ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 69 by nigel, Sat Feb 24 21:40:18 2007 UTC revision 681 by ph10, Tue Sep 6 09:16:32 2011 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
4    Version 8.20
5    ------------
7    1. Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
8       a POSIX class. After further experiments with Perl, which convinced me that
9       Perl has bugs and confusions, a closing square bracket is no longer allowed
10       in a POSIX name.
12    2. If a pattern such as /(a)b|ac/ is matched against "ac", there is no captured
13       substring, but while checking the failing first alternative, substring 1 is
14       temporarily captured. If the output vector supplied to pcre_exec() was not
15       big enough for this capture, the yield of the function was still zero
16       ("insufficient space for captured substrings"). This cannot be totally fixed
17       without adding another stack variable, which seems a lot of expense for a
18       edge case. However, I have improved the situation in cases such as
19       /(a)(b)x|abc/ matched against "abc", where the return code indicates that
20       fewer than the maximum number of slots in the ovector have been set.
22    3. Related to (2) above: when there are more back references in a pattern than
23       slots in the output vector, pcre_exec() uses temporary memory during
24       matching, and copies in the captures as far as possible afterwards. It was
25       using the entire output vector, but this conflicts with the specification
26       that only 2/3 is used for passing back captured substrings. Now it uses only
27       the first 2/3, for compatibility. This is, of course, another edge case.
29    4. Zoltan Herczeg's just-in-time compiler support has been integrated into the
30       main code base, and can be used by building with --enable-jit.
32    5. When the number of matches in a pcre_dfa_exec() run exactly filled the
33       ovector, the return from the function was zero, implying that there were
34       other matches that did not fit. The correct "exactly full" value is now
35       returned.
37    6. If a subpattern that was called recursively or as a subroutine contained
38       (*PRUNE) or any other control that caused it to give a non-standard return,
39       invalid errors such as "Error -26 (nested recursion at the same subject
40       position)" or even infinite loops could occur.
43    Version 8.13 16-Aug-2011
44    ------------------------
46    1.  The Unicode data tables have been updated to Unicode 6.0.0.
48    2.  Two minor typos in pcre_internal.h have been fixed.
50    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
51        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
52        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
54    4.  There were a number of related bugs in the code for matching backrefences
55        caselessly in UTF-8 mode when codes for the characters concerned were
56        different numbers of bytes. For example, U+023A and U+2C65 are an upper
57        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
58        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
59        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
60        2-byte code at the end of the subject (it thought there wasn't enough data
61        left).
63    5.  Comprehensive information about what went wrong is now returned by
64        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
65        as the output vector has at least 2 elements. The offset of the start of
66        the failing character and a reason code are placed in the vector.
68    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
69        now returned is for the first byte of the failing character, instead of the
70        last byte inspected. This is an incompatible change, but I hope it is small
71        enough not to be a problem. It makes the returned offset consistent with
72        pcre_exec() and pcre_dfa_exec().
74    7.  pcretest now gives a text phrase as well as the error number when
75        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
76        failure, the offset and reason code are output.
78    8.  When \R was used with a maximizing quantifier it failed to skip backwards
79        over a \r\n pair if the subsequent match failed. Instead, it just skipped
80        back over a single character (\n). This seems wrong (because it treated the
81        two characters as a single entity when going forwards), conflicts with the
82        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
83        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
84        has been changed.
86    9.  Some internal refactoring has changed the processing so that the handling
87        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
88        time (the PCRE_DOTALL option was changed this way some time ago: version
89        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
90        which was always a bit of a fudge. It also means that there is one less
91        argument for the match() function, which reduces its stack requirements
92        slightly. This change also fixes an incompatibility with Perl: the pattern
93        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
95    10. More internal refactoring has drastically reduced the number of recursive
96        calls to match() for possessively repeated groups such as (abc)++ when
97        using pcre_exec().
99    11. While implementing 10, a number of bugs in the handling of groups were
100        discovered and fixed:
102        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
103        (a|)*(?1) gave a compile-time internal error.
104        ((a|)+)+  did not notice that the outer group could match an empty string.
