, which depends on the LINK_SIZE setting.
1861    10. Fixed a bug in the documentation for get/copy named substring when
1862        duplicate names are permitted. If none of the named substrings are set, the
1863        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1864        empty string.
1866    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1867        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1868        because the ] is interpreted as the first data character and the
1869        terminating ] is not found. PCRE has been made compatible with Perl in this
1870        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1871        cause memory overwriting.
1873    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1874        string has been matched (to stop an infinite loop). It was not recognizing
1875        a conditional subpattern that could match an empty string if that
1876        subpattern was within another subpattern. For example, it looped when
1877        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
1878        condition was not nested. This bug has been fixed.
1880    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1881        past the start of the subject in the presence of bytes with the top bit
1882        set, for example "\x8aBCD".
1884    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1885        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1887    14. Optimized (?!) to (*FAIL).
1889    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1890        This restricts code points to be within the range 0 to 0x10FFFF, excluding
1891        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1892        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1893        does: it's just the validity check that is more restrictive.
1895    16. Inserted checks for integer overflows during escape sequence (backslash)
1896        processing, and also fixed erroneous offset values for syntax errors during
1897        backslash processing.
1899    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1900        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1902    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1903        caused an overrun.
1905    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1906        something other than just ASCII characters) inside a group that had an
1907        unlimited repeat caused a loop at compile time (while checking to see
1908        whether the group could match an empty string).
1910    20. Debugging a pattern containing \p or \P could cause a crash. For example,
1911        [\P{Any}] did so. (Error in the code for printing property names.)
1913    21. An orphan \E inside a character class could cause a crash.
1915    22. A repeated capturing bracket such as (A)? could cause a wild memory
1916        reference during compilation.
1918    23. There are several functions in pcre_compile() that scan along a compiled
1919        expression for various reasons (e.g. to see if it's fixed length for look
1920        behind). There were bugs in these functions when a repeated \p or \P was
1921        present in the pattern. These operators have additional parameters compared
1922        with \d, etc, and these were not being taken into account when moving along
1923        the compiled data. Specifically:
1925        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1926            length.
1928        (b) An item such as \pL+ within a repeated group could cause crashes or
1929            loops.
1931        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1932            "reference to non-existent subpattern" error.
1934        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1936    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1937        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1939    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1941    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1942        character were causing crashes (broken optimization).
1944    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1945        \p or \P) caused a compile-time loop.
1947    28. More problems have arisen in unanchored patterns when CRLF is a valid line
1948        break. For example, the unstudied pattern [\r\n]A does not match the string
1949        "\r\nA" because change 7.0/46 below moves the current point on by two
1950        characters after failing to match at the start. However, the pattern \nA
1951        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
1952        the same is true. There doesn't seem any very clean way out of this, but
1953        what I have chosen to do makes the common cases work: PCRE now takes note
1954        of whether there can be an explicit match for \r or \n anywhere in the
1955        pattern, and if so, 7.0/46 no longer applies. As part of this change,
1956        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1957        pattern has explicit CR or LF references.
1959    29. Added (*CR) etc for changing newline setting at start of pattern.
1962    Version 7.2 19-Jun-07
1963    ---------------------
1965     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
1966        which is apparently normally available under Windows.
1968     2. Re-jig the pcregrep tests with different newline settings in an attempt
1969        to make them independent of the local environment's newline setting.
1971     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
1973     4. Some of the "internals" tests were previously cut out when the link size
1974        was not 2, because the output contained actual offsets. The recent new
1975        "Z" feature of pcretest means that these can be cut out, making the tests
1976        usable with all link sizes.
1978     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
1979        stack recursion. This gives a massive performance boost under BSD, but just
1980        a small improvement under Linux. However, it saves one field in the frame
1981        in all cases.
1983     6. Added more features from the forthcoming Perl 5.10:
1985        (a) (?-n) (where n is a string of digits) is a relative subroutine or
1986            recursion call. It refers to the nth most recently opened parentheses.
1988        (b) (?+n) is also a relative subroutine call; it refers to the nth next
1989            to be opened parentheses.
1991        (c) Conditions that refer to capturing parentheses can be specified
1992            relatively, for example, (?(-2)... or (?(+3)...
1994        (d) \K resets the start of the current match so that everything before
1995            is not part of it.
1997        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1999        (f) \g{name} is another synonym - part of Perl 5.10's unification of
2000            reference syntax.
2002        (g) (?| introduces a group in which the numbering of parentheses in each
2003            alternative starts with the same number.
2005        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
2007     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
2010     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
2011        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
2012        for detecting groups that can match an empty string.
2014     9. A pattern with a very large number of alternatives (more than several
2015        hundred) was running out of internal workspace during the pre-compile
2016        phase, where pcre_compile() figures out how much memory will be needed. A
2017        bit of new cunning has reduced the workspace needed for groups with
2018        alternatives. The 1000-alternative test pattern now uses 12 bytes of
2019        workspace instead of running out of the 4096 that are available.
2021    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
2023    11. Applied patch from Google to remove an optimization that didn't quite work.
2024        The report of the bug said:
2026          pcrecpp::RE("a*").FullMatch("aaa") matches, while
2027          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
2028          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
2030    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
2031        it matched the wrong number of bytes.
2034    Version 7.1 24-Apr-07
2035    ---------------------
2037     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
2038        that is more "standard", making use of automake and other Autotools. There
2039        is some re-arrangement of the files and adjustment of comments consequent
2040        on this.
2042     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
2043        for recursive directory scanning broke on some systems because the files
2044        are not scanned in any specific order and on different systems the order
2045        was different. A call to "sort" has been inserted into RunGrepTest for the
2046        approprate test as a short-term fix. In the longer term there may be an
2047        alternative.
2049     3. I had an email from Eric Raymond about problems translating some of PCRE's
2050        man pages to HTML (despite the fact that I distribute HTML pages, some
2051        people do their own conversions for various reasons). The problems
2052        concerned the use of low-level troff macros .br and .in. I have therefore
2053        removed all such uses from the man pages (some were redundant, some could
2054        be replaced by .nf/.fi pairs). The 132html script that I use to generate
2055        HTML has been updated to handle .nf/.fi and to complain if it encounters
2056        .br or .in.
2058     4. Updated comments in configure.ac that get placed in config.h.in and also
2059        arranged for config.h to be included in the distribution, with the name
2060        config.h.generic, for the benefit of those who have to compile without
2061        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
2063     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
2064        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
2065        makevp.bat for the current PCRE, using the additional files
2066        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
2068     6. A Windows user reported a minor discrepancy with test 2, which turned out
2069        to be caused by a trailing space on an input line that had got lost in his
2070        copy. The trailing space was an accident, so I've just removed it.
2072     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
2073        that is needed.
2075     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
2076        as "const" (a) because they are and (b) because it helps the PHP
2077        maintainers who have recently made a script to detect big data structures
2078        in the php code that should be moved to the .rodata section. I remembered
2079        to update Builducptable as well, so it won't revert if ucptable.h is ever
2080        re-created.
2082     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
2083        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
2084        order to be able to cut out the UTF-8 tables in the latter when UTF-8
2085        support is not required. This saves 1.5-2K of code, which is important in
2086        some applications.
2088        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
2089        so as not to refer to the tables, even though these functions will never be
2090        called when UTF-8 support is disabled. Otherwise there are problems with a
2091        shared library.
2093    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
2095        (a) It was defining its arguments as char * instead of void *.
2097        (b) It was assuming that all moves were upwards in memory; this was true
2098            a long time ago when I wrote it, but is no longer the case.
2100        The emulated memove() is provided for those environments that have neither
2101        memmove() nor bcopy(). I didn't think anyone used it these days, but that
2102        is clearly not the case, as these two bugs were recently reported.
2104    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
2105        and Detrail to create the HTML documentation, the .txt form of the man
2106        pages, and it removes trailing spaces from listed files. It also creates
2107        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
2108        case, it wraps all the #defines with #ifndefs. This script should be run
2109        before "make dist".
