ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 85 by nigel, Sat Feb 24 21:41:13 2007 UTC revision 916 by ph10, Wed Feb 15 09:50:53 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 give a
31            partial match for the string "\r".
34    Version 8.30 04-February-2012
35    -----------------------------
37    1.  Renamed "isnumber" as "is_a_number" because in some Mac environments this
38        name is defined in ctype.h.
40    2.  Fixed a bug in fixed-length calculation for lookbehinds that would show up
41        only in quite long subpatterns.
43    3.  Removed the function pcre_info(), which has been obsolete and deprecated
44        since it was replaced by pcre_fullinfo() in February 2000.
46    4.  For a non-anchored pattern, if (*SKIP) was given with a name that did not
47        match a (*MARK), and the match failed at the start of the subject, a
48        reference to memory before the start of the subject could occur. This bug
49        was introduced by fix 17 of release 8.21.
51    5.  A reference to an unset group with zero minimum repetition was giving
52        totally wrong answers (in non-JavaScript-compatibility mode). For example,
53        /(another)?(\1?)test/ matched against "hello world test". This bug was
54        introduced in release 8.13.
56    6.  Add support for 16-bit character strings (a large amount of work involving
57        many changes and refactorings).
59    7.  RunGrepTest failed on msys because \r\n was replaced by whitespace when the
60        command "pattern=`printf 'xxx\r\njkl'`" was run. The pattern is now taken
61        from a file.
63    8.  Ovector size of 2 is also supported by JIT based pcre_exec (the ovector size
64        rounding is not applied in this particular case).
66    9.  The invalid Unicode surrogate codepoints U+D800 to U+DFFF are now rejected
67        if they appear, or are escaped, in patterns.
69    10. Get rid of a number of -Wunused-but-set-variable warnings.
71    11. The pattern /(?=(*:x))(q|)/ matches an empty string, and returns the mark
72        "x". The similar pattern /(?=(*:x))((*:y)q|)/ did not return a mark at all.
73        Oddly, Perl behaves the same way. PCRE has been fixed so that this pattern
74        also returns the mark "x". This bug applied to capturing parentheses,
75        non-capturing parentheses, and atomic parentheses. It also applied to some
76        assertions.
78    12. Stephen Kelly's patch to CMakeLists.txt allows it to parse the version
79        information out of configure.ac instead of relying on pcre.h.generic, which
80        is not stored in the repository.
82    13. Applied Dmitry V. Levin's patch for a more portable method for linking with
83        -lreadline.
85    14. ZH added PCRE_CONFIG_JITTARGET; added its output to pcretest -C.
87    15. Applied Graycode's patch to put the top-level frame on the stack rather
88        than the heap when not using the stack for recursion. This gives a
89        performance improvement in many cases when recursion is not deep.
91    16. Experimental code added to "pcretest -C" to output the stack frame size.
94    Version 8.21 12-Dec-2011
95    ------------------------
97    1.  Updating the JIT compiler.
99    2.  JIT compiler now supports OP_NCREF, OP_RREF and OP_NRREF. New test cases
100        are added as well.
102    3.  Fix cache-flush issue on PowerPC (It is still an experimental JIT port).
103        PCRE_EXTRA_TABLES is not suported by JIT, and should be checked before
104        calling _pcre_jit_exec. Some extra comments are added.
106    4.  (*MARK) settings inside atomic groups that do not contain any capturing
107        parentheses, for example, (?>a(*:m)), were not being passed out. This bug
108        was introduced by change 18 for 8.20.
110    5.  Supporting of \x, \U and \u in JavaScript compatibility mode based on the
111        ECMA-262 standard.
113    6.  Lookbehinds such as (?<=a{2}b) that contained a fixed repetition were
114        erroneously being rejected as "not fixed length" if PCRE_CASELESS was set.
115        This bug was probably introduced by change 9 of 8.13.
117    7.  While fixing 6 above, I noticed that a number of other items were being
118        incorrectly rejected as "not fixed length". This arose partly because newer
119        opcodes had not been added to the fixed-length checking code. I have (a)
120        corrected the bug and added tests for these items, and (b) arranged for an
121        error to occur if an unknown opcode is encountered while checking for fixed
122        length instead of just assuming "not fixed length". The items that were
123        rejected were: (*ACCEPT), (*COMMIT), (*FAIL), (*MARK), (*PRUNE), (*SKIP),
124        (*THEN), \h, \H, \v, \V, and single character negative classes with fixed
125        repetitions, e.g. [^a]{3}, with and without PCRE_CASELESS.
127    8.  A possessively repeated conditional subpattern such as (?(?=c)c|d)++ was
128        being incorrectly compiled and would have given unpredicatble results.
130    9.  A possessively repeated subpattern with minimum repeat count greater than
131        one behaved incorrectly. For example, (A){2,}+ behaved as if it was
132        (A)(A)++ which meant that, after a subsequent mismatch, backtracking into
133        the first (A) could occur when it should not.
135    10. Add a cast and remove a redundant test from the code.
137    11. JIT should use pcre_malloc/pcre_free for allocation.
139    12. Updated pcre-config so that it no longer shows -L/usr/lib, which seems
140        best practice nowadays, and helps with cross-compiling. (If the exec_prefix
141        is anything other than /usr, -L is still shown).
143    13. In non-UTF-8 mode, \C is now supported in lookbehinds and DFA matching.
145    14. Perl does not support \N without a following name in a [] class; PCRE now
146        also gives an error.
148    15. If a forward reference was repeated with an upper limit of around 2000,
149        it caused the error "internal error: overran compiling workspace". The
150        maximum number of forward references (including repeats) was limited by the
151        internal workspace, and dependent on the LINK_SIZE. The code has been
152        rewritten so that the workspace expands (via pcre_malloc) if necessary, and
153        the default depends on LINK_SIZE. There is a new upper limit (for safety)
154        of around 200,000 forward references. While doing this, I also speeded up
155        the filling in of repeated forward references.
157    16. A repeated forward reference in a pattern such as (a)(?2){2}(.) was
158        incorrectly expecting the subject to contain another "a" after the start.
160    17. When (*SKIP:name) is activated without a corresponding (*MARK:name) earlier
161        in the match, the SKIP should be ignored. This was not happening; instead
162        the SKIP was being treated as NOMATCH. For patterns such as
163        /A(*MARK:A)A+(*SKIP:B)Z|AAC/ this meant that the AAC branch was never
164        tested.
166    18. The behaviour of (*MARK), (*PRUNE), and (*THEN) has been reworked and is
167        now much more compatible with Perl, in particular in cases where the result
168        is a non-match for a non-anchored pattern. For example, if
169        /b(*:m)f|a(*:n)w/ is matched against "abc", the non-match returns the name
170        "m", where previously it did not return a name. A side effect of this
171        change is that for partial matches, the last encountered mark name is
172        returned, as for non matches. A number of tests that were previously not
173        Perl-compatible have been moved into the Perl-compatible test files. The
174        refactoring has had the pleasing side effect of removing one argument from
175        the match() function, thus reducing its stack requirements.
177    19. If the /S+ option was used in pcretest to study a pattern using JIT,
178        subsequent uses of /S (without +) incorrectly behaved like /S+.
180    21. Retrieve executable code size support for the JIT compiler and fixing
181        some warnings.
183    22. A caseless match of a UTF-8 character whose other case uses fewer bytes did
184        not work when the shorter character appeared right at the end of the
185        subject string.
187    23. Added some (int) casts to non-JIT modules to reduce warnings on 64-bit
188        systems.
190    24. Added PCRE_INFO_JITSIZE to pass on the value from (21) above, and also
191        output it when the /M option is used in pcretest.
193    25. The CheckMan script was not being included in the distribution. Also, added
194        an explicit "perl" to run Perl scripts from the PrepareRelease script
195        because this is reportedly needed in Windows.
197    26. If study data was being save in a file and studying had not found a set of
198        "starts with" bytes for the pattern, the data written to the file (though
199        never used) was taken from uninitialized memory and so caused valgrind to
200        complain.
202    27. Updated RunTest.bat as provided by Sheri Pierce.
204    28. Fixed a possible uninitialized memory bug in pcre_jit_compile.c.
206    29. Computation of memory usage for the table of capturing group names was
207        giving an unnecessarily large value.
210    Version 8.20 21-Oct-2011
211    ------------------------
213    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
214        a POSIX class. After further experiments with Perl, which convinced me that
215        Perl has bugs and confusions, a closing square bracket is no longer allowed
216        in a POSIX name. This bug also affected patterns with classes that started
217        with full stops.
219    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
220        captured substring, but while checking the failing first alternative,
221        substring 1 is temporarily captured. If the output vector supplied to
222        pcre_exec() was not big enough for this capture, the yield of the function
223        was still zero ("insufficient space for captured substrings"). This cannot
224        be totally fixed without adding another stack variable, which seems a lot
225        of expense for a edge case. However, I have improved the situation in cases
226        such as /(a)(b)x|abc/ matched against "abc", where the return code
227        indicates that fewer than the maximum number of slots in the ovector have
228        been set.
230    3.  Related to (2) above: when there are more back references in a pattern than
231        slots in the output vector, pcre_exec() uses temporary memory during
232        matching, and copies in the captures as far as possible afterwards. It was
233        using the entire output vector, but this conflicts with the specification
234        that only 2/3 is used for passing back captured substrings. Now it uses
235        only the first 2/3, for compatibility. This is, of course, another edge
236        case.
238    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
239        main code base, and can be used by building with --enable-jit. When this is
240        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
241        runtime --no-jit option is given.
243    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
244        ovector, the return from the function was zero, implying that there were
245        other matches that did not fit. The correct "exactly full" value is now
246        returned.
248    6.  If a subpattern that was called recursively or as a subroutine contained
249        (*PRUNE) or any other control that caused it to give a non-standard return,
250        invalid errors such as "Error -26 (nested recursion at the same subject
251        position)" or even infinite loops could occur.
253    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
254        computing the minimum length on reaching *ACCEPT, and so ended up with the
255        wrong value of 1 rather than 0. Further investigation indicates that
256        computing a minimum subject length in the presence of *ACCEPT is difficult
257        (think back references, subroutine calls), and so I have changed the code
258        so that no minimum is registered for a pattern that contains *ACCEPT.
260    8.  If (*THEN) was present in the first (true) branch of a conditional group,
261        it was not handled as intended. [But see 16 below.]
263    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
264        Sheri Pierce.
266    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
267        the first byte in a match must be "a".
269    11. Change 17 for 8.13 increased the recursion depth for patterns like
270        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
271        pattern contains any instances of (*THEN). If it does not, the old
272        optimizations are restored. It would be nice to do this on a per-group
273        basis, but at the moment that is not feasible.
275    12. In some environments, the output of pcretest -C is CRLF terminated. This
276        broke RunTest's code that checks for the link size. A single white space
277        character after the value is now allowed for.
279    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
280        For "fr", it uses the Windows-specific input and output files.
282    14. If (*THEN) appeared in a group that was called recursively or as a
283        subroutine, it did not work as intended. [But see next item.]
285    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
286        pattern fragments (but not containing any | characters). If A and B are
287        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
288        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
289        D. In other words, Perl considers parentheses that do not contain any |
290        characters to be part of a surrounding alternative, whereas PCRE was
291        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
292        differently. PCRE now behaves in the same way as Perl, except in the case
293        of subroutine/recursion calls such as (?1) which have in any case always
294        been different (but PCRE had them first :-).
296    16. Related to 15 above: Perl does not treat the | in a conditional group as
297        creating alternatives. Such a group is treated in the same way as an
298        ordinary group without any | characters when processing (*THEN). PCRE has
299        been changed to match Perl's behaviour.
301    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
302        RunGrepTest script failed.
304    18. Change 22 for version 13 caused atomic groups to use more stack. This is
305        inevitable for groups that contain captures, but it can lead to a lot of
306        stack use in large patterns. The old behaviour has been restored for atomic
307        groups that do not contain any capturing parentheses.
309    19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not
310        suppress the check for a minimum subject length at run time. (If it was
311        given to pcre_exec() or pcre_dfa_exec() it did work.)
313    20. Fixed an ASCII-dependent infelicity in pcretest that would have made it
314        fail to work when decoding hex characters in data strings in EBCDIC
315        environments.
317    21. It appears that in at least one Mac OS environment, the isxdigit() function
318        is implemented as a macro that evaluates to its argument more than once,
319        contravening the C 90 Standard (I haven't checked a later standard). There
320        was an instance in pcretest which caused it to go wrong when processing
321        \x{...} escapes in subject strings. The has been rewritten to avoid using
322        things like p++ in the argument of isxdigit().