105        (^a|^)+   was not marked as anchored.
106        (.*a|.*)+ was not marked as matching at start or after a newline.
108    12. Yet more internal refactoring has removed another argument from the match()
109        function. Special calls to this function are now indicated by setting a
110        value in a variable in the "match data" data block.
112    13. Be more explicit in pcre_study() instead of relying on "default" for
113        opcodes that mean there is no starting character; this means that when new
114        ones are added and accidentally left out of pcre_study(), testing should
115        pick them up.
117    14. The -s option of pcretest has been documented for ages as being an old
118        synonym of -m (show memory usage). I have changed it to mean "force study
119        for every regex", that is, assume /S for every regex. This is similar to -i
120        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
121        using it. It makes it easier to run collections of tests with and without
122        study enabled, and thereby test pcre_study() more easily. All the standard
123        tests are now run with and without -s (but some patterns can be marked as
124        "never study" - see 20 below).
126    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
127        restoration of the capturing data to the outer values was not happening
128        correctly.
130    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
131        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
132        pattern had matched an empty string, and so incorrectly returned a no
133        match.
135    17. There was optimizing code for the last branch of non-capturing parentheses,
136        and also for the obeyed branch of a conditional subexpression, which used
137        tail recursion to cut down on stack usage. Unfortunately, now that there is
138        the possibility of (*THEN) occurring in these branches, tail recursion is
139        no longer possible because the return has to be checked for (*THEN). These
140        two optimizations have therefore been removed.
142    18. If a pattern containing \R was studied, it was assumed that \R always
143        matched two bytes, thus causing the minimum subject length to be
144        incorrectly computed because \R can also match just one byte.
146    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
147        was incorrectly computed.
149    20. If /S is present twice on a test pattern in pcretest input, it now
150        *disables* studying, thereby overriding the use of -s on the command line
151        (see 14 above). This is necessary for one or two tests to keep the output
152        identical in both cases.
154    21. When (*ACCEPT) was used in an assertion that matched an empty string and
155        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
157    22. When an atomic group that contained a capturing parenthesis was
158        successfully matched, but the branch in which it appeared failed, the
159        capturing was not being forgotten if a higher numbered group was later
160        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
161        group 1 to "a", when in fact it should be unset. This applied to multi-
162        branched capturing and non-capturing groups, repeated or not, and also to
163        positive assertions (capturing in negative assertions does not happen
164        in PCRE) and also to nested atomic groups.
166    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
167        subject after a captured substring, to make it easier to tell which of a
168        number of identical substrings has been captured.
170    24. The way atomic groups are processed by pcre_exec() has been changed so that
171        if they are repeated, backtracking one repetition now resets captured
172        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
173        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
174        "aaa". Previously, it would have been "a". As part of this code
175        refactoring, the way recursive calls are handled has also been changed.
177    25. If an assertion condition captured any substrings, they were not passed
178        back unless some other capturing happened later. For example, if
179        (?(?=(a))a) was matched against "a", no capturing was returned.
181    26. When studying a pattern that contained subroutine calls or assertions,
182        the code for finding the minimum length of a possible match was handling
183        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
184        group 1 called group 2 while simultaneously a separate group 2 called group
185        1). A stack overflow occurred in this case. I have fixed this by limiting
186        the recursion depth to 10.
188    27. Updated RunTest.bat in the distribution to the version supplied by Tom
189        Fortmann. This supports explicit test numbers on the command line, and has
190        argument validation and error reporting.
192    28. An instance of \X with an unlimited repeat could fail if at any point the
193        first character it looked at was a mark character.
195    29. Some minor code refactoring concerning Unicode properties and scripts
196        should reduce the stack requirement of match() slightly.
198    30. Added the '=' option to pcretest to check the setting of unused capturing
199        slots at the end of the pattern, which are documented as being -1, but are
200        not included in the return count.
202    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
203        compiled something random. Now it gives a compile-time error (as does
204        Perl).
206    32. A *MARK encountered during the processing of a positive assertion is now
207        recorded and passed back (compatible with Perl).