2111    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
2112        with Unicode property support.
2114        (a) For a maximizing quantifier, if the two different cases of the
2115            character were of different lengths in their UTF-8 codings (there are
2116            some cases like this - I found 11), and the matching function had to
2117            back up over a mixture of the two cases, it incorrectly assumed they
2118            were both the same length.
2120        (b) When PCRE was configured to use the heap rather than the stack for
2121            recursion during matching, it was not correctly preserving the data for
2122            the other case of a UTF-8 character when checking ahead for a match
2123            while processing a minimizing repeat. If the check also involved
2124            matching a wide character, but failed, corruption could cause an
2125            erroneous result when trying to check for a repeat of the original
2126            character.
2128    13. Some tidying changes to the testing mechanism:
2130        (a) The RunTest script now detects the internal link size and whether there
2131            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
2132            values substituted by "configure". (The RunGrepTest script already did
2133            this for UTF-8.) The configure.ac script no longer substitutes the
2134            relevant variables.
2136        (b) The debugging options /B and /D in pcretest show the compiled bytecode
2137            with length and offset values. This means that the output is different
2138            for different internal link sizes. Test 2 is skipped for link sizes
2139            other than 2 because of this, bypassing the problem. Unfortunately,
2140            there was also a test in test 3 (the locale tests) that used /B and
2141            failed for link sizes other than 2. Rather than cut the whole test out,
2142            I have added a new /Z option to pcretest that replaces the length and
2143            offset values with spaces. This is now used to make test 3 independent
2144            of link size. (Test 2 will be tidied up later.)
2146    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
2147        segmentation fault instead of returning the appropriate error message.
2149    15. In multiline mode when the newline sequence was set to "any", the pattern
2150        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
2151        This doesn't seem right; it now treats the CRLF combination as the line
2152        ending, and so does not match in that case. It's only a pattern such as ^$
2153        that would hit this one: something like ^ABC$ would have failed after \r
2154        and then tried again after \r\n.
2156    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
2157        in an attempt to make files that differ only in their line terminators
2158        compare equal. This works on Linux.
2160    17. Under certain error circumstances pcregrep might try to free random memory
2161        as it exited. This is now fixed, thanks to valgrind.
2163    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
2164        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
2165        was because its rules for how to advance for /g after matching an empty
2166        string at the end of a line did not allow for this case. They now check for
2167        it specially.
2169    20. pcretest is supposed to handle patterns and data of any length, by
2170        extending its buffers when necessary. It was getting this wrong when the
2171        buffer for a data line had to be extended.
2173    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
2174        CRLF as a newline sequence.
2176    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
2177        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
2178        I have nevertheless tidied it up.
2180    23. Added some casts to kill warnings from HP-UX ia64 compiler.
2182    24. Added a man page for pcre-config.
2185    Version 7.0 19-Dec-06
2186    ---------------------
2188     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
2189        moving to gcc 4.1.1.
2191     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
2192        sys/time.h, which is documented as needed for this function. It doesn't
2193        seem to matter on Linux, but it showed up on some releases of OS X.
2195     3. It seems that there are systems where bytes whose values are greater than
2196        127 match isprint() in the "C" locale. The "C" locale should be the
2197        default when a C program starts up. In most systems, only ASCII printing
2198        characters match isprint(). This difference caused the output from pcretest
2199        to vary, making some of the tests fail. I have changed pcretest so that:
2201        (a) When it is outputting text in the compiled version of a pattern, bytes
2202            other than 32-126 are always shown as hex escapes.
2204        (b) When it is outputting text that is a matched part of a subject string,
2205            it does the same, unless a different locale has been set for the match
2206            (using the /L modifier). In this case, it uses isprint() to decide.
2208     4. Fixed a major bug that caused incorrect computation of the amount of memory
2209        required for a compiled pattern when options that changed within the
2210        pattern affected the logic of the preliminary scan that determines the
2211        length. The relevant options are -x, and -i in UTF-8 mode. The result was
2212        that the computed length was too small. The symptoms of this bug were
2213        either the PCRE error "internal error: code overflow" from pcre_compile(),
2214        or a glibc crash with a message such as "pcretest: free(): invalid next
2215        size (fast)". Examples of patterns that provoked this bug (shown in
2216        pcretest format) are:
2218          /(?-x: )/x
2219          /(?x)(?-x: \s*#\s*)/
2220          /((?i)[\x{c0}])/8
2221          /(?i:[\x{c0}])/8
2223        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
2224        is now done differently.
2226     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
2227        wrapper classes; (b) implement a new function in the C++ scanner that is
2228        more efficient than the old way of doing things because it avoids levels of
2229        recursion in the regex matching; (c) add a paragraph to the documentation
2230        for the FullMatch() function.
2232     6. The escape sequence \n was being treated as whatever was defined as
2233        "newline". Not only was this contrary to the documentation, which states
2234        that \n is character 10 (hex 0A), but it also went horribly wrong when
2235        "newline" was defined as CRLF. This has been fixed.
2237     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
2238        was being set to -1 for the "end of line" case (supposedly a value that no
2239        character can have). Though this value is never used (the check for end of
2240        line is "zero bytes in current character"), it caused compiler complaints.
2241        I've changed it to 0xffffffff.
2243     8. In pcre_version.c, the version string was being built by a sequence of
2244        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
2245        string (as it is for production releases) called a macro with an empty
2246        argument. The C standard says the result of this is undefined. The gcc
2247        compiler treats it as an empty string (which was what was wanted) but it is
2248        reported that Visual C gives an error. The source has been hacked around to
2249        avoid this problem.
2251     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
2252        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
2253        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
2254        of them did).
2256    10. Originally, pcretest opened its input and output without "b"; then I was
2257        told that "b" was needed in some environments, so it was added for release
2258        5.0 to both the input and output. (It makes no difference on Unix-like
2259        systems.) Later I was told that it is wrong for the input on Windows. I've
2260        now abstracted the modes into two macros, to make it easier to fiddle with
2261        them, and removed "b" from the input mode under Windows.
2263    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
2265    12. Added -help and --help to pcretest as an official way of being reminded
2266        of the options.
2268    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
2269        and pcrecpp.cc because they annoy compilers at high warning levels.
2271    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
2273    15. Fixed an occurrence of == in configure.ac that should have been = (shell
2274        scripts are not C programs :-) and which was not noticed because it works
2275        on Linux.
2277    16. pcretest is supposed to handle any length of pattern and data line (as one
2278        line or as a continued sequence of lines) by extending its input buffer if
2279        necessary. This feature was broken for very long pattern lines, leading to
2280        a string of junk being passed to pcre_compile() if the pattern was longer
2281        than about 50K.
2283    17. I have done a major re-factoring of the way pcre_compile() computes the
2284        amount of memory needed for a compiled pattern. Previously, there was code
2285        that made a preliminary scan of the pattern in order to do this. That was
2286        OK when PCRE was new, but as the facilities have expanded, it has become
2287        harder and harder to keep it in step with the real compile phase, and there
2288        have been a number of bugs (see for example, 4 above). I have now found a
2289        cunning way of running the real compile function in a "fake" mode that
2290        enables it to compute how much memory it would need, while actually only
2291        ever using a few hundred bytes of working memory and without too many
2292        tests of the mode. This should make future maintenance and development
2293        easier. A side effect of this work is that the limit of 200 on the nesting
2294        depth of parentheses has been removed (though this was never a serious
2295        limitation, I suspect). However, there is a downside: pcre_compile() now
2296        runs more slowly than before (30% or more, depending on the pattern). I
2297        hope this isn't a big issue. There is no effect on runtime performance.
2299    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
2300        newline (only possible for the last line of a file) and it was a
2301        pattern that set a locale (followed by /Lsomething), pcretest crashed.