325    Version 8.13 16-Aug-2011
326    ------------------------
328    1.  The Unicode data tables have been updated to Unicode 6.0.0.
330    2.  Two minor typos in pcre_internal.h have been fixed.
332    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
333        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
334        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
336    4.  There were a number of related bugs in the code for matching backrefences
337        caselessly in UTF-8 mode when codes for the characters concerned were
338        different numbers of bytes. For example, U+023A and U+2C65 are an upper
339        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
340        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
341        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
342        2-byte code at the end of the subject (it thought there wasn't enough data
343        left).
345    5.  Comprehensive information about what went wrong is now returned by
346        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
347        as the output vector has at least 2 elements. The offset of the start of
348        the failing character and a reason code are placed in the vector.
350    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
351        now returned is for the first byte of the failing character, instead of the
352        last byte inspected. This is an incompatible change, but I hope it is small
353        enough not to be a problem. It makes the returned offset consistent with
354        pcre_exec() and pcre_dfa_exec().
356    7.  pcretest now gives a text phrase as well as the error number when
357        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
358        failure, the offset and reason code are output.
360    8.  When \R was used with a maximizing quantifier it failed to skip backwards
361        over a \r\n pair if the subsequent match failed. Instead, it just skipped
362        back over a single character (\n). This seems wrong (because it treated the
363        two characters as a single entity when going forwards), conflicts with the
364        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
365        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
366        has been changed.
368    9.  Some internal refactoring has changed the processing so that the handling
369        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
370        time (the PCRE_DOTALL option was changed this way some time ago: version
371        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
372        which was always a bit of a fudge. It also means that there is one less
373        argument for the match() function, which reduces its stack requirements
374        slightly. This change also fixes an incompatibility with Perl: the pattern
375        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
377    10. More internal refactoring has drastically reduced the number of recursive
378        calls to match() for possessively repeated groups such as (abc)++ when
379        using pcre_exec().
381    11. While implementing 10, a number of bugs in the handling of groups were
382        discovered and fixed:
384        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
385        (a|)*(?1) gave a compile-time internal error.
386        ((a|)+)+  did not notice that the outer group could match an empty string.
387        (^a|^)+   was not marked as anchored.
388        (.*a|.*)+ was not marked as matching at start or after a newline.
390    12. Yet more internal refactoring has removed another argument from the match()
391        function. Special calls to this function are now indicated by setting a
392        value in a variable in the "match data" data block.
394    13. Be more explicit in pcre_study() instead of relying on "default" for
395        opcodes that mean there is no starting character; this means that when new
396        ones are added and accidentally left out of pcre_study(), testing should
397        pick them up.
399    14. The -s option of pcretest has been documented for ages as being an old
400        synonym of -m (show memory usage). I have changed it to mean "force study
401        for every regex", that is, assume /S for every regex. This is similar to -i
402        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
403        using it. It makes it easier to run collections of tests with and without
404        study enabled, and thereby test pcre_study() more easily. All the standard
405        tests are now run with and without -s (but some patterns can be marked as
406        "never study" - see 20 below).
408    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
409        restoration of the capturing data to the outer values was not happening
410        correctly.
412    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
413        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
414        pattern had matched an empty string, and so incorrectly returned a no
415        match.
417    17. There was optimizing code for the last branch of non-capturing parentheses,
418        and also for the obeyed branch of a conditional subexpression, which used
419        tail recursion to cut down on stack usage. Unfortunately, now that there is
420        the possibility of (*THEN) occurring in these branches, tail recursion is
421        no longer possible because the return has to be checked for (*THEN). These
422        two optimizations have therefore been removed. [But see 8.20/11 above.]
424    18. If a pattern containing \R was studied, it was assumed that \R always
425        matched two bytes, thus causing the minimum subject length to be
426        incorrectly computed because \R can also match just one byte.
428    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
429        was incorrectly computed.
431    20. If /S is present twice on a test pattern in pcretest input, it now
432        *disables* studying, thereby overriding the use of -s on the command line
433        (see 14 above). This is necessary for one or two tests to keep the output
434        identical in both cases.
436    21. When (*ACCEPT) was used in an assertion that matched an empty string and
437        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
439    22. When an atomic group that contained a capturing parenthesis was
440        successfully matched, but the branch in which it appeared failed, the
441        capturing was not being forgotten if a higher numbered group was later
442        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
443        group 1 to "a", when in fact it should be unset. This applied to multi-
444        branched capturing and non-capturing groups, repeated or not, and also to
445        positive assertions (capturing in negative assertions does not happen
446        in PCRE) and also to nested atomic groups.
448    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
449        subject after a captured substring, to make it easier to tell which of a
450        number of identical substrings has been captured.
452    24. The way atomic groups are processed by pcre_exec() has been changed so that
453        if they are repeated, backtracking one repetition now resets captured
454        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
455        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
456        "aaa". Previously, it would have been "a". As part of this code
457        refactoring, the way recursive calls are handled has also been changed.
459    25. If an assertion condition captured any substrings, they were not passed
460        back unless some other capturing happened later. For example, if
461        (?(?=(a))a) was matched against "a", no capturing was returned.
463    26. When studying a pattern that contained subroutine calls or assertions,
464        the code for finding the minimum length of a possible match was handling
465        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
466        group 1 called group 2 while simultaneously a separate group 2 called group
467        1). A stack overflow occurred in this case. I have fixed this by limiting
468        the recursion depth to 10.
470    27. Updated RunTest.bat in the distribution to the version supplied by Tom
471        Fortmann. This supports explicit test numbers on the command line, and has
472        argument validation and error reporting.
474    28. An instance of \X with an unlimited repeat could fail if at any point the
475        first character it looked at was a mark character.
477    29. Some minor code refactoring concerning Unicode properties and scripts
478        should reduce the stack requirement of match() slightly.
480    30. Added the '=' option to pcretest to check the setting of unused capturing
481        slots at the end of the pattern, which are documented as being -1, but are
482        not included in the return count.
484    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
485        compiled something random. Now it gives a compile-time error (as does
486        Perl).
488    32. A *MARK encountered during the processing of a positive assertion is now
489        recorded and passed back (compatible with Perl).
491    33. If --only-matching or --colour was set on a pcregrep call whose pattern
492        had alternative anchored branches, the search for a second match in a line
493        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
494        matched the line "0102" twice. The same bug affected patterns that started
495        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
496        twice.
498    34. Previously, PCRE did not allow quantification of assertions. However, Perl
499        does, and because of capturing effects, quantifying parenthesized
500        assertions may at times be useful. Quantifiers are now allowed for
501        parenthesized assertions.
503    35. A minor code tidy in pcre_compile() when checking options for \R usage.
505    36. \g was being checked for fancy things in a character class, when it should
506        just be a literal "g".
508    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
509        appearance of a nested POSIX class supersedes an apparent external class.
510        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
511        unescaped square brackets may also appear as part of class names. For
512        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
513        more like Perl. (But see 8.20/1 above.)
515    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
516        was because it thought it was \N{name}, which is not supported).
518    39. Add minix to OS list not supporting the -S option in pcretest.
520    40. PCRE tries to detect cases of infinite recursion at compile time, but it
521        cannot analyze patterns in sufficient detail to catch mutual recursions
522        such as ((?1))((?2)). There is now a runtime test that gives an error if a
523        subgroup is called recursively as a subpattern for a second time at the
524        same position in the subject string. In previous releases this might have
525        been caught by the recursion limit, or it might have run out of stack.
527    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
528        happen only once. PCRE was, however incorrectly giving a compile time error
529        "recursive call could loop indefinitely" because it cannot analyze the
530        pattern in sufficient detail. The compile time test no longer happens when
531        PCRE is compiling a conditional subpattern, but actual runaway loops are
532        now caught at runtime (see 40 above).
534    42. It seems that Perl allows any characters other than a closing parenthesis
535        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
536        has been changed to be the same.
538    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
539        as not to get warnings when autogen.sh is called. Also changed
540        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
542    44. To help people who use pcregrep to scan files containing exceedingly long
543        lines, the following changes have been made:
545        (a) The default value of the buffer size parameter has been increased from
546            8K to 20K. (The actual buffer used is three times this size.)
548        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
549            PCRE is built.
551        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
552            to be set at run time.
554        (d) Numerical values in pcregrep options can be followed by K or M, for
555            example --buffer-size=50K.
557        (e) If a line being scanned overflows pcregrep's buffer, an error is now
558            given and the return code is set to 2.
560    45. Add a pointer to the latest mark to the callout data block.
562    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
563        partial match of an empty string instead of no match. This was specific to
564        the use of ".".
566    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
567        complete match instead of a partial match. This bug was dependent on both
568        the PCRE_UTF8 and PCRE_DOTALL options being set.
570    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
571        starting byte set, because \b was not being ignored.
574    Version 8.12 15-Jan-2011
575    ------------------------
577    1.  Fixed some typos in the markup of the man pages, and wrote a script that
578        checks for such things as part of the documentation building process.
580    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
581        --match-limit and --recursion-limit options (added for 8.11). In
582        particular, this made one of the standard tests fail. (The integer value
583        went into the wrong half of a long int.)
585    3.  If the --colour option was given to pcregrep with -v (invert match), it
586        did strange things, either producing crazy output, or crashing. It should,
587        of course, ignore a request for colour when reporting lines that do not
588        match.
590    4.  Another pcregrep bug caused similar problems if --colour was specified with
591        -M (multiline) and the pattern match finished with a line ending.
593    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
594        matched in multiline mode, the following line was shown as part of the
595        match. This seems wrong, so I have changed it.
597    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
598        the check for further matches in the same line (so they could be coloured)
599        to overrun the end of the current line. If another match was found, it was
600        incorrectly shown (and then shown again when found in the next line).
602    7.  If pcregrep was compiled under Windows, there was a reference to the
603        function pcregrep_exit() before it was defined. I am assuming this was
604        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
605        reported by a user. I've moved the definition above the reference.
608    Version 8.11 10-Dec-2010
609    ------------------------
611    1.  (*THEN) was not working properly if there were untried alternatives prior
612        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
613        backtracked to try for "b" instead of moving to the next alternative branch
614        at the same level (in this case, to look for "c"). The Perl documentation
615        is clear that when (*THEN) is backtracked onto, it goes to the "next
616        alternative in the innermost enclosing group".
618    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
619        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
620        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
621        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
622        (*THEN).
624    3.  If \s appeared in a character class, it removed the VT character from
625        the class, even if it had been included by some previous item, for example
626        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
627        of \s, but is part of the POSIX "space" class.)
629    4.  A partial match never returns an empty string (because you can always
630        match an empty string at the end of the subject); however the checking for
631        an empty string was starting at the "start of match" point. This has been
632        changed to the "earliest inspected character" point, because the returned
633        data for a partial match starts at this character. This means that, for
634        example, /(?<=abc)def/ gives a partial match for the subject "abc"
635        (previously it gave "no match").
637    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
638        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
639        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
640        has an implication that the given string is incomplete (because a partial
641        match is preferred over a full match). For this reason, these items now
642        give a partial match in this situation. [Aside: previously, the one case
643        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
644        match rather than a full match, which was wrong by the old rules, but is
645        now correct.]
647    6.  There was a bug in the handling of #-introduced comments, recognized when
648        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
649        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
650        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
651        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
652        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
653        places in pcre_compile().
655    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
656        comments when looking ahead for named forward references to subpatterns,
657        the only newline sequence it recognized was NL. It now handles newlines
658        according to the set newline convention.
660    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
661        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
662        cater for a lack of strerror(). These oversights have been fixed.
664    9.  Added --match-limit and --recursion-limit to pcregrep.
666    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
668    11. When the -o option was used, pcregrep was setting a return code of 1, even
669        when matches were found, and --line-buffered was not being honoured.
671    12. Added an optional parentheses number to the -o and --only-matching options
672        of pcregrep.
674    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
675        can match an empty string. The code to do it in pcretest and pcredemo
676        needed fixing:
678        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
679            only one byte after an empty string match just before CRLF (this case
680            just got forgotten; "any" and "anycrlf" were OK).
682        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
683            mode when an empty string match preceded an ASCII character followed by
684            a non-ASCII character. (The code for advancing by one character rather
685            than one byte was nonsense.)
687        (c) The pcredemo.c sample program did not have any code at all to handle
688            the cases when CRLF is a valid newline sequence.
690    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
691        as a starting offset was within the subject string. There is now a new
692        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
693        negative or greater than the length of the string. In order to test this,
694        pcretest is extended to allow the setting of negative starting offsets.
696    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
697        starting offset points to the beginning of a UTF-8 character was
698        unnecessarily clumsy. I tidied it up.