209    33. If --only-matching or --colour was set on a pcregrep call whose pattern
210        had alternative anchored branches, the search for a second match in a line
211        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
212        matched the line "0102" twice. The same bug affected patterns that started
213        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
214        twice.
216    34. Previously, PCRE did not allow quantification of assertions. However, Perl
217        does, and because of capturing effects, quantifying parenthesized
218        assertions may at times be useful. Quantifiers are now allowed for
219        parenthesized assertions.
221    35. A minor code tidy in pcre_compile() when checking options for \R usage.
223    36. \g was being checked for fancy things in a character class, when it should
224        just be a literal "g".
226    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
227        appearance of a nested POSIX class supersedes an apparent external class.
228        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
229        unescaped square brackets may also appear as part of class names. For
230        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
231        more like Perl. (But see 8.20/1 above.)
233    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
234        was because it thought it was \N{name}, which is not supported).
236    39. Add minix to OS list not supporting the -S option in pcretest.
238    40. PCRE tries to detect cases of infinite recursion at compile time, but it
239        cannot analyze patterns in sufficient detail to catch mutual recursions
240        such as ((?1))((?2)). There is now a runtime test that gives an error if a
241        subgroup is called recursively as a subpattern for a second time at the
242        same position in the subject string. In previous releases this might have
243        been caught by the recursion limit, or it might have run out of stack.
245    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
246        happen only once. PCRE was, however incorrectly giving a compile time error
247        "recursive call could loop indefinitely" because it cannot analyze the
248        pattern in sufficient detail. The compile time test no longer happens when
249        PCRE is compiling a conditional subpattern, but actual runaway loops are
250        now caught at runtime (see 40 above).
252    42. It seems that Perl allows any characters other than a closing parenthesis
253        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
254        has been changed to be the same.
256    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
257        as not to get warnings when autogen.sh is called. Also changed
258        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
260    44. To help people who use pcregrep to scan files containing exceedingly long
261        lines, the following changes have been made:
263        (a) The default value of the buffer size parameter has been increased from
264            8K to 20K. (The actual buffer used is three times this size.)
266        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
267            PCRE is built.
269        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
270            to be set at run time.
272        (d) Numerical values in pcregrep options can be followed by K or M, for
273            example --buffer-size=50K.
275        (e) If a line being scanned overflows pcregrep's buffer, an error is now
276            given and the return code is set to 2.
278    45. Add a pointer to the latest mark to the callout data block.
280    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
281        partial match of an empty string instead of no match. This was specific to
282        the use of ".".
284    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
285        complete match instead of a partial match. This bug was dependent on both
286        the PCRE_UTF8 and PCRE_DOTALL options being set.
288    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
289        starting byte set, because \b was not being ignored.
292    Version 8.12 15-Jan-2011
293    ------------------------
295    1.  Fixed some typos in the markup of the man pages, and wrote a script that
296        checks for such things as part of the documentation building process.
298    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
299        --match-limit and --recursion-limit options (added for 8.11). In
300        particular, this made one of the standard tests fail. (The integer value
301        went into the wrong half of a long int.)
303    3.  If the --colour option was given to pcregrep with -v (invert match), it
304        did strange things, either producing crazy output, or crashing. It should,
305        of course, ignore a request for colour when reporting lines that do not
306        match.
308    4.  Another pcregrep bug caused similar problems if --colour was specified with
309        -M (multiline) and the pattern match finished with a line ending.
311    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
312        matched in multiline mode, the following line was shown as part of the
313        match. This seems wrong, so I have changed it.
315    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
316        the check for further matches in the same line (so they could be coloured)
317        to overrun the end of the current line. If another match was found, it was
318        incorrectly shown (and then shown again when found in the next line).
320    7.  If pcregrep was compiled under Windows, there was a reference to the
321        function pcregrep_exit() before it was defined. I am assuming this was
322        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
323        reported by a user. I've moved the definition above the reference.
326    Version 8.11 10-Dec-2010
327    ------------------------
329    1.  (*THEN) was not working properly if there were untried alternatives prior
330        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
331        backtracked to try for "b" instead of moving to the next alternative branch
332        at the same level (in this case, to look for "c"). The Perl documentation
333        is clear that when (*THEN) is backtracked onto, it goes to the "next
334        alternative in the innermost enclosing group".