2303    19. Added additional timing features to pcretest. (1) The -tm option now times
2304        matching only, not compiling. (2) Both -t and -tm can be followed, as a
2305        separate command line item, by a number that specifies the number of
2306        repeats to use when timing. The default is 50000; this gives better
2307        precision, but takes uncomfortably long for very large patterns.
2309    20. Extended pcre_study() to be more clever in cases where a branch of a
2310        subpattern has no definite first character. For example, (a*|b*)[cd] would
2311        previously give no result from pcre_study(). Now it recognizes that the
2312        first character must be a, b, c, or d.
2314    21. There was an incorrect error "recursive call could loop indefinitely" if
2315        a subpattern (or the entire pattern) that was being tested for matching an
2316        empty string contained only one non-empty item after a nested subpattern.
2317        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
2318        incorrectly, because the \d was being skipped in the check.
2320    22. The pcretest program now has a new pattern option /B and a command line
2321        option -b, which is equivalent to adding /B to every pattern. This causes
2322        it to show the compiled bytecode, without the additional information that
2323        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
2324        is the same as /B/I).
2326    23. A new optimization is now able automatically to treat some sequences such
2327        as a*b as a*+b. More specifically, if something simple (such as a character
2328        or a simple class like \d) has an unlimited quantifier, and is followed by
2329        something that cannot possibly match the quantified thing, the quantifier
2330        is automatically "possessified".
2332    24. A recursive reference to a subpattern whose number was greater than 39
2333        went wrong under certain circumstances in UTF-8 mode. This bug could also
2334        have affected the operation of pcre_study().
2336    25. Realized that a little bit of performance could be had by replacing
2337        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
2339    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
2341    27. Possessive quantifiers such as a++ were previously implemented by turning
2342        them into atomic groups such as ($>a+). Now they have their own opcodes,
2343        which improves performance. This includes the automatically created ones
2344        from 23 above.
2346    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
2347        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
2348        the first matched character to be a colon. This applied both to named and
2349        numbered groups.
2351    29. The ucpinternal.h header file was missing its idempotency #ifdef.
2353    30. I was sent a "project" file called libpcre.a.dev which I understand makes
2354        building PCRE on Windows easier, so I have included it in the distribution.
2356    31. There is now a check in pcretest against a ridiculously large number being
2357        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
2358        loop, the loop is abandoned.
2360    32. Forward references to subpatterns in conditions such as (?(2)...) where
2361        subpattern 2 is defined later cause pcre_compile() to search forwards in
2362        the pattern for the relevant set of parentheses. This search went wrong
2363        when there were unescaped parentheses in a character class, parentheses
2364        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
2366    33. "Subroutine" calls and backreferences were previously restricted to
2367        referencing subpatterns earlier in the regex. This restriction has now
2368        been removed.
2370    34. Added a number of extra features that are going to be in Perl 5.10. On the
2371        whole, these are just syntactic alternatives for features that PCRE had
2372        previously implemented using the Python syntax or my own invention. The
2373        other formats are all retained for compatibility.
2375        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
2376            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
2377            also .NET compatible.
2379        (b) A recursion or subroutine call to a named group can now be defined as
2380            (?&name) as well as (?P>name).
2382        (c) A backreference to a named group can now be defined as \k<name> or
2383            \k'name' as well as (?P=name). The new forms, as well as being in Perl
2384            5.10, are also .NET compatible.
2386        (d) A conditional reference to a named group can now use the syntax
2387            (?(<name>) or (?('name') as well as (?(name).
2389        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
2390            groups (named and numbered) that are never evaluated inline, but can be
2391            called as "subroutines" from elsewhere. In effect, the DEFINE condition
2392            is always false. There may be only one alternative in such a group.
2394        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
2395            as the simple (?(R). The condition is true only if the most recent
2396            recursion is that of the given number or name. It does not search out
2397            through the entire recursion stack.
2399        (g) The escape \gN or \g{N} has been added, where N is a positive or
2400            negative number, specifying an absolute or relative reference.
2402    35. Tidied to get rid of some further signed/unsigned compiler warnings and
2403        some "unreachable code" warnings.
2405    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
2406        things, this adds five new scripts.
2408    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
2409        There were also incompatibilities regarding the handling of \Q..\E inside
2410        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
2411        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
2413    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
2414        matches an empty string, and forcibly breaks the loop. There were bugs in
2415        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
2416        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
2417        separate and independent bugs (that affected different cases) have been
2418        fixed.
2420    39. Refactored the code to abolish the use of different opcodes for small
2421        capturing bracket numbers. This is a tidy that I avoided doing when I
2422        removed the limit on the number of capturing brackets for 3.5 back in 2001.
2423        The new approach is not only tidier, it makes it possible to reduce the
2424        memory needed to fix the previous bug (38).
2426    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
2427        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
2428        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
2429        mode.
2431    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
2432        report.
2434    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
2435        copy construction and assignment in the C++ wrapper.
2437    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
2438        couple of bugs that could have given wrong results in the "--newline=crlf"
2439        case.
2441    44. Added a number of casts and did some reorganization of signed/unsigned int
2442        variables following suggestions from Dair Grant. Also renamed the variable
2443        "this" as "item" because it is a C++ keyword.
2445    45. Arranged for dftables to add
2447          #include "pcre_internal.h"
2449        to pcre_chartables.c because without it, gcc 4.x may remove the array
2450        definition from the final binary if PCRE is built into a static library and
2451        dead code stripping is activated.
2453    46. For an unanchored pattern, if a match attempt fails at the start of a
2454        newline sequence, and the newline setting is CRLF or ANY, and the next two
2455        characters are CRLF, advance by two characters instead of one.
2458    Version 6.7 04-Jul-06
2459    ---------------------
2461     1. In order to handle tests when input lines are enormously long, pcretest has
2462        been re-factored so that it automatically extends its buffers when
2463        necessary. The code is crude, but this _is_ just a test program. The
2464        default size has been increased from 32K to 50K.
2466     2. The code in pcre_study() was using the value of the re argument before
2467        testing it for NULL. (Of course, in any sensible call of the function, it
2468        won't be NULL.)
2470     3. The memmove() emulation function in pcre_internal.h, which is used on
2471        systems that lack both memmove() and bcopy() - that is, hardly ever -
2472        was missing a "static" storage class specifier.
2474     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
2475        containing an extended class (one that cannot be represented by a bitmap
2476        because it contains high-valued characters or Unicode property items, e.g.
2477        [\pZ]). Almost always one would set UTF-8 mode when processing such a
2478        pattern, but PCRE should not loop if you do not (it no longer does).
2479        [Detail: two cases were found: (a) a repeated subpattern containing an
2480        extended class; (b) a recursive reference to a subpattern that followed a
2481        previous extended class. It wasn't skipping over the extended class
2482        correctly when UTF-8 mode was not set.]
2484     5. A negated single-character class was not being recognized as fixed-length
2485        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
2486        compile error "lookbehind assertion is not fixed length".
2488     6. The RunPerlTest auxiliary script was showing an unexpected difference
2489        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
2490        write a Perl script that can interpret lines of an input file either as
2491        byte characters or as UTF-8, which is what "perltest" was being required to
2492        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
2493        can't do is switch easily at run time between having the "use utf8;" pragma
2494        or not. In the end, I fudged it by using the RunPerlTest script to insert
2495        "use utf8;" explicitly for the UTF-8 tests.
2497     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
2498        the end of the subject string, contrary to the documentation and to what
2499        Perl does. This was true of both matching functions. Now it matches only at
2500        the start of the subject and immediately after *internal* newlines.
2502     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
2503        a pointer to an int instead of a pointer to an unsigned long int. This
2504        caused problems on 64-bit systems.
2506     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
2507        instance of the 'standard' template library not being so standard".