700    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
701        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
703    17. Nobody had reported that the --include_dir option, which was added in
704        release 7.7 should have been called --include-dir (hyphen, not underscore)
705        for compatibility with GNU grep. I have changed it to --include-dir, but
706        left --include_dir as an undocumented synonym, and the same for
707        --exclude-dir, though that is not available in GNU grep, at least as of
708        release 2.5.4.
710    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
711        characters from a string of bytes) have been redefined so as not to use
712        loops, in order to improve performance in some environments. At the same
713        time, I abstracted some of the common code into auxiliary macros to save
714        repetition (this should not affect the compiled code).
716    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
717        compile-time error is now given if \c is not followed by an ASCII
718        character, that is, a byte less than 128. (In EBCDIC mode, the code is
719        different, and any byte value is allowed.)
721    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
722        START_OPTIMIZE option, which is now allowed at compile time - but just
723        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
724        to pcregrep and other applications that have no direct access to PCRE
725        options. The new /Y option in pcretest sets this option when calling
726        pcre_compile().
728    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
729        back references. Groups containing recursive back references were forced to
730        be atomic by that change, but in the case of named groups, the amount of
731        memory required was incorrectly computed, leading to "Failed: internal
732        error: code overflow". This has been fixed.
734    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
735        pcretest.c, to avoid build problems in some Borland environments.
738    Version 8.10 25-Jun-2010
739    ------------------------
741    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
742        THEN.
744    2.  (*ACCEPT) was not working when inside an atomic group.
746    3.  Inside a character class, \B is treated as a literal by default, but
747        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
748        causes the error). The code is unchanged, but I tidied the documentation.
750    4.  Inside a character class, PCRE always treated \R and \X as literals,
751        whereas Perl faults them if its -w option is set. I have changed PCRE so
752        that it faults them when PCRE_EXTRA is set.
754    5.  Added support for \N, which always matches any character other than
755        newline. (It is the same as "." when PCRE_DOTALL is not set.)
757    6.  When compiling pcregrep with newer versions of gcc which may have
758        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
759        declared with attribute warn_unused_result" were given. Just casting the
760        result to (void) does not stop the warnings; a more elaborate fudge is
761        needed. I've used a macro to implement this.
763    7.  Minor change to pcretest.c to avoid a compiler warning.
765    8.  Added four artifical Unicode properties to help with an option to make
766        \s etc use properties (see next item). The new properties are: Xan
767        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
769    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
770        use Unicode properties. (*UCP) at the start of a pattern can be used to set
771        this option. Modified pcretest to add /W to test this facility. Added
772        REG_UCP to make it available via the POSIX interface.
774    10. Added --line-buffered to pcregrep.
776    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
777        studied, and the match started with a letter with a code point greater than
778        127 whose first byte was different to the first byte of the other case of
779        the letter, the other case of this starting letter was not recognized
780        (#976).
782    12. If a pattern that was studied started with a repeated Unicode property
783        test, for example, \p{Nd}+, there was the theoretical possibility of
784        setting up an incorrect bitmap of starting bytes, but fortunately it could
785        not have actually happened in practice until change 8 above was made (it
786        added property types that matched character-matching opcodes).
788    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
789        possible starting bytes for non-anchored patterns.
791    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
792        \R, and also a number of cases that involve Unicode properties, both
793        explicit and implicit when PCRE_UCP is set.
795    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
796        input, it could crash or give wrong results if characters with values
797        greater than 0xc0 were present in the subject string. (Detail: it assumed
798        UTF-8 input when processing these items.)
800    16. Added a lot of (int) casts to avoid compiler warnings in systems where
801        size_t is 64-bit (#991).
803    17. Added a check for running out of memory when PCRE is compiled with
804        --disable-stack-for-recursion (#990).
806    18. If the last data line in a file for pcretest does not have a newline on
807        the end, a newline was missing in the output.
809    19. The default pcre_chartables.c file recognizes only ASCII characters (values
810        less than 128) in its various bitmaps. However, there is a facility for
811        generating tables according to the current locale when PCRE is compiled. It
812        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
813        characters, are recognized by isspace() and therefore were getting set in
814        these tables, and indeed these tables seem to approximate to ISO 8859. This
815        caused a problem in UTF-8 mode when pcre_study() was used to create a list
816        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
817        which of course cannot start UTF-8 characters. I have changed the code so
818        that only real ASCII characters (less than 128) and the correct starting
819        bytes for UTF-8 encodings are set for characters greater than 127 when in
820        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
821        altogether.)
823    20. Added the /T option to pcretest so as to be able to run tests with non-
824        standard character tables, thus making it possible to include the tests
825        used for 19 above in the standard set of tests.
827    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
828        reference to a subpattern the other side of a comment that contains an
829        opening parenthesis caused either an internal compiling error, or a
830        reference to the wrong subpattern.
833    Version 8.02 19-Mar-2010
834    ------------------------
836    1.  The Unicode data tables have been updated to Unicode 5.2.0.
838    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
839        configured.
841    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
842        original author of that file, following a query about its status.
844    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
845        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
847    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
848        quantifier applied to a forward-referencing subroutine call, could compile
849        incorrect code or give the error "internal error: previously-checked
850        referenced subpattern not found".
852    6.  Both MS Visual Studio and Symbian OS have problems with initializing
853        variables to point to external functions. For these systems, therefore,
854        pcre_malloc etc. are now initialized to local functions that call the
855        relevant global functions.
857    7.  There were two entries missing in the vectors called coptable and poptable
858        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
859        I've fixed the data, and added a kludgy way of testing at compile time that
860        the lengths are correct (equal to the number of opcodes).
862    8.  Following on from 7, I added a similar kludge to check the length of the
863        eint vector in pcreposix.c.
865    9.  Error texts for pcre_compile() are held as one long string to avoid too
866        much relocation at load time. To find a text, the string is searched,
867        counting zeros. There was no check for running off the end of the string,
868        which could happen if a new error number was added without updating the
869        string.
871    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
873    11. \K was not working if it appeared in an atomic group or in a group that
874        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
875        \K is "not well defined" if used in an assertion. PCRE now accepts it if
876        the assertion is positive, but not if it is negative.
878    12. Change 11 fortuitously reduced the size of the stack frame used in the
879        "match()" function of pcre_exec.c by one pointer. Forthcoming
880        implementation of support for (*MARK) will need an extra pointer on the
881        stack; I have reserved it now, so that the stack frame size does not
882        decrease.
884    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
885        item in branch that calls a recursion is a subroutine call - as in the
886        second branch in the above example - was incorrectly given the compile-
887        time error "recursive call could loop indefinitely" because pcre_compile()
888        was not correctly checking the subroutine for matching a non-empty string.
890    14. The checks for overrunning compiling workspace could trigger after an
891        overrun had occurred. This is a "should never occur" error, but it can be
892        triggered by pathological patterns such as hundreds of nested parentheses.
893        The checks now trigger 100 bytes before the end of the workspace.
895    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
898    Version 8.01 19-Jan-2010
899    ------------------------
901    1.  If a pattern contained a conditional subpattern with only one branch (in
902        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
903        computed the wrong minimum data length (which is of course zero for such
904        subpatterns). This could cause incorrect "no match" results.
906    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
907        the pattern is reset in the first branch, pcre_compile() failed with
908        "internal error: code overflow at offset...". This happened only when
909        the reset was to the original external option setting. (An optimization
910        abstracts leading options settings into an external setting, which was the
911        cause of this.)
913    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
914        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
915        assertion pattern did not match (meaning that the assertion was true), it
916        was incorrectly treated as false if the SKIP had been reached during the
917        matching. This also applied to assertions used as conditions.
919    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
920        assertion subpattern, including such a pattern used as a condition,
921        unpredictable results occurred, instead of the error return
924    5.  The C++ GlobalReplace function was not working like Perl for the special
925        situation when an empty string is matched. It now does the fancy magic
926        stuff that is necessary.
928    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
929        removed. (These were left over from very, very early versions of PCRE.)
931    7.  Some cosmetic changes to the code to make life easier when compiling it
932        as part of something else:
934        (a) Change DEBUG to PCRE_DEBUG.
936        (b) In pcre_compile(), rename the member of the "branch_chain" structure
937            called "current" as "current_branch", to prevent a collision with the
938            Linux macro when compiled as a kernel module.
940        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
941            prevent a collision with the Linux macro when compiled as a kernel
942            module.
944    8.  In pcre_compile() there are some checks for integer overflows that used to
945        cast potentially large values to (double). This has been changed to that
946        when building, a check for int64_t is made, and if it is found, it is used
947        instead, thus avoiding the use of floating point arithmetic. (There is no
948        other use of FP in PCRE.) If int64_t is not found, the fallback is to
949        double.
951    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
952        2005 (difference between two addresses compared to an unsigned value).
954    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
955        custom one, because of the following reported problem in Windows:
957          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
958              under Win32.
959          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
960              therefore missing the function definition.
961          - The compiler thus generates a "C" signature for the test function.
962          - The linker fails to find the "C" function.
963          - PCRE fails to configure if asked to do so against libbz2.
965    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
966        messages were output:
968          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
969          rerunning libtoolize, to keep the correct libtool macros in-tree.
970          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
972        I have done both of these things.
974    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
975        most of the time, it *can* run out if it is given a pattern that contains a
976        runaway infinite recursion. I updated the discussion in the pcrestack man
977        page.
979    13. Now that we have gone to the x.xx style of version numbers, the minor
980        version may start with zero. Using 08 or 09 is a bad idea because users
981        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
982        interpreted as invalid octal numbers. I've updated the previous comment in
983        configure.ac, and also added a check that gives an error if 08 or 09 are
984        used.
986    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
987        causing partial matching to fail when the end of the subject matched \W
988        in a UTF-8 pattern where \W was quantified with a minimum of 3.
990    15. There were some discrepancies between the declarations in pcre_internal.h
991        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
992        their definitions. The declarations used "const uschar *" and the
993        definitions used USPTR. Even though USPTR is normally defined as "const
994        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
995        reported that: "This difference in casting confuses some C++ compilers, for
996        example, SunCC recognizes above declarations as different functions and
997        generates broken code for hbpcre." I have changed the declarations to use
998        USPTR.
1000    16. GNU libtool is named differently on some systems. The autogen.sh script now
1001        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
1002        (FreeBSD).
1004    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
1005        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
1006        comment: "Figure out how to create a longlong from a string: strtoll and
1007        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
1008        instance, but it only takes 2 args instead of 3!"
1010    18. A subtle bug concerned with back references has been fixed by a change of
1011        specification, with a corresponding code fix. A pattern such as
1012        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
1013        refers, was giving matches when it shouldn't. For example, xa=xaaa would
1014        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
1015        same bug. Such groups have to be quantified to be useful, or contained
1016        inside another quantified group. (If there's no repetition, the reference
1017        can never match.) The problem arises because, having left the group and
1018        moved on to the rest of the pattern, a later failure that backtracks into
1019        the group uses the captured value from the final iteration of the group
1020        rather than the correct earlier one. I have fixed this in PCRE by forcing
1021        any group that contains a reference to itself to be an atomic group; that
1022        is, there cannot be any backtracking into it once it has completed. This is
1023        similar to recursive and subroutine calls.
1026    Version 8.00 19-Oct-09
1027    ----------------------
1029    1.  The table for translating pcre_compile() error codes into POSIX error codes
1030        was out-of-date, and there was no check on the pcre_compile() error code
1031        being within the table. This could lead to an OK return being given in
1032        error.
1034    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
1035        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
1036        in a Windows environment.
1038    3.  The pcregrep --count option prints the count for each file even when it is
1039        zero, as does GNU grep. However, pcregrep was also printing all files when
1040        --files-with-matches was added. Now, when both options are given, it prints
1041        counts only for those files that have at least one match. (GNU grep just
1042        prints the file name in this circumstance, but including the count seems
1043        more useful - otherwise, why use --count?) Also ensured that the
1044        combination -clh just lists non-zero counts, with no names.
1046    4.  The long form of the pcregrep -F option was incorrectly implemented as
1047        --fixed_strings instead of --fixed-strings. This is an incompatible change,
1048        but it seems right to fix it, and I didn't think it was worth preserving
1049        the old behaviour.
1051    5.  The command line items --regex=pattern and --regexp=pattern were not
1052        recognized by pcregrep, which required --regex pattern or --regexp pattern
1053        (with a space rather than an '='). The man page documented the '=' forms,
1054        which are compatible with GNU grep; these now work.