336    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
337        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
338        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
339        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
340        (*THEN).
342    3.  If \s appeared in a character class, it removed the VT character from
343        the class, even if it had been included by some previous item, for example
344        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
345        of \s, but is part of the POSIX "space" class.)
347    4.  A partial match never returns an empty string (because you can always
348        match an empty string at the end of the subject); however the checking for
349        an empty string was starting at the "start of match" point. This has been
350        changed to the "earliest inspected character" point, because the returned
351        data for a partial match starts at this character. This means that, for
352        example, /(?<=abc)def/ gives a partial match for the subject "abc"
353        (previously it gave "no match").
355    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
356        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
357        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
358        has an implication that the given string is incomplete (because a partial
359        match is preferred over a full match). For this reason, these items now
360        give a partial match in this situation. [Aside: previously, the one case
361        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
362        match rather than a full match, which was wrong by the old rules, but is
363        now correct.]
365    6.  There was a bug in the handling of #-introduced comments, recognized when
366        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
367        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
368        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
369        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
370        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
371        places in pcre_compile().
373    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
374        comments when looking ahead for named forward references to subpatterns,
375        the only newline sequence it recognized was NL. It now handles newlines
376        according to the set newline convention.
378    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
379        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
380        cater for a lack of strerror(). These oversights have been fixed.
382    9.  Added --match-limit and --recursion-limit to pcregrep.
384    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
386    11. When the -o option was used, pcregrep was setting a return code of 1, even
387        when matches were found, and --line-buffered was not being honoured.
389    12. Added an optional parentheses number to the -o and --only-matching options
390        of pcregrep.
392    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
393        can match an empty string. The code to do it in pcretest and pcredemo
394        needed fixing:
396        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
397            only one byte after an empty string match just before CRLF (this case
398            just got forgotten; "any" and "anycrlf" were OK).
400        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
401            mode when an empty string match preceded an ASCII character followed by
402            a non-ASCII character. (The code for advancing by one character rather
403            than one byte was nonsense.)
405        (c) The pcredemo.c sample program did not have any code at all to handle
406            the cases when CRLF is a valid newline sequence.
408    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
409        as a starting offset was within the subject string. There is now a new
410        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
411        negative or greater than the length of the string. In order to test this,
412        pcretest is extended to allow the setting of negative starting offsets.
414    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
415        starting offset points to the beginning of a UTF-8 character was
416        unnecessarily clumsy. I tidied it up.
418    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
419        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
421    17. Nobody had reported that the --include_dir option, which was added in
422        release 7.7 should have been called --include-dir (hyphen, not underscore)
423        for compatibility with GNU grep. I have changed it to --include-dir, but
424        left --include_dir as an undocumented synonym, and the same for
425        --exclude-dir, though that is not available in GNU grep, at least as of
426        release 2.5.4.
428    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
429        characters from a string of bytes) have been redefined so as not to use
430        loops, in order to improve performance in some environments. At the same
431        time, I abstracted some of the common code into auxiliary macros to save
432        repetition (this should not affect the compiled code).
434    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
435        compile-time error is now given if \c is not followed by an ASCII
436        character, that is, a byte less than 128. (In EBCDIC mode, the code is
437        different, and any byte value is allowed.)
439    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
440        START_OPTIMIZE option, which is now allowed at compile time - but just
441        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
442        to pcregrep and other applications that have no direct access to PCRE
443        options. The new /Y option in pcretest sets this option when calling
444        pcre_compile().
446    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
447        back references. Groups containing recursive back references were forced to
448        be atomic by that change, but in the case of named groups, the amount of
449        memory required was incorrectly computed, leading to "Failed: internal
450        error: code overflow". This has been fixed.
452    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
453        pcretest.c, to avoid build problems in some Borland environments.
456    Version 8.10 25-Jun-2010
457    ------------------------
459    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
460        THEN.
462    2.  (*ACCEPT) was not working when inside an atomic group.