2509    10. There was no check on the number of named subpatterns nor the maximum
2510        length of a subpattern name. The product of these values is used to compute
2511        the size of the memory block for a compiled pattern. By supplying a very
2512        long subpattern name and a large number of named subpatterns, the size
2513        computation could be caused to overflow. This is now prevented by limiting
2514        the length of names to 32 characters, and the number of named subpatterns
2515        to 10,000.
2517    11. Subpatterns that are repeated with specific counts have to be replicated in
2518        the compiled pattern. The size of memory for this was computed from the
2519        length of the subpattern and the repeat count. The latter is limited to
2520        65535, but there was no limit on the former, meaning that integer overflow
2521        could in principle occur. The compiled length of a repeated subpattern is
2522        now limited to 30,000 bytes in order to prevent this.
2524    12. Added the optional facility to have named substrings with the same name.
2526    13. Added the ability to use a named substring as a condition, using the
2527        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
2528        are numbers (not recommended). Forward references are permitted.
2530    14. Added forward references in named backreferences (if you see what I mean).
2532    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
2533        pattern could run off the end of the subject. For example, the pattern
2534        "(?s)(.{1,5})"8 did this with the subject "ab".
2536    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
2537        PCRE_CASELESS was set when matching characters that were quantified with ?
2538        or *.
2540    17. A character class other than a single negated character that had a minimum
2541        but no maximum quantifier - for example [ab]{6,} - was not handled
2542        correctly by pce_dfa_exec(). It would match only one character.
2544    18. A valid (though odd) pattern that looked like a POSIX character
2545        class but used an invalid character after [ (for example [[,abc,]]) caused
2546        pcre_compile() to give the error "Failed: internal error: code overflow" or
2547        in some cases to crash with a glibc free() error. This could even happen if
2548        the pattern terminated after [[ but there just happened to be a sequence of
2549        letters, a binary zero, and a closing ] in the memory that followed.
2551    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
2552        over the years. Originally (before any Unicode support), just the bottom 8
2553        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
2554        output from "man perlunicode" includes this:
2556          The regular expression compiler produces polymorphic opcodes.  That
2557          is, the pattern adapts to the data and automatically switches to
2558          the Unicode character scheme when presented with Unicode data--or
2559          instead uses a traditional byte scheme when presented with byte
2560          data.
2562        Sadly, a wide octal escape does not cause a switch, and in a string with
2563        no other multibyte characters, these octal escapes are treated as before.
2564        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
2565        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
2566        Unicode string.
2568        I have not perpetrated such confusion in PCRE. Up till now, it took just
2569        the bottom 8 bits, as in old Perl. I have now made octal escapes with
2570        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
2571        translate to the appropriate multibyte character.
2573    29. Applied some refactoring to reduce the number of warnings from Microsoft
2574        and Borland compilers. This has included removing the fudge introduced
2575        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
2576        a warning about an unused variable.
2578    21. PCRE has not included VT (character 0x0b) in the set of whitespace
2579        characters since release 4.0, because Perl (from release 5.004) does not.
2580        [Or at least, is documented not to: some releases seem to be in conflict
2581        with the documentation.] However, when a pattern was studied with
2582        pcre_study() and all its branches started with \s, PCRE still included VT
2583        as a possible starting character. Of course, this did no harm; it just
2584        caused an unnecessary match attempt.
2586    22. Removed a now-redundant internal flag bit that recorded the fact that case
2587        dependency changed within the pattern. This was once needed for "required
2588        byte" processing, but is no longer used. This recovers a now-scarce options
2589        bit. Also moved the least significant internal flag bit to the most-
2590        significant bit of the word, which was not previously used (hangover from
2591        the days when it was an int rather than a uint) to free up another bit for
2592        the future.
2594    23. Added support for CRLF line endings as well as CR and LF. As well as the
2595        default being selectable at build time, it can now be changed at runtime
2596        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
2597        specify that it is scanning data with non-default line endings.
2599    24. Changed the definition of CXXLINK to make it agree with the definition of
2600        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
2602    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
2603        recursions. This makes a big different to stack usage for some patterns.
2605    26. If a subpattern containing a named recursion or subroutine reference such
2606        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
2607        the space required for the compiled pattern went wrong and gave too small a
2608        value. Depending on the environment, this could lead to "Failed: internal
2609        error: code overflow at offset 49" or "glibc detected double free or
2610        corruption" errors.
2612    27. Applied patches from Google (a) to support the new newline modes and (b) to
2613        advance over multibyte UTF-8 characters in GlobalReplace.
2615    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
2616        difference for some implementation of PCRE in some Windows version.
2618    29. Added some extra testing facilities to pcretest:
2620        \q<number>   in a data line sets the "match limit" value
2621        \Q<number>   in a data line sets the "match recursion limt" value
2622        -S <number>  sets the stack size, where <number> is in megabytes
2624        The -S option isn't available for Windows.
2627    Version 6.6 06-Feb-06
2628    ---------------------
2630     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
2631        in pcreposix.h. I have copied the definition from pcre.h.
2633     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
2634        because pcre.h is no longer a built file.
2636     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
2637        not normally included in the compiled code.
2640    Version 6.5 01-Feb-06
2641    ---------------------
2643     1. When using the partial match feature with pcre_dfa_exec(), it was not
2644        anchoring the second and subsequent partial matches at the new starting
2645        point. This could lead to incorrect results. For example, with the pattern
2646        /1234/, partially matching against "123" and then "a4" gave a match.
2648     2. Changes to pcregrep:
2650        (a) All non-match returns from pcre_exec() were being treated as failures
2651            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
2652            error message is output. Some extra information is given for the
2653            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
2654            probably the only errors that are likely to be caused by users (by
2655            specifying a regex that has nested indefinite repeats, for instance).
2656            If there are more than 20 of these errors, pcregrep is abandoned.
2658        (b) A binary zero was treated as data while matching, but terminated the
2659            output line if it was written out. This has been fixed: binary zeroes
2660            are now no different to any other data bytes.
2662        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
2663            used to set a locale for matching. The --locale=xxxx long option has
2664            been added (no short equivalent) to specify a locale explicitly on the
2665            pcregrep command, overriding the environment variables.
2667        (d) When -B was used with -n, some line numbers in the output were one less
2668            than they should have been.
2670        (e) Added the -o (--only-matching) option.
2672        (f) If -A or -C was used with -c (count only), some lines of context were
2673            accidentally printed for the final match.
2675        (g) Added the -H (--with-filename) option.
2677        (h) The combination of options -rh failed to suppress file names for files
2678            that were found from directory arguments.
2680        (i) Added the -D (--devices) and -d (--directories) options.
2682        (j) Added the -F (--fixed-strings) option.
2684        (k) Allow "-" to be used as a file name for -f as well as for a data file.
2686        (l) Added the --colo(u)r option.
2688        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
2689            is not present by default.
2691     3. A nasty bug was discovered in the handling of recursive patterns, that is,
2692        items such as (?R) or (?1), when the recursion could match a number of
2693        alternatives. If it matched one of the alternatives, but subsequently,
2694        outside the recursion, there was a failure, the code tried to back up into
2695        the recursion. However, because of the way PCRE is implemented, this is not
2696        possible, and the result was an incorrect result from the match.
2698        In order to prevent this happening, the specification of recursion has
2699        been changed so that all such subpatterns are automatically treated as
2700        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
2702     4. I had overlooked the fact that, in some locales, there are characters for
2703        which isalpha() is true but neither isupper() nor islower() are true. In
2704        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
2705        and ordfeminine) are like this. This affected the treatment of \w and \W
2706        when they appeared in character classes, but not when they appeared outside
2707        a character class. The bit map for "word" characters is now created
2708        separately from the results of isalnum() instead of just taking it from the
2709        upper, lower, and digit maps. (Plus the underscore character, of course.)
2711     5. The above bug also affected the handling of POSIX character classes such as
2712        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
2713        permanent tables. Instead, the bit maps for such a class were previously
2714        created as the appropriate unions of the upper, lower, and digit bitmaps.