1056    6.  No libpcreposix.pc file was created for pkg-config; there was just
1057        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
1059    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
1060        when UCP support is not needed, by modifying the Python script that
1061        generates it from Unicode data files. This should not matter if the module
1062        is correctly used as a library, but I received one complaint about 50K of
1063        unwanted data. My guess is that the person linked everything into his
1064        program rather than using a library. Anyway, it does no harm.
1066    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
1067        was a minimum greater than 1 for a wide character in a possessive
1068        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
1069        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
1070        character. Chaos in the form of incorrect output or a compiling loop could
1071        result.
1073    9.  The restrictions on what a pattern can contain when partial matching is
1074        requested for pcre_exec() have been removed. All patterns can now be
1075        partially matched by this function. In addition, if there are at least two
1076        slots in the offset vector, the offset of the earliest inspected character
1077        for the match and the offset of the end of the subject are set in them when
1078        PCRE_ERROR_PARTIAL is returned.
1080    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
1081        synonymous with PCRE_PARTIAL, for backwards compatibility, and
1082        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
1083        and may be more useful for multi-segment matching.
1085    11. Partial matching with pcre_exec() is now more intuitive. A partial match
1086        used to be given if ever the end of the subject was reached; now it is
1087        given only if matching could not proceed because another character was
1088        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
1089        string "Z", which now yields "no match" instead of "partial match". In the
1090        case of pcre_dfa_exec(), "no match" is given if every matching path for the
1091        final character ended with (*FAIL).
1093    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
1094        if the pattern had a "must contain" character that was already found in the
1095        earlier partial match, unless partial matching was again requested. For
1096        example, with the pattern /dog.(body)?/, the "must contain" character is
1097        "g". If the first part-match was for the string "dog", restarting with
1098        "sbody" failed. This bug has been fixed.
1100    13. The string returned by pcre_dfa_exec() after a partial match has been
1101        changed so that it starts at the first inspected character rather than the
1102        first character of the match. This makes a difference only if the pattern
1103        starts with a lookbehind assertion or \b or \B (\K is not supported by
1104        pcre_dfa_exec()). It's an incompatible change, but it makes the two
1105        matching functions compatible, and I think it's the right thing to do.
1107    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
1108        so that the demonstration program is easily available in environments where
1109        PCRE has not been installed from source.
1111    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
1112        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
1113        library.
1115    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
1116        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
1117        is not the first non-POSIX option to be added. Clearly some people find
1118        these options useful.
1120    17. If a caller to the POSIX matching function regexec() passes a non-zero
1121        value for nmatch with a NULL value for pmatch, the value of
1122        nmatch is forced to zero.
1124    18. RunGrepTest did not have a test for the availability of the -u option of
1125        the diff command, as RunTest does. It now checks in the same way as
1126        RunTest, and also checks for the -b option.
1128    19. If an odd number of negated classes containing just a single character
1129        interposed, within parentheses, between a forward reference to a named
1130        subpattern and the definition of the subpattern, compilation crashed with
1131        an internal error, complaining that it could not find the referenced
1132        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
1133        [The bug was that it was starting one character too far in when skipping
1134        over the character class, thus treating the ] as data rather than
1135        terminating the class. This meant it could skip too much.]
1137    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
1138        /g option in pcretest when the pattern contains \K, which makes it possible
1139        to have an empty string match not at the start, even when the pattern is
1140        anchored. Updated pcretest and pcredemo to use this option.
1142    21. If the maximum number of capturing subpatterns in a recursion was greater
1143        than the maximum at the outer level, the higher number was returned, but
1144        with unset values at the outer level. The correct (outer level) value is
1145        now given.
1147    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
1148        PCRE did not set those parentheses (unlike Perl). I have now found a way to
1149        make it do so. The string so far is captured, making this feature
1150        compatible with Perl.
1152    23. The tests have been re-organized, adding tests 11 and 12, to make it
1153        possible to check the Perl 5.10 features against Perl 5.10.
1155    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
1156        pattern matches a fixed length string. PCRE did not allow this; now it
1157        does. Neither allows recursion.
1159    25. I finally figured out how to implement a request to provide the minimum
1160        length of subject string that was needed in order to match a given pattern.
1161        (It was back references and recursion that I had previously got hung up
1162        on.) This code has now been added to pcre_study(); it finds a lower bound
1163        to the length of subject needed. It is not necessarily the greatest lower
1164        bound, but using it to avoid searching strings that are too short does give
1165        some useful speed-ups. The value is available to calling programs via
1166        pcre_fullinfo().
1168    26. While implementing 25, I discovered to my embarrassment that pcretest had
1169        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
1170        study optimizations had never been tested with that matching function.
1171        Oops. What is worse, even when it was passed study data, there was a bug in
1172        pcre_dfa_exec() that meant it never actually used it. Double oops. There
1173        were also very few tests of studied patterns with pcre_dfa_exec().
1175    27. If (?| is used to create subpatterns with duplicate numbers, they are now
1176        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
1177        on the other side of the coin, they are no longer allowed to have different
1178        names, because these cannot be distinguished in PCRE, and this has caused
1179        confusion. (This is a difference from Perl.)
1181    28. When duplicate subpattern names are present (necessarily with different
1182        numbers, as required by 27 above), and a test is made by name in a
1183        conditional pattern, either for a subpattern having been matched, or for
1184        recursion in such a pattern, all the associated numbered subpatterns are
1185        tested, and the overall condition is true if the condition is true for any
1186        one of them. This is the way Perl works, and is also more like the way
1187        testing by number works.
1190    Version 7.9 11-Apr-09
1191    ---------------------
1193    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
1194        (pcretest), all targets were linked against these libraries. This included
1195        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
1196        libraries. This caused unwanted dependencies to be created. This problem
1197        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
1198        pcretest is linked with readline.
1200    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
1201        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
1202        moved outside it again, because FALSE and TRUE are already defined in AIX,
1203        but BOOL is not.
1205    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
1206        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
1208    4.  The pcregrep documentation said spaces were inserted as well as colons (or
1209        hyphens) following file names and line numbers when outputting matching
1210        lines. This is not true; no spaces are inserted. I have also clarified the
1211        wording for the --colour (or --color) option.
1213    5.  In pcregrep, when --colour was used with -o, the list of matching strings
1214        was not coloured; this is different to GNU grep, so I have changed it to be
1215        the same.
1217    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
1218        each matching line was coloured. Now it goes on to look for further matches
1219        of any of the test patterns, which is the same behaviour as GNU grep.
1221    7.  A pattern that could match an empty string could cause pcregrep to loop; it
1222        doesn't make sense to accept an empty string match in pcregrep, so I have
1223        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
1224        seems to be how GNU grep behaves.
1226    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
1227        start or after a newline", because the conditional assertion was not being
1228        correctly handled. The rule now is that both the assertion and what follows
1229        in the first alternative must satisfy the test.
1231    9.  If auto-callout was enabled in a pattern with a conditional group whose
1232        condition was an assertion, PCRE could crash during matching, both with
1233        pcre_exec() and pcre_dfa_exec().
1235    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
1236        used for matching.
1238    11. Unicode property support in character classes was not working for
1239        characters (bytes) greater than 127 when not in UTF-8 mode.
1241    12. Added the -M command line option to pcretest.
1243    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
1245    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
1247    16. Added comments and documentation about mis-use of no_arg in the C++
1248        wrapper.
1250    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
1251        from Martin Jerabek that uses macro names for all relevant character and
1252        string constants.
1254    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
1255        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
1256        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
1257        these, but not everybody uses configure.
1259    19. A conditional group that had only one branch was not being correctly
1260        recognized as an item that could match an empty string. This meant that an
1261        enclosing group might also not be so recognized, causing infinite looping
1262        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
1263        with the subject "ab", where knowledge that the repeated group can match
1264        nothing is needed in order to break the loop.
1266    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
1267        exec(), but without supplying a callout function, matching went wrong.
1269    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
1270        leak if the size of the offset vector was greater than 30. When the vector
1271        is smaller, the saved offsets during recursion go onto a local stack
1272        vector, but for larger vectors malloc() is used. It was failing to free
1273        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
1274        error, in fact).
1276    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
1277        heapframe that is used only when UTF-8 support is enabled. This caused no
1278        problem, but was untidy.
1280    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
1281        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
1282        included within another project.
1284    24. Steven Van Ingelgem's patches to add more options to the CMake support,
1285        slightly modified by me:
1287          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
1288              not building pcregrep.
1290          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
1291              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
1293    25. Forward references, both numeric and by name, in patterns that made use of
1294        duplicate group numbers, could behave incorrectly or give incorrect errors,
1295        because when scanning forward to find the reference group, PCRE was not
1296        taking into account the duplicate group numbers. A pattern such as
1297        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
1299    26. Changed a few more instances of "const unsigned char *" to USPTR, making
1300        the feature of a custom pointer more persuasive (as requested by a user).
1302    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
1303        pcretest.c, inside #ifndefs, because it seems they are sometimes already
1304        pre-defined.
1306    28. Added support for (*UTF8) at the start of a pattern.
1308    29. Arrange for flags added by the "release type" setting in CMake to be shown
1309        in the configuration summary.
1312    Version 7.8 05-Sep-08
1313    ---------------------
1315    1.  Replaced UCP searching code with optimized version as implemented for Ad
1316        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
1317        stage table and inline lookup instead of a function, giving speed ups of 2
1318        to 5 times on some simple patterns that I tested. Permission was given to
1319        distribute the MultiStage2.py script that generates the tables (it's not in
1320        the tarball, but is in the Subversion repository).
1322    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
1323        scripts.
1325    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
1326        a group with a zero qualifier. The result of the study could be incorrect,
1327        or the function might crash, depending on the pattern.
1329    4.  Caseless matching was not working for non-ASCII characters in back
1330        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
1331        It now works when Unicode Property Support is available.
1333    5.  In pcretest, an escape such as \x{de} in the data was always generating
1334        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
1335        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
1336        truncation.
1338    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
1340    7.  Added two (int) casts to pcregrep when printing the difference of two
1341        pointers, in case they are 64-bit values.
1343    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
1344        test 2 if it fails.
1346    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
1347        and a #define of that name to empty if it is not externally set. This is to
1348        allow users of MSVC to set it if necessary.
1350    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
1351        the convenience functions in the pcre_get.c source file.
1353    11. An option change at the start of a pattern that had top-level alternatives
1354        could cause overwriting and/or a crash. This command provoked a crash in
1355        some environments:
1357          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
1359        This potential security problem was recorded as CVE-2008-2371.
1361    12. For a pattern where the match had to start at the beginning or immediately
1362        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
1363        pcre_dfa_exec() could read past the end of the passed subject if there was
1364        no match. To help with detecting such bugs (e.g. with valgrind), I modified
1365        pcretest so that it places the subject at the end of its malloc-ed buffer.
1367    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
1368        exec() might read past the end of the data buffer in UTF-8 mode.
1370    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
1371        the data contained the byte 0x85 as part of a UTF-8 character within its
1372        first line. This applied both to normal and DFA matching.
1374    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
1375        /^[^d]*?$/8 failed to match "abc".
1377    16. Added a missing copyright notice to pcrecpp_internal.h.
1379    17. Make it more clear in the documentation that values returned from
1380        pcre_exec() in ovector are byte offsets, not character counts.
1382    18. Tidied a few places to stop certain compilers from issuing warnings.
1384    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
1385        supplied by Stefan Weber. I made a further small update for 7.8 because
1386        there is a change of source arrangements: the pcre_searchfuncs.c module is
1387        replaced by pcre_ucd.c.
1390    Version 7.7 07-May-08
1391    ---------------------
1393    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
1394        a string to a long long, pretend we don't even have a long long." This is
1395        done by checking for the strtoq, strtoll, and _strtoi64 functions.
1397    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
1398        pre-7.6 versions, which defined a global no_arg variable instead of putting
1399        it in the RE class. (See also #8 below.)
1401    3.  Remove a line of dead code, identified by coverity and reported by Nuno
1402        Lopes.
1404    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
1406        (1) The include/exclude patterns were being applied to the whole pathnames
1407            of files, instead of just to the final components.
1409        (2) If there was more than one level of directory, the subdirectories were
1410            skipped unless they satisfied the include/exclude conditions. This is
1411            inconsistent with GNU grep (and could even be seen as contrary to the
1412            pcregrep specification - which I improved to make it absolutely clear).
1413            The action now is always to scan all levels of directory, and just
1414            apply the include/exclude patterns to regular files.
1416    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
1417        --exclude_dir in the tests to avoid scanning .svn directories.
1419    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
1420        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
1421        doesn't support NULs in patterns.
1423    7.  Added some missing "const"s to declarations of static tables in
1424        pcre_compile.c and pcre_dfa_exec.c.