464    3.  Inside a character class, \B is treated as a literal by default, but
465        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
466        causes the error). The code is unchanged, but I tidied the documentation.
468    4.  Inside a character class, PCRE always treated \R and \X as literals,
469        whereas Perl faults them if its -w option is set. I have changed PCRE so
470        that it faults them when PCRE_EXTRA is set.
472    5.  Added support for \N, which always matches any character other than
473        newline. (It is the same as "." when PCRE_DOTALL is not set.)
475    6.  When compiling pcregrep with newer versions of gcc which may have
476        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
477        declared with attribute warn_unused_result" were given. Just casting the
478        result to (void) does not stop the warnings; a more elaborate fudge is
479        needed. I've used a macro to implement this.
481    7.  Minor change to pcretest.c to avoid a compiler warning.
483    8.  Added four artifical Unicode properties to help with an option to make
484        \s etc use properties (see next item). The new properties are: Xan
485        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
487    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
488        use Unicode properties. (*UCP) at the start of a pattern can be used to set
489        this option. Modified pcretest to add /W to test this facility. Added
490        REG_UCP to make it available via the POSIX interface.
492    10. Added --line-buffered to pcregrep.
494    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
495        studied, and the match started with a letter with a code point greater than
496        127 whose first byte was different to the first byte of the other case of
497        the letter, the other case of this starting letter was not recognized
498        (#976).
500    12. If a pattern that was studied started with a repeated Unicode property
501        test, for example, \p{Nd}+, there was the theoretical possibility of
502        setting up an incorrect bitmap of starting bytes, but fortunately it could
503        not have actually happened in practice until change 8 above was made (it
504        added property types that matched character-matching opcodes).
506    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
507        possible starting bytes for non-anchored patterns.
509    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
510        \R, and also a number of cases that involve Unicode properties, both
511        explicit and implicit when PCRE_UCP is set.
513    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
514        input, it could crash or give wrong results if characters with values
515        greater than 0xc0 were present in the subject string. (Detail: it assumed
516        UTF-8 input when processing these items.)
518    16. Added a lot of (int) casts to avoid compiler warnings in systems where
519        size_t is 64-bit (#991).
521    17. Added a check for running out of memory when PCRE is compiled with
522        --disable-stack-for-recursion (#990).
524    18. If the last data line in a file for pcretest does not have a newline on
525        the end, a newline was missing in the output.
527    19. The default pcre_chartables.c file recognizes only ASCII characters (values
528        less than 128) in its various bitmaps. However, there is a facility for
529        generating tables according to the current locale when PCRE is compiled. It
530        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
531        characters, are recognized by isspace() and therefore were getting set in
532        these tables, and indeed these tables seem to approximate to ISO 8859. This
533        caused a problem in UTF-8 mode when pcre_study() was used to create a list
534        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
535        which of course cannot start UTF-8 characters. I have changed the code so
536        that only real ASCII characters (less than 128) and the correct starting
537        bytes for UTF-8 encodings are set for characters greater than 127 when in
538        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
539        altogether.)
541    20. Added the /T option to pcretest so as to be able to run tests with non-
542        standard character tables, thus making it possible to include the tests
543        used for 19 above in the standard set of tests.
545    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
546        reference to a subpattern the other side of a comment that contains an
547        opening parenthesis caused either an internal compiling error, or a
548        reference to the wrong subpattern.
551    Version 8.02 19-Mar-2010
552    ------------------------
554    1.  The Unicode data tables have been updated to Unicode 5.2.0.
556    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
557        configured.
559    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
560        original author of that file, following a query about its status.
562    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
563        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
565    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
566        quantifier applied to a forward-referencing subroutine call, could compile
567        incorrect code or give the error "internal error: previously-checked
568        referenced subpattern not found".
570    6.  Both MS Visual Studio and Symbian OS have problems with initializing
571        variables to point to external functions. For these systems, therefore,
572        pcre_malloc etc. are now initialized to local functions that call the
573        relevant global functions.
575    7.  There were two entries missing in the vectors called coptable and poptable
576        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
577        I've fixed the data, and added a kludgy way of testing at compile time that
578        the lengths are correct (equal to the number of opcodes).