2715        Now they are created by subtraction from the [[:word:]] class, which has
2716        its own bitmap.
2718     6. The [[:blank:]] character class matches horizontal, but not vertical space.
2719        It is created by subtracting the vertical space characters (\x09, \x0a,
2720        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
2721        subtraction was done in the overall bitmap for a character class, meaning
2722        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
2723        be recognized. This bug has been fixed.
2725     7. Patches from the folks at Google:
2727          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
2728          real life, but is still worth protecting against".
2730          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
2731          regular expressions".
2733          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
2734          have it.
2736          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
2737          "configure" and the latter not, in order to fix a problem somebody had
2738          with compiling the Arg class on HP-UX.
2740          (e) Improve the error-handling of the C++ wrapper a little bit.
2742          (f) New tests for checking recursion limiting.
2744     8. The pcre_memmove() function, which is used only if the environment does not
2745        have a standard memmove() function (and is therefore rarely compiled),
2746        contained two bugs: (a) use of int instead of size_t, and (b) it was not
2747        returning a result (though PCRE never actually uses the result).
2749     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
2750        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
2751        returned instead of calling malloc() with an overflowing number that would
2752        most likely cause subsequent chaos.
2754    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
2756    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
2757        with this option is matched, the nmatch and pmatch options of regexec() are
2758        ignored.
2760    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
2761        provided in case anyone wants to the the POSIX interface with UTF-8
2762        strings.
2764    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
2765        C++ linking (needed for some HP-UX environments).
2767    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
2768        (unused parameter) and in the pcre_printint() function (omitted "default"
2769        switch label when the default is to do nothing).
2771    15. Added some code to make it possible, when PCRE is compiled as a C++
2772        library, to replace subject pointers for pcre_exec() with a smart pointer
2773        class, thus making it possible to process discontinuous strings.
2775    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
2776        much the same function. They were added by different people who were trying
2777        to make PCRE easy to compile on non-Unix systems. It has been suggested
2778        that PCRE_EXPORT be abolished now that there is more automatic apparatus
2779        for compiling on Windows systems. I have therefore replaced it with
2780        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
2781        defaults to "extern" for C or "extern C" for C++, which works fine on
2782        Unix-like systems. It is now possible to override the value of PCRE_DATA_
2783        SCOPE with something explicit in config.h. In addition:
2785        (a) pcreposix.h still had just "extern" instead of either of these macros;
2786            I have replaced it with PCRE_DATA_SCOPE.
2788        (b) Functions such as _pcre_xclass(), which are internal to the library,
2789            but external in the C sense, all had PCRE_EXPORT in their definitions.
2790            This is apparently wrong for the Windows case, so I have removed it.
2791            (It makes no difference on Unix-like systems.)
2793    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
2794        of recursive calls to match(). This is different to MATCH_LIMIT because
2795        that limits the total number of calls to match(), not all of which increase
2796        the depth of recursion. Limiting the recursion depth limits the amount of
2797        stack (or heap if NO_RECURSE is set) that is used. The default can be set
2798        when PCRE is compiled, and changed at run time. A patch from Google adds
2799        this functionality to the C++ interface.
2801    18. Changes to the handling of Unicode character properties:
2803        (a) Updated the table to Unicode 4.1.0.
2805        (b) Recognize characters that are not in the table as "Cn" (undefined).
2807        (c) I revised the way the table is implemented to a much improved format
2808            which includes recognition of ranges. It now supports the ranges that
2809            are defined in UnicodeData.txt, and it also amalgamates other
2810            characters into ranges. This has reduced the number of entries in the
2811            table from around 16,000 to around 3,000, thus reducing its size
2812            considerably. I realized I did not need to use a tree structure after
2813            all - a binary chop search is just as efficient. Having reduced the
2814            number of entries, I extended their size from 6 bytes to 8 bytes to
2815            allow for more data.
2817        (d) Added support for Unicode script names via properties such as \p{Han}.
2819    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
2820        matching that character.
2822    20. When matching a repeated Unicode property with a minimum greater than zero,
2823        (for example \pL{2,}), PCRE could look past the end of the subject if it
2824        reached it while seeking the minimum number of characters. This could
2825        happen only if some of the characters were more than one byte long, because
2826        there is a check for at least the minimum number of bytes.
2828    21. Refactored the implementation of \p and \P so as to be more general, to
2829        allow for more different types of property in future. This has changed the
2830        compiled form incompatibly. Anybody with saved compiled patterns that use
2831        \p or \P will have to recompile them.
2833    22. Added "Any" and "L&" to the supported property types.
2835    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
2836        but give a compile time error if the value is greater than 0xff.
2838    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
2839        accidentally not being installed or uninstalled.
2841    25. The pcre.h file was built from pcre.h.in, but the only changes that were
2842        made were to insert the current release number. This seemed silly, because
2843        it made things harder for people building PCRE on systems that don't run
2844        "configure". I have turned pcre.h into a distributed file, no longer built
2845        by "configure", with the version identification directly included. There is
2846        no longer a pcre.h.in file.
2848        However, this change necessitated a change to the pcre-config script as
2849        well. It is built from pcre-config.in, and one of the substitutions was the
2850        release number. I have updated configure.ac so that ./configure now finds
2851        the release number by grepping pcre.h.
2853    26. Added the ability to run the tests under valgrind.
2856    Version 6.4 05-Sep-05
2857    ---------------------
2859     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
2860        "--" to be printed when multiple files were scanned, even when none of the
2861        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
2862        consider it to be a bug, and have restored the previous behaviour.
2864     2. A couple of code tidies to get rid of compiler warnings.
2866     3. The pcretest program used to cheat by referring to symbols in the library
2867        whose names begin with _pcre_. These are internal symbols that are not
2868        really supposed to be visible externally, and in some environments it is
2869        possible to suppress them. The cheating is now confined to including
2870        certain files from the library's source, which is a bit cleaner.
2872     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
2873        file's purpose clearer.
2875     5. Reorganized pcre_ucp_findchar().
2878    Version 6.3 15-Aug-05
2879    ---------------------
2881     1. The file libpcre.pc.in did not have general read permission in the tarball.
2883     2. There were some problems when building without C++ support:
2885        (a) If C++ support was not built, "make install" and "make test" still
2886            tried to test it.
2888        (b) There were problems when the value of CXX was explicitly set. Some
2889            changes have been made to try to fix these, and ...
2891        (c) --disable-cpp can now be used to explicitly disable C++ support.
2893        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
2894            backslash in a target when C++ was disabled. This confuses some
2895            versions of "make", apparently. Using an intermediate variable solves
2896            this. (Same for CPP_LOBJ.)
2898     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
2899        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
2900        necessary on certain architectures.
2902     4. Added a setting of -export-symbols-regex to the link command to remove
2903        those symbols that are exported in the C sense, but actually are local
2904        within the library, and not documented. Their names all begin with
2905        "_pcre_". This is not a perfect job, because (a) we have to except some
2906        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
2907        available (and never for static libraries). I have made a note to try to
2908        find a way round (a) in the future.
2911    Version 6.2 01-Aug-05
2912    ---------------------
2914     1. There was no test for integer overflow of quantifier values. A construction
2915        such as {1111111111111111} would give undefined results. What is worse, if
2916        a minimum quantifier for a parenthesized subpattern overflowed and became
2917        negative, the calculation of the memory size went wrong. This could have
2918        led to memory overwriting.
2920     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
2922     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
2923        operating environments where this matters.
2925     4. Applied Giuseppe Maxia's patch to add additional features for controlling
2926        PCRE options from within the C++ wrapper.
2928     5. Named capturing subpatterns were not being correctly counted when a pattern
2929        was compiled. This caused two problems: (a) If there were more than 100
2930        such subpatterns, the calculation of the memory needed for the whole
2931        compiled pattern went wrong, leading to an overflow error. (b) Numerical
2932        back references of the form \12, where the number was greater than 9, were
2933        not recognized as back references, even though there were sufficient
2934        previous subpatterns.