1426    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
1427        caused by fix #2  above. (Subsequently also a second patch to fix the
1428        first patch. And a third patch - this was a messy problem.)
1430    9.  Applied Craig's patch to remove the use of push_back().
1432    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
1433        matching function regexec().
1435    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
1436        which, however, unlike Perl's \g{...}, are subroutine calls, not back
1437        references. PCRE supports relative numbers with this syntax (I don't think
1438        Oniguruma does).
1440    12. Previously, a group with a zero repeat such as (...){0} was completely
1441        omitted from the compiled regex. However, this means that if the group
1442        was called as a subroutine from elsewhere in the pattern, things went wrong
1443        (an internal error was given). Such groups are now left in the compiled
1444        pattern, with a new opcode that causes them to be skipped at execution
1445        time.
1447    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
1448        to the way PCRE behaves:
1450        (a) A lone ] character is dis-allowed (Perl treats it as data).
1452        (b) A back reference to an unmatched subpattern matches an empty string
1453            (Perl fails the current match path).
1455        (c) A data ] in a character class must be notated as \] because if the
1456            first data character in a class is ], it defines an empty class. (In
1457            Perl it is not possible to have an empty class.) The empty class []
1458            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
1459            The negative empty class [^] matches any one character, independently
1460            of the DOTALL setting.
1462    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
1463        non-existent subpattern following a character class starting with ']' and
1464        containing () gave an internal compiling error instead of "reference to
1465        non-existent subpattern". Fortunately, when the pattern did exist, the
1466        compiled code was correct. (When scanning forwards to check for the
1467        existencd of the subpattern, it was treating the data ']' as terminating
1468        the class, so got the count wrong. When actually compiling, the reference
1469        was subsequently set up correctly.)
1471    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
1472        it was being rejected as not supported by pcre_dfa_exec(), even though
1473        other assertions are supported. I have made pcre_dfa_exec() support
1474        (*FAIL).
1476    16. The implementation of 13c above involved the invention of a new opcode,
1477        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
1478        cannot be changed at match time, I realized I could make a small
1479        improvement to matching performance by compiling OP_ALLANY instead of
1480        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
1481        on the OP_ANY path.
1483    17. Compiling pcretest on Windows with readline support failed without the
1484        following two fixes: (1) Make the unistd.h include conditional on
1485        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
1487    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
1488        ncurses library to be included for pcretest when ReadLine support is
1489        requested, but also to allow for it to be overridden. This patch came from
1490        Daniel Bergström.
1492    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
1493        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
1494        any errors with the current Unicode tables. Thanks to Peter Kankowski for
1495        spotting this.
1498    Version 7.6 28-Jan-08
1499    ---------------------
1501    1.  A character class containing a very large number of characters with
1502        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
1503        overflow.
1505    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
1506        HAVE_LONG_LONG is not defined.
1508    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
1509        bring them up to date and include new features. This patch includes:
1511        - Fixed PH's badly added libz and libbz2 support.
1512        - Fixed a problem with static linking.
1513        - Added pcredemo. [But later removed - see 7 below.]
1514        - Fixed dftables problem and added an option.
1515        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
1516            HAVE_LONG_LONG.
1517        - Added readline support for pcretest.
1518        - Added an listing of the option settings after cmake has run.
1520    4.  A user submitted a patch to Makefile that makes it easy to create
1521        "pcre.dll" under mingw when using Configure/Make. I added stuff to
1522        Makefile.am that cause it to include this special target, without
1523        affecting anything else. Note that the same mingw target plus all
1524        the other distribution libraries and programs are now supported
1525        when configuring with CMake (see 6 below) instead of with
1526        Configure/Make.
1528    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
1529        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
1530        exported in the Windows port". It has not yet been confirmed that the patch
1531        solves the problem, but it does no harm.
1533    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
1534        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
1535        with CMake, and also correct the comment about stack recursion.
1537    7.  Remove the automatic building of pcredemo from the ./configure system and
1538        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
1539        of a program that users should build themselves after PCRE is installed, so
1540        building it automatically is not really right. What is more, it gave
1541        trouble in some build environments.
1543    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
1546    Version 7.5 10-Jan-08
1547    ---------------------
1549    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
1550        values in parens when parsing an RE using the C++ wrapper."
1552    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
1553        Characters greater than 255 were excluded from the class instead of being
1554        included.
1556    3.  The same bug as (2) above applied to negated POSIX classes such as
1557        [:^space:].
1559    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
1560        defined or documented. It seems to have been a typo for PCRE_STATIC, so
1561        I have changed it.
1563    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
1564        first named subpattern) and a construct such as (?&a) would reference the
1565        first named subpattern whose name started with "a" (in other words, the
1566        length check was missing). Both these problems are fixed. "Subpattern name
1567        expected" is now given for (?&) (a zero-length name), and this patch also
1568        makes it give the same error for \k'' (previously it complained that that
1569        was a reference to a non-existent subpattern).
1571    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
1572        this is right because (?- can be followed by option settings as well as by
1573        digits. I have, however, made the messages clearer.
1575    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
1576        than the number used in the conditional) now cause a compile-time error.
1577        This is actually not compatible with Perl, which accepts such patterns, but
1578        treats the conditional as always being FALSE (as PCRE used to), but it
1579        seems to me that giving a diagnostic is better.
1581    8.  Change "alphameric" to the more common word "alphanumeric" in comments
1582        and messages.
1584    9.  Fix two occurrences of "backslash" in comments that should have been
1585        "backspace".
1587    10. Remove two redundant lines of code that can never be obeyed (their function
1588        was moved elsewhere).
1590    11. The program that makes PCRE's Unicode character property table had a bug
1591        which caused it to generate incorrect table entries for sequences of
1592        characters that have the same character type, but are in different scripts.
1593        It amalgamated them into a single range, with the script of the first of
1594        them. In other words, some characters were in the wrong script. There were
1595        thirteen such cases, affecting characters in the following ranges:
1597          U+002b0 - U+002c1
1598          U+0060c - U+0060d
1599          U+0061e - U+00612
1600          U+0064b - U+0065e
1601          U+0074d - U+0076d
1602          U+01800 - U+01805
1603          U+01d00 - U+01d77
1604          U+01d9b - U+01dbf
1605          U+0200b - U+0200f
1606          U+030fc - U+030fe
1607          U+03260 - U+0327f
1608          U+0fb46 - U+0fbb1
1609          U+10450 - U+1049d
1611    12. The -o option (show only the matching part of a line) for pcregrep was not
1612        compatible with GNU grep in that, if there was more than one match in a
1613        line, it showed only the first of them. It now behaves in the same way as
1614        GNU grep.
1616    13. If the -o and -v options were combined for pcregrep, it printed a blank
1617        line for every non-matching line. GNU grep prints nothing, and pcregrep now
1618        does the same. The return code can be used to tell if there were any
1619        non-matching lines.
1621    14. Added --file-offsets and --line-offsets to pcregrep.
1623    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
1624        infinitely looping recursion. The bug was that positive lookaheads were not
1625        being skipped when checking for a possible empty match (negative lookaheads
1626        and both kinds of lookbehind were skipped).
1628    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
1629        inclusion of <windows.h> to before rather than after the definition of
1630        INVALID_FILE_ATTRIBUTES (patch from David Byron).
1632    17. Specifying a possessive quantifier with a specific limit for a Unicode
1633        character property caused pcre_compile() to compile bad code, which led at
1634        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
1635        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
1636        caused the error; without that there was no problem.
1638    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
1640    19. Added --enable-pcretest-libreadline.
1642    20. In pcrecpp.cc, the variable 'count' was incremented twice in
1643        RE::GlobalReplace(). As a result, the number of replacements returned was
1644        double what it should be. I removed one of the increments, but Craig sent a
1645        later patch that removed the other one (the right fix) and added unit tests
1646        that check the return values (which was not done before).
1648    21. Several CMake things:
1650        (1) Arranged that, when cmake is used on Unix, the libraries end up with
1651            the names libpcre and libpcreposix, not just pcre and pcreposix.
1653        (2) The above change means that pcretest and pcregrep are now correctly
1654            linked with the newly-built libraries, not previously installed ones.
1658    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
1659        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
1660        UTF-8 newline character). The key issue is that the pattern starts .*;
1661        this means that the match must be either at the beginning, or after a
1662        newline. The bug was in the code for advancing after a failed match and
1663        checking that the new position followed a newline. It was not taking
1664        account of UTF-8 characters correctly.
1666    23. PCRE was behaving differently from Perl in the way it recognized POSIX
1667        character classes. PCRE was not treating the sequence [:...:] as a
1668        character class unless the ... were all letters. Perl, however, seems to
1669        allow any characters between [: and :], though of course it rejects as
1670        unknown any "names" that contain non-letters, because all the known class
1671        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
1672        for example, whereas PCRE did not - it did not recognize a POSIX character
1673        class. This seemed a bit dangerous, so the code has been changed to be
1674        closer to Perl. The behaviour is not identical to Perl, because PCRE will
1675        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
1676        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
1677        Perl does, and where it didn't before.
1679    24. Rewrite so as to remove the single use of %n from pcregrep because in some
1680        Windows environments %n is disabled by default.
1683    Version 7.4 21-Sep-07
1684    ---------------------
1686    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
1687        means that a class such as [\s] counted as "explicit reference to CR or
1688        LF". That isn't really right - the whole point of the change was to try to
1689        help when there was an actual mention of one of the two characters. So now
1690        the change happens only if \r or \n (or a literal CR or LF) character is
1691        encountered.
1693    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
1694        of both had grown to the point where there were only 3 bits left.
1695        Fortunately, there was spare space in the data structure, and so I have
1696        moved the internal flags into a new 16-bit field to free up more option
1697        bits.
1699    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
1700        but did not set the internal JCHANGED flag - either of these is enough to
1701        control the way the "get" function works - but the PCRE_INFO_JCHANGED
1702        facility is supposed to tell if (?J) was ever used, so now (?J) at the
1703        start sets both bits.
1705    4.  Added options (at build time, compile time, exec time) to change \R from
1706        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
1708    5.  doc/pcresyntax.html was missing from the distribution.
1710    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
1711        compatibility, even though it is no longer used.
1713    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
1714        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
1715        windows.h file is present (where different names are used). [This was
1716        reversed later after testing - see 16 below.]
1718    8.  Changed all #include <config.h> to #include "config.h". There were also
1719        some further <pcre.h> cases that I changed to "pcre.h".
1721    9.  When pcregrep was used with the --colour option, it missed the line ending
1722        sequence off the lines that it output.
1724    10. It was pointed out to me that arrays of string pointers cause lots of
1725        relocations when a shared library is dynamically loaded. A technique of
1726        using a single long string with a table of offsets can drastically reduce
1727        these. I have refactored PCRE in four places to do this. The result is
1728        dramatic:
1730          Originally:                          290
1731          After changing UCP table:            187
1732          After changing error message table:   43
1733          After changing table of "verbs"       36
1734          After changing table of Posix names   22
1736        Thanks to the folks working on Gregex for glib for this insight.
1738    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
1739        unicode-properties was also set.
1741    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
1743    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
1744        checked only for CRLF.
1746    14. Added casts to pcretest.c to avoid compiler warnings.
1748    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
1750    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
1751        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
1752        entirely. This removes changes made in 7 above.
1754    17. The CMake files have been updated, and there is now more information about
1755        building with CMake in the NON-UNIX-USE document.
1758    Version 7.3 28-Aug-07
1759    ---------------------
1761     1. In the rejigging of the build system that eventually resulted in 7.1, the
1762        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
1763        brackets there is not right, since it causes compilers to look for an
1764        installed pcre.h, not the version that is in the source that is being
1765        compiled (which of course may be different). I have changed it back to:
1767          #include "pcre.h"
1769        I have a vague recollection that the change was concerned with compiling in
1770        different directories, but in the new build system, that is taken care of
1771        by the VPATH setting the Makefile.
1773     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
1774        when the subject happened to end in the byte 0x85 (e.g. if the last
1775        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
1776        characters but of course it shouldn't be taken as a newline when it is part
1777        of another character. The bug was that, for an unlimited repeat of . in
1778        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
1779        characters when looking for a newline.
1781     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
1783     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
1784        in debug output.
1786     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
1787        long printing in the pcrecpp unittest when running under MinGW.
1789     6. ESC_K was left out of the EBCDIC table.