2936     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
2937        versions of gcc, e.g. 2.95.4.
2940    Version 6.1 21-Jun-05
2941    ---------------------
2943     1. There was one reference to the variable "posix" in pcretest.c that was not
2944        surrounded by "#if !defined NOPOSIX".
2946     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
2947        the cross-check on the old pcre_info() function, for the benefit of the
2948        cut-down version of PCRE that is currently imported into Exim.
2950     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
2951        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
2952        patterns that start (?i) but I don't think that matters. The use of (?i) is
2953        just an example; this all applies to the other options as well.
2955     4. Since libtool seems to echo the compile commands it is issuing, the output
2956        from "make" can be reduced a bit by putting "@" in front of each libtool
2957        compile command.
2959     5. Patch from the folks at Google for configure.in to be a bit more thorough
2960        in checking for a suitable C++ installation before trying to compile the
2961        C++ stuff. This should fix a reported problem when a compiler was present,
2962        but no suitable headers.
2964     6. The man pages all had just "PCRE" as their title. I have changed them to
2965        be the relevant file name. I have also arranged that these names are
2966        retained in the file doc/pcre.txt, which is a concatenation in text format
2967        of all the man pages except the little individual ones for each function.
2969     7. The NON-UNIX-USE file had not been updated for the different set of source
2970        files that come with release 6. I also added a few comments about the C++
2971        wrapper.
2974    Version 6.0 07-Jun-05
2975    ---------------------
2977     1. Some minor internal re-organization to help with my DFA experiments.
2979     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
2980        didn't matter for the library itself when fully configured, but did matter
2981        when compiling without UCP support, or within Exim, where the ucp files are
2982        not imported.
2984     3. Refactoring of the library code to split up the various functions into
2985        different source modules. The addition of the new DFA matching code (see
2986        below) to a single monolithic source would have made it really too
2987        unwieldy, quite apart from causing all the code to be include in a
2988        statically linked application, when only some functions are used. This is
2989        relevant even without the DFA addition now that patterns can be compiled in
2990        one application and matched in another.
2992        The downside of splitting up is that there have to be some external
2993        functions and data tables that are used internally in different modules of
2994        the library but which are not part of the API. These have all had their
2995        names changed to start with "_pcre_" so that they are unlikely to clash
2996        with other external names.
2998     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
2999        a different (DFA) algorithm. Although it is slower than the original
3000        function, it does have some advantages for certain types of matching
3001        problem.
3003     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
3004        including restarting after a partial match.
3006     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
3007        defined when compiling for Windows was sent to me. I have put it into the
3008        code, though I have no means of testing or verifying it.
3010     7. Added the pcre_refcount() auxiliary function.
3012     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
3013        match before or at the first newline in the subject string. In pcretest,
3014        the /f option on a pattern can be used to set this.
3016     9. A repeated \w when used in UTF-8 mode with characters greater than 256
3017        would behave wrongly. This has been present in PCRE since release 4.0.
3019    10. A number of changes to the pcregrep command:
3021        (a) Refactored how -x works; insert ^(...)$ instead of setting
3022            PCRE_ANCHORED and checking the length, in preparation for adding
3023            something similar for -w.
3025        (b) Added the -w (match as a word) option.
3027        (c) Refactored the way lines are read and buffered so as to have more
3028            than one at a time available.
3030        (d) Implemented a pcregrep test script.
3032        (e) Added the -M (multiline match) option. This allows patterns to match
3033            over several lines of the subject. The buffering ensures that at least
3034            8K, or the rest of the document (whichever is the shorter) is available
3035            for matching (and similarly the previous 8K for lookbehind assertions).
3037        (f) Changed the --help output so that it now says
3039              -w, --word-regex(p)
3041            instead of two lines, one with "regex" and the other with "regexp"
3042            because that confused at least one person since the short forms are the
3043            same. (This required a bit of code, as the output is generated
3044            automatically from a table. It wasn't just a text change.)
3046        (g) -- can be used to terminate pcregrep options if the next thing isn't an
3047            option but starts with a hyphen. Could be a pattern or a path name
3048            starting with a hyphen, for instance.
3050        (h) "-" can be given as a file name to represent stdin.
3052        (i) When file names are being printed, "(standard input)" is used for
3053            the standard input, for compatibility with GNU grep. Previously
3054            "<stdin>" was used.
3056        (j) The option --label=xxx can be used to supply a name to be used for
3057            stdin when file names are being printed. There is no short form.
3059        (k) Re-factored the options decoding logic because we are going to add
3060            two more options that take data. Such options can now be given in four
3061            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
3063        (l) Added the -A, -B, and -C options for requesting that lines of context
3064            around matches be printed.
3066        (m) Added the -L option to print the names of files that do not contain
3067            any matching lines, that is, the complement of -l.
3069        (n) The return code is 2 if any file cannot be opened, but pcregrep does
3070            continue to scan other files.
3072        (o) The -s option was incorrectly implemented. For compatibility with other
3073            greps, it now suppresses the error message for a non-existent or non-
3074            accessible file (but not the return code). There is a new option called
3075            -q that suppresses the output of matching lines, which was what -s was
3076            previously doing.
3078        (p) Added --include and --exclude options to specify files for inclusion
3079            and exclusion when recursing.
3081    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
3082        Hopefully, it now does.
3084    12. Missing cast in pcre_study().
3086    13. Added an "uninstall" target to the makefile.
3088    14. Replaced "extern" in the function prototypes in Makefile.in with
3089        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
3090        world, but is set differently for Windows.
3092    15. Added a second compiling function called pcre_compile2(). The only
3093        difference is that it has an extra argument, which is a pointer to an
3094        integer error code. When there is a compile-time failure, this is set
3095        non-zero, in addition to the error test pointer being set to point to an
3096        error message. The new argument may be NULL if no error number is required
3097        (but then you may as well call pcre_compile(), which is now just a
3098        wrapper). This facility is provided because some applications need a
3099        numeric error indication, but it has also enabled me to tidy up the way
3100        compile-time errors are handled in the POSIX wrapper.
3102    16. Added VPATH=.libs to the makefile; this should help when building with one
3103        prefix path and installing with another. (Or so I'm told by someone who
3104        knows more about this stuff than I do.)
3106    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
3107        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
3108        match everything, including newlines. This is not POSIX-compatible, but
3109        somebody wanted the feature. From pcretest it can be activated by using
3110        both the P and the s flags.
3112    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
3114    19. libpcre.pc was being incorrectly installed as executable.
3116    20. A couple of places in pcretest check for end-of-line by looking for '\n';
3117        it now also looks for '\r' so that it will work unmodified on Windows.
3119    21. Added Google's contributed C++ wrapper to the distribution.
3121    22. Added some untidy missing memory free() calls in pcretest, to keep
3122        Electric Fence happy when testing.
3126    Version 5.0 13-Sep-04
3127    ---------------------
3129     1. Internal change: literal characters are no longer packed up into items
3130        containing multiple characters in a single byte-string. Each character
3131        is now matched using a separate opcode. However, there may be more than one
3132        byte in the character in UTF-8 mode.
3134     2. The pcre_callout_block structure has two new fields: pattern_position and
3135        next_item_length. These contain the offset in the pattern to the next match
3136        item, and its length, respectively.
3138     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
3139        insertion of callouts before each pattern item. Added the /C option to
3140        pcretest to make use of this.
3142     4. On the advice of a Windows user, the lines
3144          #if defined(_WIN32) || defined(WIN32)
3145          _setmode( _fileno( stdout ), 0x8000 );
3146          #endif  /* defined(_WIN32) || defined(WIN32) */
3148        have been added to the source of pcretest. This apparently does useful
3149        magic in relation to line terminators.
3151     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
3152        for the benefit of those environments where the "b" makes a difference.