1791     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
1792        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
1793        limit also applies to "virtual nesting" when a pattern is recursive, and in
1794        this case 1000 isn't so big. I have been able to remove this limit at the
1795        expense of backing off one optimization in certain circumstances. Normally,
1796        when pcre_exec() would call its internal match() function recursively and
1797        immediately return the result unconditionally, it uses a "tail recursion"
1798        feature to save stack. However, when a subpattern that can match an empty
1799        string has an unlimited repetition quantifier, it no longer makes this
1800        optimization. That gives it a stack frame in which to save the data for
1801        checking that an empty string has been matched. Previously this was taken
1802        from the 1000-entry workspace that had been reserved. So now there is no
1803        explicit limit, but more stack is used.
1805     8. Applied Daniel's patches to solve problems with the import/export magic
1806        syntax that is required for Windows, and which was going wrong for the
1807        pcreposix and pcrecpp parts of the library. These were overlooked when this
1808        problem was solved for the main library.
1810     9. There were some crude static tests to avoid integer overflow when computing
1811        the size of patterns that contain repeated groups with explicit upper
1812        limits. As the maximum quantifier is 65535, the maximum group length was
1813        set at 30,000 so that the product of these two numbers did not overflow a
1814        32-bit integer. However, it turns out that people want to use groups that
1815        are longer than 30,000 bytes (though not repeat them that many times).
1816        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1817        made it possible to implement the integer overflow checks in a much more
1818        dynamic way, which I have now done. The artificial limitation on group
1819        length has been removed - we now have only the limit on the total length of
1820        the compiled pattern, which depends on the LINK_SIZE setting.
1822    10. Fixed a bug in the documentation for get/copy named substring when
1823        duplicate names are permitted. If none of the named substrings are set, the
1824        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1825        empty string.
1827    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1828        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1829        because the ] is interpreted as the first data character and the
1830        terminating ] is not found. PCRE has been made compatible with Perl in this
1831        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1832        cause memory overwriting.
1834    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1835        string has been matched (to stop an infinite loop). It was not recognizing
1836        a conditional subpattern that could match an empty string if that
1837        subpattern was within another subpattern. For example, it looped when
1838        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
1839        condition was not nested. This bug has been fixed.
1841    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1842        past the start of the subject in the presence of bytes with the top bit
1843        set, for example "\x8aBCD".
1845    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1846        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1848    14. Optimized (?!) to (*FAIL).
1850    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1851        This restricts code points to be within the range 0 to 0x10FFFF, excluding
1852        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1853        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1854        does: it's just the validity check that is more restrictive.
1856    16. Inserted checks for integer overflows during escape sequence (backslash)
1857        processing, and also fixed erroneous offset values for syntax errors during
1858        backslash processing.
1860    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1861        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1863    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1864        caused an overrun.
1866    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1867        something other than just ASCII characters) inside a group that had an
1868        unlimited repeat caused a loop at compile time (while checking to see
1869        whether the group could match an empty string).
1871    20. Debugging a pattern containing \p or \P could cause a crash. For example,
1872        [\P{Any}] did so. (Error in the code for printing property names.)
1874    21. An orphan \E inside a character class could cause a crash.
1876    22. A repeated capturing bracket such as (A)? could cause a wild memory
1877        reference during compilation.
1879    23. There are several functions in pcre_compile() that scan along a compiled
1880        expression for various reasons (e.g. to see if it's fixed length for look
1881        behind). There were bugs in these functions when a repeated \p or \P was
1882        present in the pattern. These operators have additional parameters compared
1883        with \d, etc, and these were not being taken into account when moving along
1884        the compiled data. Specifically:
1886        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1887            length.
1889        (b) An item such as \pL+ within a repeated group could cause crashes or
1890            loops.
1892        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1893            "reference to non-existent subpattern" error.
1895        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1897    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1898        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1900    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1902    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1903        character were causing crashes (broken optimization).
1905    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1906        \p or \P) caused a compile-time loop.
1908    28. More problems have arisen in unanchored patterns when CRLF is a valid line
1909        break. For example, the unstudied pattern [\r\n]A does not match the string
1910        "\r\nA" because change 7.0/46 below moves the current point on by two
1911        characters after failing to match at the start. However, the pattern \nA
1912        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
1913        the same is true. There doesn't seem any very clean way out of this, but
1914        what I have chosen to do makes the common cases work: PCRE now takes note
1915        of whether there can be an explicit match for \r or \n anywhere in the
1916        pattern, and if so, 7.0/46 no longer applies. As part of this change,
1917        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1918        pattern has explicit CR or LF references.
1920    29. Added (*CR) etc for changing newline setting at start of pattern.
1923    Version 7.2 19-Jun-07
1924    ---------------------
1926     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
1927        which is apparently normally available under Windows.
1929     2. Re-jig the pcregrep tests with different newline settings in an attempt
1930        to make them independent of the local environment's newline setting.
1932     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
1934     4. Some of the "internals" tests were previously cut out when the link size
1935        was not 2, because the output contained actual offsets. The recent new
1936        "Z" feature of pcretest means that these can be cut out, making the tests
1937        usable with all link sizes.
1939     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
1940        stack recursion. This gives a massive performance boost under BSD, but just
1941        a small improvement under Linux. However, it saves one field in the frame
1942        in all cases.
1944     6. Added more features from the forthcoming Perl 5.10:
1946        (a) (?-n) (where n is a string of digits) is a relative subroutine or
1947            recursion call. It refers to the nth most recently opened parentheses.
1949        (b) (?+n) is also a relative subroutine call; it refers to the nth next
1950            to be opened parentheses.
1952        (c) Conditions that refer to capturing parentheses can be specified
1953            relatively, for example, (?(-2)... or (?(+3)...
1955        (d) \K resets the start of the current match so that everything before
1956            is not part of it.
1958        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1960        (f) \g{name} is another synonym - part of Perl 5.10's unification of
1961            reference syntax.
1963        (g) (?| introduces a group in which the numbering of parentheses in each
1964            alternative starts with the same number.
1966        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
1968     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
1971     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
1972        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
1973        for detecting groups that can match an empty string.
1975     9. A pattern with a very large number of alternatives (more than several
1976        hundred) was running out of internal workspace during the pre-compile
1977        phase, where pcre_compile() figures out how much memory will be needed. A
1978        bit of new cunning has reduced the workspace needed for groups with
1979        alternatives. The 1000-alternative test pattern now uses 12 bytes of
1980        workspace instead of running out of the 4096 that are available.
1982    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
1984    11. Applied patch from Google to remove an optimization that didn't quite work.
1985        The report of the bug said:
1987          pcrecpp::RE("a*").FullMatch("aaa") matches, while
1988          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
1989          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
1991    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
1992        it matched the wrong number of bytes.
1995    Version 7.1 24-Apr-07
1996    ---------------------
1998     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
1999        that is more "standard", making use of automake and other Autotools. There
2000        is some re-arrangement of the files and adjustment of comments consequent
2001        on this.
2003     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
2004        for recursive directory scanning broke on some systems because the files
2005        are not scanned in any specific order and on different systems the order
2006        was different. A call to "sort" has been inserted into RunGrepTest for the
2007        approprate test as a short-term fix. In the longer term there may be an
2008        alternative.
2010     3. I had an email from Eric Raymond about problems translating some of PCRE's
2011        man pages to HTML (despite the fact that I distribute HTML pages, some
2012        people do their own conversions for various reasons). The problems
2013        concerned the use of low-level troff macros .br and .in. I have therefore
2014        removed all such uses from the man pages (some were redundant, some could
2015        be replaced by .nf/.fi pairs). The 132html script that I use to generate
2016        HTML has been updated to handle .nf/.fi and to complain if it encounters
2017        .br or .in.
2019     4. Updated comments in configure.ac that get placed in config.h.in and also
2020        arranged for config.h to be included in the distribution, with the name
2021        config.h.generic, for the benefit of those who have to compile without
2022        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
2024     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
2025        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
2026        makevp.bat for the current PCRE, using the additional files
2027        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
2029     6. A Windows user reported a minor discrepancy with test 2, which turned out
2030        to be caused by a trailing space on an input line that had got lost in his
2031        copy. The trailing space was an accident, so I've just removed it.
2033     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
2034        that is needed.
2036     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
2037        as "const" (a) because they are and (b) because it helps the PHP
2038        maintainers who have recently made a script to detect big data structures
2039        in the php code that should be moved to the .rodata section. I remembered
2040        to update Builducptable as well, so it won't revert if ucptable.h is ever
2041        re-created.
2043     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
2044        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
2045        order to be able to cut out the UTF-8 tables in the latter when UTF-8
2046        support is not required. This saves 1.5-2K of code, which is important in
2047        some applications.
2049        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
2050        so as not to refer to the tables, even though these functions will never be
2051        called when UTF-8 support is disabled. Otherwise there are problems with a
2052        shared library.
2054    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
2056        (a) It was defining its arguments as char * instead of void *.
2058        (b) It was assuming that all moves were upwards in memory; this was true
2059            a long time ago when I wrote it, but is no longer the case.
2061        The emulated memove() is provided for those environments that have neither
2062        memmove() nor bcopy(). I didn't think anyone used it these days, but that
2063        is clearly not the case, as these two bugs were recently reported.
2065    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
2066        and Detrail to create the HTML documentation, the .txt form of the man
2067        pages, and it removes trailing spaces from listed files. It also creates
2068        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
2069        case, it wraps all the #defines with #ifndefs. This script should be run
2070        before "make dist".
2072    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
2073        with Unicode property support.
2075        (a) For a maximizing quantifier, if the two different cases of the
2076            character were of different lengths in their UTF-8 codings (there are
2077            some cases like this - I found 11), and the matching function had to
2078            back up over a mixture of the two cases, it incorrectly assumed they
2079            were both the same length.
2081        (b) When PCRE was configured to use the heap rather than the stack for
2082            recursion during matching, it was not correctly preserving the data for
2083            the other case of a UTF-8 character when checking ahead for a match
2084            while processing a minimizing repeat. If the check also involved
2085            matching a wide character, but failed, corruption could cause an
2086            erroneous result when trying to check for a repeat of the original
2087            character.
2089    13. Some tidying changes to the testing mechanism:
2091        (a) The RunTest script now detects the internal link size and whether there
2092            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
2093            values substituted by "configure". (The RunGrepTest script already did
2094            this for UTF-8.) The configure.ac script no longer substitutes the
2095            relevant variables.
2097        (b) The debugging options /B and /D in pcretest show the compiled bytecode
2098            with length and offset values. This means that the output is different
2099            for different internal link sizes. Test 2 is skipped for link sizes
2100            other than 2 because of this, bypassing the problem. Unfortunately,
2101            there was also a test in test 3 (the locale tests) that used /B and
2102            failed for link sizes other than 2. Rather than cut the whole test out,
2103            I have added a new /Z option to pcretest that replaces the length and
2104            offset values with spaces. This is now used to make test 3 independent
2105            of link size. (Test 2 will be tidied up later.)
2107    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
2108        segmentation fault instead of returning the appropriate error message.
2110    15. In multiline mode when the newline sequence was set to "any", the pattern
2111        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
2112        This doesn't seem right; it now treats the CRLF combination as the line
2113        ending, and so does not match in that case. It's only a pattern such as ^$
2114        that would hit this one: something like ^ABC$ would have failed after \r
2115        and then tried again after \r\n.
2117    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
2118        in an attempt to make files that differ only in their line terminators
2119        compare equal. This works on Linux.
2121    17. Under certain error circumstances pcregrep might try to free random memory
2122        as it exited. This is now fixed, thanks to valgrind.
2124    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
2125        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
2126        was because its rules for how to advance for /g after matching an empty
2127        string at the end of a line did not allow for this case. They now check for
2128        it specially.
2130    20. pcretest is supposed to handle patterns and data of any length, by
2131        extending its buffers when necessary. It was getting this wrong when the
2132        buffer for a data line had to be extended.
2134    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
2135        CRLF as a newline sequence.
2137    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
2138        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
2139        I have nevertheless tidied it up.
2141    23. Added some casts to kill warnings from HP-UX ia64 compiler.
2143    24. Added a man page for pcre-config.
2146    Version 7.0 19-Dec-06
2147    ---------------------
2149     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
2150        moving to gcc 4.1.1.
2152     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
2153        sys/time.h, which is documented as needed for this function. It doesn't
2154        seem to matter on Linux, but it showed up on some releases of OS X.
2156     3. It seems that there are systems where bytes whose values are greater than
2157        127 match isprint() in the "C" locale. The "C" locale should be the
2158        default when a C program starts up. In most systems, only ASCII printing
2159        characters match isprint(). This difference caused the output from pcretest
2160        to vary, making some of the tests fail. I have changed pcretest so that:
2162        (a) When it is outputting text in the compiled version of a pattern, bytes
2163            other than 32-126 are always shown as hex escapes.