3154     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
3155        to know about it. I have put a hack into configure.in that adds in code
3156        to set GCC=yes if CC=icc. This seems to end up at a point in the
3157        generated configure script that is early enough to affect the setting of
3158        compiler options, which is what is needed, but I have no means of testing
3159        whether it really works. (The user who reported this had patched the
3160        generated configure script, which of course I cannot do.)
3162        LATER: After change 22 below (new libtool files), the configure script
3163        seems to know about icc (and also ecc). Therefore, I have commented out
3164        this hack in configure.in.
3166     7. Added support for pkg-config (2 patches were sent in).
3168     8. Negated POSIX character classes that used a combination of internal tables
3169        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
3170        [[:^ascii]]. Typically, they would match almost any characters. The other
3171        POSIX classes were not broken in this way.
3173     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
3174        to find the match, as PCRE was deluded into thinking that the match had to
3175        start at the start point or following a newline. The same bug applied to
3176        patterns with negative forward assertions or any backward assertions
3177        preceding ".*" at the start, unless the pattern required a fixed first
3178        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
3180    10. In UTF-8 mode, when moving forwards in the subject after a failed match
3181        starting at the last subject character, bytes beyond the end of the subject
3182        string were read.
3184    11. Renamed the variable "class" as "classbits" to make life easier for C++
3185        users. (Previously there was a macro definition, but it apparently wasn't
3186        enough.)
3188    12. Added the new field "tables" to the extra data so that tables can be passed
3189        in at exec time, or the internal tables can be re-selected. This allows
3190        a compiled regex to be saved and re-used at a later time by a different
3191        program that might have everything at different addresses.
3193    13. Modified the pcre-config script so that, when run on Solaris, it shows a
3194        -R library as well as a -L library.
3196    14. The debugging options of pcretest (-d on the command line or D on a
3197        pattern) showed incorrect output for anything following an extended class
3198        that contained multibyte characters and which was followed by a quantifier.
3200    15. Added optional support for general category Unicode character properties
3201        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
3202        support. It adds about 90K to the size of the library. The meanings of the
3203        inbuilt class escapes such as \d and \s have NOT been changed.
3205    16. Updated pcredemo.c to include calls to free() to release the memory for the
3206        compiled pattern.
3208    17. The generated file chartables.c was being created in the source directory
3209        instead of in the building directory. This caused the build to fail if the
3210        source directory was different from the building directory, and was
3211        read-only.
3213    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
3214        file. No doubt somebody will tell me if they don't make sense... Also added
3215        Dan Mooney's comments about building on OpenVMS.
3217    19. Added support for partial matching via the PCRE_PARTIAL option for
3218        pcre_exec() and the \P data escape in pcretest.
3220    20. Extended pcretest with 3 new pattern features:
3222        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
3223              write the compiled pattern to the file whose name is "rest-of-line".
3224              This is a straight binary dump of the data, with the saved pointer to
3225              the character tables forced to be NULL. The study data, if any, is
3226              written too. After writing, pcretest reads a new pattern.
3228        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
3229              compiled pattern from the given file. There must not be any
3230              occurrences of "<" in the file name (pretty unlikely); if there are,
3231              pcretest will instead treat the initial "<" as a pattern delimiter.
3232              After reading in the pattern, pcretest goes on to read data lines as
3233              usual.
3235        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
3236              and 16-bit fields in a compiled pattern, to simulate a pattern that
3237              was compiled on a host of opposite endianness.
3239    21. The pcre-exec() function can now cope with patterns that were compiled on
3240        hosts of opposite endianness, with this restriction:
3242          As for any compiled expression that is saved and used later, the tables
3243          pointer field cannot be preserved; the extra_data field in the arguments
3244          to pcre_exec() should be used to pass in a tables address if a value
3245          other than the default internal tables were used at compile time.
3247    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
3248        now diagnosed as an error. Previously, most of the time, a negative number
3249        would have been treated as zero, but if in addition "ovector" was passed as
3250        NULL, a crash could occur.
3252    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
3253        new versions from the libtool 1.5 distribution (the last one is a copy of
3254        a file called libtool.m4). This seems to have fixed the need to patch
3255        "configure" to support Darwin 1.3 (which I used to do). However, I still
3256        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
3257        workstation).
3259    24. Changed the PCRE licence to be the more standard "BSD" licence.
3262    Version 4.5 01-Dec-03
3263    ---------------------
3265     1. There has been some re-arrangement of the code for the match() function so
3266        that it can be compiled in a version that does not call itself recursively.
3267        Instead, it keeps those local variables that need separate instances for
3268        each "recursion" in a frame on the heap, and gets/frees frames whenever it
3269        needs to "recurse". Keeping track of where control must go is done by means
3270        of setjmp/longjmp. The whole thing is implemented by a set of macros that
3271        hide most of the details from the main code, and operates only if
3272        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
3273        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
3274        operating.
3276        To make it easier for callers to provide specially tailored get/free
3277        functions for this usage, two new functions, pcre_stack_malloc, and
3278        pcre_stack_free, are used. They are always called in strict stacking order,
3279        and the size of block requested is always the same.
3281        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
3282        PCRE has been compiled to use the stack or the heap for recursion. The
3283        -C option of pcretest uses this to show which version is compiled.
3285        A new data escape \S, is added to pcretest; it causes the amounts of store
3286        obtained and freed by both kinds of malloc/free at match time to be added
3287        to the output.
3289     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
3290        what's available on my current Linux desktop machine.
3292     3. When matching a UTF-8 string, the test for a valid string at the start has
3293        been extended. If start_offset is not zero, PCRE now checks that it points
3294        to a byte that is the start of a UTF-8 character. If not, it returns
3295        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
3296        this is necessary because there may be backward assertions in the pattern.
3297        When matching the same subject several times, it may save resources to use
3298        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
3300     4. The code for checking the validity of UTF-8 strings has been tightened so
3301        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
3302        containing "overlong sequences".
3304     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
3305        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
3306        should have been "&", but it just so happened that all the cases this let
3307        through by mistake were picked up later in the function.
3309     6. I had used a variable called "isblank" - this is a C99 function, causing
3310        some compilers to warn. To avoid this, I renamed it (as "blankclass").
3312     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
3313        prompting; (b) run "./pcretest /dev/null" at the start of the test script
3314        so the version is shown; (c) stop "make test" echoing "./RunTest".
3316     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
3318     9. The prototype for memmove() for systems that don't have it was using
3319        size_t, but the inclusion of the header that defines size_t was later. I've
3320        moved the #includes for the C headers earlier to avoid this.
3322    10. Added some adjustments to the code to make it easier to compiler on certain
3323        special systems:
3325          (a) Some "const" qualifiers were missing.
3326          (b) Added the macro EXPORT before all exported functions; by default this
3327              is defined to be empty.
3328          (c) Changed the dftables auxiliary program (that builds chartables.c) so
3329              that it reads its output file name as an argument instead of writing
3330              to the standard output and assuming this can be redirected.
3332    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
3333        class containing characters with values greater than 255, PCRE compilation
3334        went into a loop.
3336    12. A recursive reference to a subpattern that was within another subpattern
3337        that had a minimum quantifier of zero caused PCRE to crash. For example,
3338        (x(y(?2))z)? provoked this bug with a subject that got as far as the
3339        recursion. If the recursively-called subpattern itself had a zero repeat,
3340        that was OK.
3342    13. In pcretest, the buffer for reading a data line was set at 30K, but the
3343        buffer into which it was copied (for escape processing) was still set at
3344        1024, so long lines caused crashes.
3346    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
3347        "internal error: code overflow...". This applied to any character class
3348        that was followed by a possessive quantifier.
3350    15. Modified the Makefile to add libpcre.la as a prerequisite for
3351        libpcreposix.la because I was told this is needed for a parallel build to
3352        work.
3354    16. If a pattern that contained .* following optional items at the start was
3355        studied, the wrong optimizing data was generated, leading to matching
3356        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
3357        matching string must start with a or b or c. The correct conclusion for
3358        this pattern is that a match can start with any character.