2165        (b) When it is outputting text that is a matched part of a subject string,
2166            it does the same, unless a different locale has been set for the match
2167            (using the /L modifier). In this case, it uses isprint() to decide.
2169     4. Fixed a major bug that caused incorrect computation of the amount of memory
2170        required for a compiled pattern when options that changed within the
2171        pattern affected the logic of the preliminary scan that determines the
2172        length. The relevant options are -x, and -i in UTF-8 mode. The result was
2173        that the computed length was too small. The symptoms of this bug were
2174        either the PCRE error "internal error: code overflow" from pcre_compile(),
2175        or a glibc crash with a message such as "pcretest: free(): invalid next
2176        size (fast)". Examples of patterns that provoked this bug (shown in
2177        pcretest format) are:
2179          /(?-x: )/x
2180          /(?x)(?-x: \s*#\s*)/
2181          /((?i)[\x{c0}])/8
2182          /(?i:[\x{c0}])/8
2184        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
2185        is now done differently.
2187     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
2188        wrapper classes; (b) implement a new function in the C++ scanner that is
2189        more efficient than the old way of doing things because it avoids levels of
2190        recursion in the regex matching; (c) add a paragraph to the documentation
2191        for the FullMatch() function.
2193     6. The escape sequence \n was being treated as whatever was defined as
2194        "newline". Not only was this contrary to the documentation, which states
2195        that \n is character 10 (hex 0A), but it also went horribly wrong when
2196        "newline" was defined as CRLF. This has been fixed.
2198     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
2199        was being set to -1 for the "end of line" case (supposedly a value that no
2200        character can have). Though this value is never used (the check for end of
2201        line is "zero bytes in current character"), it caused compiler complaints.
2202        I've changed it to 0xffffffff.
2204     8. In pcre_version.c, the version string was being built by a sequence of
2205        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
2206        string (as it is for production releases) called a macro with an empty
2207        argument. The C standard says the result of this is undefined. The gcc
2208        compiler treats it as an empty string (which was what was wanted) but it is
2209        reported that Visual C gives an error. The source has been hacked around to
2210        avoid this problem.
2212     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
2213        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
2214        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
2215        of them did).
2217    10. Originally, pcretest opened its input and output without "b"; then I was
2218        told that "b" was needed in some environments, so it was added for release
2219        5.0 to both the input and output. (It makes no difference on Unix-like
2220        systems.) Later I was told that it is wrong for the input on Windows. I've
2221        now abstracted the modes into two macros, to make it easier to fiddle with
2222        them, and removed "b" from the input mode under Windows.
2224    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
2226    12. Added -help and --help to pcretest as an official way of being reminded
2227        of the options.
2229    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
2230        and pcrecpp.cc because they annoy compilers at high warning levels.
2232    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
2234    15. Fixed an occurrence of == in configure.ac that should have been = (shell
2235        scripts are not C programs :-) and which was not noticed because it works
2236        on Linux.
2238    16. pcretest is supposed to handle any length of pattern and data line (as one
2239        line or as a continued sequence of lines) by extending its input buffer if
2240        necessary. This feature was broken for very long pattern lines, leading to
2241        a string of junk being passed to pcre_compile() if the pattern was longer
2242        than about 50K.
2244    17. I have done a major re-factoring of the way pcre_compile() computes the
2245        amount of memory needed for a compiled pattern. Previously, there was code
2246        that made a preliminary scan of the pattern in order to do this. That was
2247        OK when PCRE was new, but as the facilities have expanded, it has become
2248        harder and harder to keep it in step with the real compile phase, and there
2249        have been a number of bugs (see for example, 4 above). I have now found a
2250        cunning way of running the real compile function in a "fake" mode that
2251        enables it to compute how much memory it would need, while actually only
2252        ever using a few hundred bytes of working memory and without too many
2253        tests of the mode. This should make future maintenance and development
2254        easier. A side effect of this work is that the limit of 200 on the nesting
2255        depth of parentheses has been removed (though this was never a serious
2256        limitation, I suspect). However, there is a downside: pcre_compile() now
2257        runs more slowly than before (30% or more, depending on the pattern). I
2258        hope this isn't a big issue. There is no effect on runtime performance.
2260    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
2261        newline (only possible for the last line of a file) and it was a
2262        pattern that set a locale (followed by /Lsomething), pcretest crashed.
2264    19. Added additional timing features to pcretest. (1) The -tm option now times
2265        matching only, not compiling. (2) Both -t and -tm can be followed, as a
2266        separate command line item, by a number that specifies the number of
2267        repeats to use when timing. The default is 50000; this gives better
2268        precision, but takes uncomfortably long for very large patterns.
2270    20. Extended pcre_study() to be more clever in cases where a branch of a
2271        subpattern has no definite first character. For example, (a*|b*)[cd] would
2272        previously give no result from pcre_study(). Now it recognizes that the
2273        first character must be a, b, c, or d.
2275    21. There was an incorrect error "recursive call could loop indefinitely" if
2276        a subpattern (or the entire pattern) that was being tested for matching an
2277        empty string contained only one non-empty item after a nested subpattern.
2278        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
2279        incorrectly, because the \d was being skipped in the check.
2281    22. The pcretest program now has a new pattern option /B and a command line
2282        option -b, which is equivalent to adding /B to every pattern. This causes
2283        it to show the compiled bytecode, without the additional information that
2284        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
2285        is the same as /B/I).
2287    23. A new optimization is now able automatically to treat some sequences such
2288        as a*b as a*+b. More specifically, if something simple (such as a character
2289        or a simple class like \d) has an unlimited quantifier, and is followed by
2290        something that cannot possibly match the quantified thing, the quantifier
2291        is automatically "possessified".
2293    24. A recursive reference to a subpattern whose number was greater than 39
2294        went wrong under certain circumstances in UTF-8 mode. This bug could also
2295        have affected the operation of pcre_study().
2297    25. Realized that a little bit of performance could be had by replacing
2298        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
2300    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
2302    27. Possessive quantifiers such as a++ were previously implemented by turning
2303        them into atomic groups such as ($>a+). Now they have their own opcodes,
2304        which improves performance. This includes the automatically created ones
2305        from 23 above.
2307    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
2308        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
2309        the first matched character to be a colon. This applied both to named and
2310        numbered groups.
2312    29. The ucpinternal.h header file was missing its idempotency #ifdef.
2314    30. I was sent a "project" file called libpcre.a.dev which I understand makes
2315        building PCRE on Windows easier, so I have included it in the distribution.
2317    31. There is now a check in pcretest against a ridiculously large number being
2318        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
2319        loop, the loop is abandoned.
2321    32. Forward references to subpatterns in conditions such as (?(2)...) where
2322        subpattern 2 is defined later cause pcre_compile() to search forwards in
2323        the pattern for the relevant set of parentheses. This search went wrong
2324        when there were unescaped parentheses in a character class, parentheses
2325        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
2327    33. "Subroutine" calls and backreferences were previously restricted to
2328        referencing subpatterns earlier in the regex. This restriction has now
2329        been removed.
2331    34. Added a number of extra features that are going to be in Perl 5.10. On the
2332        whole, these are just syntactic alternatives for features that PCRE had
2333        previously implemented using the Python syntax or my own invention. The
2334        other formats are all retained for compatibility.
2336        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
2337            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
2338            also .NET compatible.
2340        (b) A recursion or subroutine call to a named group can now be defined as
2341            (?&name) as well as (?P>name).
2343        (c) A backreference to a named group can now be defined as \k<name> or
2344            \k'name' as well as (?P=name). The new forms, as well as being in Perl
2345            5.10, are also .NET compatible.
2347        (d) A conditional reference to a named group can now use the syntax
2348            (?(<name>) or (?('name') as well as (?(name).
2350        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
2351            groups (named and numbered) that are never evaluated inline, but can be
2352            called as "subroutines" from elsewhere. In effect, the DEFINE condition
2353            is always false. There may be only one alternative in such a group.
2355        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
2356            as the simple (?(R). The condition is true only if the most recent
2357            recursion is that of the given number or name. It does not search out
2358            through the entire recursion stack.
2360        (g) The escape \gN or \g{N} has been added, where N is a positive or
2361            negative number, specifying an absolute or relative reference.
2363    35. Tidied to get rid of some further signed/unsigned compiler warnings and
2364        some "unreachable code" warnings.
2366    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
2367        things, this adds five new scripts.
2369    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
2370        There were also incompatibilities regarding the handling of \Q..\E inside
2371        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
2372        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
2374    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
2375        matches an empty string, and forcibly breaks the loop. There were bugs in
2376        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
2377        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
2378        separate and independent bugs (that affected different cases) have been
2379        fixed.
2381    39. Refactored the code to abolish the use of different opcodes for small
2382        capturing bracket numbers. This is a tidy that I avoided doing when I
2383        removed the limit on the number of capturing brackets for 3.5 back in 2001.
2384        The new approach is not only tidier, it makes it possible to reduce the
2385        memory needed to fix the previous bug (38).
2387    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
2388        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
2389        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
2390        mode.
2392    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
2393        report.
2395    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
2396        copy construction and assignment in the C++ wrapper.
2398    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
2399        couple of bugs that could have given wrong results in the "--newline=crlf"
2400        case.
2402    44. Added a number of casts and did some reorganization of signed/unsigned int
2403        variables following suggestions from Dair Grant. Also renamed the variable
2404        "this" as "item" because it is a C++ keyword.
2406    45. Arranged for dftables to add
2408          #include "pcre_internal.h"
2410        to pcre_chartables.c because without it, gcc 4.x may remove the array
2411        definition from the final binary if PCRE is built into a static library and
2412        dead code stripping is activated.
2414    46. For an unanchored pattern, if a match attempt fails at the start of a
2415        newline sequence, and the newline setting is CRLF or ANY, and the next two
2416        characters are CRLF, advance by two characters instead of one.
2419    Version 6.7 04-Jul-06
2420    ---------------------
2422     1. In order to handle tests when input lines are enormously long, pcretest has
2423        been re-factored so that it automatically extends its buffers when
2424        necessary. The code is crude, but this _is_ just a test program. The
2425        default size has been increased from 32K to 50K.
2427     2. The code in pcre_study() was using the value of the re argument before
2428        testing it for NULL. (Of course, in any sensible call of the function, it
2429        won't be NULL.)
2431     3. The memmove() emulation function in pcre_internal.h, which is used on
2432        systems that lack both memmove() and bcopy() - that is, hardly ever -
2433        was missing a "static" storage class specifier.
2435     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
2436        containing an extended class (one that cannot be represented by a bitmap
2437        because it contains high-valued characters or Unicode property items, e.g.
2438        [\pZ]). Almost always one would set UTF-8 mode when processing such a
2439        pattern, but PCRE should not loop if you do not (it no longer does).
2440        [Detail: two cases were found: (a) a repeated subpattern containing an
2441        extended class; (b) a recursive reference to a subpattern that followed a
2442        previous extended class. It wasn't skipping over the extended class
2443        correctly when UTF-8 mode was not set.]
2445     5. A negated single-character class was not being recognized as fixed-length
2446        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
2447        compile error "lookbehind assertion is not fixed length".
2449     6. The RunPerlTest auxiliary script was showing an unexpected difference
2450        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
2451        write a Perl script that can interpret lines of an input file either as
2452        byte characters or as UTF-8, which is what "perltest" was being required to
2453        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
2454        can't do is switch easily at run time between having the "use utf8;" pragma
2455        or not. In the end, I fudged it by using the RunPerlTest script to insert
2456        "use utf8;" explicitly for the UTF-8 tests.
2458     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
2459        the end of the subject string, contrary to the documentation and to what
2460        Perl does. This was true of both matching functions. Now it matches only at
2461        the start of the subject and immediately after *internal* newlines.
2463     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
2464        a pointer to an int instead of a pointer to an unsigned long int. This
2465        caused problems on 64-bit systems.
2467     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
2468        instance of the 'standard' template library not being so standard".
2470    10. There was no check on the number of named subpatterns nor the maximum
2471        length of a subpattern name. The product of these values is used to compute
2472        the size of the memory block for a compiled pattern. By supplying a very
2473        long subpattern name and a large number of named subpatterns, the size
2474        computation could be caused to overflow. This is now prevented by limiting
2475        the length of names to 32 characters, and the number of named subpatterns
2476        to 10,000.
2478    11. Subpatterns that are repeated with specific counts have to be replicated in
2479        the compiled pattern. The size of memory for this was computed from the
2480        length of the subpattern and the repeat count. The latter is limited to
2481        65535, but there was no limit on the former, meaning that integer overflow
2482        could in principle occur. The compiled length of a repeated subpattern is
2483        now limited to 30,000 bytes in order to prevent this.
2485    12. Added the optional facility to have named substrings with the same name.
2487    13. Added the ability to use a named substring as a condition, using the
2488        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
2489        are numbers (not recommended). Forward references are permitted.
2491    14. Added forward references in named backreferences (if you see what I mean).
2493    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
2494        pattern could run off the end of the subject. For example, the pattern
2495        "(?s)(.{1,5})"8 did this with the subject "ab".
2497    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
2498        PCRE_CASELESS was set when matching characters that were quantified with ?
2499        or *.
2501    17. A character class other than a single negated character that had a minimum
2502        but no maximum quantifier - for example [ab]{6,} - was not handled
2503        correctly by pce_dfa_exec(). It would match only one character.
2505    18. A valid (though odd) pattern that looked like a POSIX character
2506        class but used an invalid character after [ (for example [[,abc,]]) caused
2507        pcre_compile() to give the error "Failed: internal error: code overflow" or
2508        in some cases to crash with a glibc free() error. This could even happen if
2509        the pattern terminated after [[ but there just happened to be a sequence of
2510        letters, a binary zero, and a closing ] in the memory that followed.
2512    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
2513        over the years. Originally (before any Unicode support), just the bottom 8
2514        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
2515        output from "man perlunicode" includes this:
2517          The regular expression compiler produces polymorphic opcodes.  That
2518          is, the pattern adapts to the data and automatically switches to
2519          the Unicode character scheme when presented with Unicode data--or
2520          instead uses a traditional byte scheme when presented with byte
2521          data.
2523        Sadly, a wide octal escape does not cause a switch, and in a string with
2524        no other multibyte characters, these octal escapes are treated as before.
2525        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
2526        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
2527        Unicode string.
2529        I have not perpetrated such confusion in PCRE. Up till now, it took just
2530        the bottom 8 bits, as in old Perl. I have now made octal escapes with
2531        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
2532        translate to the appropriate multibyte character.
2534    29. Applied some refactoring to reduce the number of warnings from Microsoft
2535        and Borland compilers. This has included removing the fudge introduced
2536        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
2537        a warning about an unused variable.
2539    21. PCRE has not included VT (character 0x0b) in the set of whitespace
2540        characters since release 4.0, because Perl (from release 5.004) does not.
2541        [Or at least, is documented not to: some releases seem to be in conflict
2542        with the documentation.] However, when a pattern was studied with
2543        pcre_study() and all its branches started with \s, PCRE still included VT
2544        as a possible starting character. Of course, this did no harm; it just
2545        caused an unnecessary match attempt.
2547    22. Removed a now-redundant internal flag bit that recorded the fact that case
2548        dependency changed within the pattern. This was once needed for "required
2549        byte" processing, but is no longer used. This recovers a now-scarce options
2550        bit. Also moved the least significant internal flag bit to the most-
2551        significant bit of the word, which was not previously used (hangover from
2552        the days when it was an int rather than a uint) to free up another bit for
2553        the future.
2555    23. Added support for CRLF line endings as well as CR and LF. As well as the
2556        default being selectable at build time, it can now be changed at runtime
2557        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
2558        specify that it is scanning data with non-default line endings.
2560    24. Changed the definition of CXXLINK to make it agree with the definition of
2561        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
2563    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
2564        recursions. This makes a big different to stack usage for some patterns.
2566    26. If a subpattern containing a named recursion or subroutine reference such
2567        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
2568        the space required for the compiled pattern went wrong and gave too small a
2569        value. Depending on the environment, this could lead to "Failed: internal
2570        error: code overflow at offset 49" or "glibc detected double free or
2571        corruption" errors.
2573    27. Applied patches from Google (a) to support the new newline modes and (b) to
2574        advance over multibyte UTF-8 characters in GlobalReplace.
2576    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
2577        difference for some implementation of PCRE in some Windows version.
2579    29. Added some extra testing facilities to pcretest:
2581        \q<number>   in a data line sets the "match limit" value
2582        \Q<number>   in a data line sets the "match recursion limt" value
2583        -S <number>  sets the stack size, where <number> is in megabytes
2585        The -S option isn't available for Windows.
2588    Version 6.6 06-Feb-06
2589    ---------------------
2591     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
2592        in pcreposix.h. I have copied the definition from pcre.h.
2594     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
2595        because pcre.h is no longer a built file.
2597     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
2598        not normally included in the compiled code.
2601    Version 6.5 01-Feb-06
2602    ---------------------
2604     1. When using the partial match feature with pcre_dfa_exec(), it was not
2605        anchoring the second and subsequent partial matches at the new starting
2606        point. This could lead to incorrect results. For example, with the pattern
2607        /1234/, partially matching against "123" and then "a4" gave a match.
2609     2. Changes to pcregrep:
2611        (a) All non-match returns from pcre_exec() were being treated as failures
2612            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
2613            error message is output. Some extra information is given for the
2614            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
2615            probably the only errors that are likely to be caused by users (by
2616            specifying a regex that has nested indefinite repeats, for instance).
2617            If there are more than 20 of these errors, pcregrep is abandoned.
2619        (b) A binary zero was treated as data while matching, but terminated the
2620            output line if it was written out. This has been fixed: binary zeroes
2621            are now no different to any other data bytes.
2623        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
2624            used to set a locale for matching. The --locale=xxxx long option has
2625            been added (no short equivalent) to specify a locale explicitly on the
2626            pcregrep command, overriding the environment variables.
2628        (d) When -B was used with -n, some line numbers in the output were one less
2629            than they should have been.
2631        (e) Added the -o (--only-matching) option.
2633        (f) If -A or -C was used with -c (count only), some lines of context were
2634            accidentally printed for the final match.
2636        (g) Added the -H (--with-filename) option.
2638        (h) The combination of options -rh failed to suppress file names for files
2639            that were found from directory arguments.
2641        (i) Added the -D (--devices) and -d (--directories) options.
2643        (j) Added the -F (--fixed-strings) option.
2645        (k) Allow "-" to be used as a file name for -f as well as for a data file.
2647        (l) Added the --colo(u)r option.
2649        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
2650            is not present by default.
2652     3. A nasty bug was discovered in the handling of recursive patterns, that is,
2653        items such as (?R) or (?1), when the recursion could match a number of
2654        alternatives. If it matched one of the alternatives, but subsequently,
2655        outside the recursion, there was a failure, the code tried to back up into
2656        the recursion. However, because of the way PCRE is implemented, this is not
2657        possible, and the result was an incorrect result from the match.
2659        In order to prevent this happening, the specification of recursion has
2660        been changed so that all such subpatterns are automatically treated as
2661        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
2663     4. I had overlooked the fact that, in some locales, there are characters for
2664        which isalpha() is true but neither isupper() nor islower() are true. In
2665        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
2666        and ordfeminine) are like this. This affected the treatment of \w and \W
2667        when they appeared in character classes, but not when they appeared outside
2668        a character class. The bit map for "word" characters is now created
2669        separately from the results of isalnum() instead of just taking it from the
2670        upper, lower, and digit maps. (Plus the underscore character, of course.)
2672     5. The above bug also affected the handling of POSIX character classes such as
2673        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
2674        permanent tables. Instead, the bit maps for such a class were previously
2675        created as the appropriate unions of the upper, lower, and digit bitmaps.
2676        Now they are created by subtraction from the [[:word:]] class, which has
2677        its own bitmap.
2679     6. The [[:blank:]] character class matches horizontal, but not vertical space.
2680        It is created by subtracting the vertical space characters (\x09, \x0a,
2681        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
2682        subtraction was done in the overall bitmap for a character class, meaning
2683        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
2684        be recognized. This bug has been fixed.
2686     7. Patches from the folks at Google:
2688          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
2689          real life, but is still worth protecting against".
2691          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
2692          regular expressions".
2694          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
2695          have it.
2697          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
2698          "configure" and the latter not, in order to fix a problem somebody had
2699          with compiling the Arg class on HP-UX.
2701          (e) Improve the error-handling of the C++ wrapper a little bit.
2703          (f) New tests for checking recursion limiting.
2705     8. The pcre_memmove() function, which is used only if the environment does not
2706        have a standard memmove() function (and is therefore rarely compiled),
2707        contained two bugs: (a) use of int instead of size_t, and (b) it was not
2708        returning a result (though PCRE never actually uses the result).
2710     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
2711        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
2712        returned instead of calling malloc() with an overflowing number that would
2713        most likely cause subsequent chaos.
2715    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
2717    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
2718        with this option is matched, the nmatch and pmatch options of regexec() are
2719        ignored.
2721    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
2722        provided in case anyone wants to the the POSIX interface with UTF-8
2723        strings.
2725    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
2726        C++ linking (needed for some HP-UX environments).
2728    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
2729        (unused parameter) and in the pcre_printint() function (omitted "default"
2730        switch label when the default is to do nothing).
2732    15. Added some code to make it possible, when PCRE is compiled as a C++
2733        library, to replace subject pointers for pcre_exec() with a smart pointer
2734        class, thus making it possible to process discontinuous strings.
2736    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
2737        much the same function. They were added by different people who were trying
2738        to make PCRE easy to compile on non-Unix systems. It has been suggested
2739        that PCRE_EXPORT be abolished now that there is more automatic apparatus
2740        for compiling on Windows systems. I have therefore replaced it with
2741        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
2742        defaults to "extern" for C or "extern C" for C++, which works fine on
2743        Unix-like systems. It is now possible to override the value of PCRE_DATA_
2744        SCOPE with something explicit in config.h. In addition:
2746        (a) pcreposix.h still had just "extern" instead of either of these macros;
2747            I have replaced it with PCRE_DATA_SCOPE.
2749        (b) Functions such as _pcre_xclass(), which are internal to the library,
2750            but external in the C sense, all had PCRE_EXPORT in their definitions.
2751            This is apparently wrong for the Windows case, so I have removed it.
2752            (It makes no difference on Unix-like systems.)
2754    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
2755        of recursive calls to match(). This is different to MATCH_LIMIT because
2756        that limits the total number of calls to match(), not all of which increase
2757        the depth of recursion. Limiting the recursion depth limits the amount of
2758        stack (or heap if NO_RECURSE is set) that is used. The default can be set
2759        when PCRE is compiled, and changed at run time. A patch from Google adds
2760        this functionality to the C++ interface.
2762    18. Changes to the handling of Unicode character properties:
2764        (a) Updated the table to Unicode 4.1.0.
2766        (b) Recognize characters that are not in the table as "Cn" (undefined).
2768        (c) I revised the way the table is implemented to a much improved format
2769            which includes recognition of ranges. It now supports the ranges that
2770            are defined in UnicodeData.txt, and it also amalgamates other
2771            characters into ranges. This has reduced the number of entries in the
2772            table from around 16,000 to around 3,000, thus reducing its size
2773            considerably. I realized I did not need to use a tree structure after
2774            all - a binary chop search is just as efficient. Having reduced the
2775            number of entries, I extended their size from 6 bytes to 8 bytes to
2776            allow for more data.
2778        (d) Added support for Unicode script names via properties such as \p{Han}.
2780    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
2781        matching that character.
2783    20. When matching a repeated Unicode property with a minimum greater than zero,
2784        (for example \pL{2,}), PCRE could look past the end of the subject if it
2785        reached it while seeking the minimum number of characters. This could
2786        happen only if some of the characters were more than one byte long, because
2787        there is a check for at least the minimum number of bytes.
2789    21. Refactored the implementation of \p and \P so as to be more general, to
2790        allow for more different types of property in future. This has changed the
2791        compiled form incompatibly. Anybody with saved compiled patterns that use
2792        \p or \P will have to recompile them.
2794    22. Added "Any" and "L&" to the supported property types.
2796    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
2797        but give a compile time error if the value is greater than 0xff.
2799    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
2800        accidentally not being installed or uninstalled.
2802    25. The pcre.h file was built from pcre.h.in, but the only changes that were
2803        made were to insert the current release number. This seemed silly, because
2804        it made things harder for people building PCRE on systems that don't run
2805        "configure". I have turned pcre.h into a distributed file, no longer built
2806        by "configure", with the version identification directly included. There is
2807        no longer a pcre.h.in file.
2809        However, this change necessitated a change to the pcre-config script as
2810        well. It is built from pcre-config.in, and one of the substitutions was the
2811        release number. I have updated configure.ac so that ./configure now finds
2812        the release number by grepping pcre.h.
2814    26. Added the ability to run the tests under valgrind.
2817  Version 6.4 05-Sep-05  Version 6.4 05-Sep-05
2818  ---------------------  ---------------------

Removed from v.85  
changed lines
  Added in v.916

  ViewVC Help
Powered by ViewVC 1.1.5