3361    Version 4.4 13-Aug-03
3362    ---------------------
3364     1. In UTF-8 mode, a character class containing characters with values between
3365        127 and 255 was not handled correctly if the compiled pattern was studied.
3366        In fixing this, I have also improved the studying algorithm for such
3367        classes (slightly).
3369     2. Three internal functions had redundant arguments passed to them. Removal
3370        might give a very teeny performance improvement.
3372     3. Documentation bug: the value of the capture_top field in a callout is *one
3373        more than* the number of the hightest numbered captured substring.
3375     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
3376        in incorrectly linking with a previously installed version. They now link
3377        explicitly with libpcre.la.
3379     5. configure.in no longer needs to recognize Cygwin specially.
3381     6. A problem in pcre.in for Windows platforms is fixed.
3383     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
3384        pcretest, it used to include the size of the study block as part of its
3385        output. Unfortunately, the structure contains a field that has a different
3386        size on different hardware architectures. This meant that the tests that
3387        showed this size failed. As the block is currently always of a fixed size,
3388        this information isn't actually particularly useful in pcretest output, so
3389        I have just removed it.
3391     8. Three pre-processor statements accidentally did not start in column 1.
3392        Sadly, there are *still* compilers around that complain, even though
3393        standard C has not required this for well over a decade. Sigh.
3395     9. In pcretest, the code for checking callouts passed small integers in the
3396        callout_data field, which is a void * field. However, some picky compilers
3397        complained about the casts involved for this on 64-bit systems. Now
3398        pcretest passes the address of the small integer instead, which should get
3399        rid of the warnings.
3401    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
3402        both compile and run time, and gives an error if an invalid UTF-8 sequence
3403        is found. There is a option for disabling this check in cases where the
3404        string is known to be correct and/or the maximum performance is wanted.
3406    11. In response to a bug report, I changed one line in Makefile.in from
3408            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
3409        to
3410            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
3412        to look similar to other lines, but I have no way of telling whether this
3413        is the right thing to do, as I do not use Windows. No doubt I'll get told
3414        if it's wrong...
3417    Version 4.3 21-May-03
3418    ---------------------
3420    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
3421       Makefile.
3423    2. Some refactoring to improve the quality of the code:
3425       (i)   The utf8_table... variables are now declared "const".
3427       (ii)  The code for \cx, which used the "case flipping" table to upper case
3428             lower case letters, now just substracts 32. This is ASCII-specific,
3429             but the whole concept of \cx is ASCII-specific, so it seems
3430             reasonable.
3432       (iii) PCRE was using its character types table to recognize decimal and
3433             hexadecimal digits in the pattern. This is silly, because it handles
3434             only 0-9, a-f, and A-F, but the character types table is locale-
3435             specific, which means strange things might happen. A private
3436             table is now used for this - though it costs 256 bytes, a table is
3437             much faster than multiple explicit tests. Of course, the standard
3438             character types table is still used for matching digits in subject
3439             strings against \d.
3441       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
3442             ending in _t are). So I've renamed it as ESC_tee.
3444    3. The first argument for regexec() in the POSIX wrapper should have been
3445       defined as "const".
3447    4. Changed pcretest to use malloc() for its buffers so that they can be
3448       Electric Fenced for debugging.
3450    5. There were several places in the code where, in UTF-8 mode, PCRE would try
3451       to read one or more bytes before the start of the subject string. Often this
3452       had no effect on PCRE's behaviour, but in some circumstances it could
3453       provoke a segmentation fault.
3455    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
3456       to try to read one or more bytes before the start of the subject string.
3458    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
3459       UTF-8 support could misbehave in various ways if the subject string
3460       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
3461       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
3462       back over UTF-8 characters.)
3465    Version 4.2 14-Apr-03
3466    ---------------------
3468    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
3470    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
3471         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
3472         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
3473         [WIN_PREFIX]: new variable, "cyg" for Cygwin
3474         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
3475           and BUILD_EXEEXT
3476         Note: automatic setting of the BUILD variables is not yet working
3477         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
3478           compile-time but not at link-time
3479         [LINK]: use for linking executables only
3480         make different versions for Windows and non-Windows
3481         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
3482           libraries
3483         [LINK_FOR_BUILD]: new variable
3484         [OBJEXT]: use throughout
3485         [EXEEXT]: use throughout
3486         <winshared>: new target
3487         <wininstall>: new target
3488         <dftables.o>: use native compiler
3489         <dftables>: use native linker
3490         <install>: handle Windows platform correctly
3491         <clean>: ditto
3492         <check>: ditto
3493         copy DLL to top builddir before testing
3495       As part of these changes, -no-undefined was removed again. This was reported
3496       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
3497       in any case.
3499    3. Some tidies to get rid of compiler warnings:
3501       . In the match_data structure, match_limit was an unsigned long int, whereas
3502         match_call_count was an int. I've made them both unsigned long ints.
3504       . In pcretest the fact that a const uschar * doesn't automatically cast to
3505         a void * provoked a warning.
3507       . Turning on some more compiler warnings threw up some "shadow" variables
3508         and a few more missing casts.
3510    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
3511       option, a class that contained a single character with a value between 128
3512       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
3514    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
3515       option, a class that contained several characters, but with at least one
3516       whose value was between 128 and 255 caused PCRE to crash.
3519    Version 4.1 12-Mar-03
3520    ---------------------
3522    1. Compiling with gcc -pedantic found a couple of places where casts were
3523    needed, and a string in dftables.c that was longer than standard compilers are
3524    required to support.
3526    2. Compiling with Sun's compiler found a few more places where the code could
3527    be tidied up in order to avoid warnings.
3529    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
3530    first of these names is deprecated in the latest Autoconf in favour of the name
3531    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
3532    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
3533    analogy I have changed it to CFLAGS_FOR_BUILD.
3535    4. Added -no-undefined to the linking command in the Makefile, because this is
3536    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
3537    linking step for the pcreposix library.
3539    5. PCRE was failing to diagnose the case of two named groups with the same
3540    name.
3542    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
3543    literal character that is needed in the subject for a match, and scans along to
3544    ensure that it is present before embarking on the full matching process. This
3545    saves time in cases of nested unlimited repeats that are never going to match.
3546    Problem: the scan can take a lot of time if the subject is very long (e.g.
3547    megabytes), thus penalizing straightforward matches. It is now done only if the
3548    amount of subject to be scanned is less than 1000 bytes.
3550    7. A lesser problem with the same optimization is that it was recording the
3551    first character of an anchored pattern as "needed", thus provoking a search
3552    right along the subject, even when the first match of the pattern was going to
3553    fail. The "needed" character is now not set for anchored patterns, unless it
3554    follows something in the pattern that is of non-fixed length. Thus, it still
3555    fulfils its original purpose of finding quick non-matches in cases of nested
3556    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
3559    Version 4.0 17-Feb-03
3560    ---------------------
3562    1. If a comment in an extended regex that started immediately after a meta-item
3563    extended to the end of string, PCRE compiled incorrect data. This could lead to
3564    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
3566    2. Moved to autoconf 2.53 and libtool 1.4.2.
3568    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
3569    the special perltest8 script is no longer needed - all the tests can be run
3570    from a single perltest script.
3572    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
3573    by \s. It has now been removed in PCRE. This means it isn't recognized as
3574    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
3575    class [:space:] *does* include VT, thereby creating a mess.
3577    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
3578    space and tab.
3580    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
3581    its new features into the main test script, reducing the number of scripts.
3583    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
3584    were backward compatible, and made the (?i) apply to the whole pattern, as if
3585    /i were given. Now it behaves more logically, and applies the option setting
3586    only to what follows. PCRE has been changed to follow suit. However, if it
3587    finds options settings right at the start of the pattern, it extracts them into
3588    the global options, as before. Thus, they show up in the info data.