580    8.  Following on from 7, I added a similar kludge to check the length of the
581        eint vector in pcreposix.c.
583    9.  Error texts for pcre_compile() are held as one long string to avoid too
584        much relocation at load time. To find a text, the string is searched,
585        counting zeros. There was no check for running off the end of the string,
586        which could happen if a new error number was added without updating the
587        string.
589    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
591    11. \K was not working if it appeared in an atomic group or in a group that
592        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
593        \K is "not well defined" if used in an assertion. PCRE now accepts it if
594        the assertion is positive, but not if it is negative.
596    12. Change 11 fortuitously reduced the size of the stack frame used in the
597        "match()" function of pcre_exec.c by one pointer. Forthcoming
598        implementation of support for (*MARK) will need an extra pointer on the
599        stack; I have reserved it now, so that the stack frame size does not
600        decrease.
602    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
603        item in branch that calls a recursion is a subroutine call - as in the
604        second branch in the above example - was incorrectly given the compile-
605        time error "recursive call could loop indefinitely" because pcre_compile()
606        was not correctly checking the subroutine for matching a non-empty string.
608    14. The checks for overrunning compiling workspace could trigger after an
609        overrun had occurred. This is a "should never occur" error, but it can be
610        triggered by pathological patterns such as hundreds of nested parentheses.
611        The checks now trigger 100 bytes before the end of the workspace.
613    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
616    Version 8.01 19-Jan-2010
617    ------------------------
619    1.  If a pattern contained a conditional subpattern with only one branch (in
620        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
621        computed the wrong minimum data length (which is of course zero for such
622        subpatterns). This could cause incorrect "no match" results.
624    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
625        the pattern is reset in the first branch, pcre_compile() failed with
626        "internal error: code overflow at offset...". This happened only when
627        the reset was to the original external option setting. (An optimization
628        abstracts leading options settings into an external setting, which was the
629        cause of this.)
631    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
632        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
633        assertion pattern did not match (meaning that the assertion was true), it
634        was incorrectly treated as false if the SKIP had been reached during the
635        matching. This also applied to assertions used as conditions.
637    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
638        assertion subpattern, including such a pattern used as a condition,
639        unpredictable results occurred, instead of the error return
642    5.  The C++ GlobalReplace function was not working like Perl for the special
643        situation when an empty string is matched. It now does the fancy magic
644        stuff that is necessary.
646    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
647        removed. (These were left over from very, very early versions of PCRE.)
649    7.  Some cosmetic changes to the code to make life easier when compiling it
650        as part of something else:
652        (a) Change DEBUG to PCRE_DEBUG.
654        (b) In pcre_compile(), rename the member of the "branch_chain" structure
655            called "current" as "current_branch", to prevent a collision with the
656            Linux macro when compiled as a kernel module.
658        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
659            prevent a collision with the Linux macro when compiled as a kernel
660            module.
662    8.  In pcre_compile() there are some checks for integer overflows that used to
663        cast potentially large values to (double). This has been changed to that
664        when building, a check for int64_t is made, and if it is found, it is used
665        instead, thus avoiding the use of floating point arithmetic. (There is no
666        other use of FP in PCRE.) If int64_t is not found, the fallback is to
667        double.
669    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
670        2005 (difference between two addresses compared to an unsigned value).
672    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
673        custom one, because of the following reported problem in Windows:
675          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
676              under Win32.
677          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
678              therefore missing the function definition.
679          - The compiler thus generates a "C" signature for the test function.
680          - The linker fails to find the "C" function.
681          - PCRE fails to configure if asked to do so against libbz2.
683    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
684        messages were output:
686          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
687          rerunning libtoolize, to keep the correct libtool macros in-tree.
688          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
690        I have done both of these things.
692    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
693        most of the time, it *can* run out if it is given a pattern that contains a
694        runaway infinite recursion. I updated the discussion in the pcrestack man
695        page.
697    13. Now that we have gone to the x.xx style of version numbers, the minor
698        version may start with zero. Using 08 or 09 is a bad idea because users
699        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
700        interpreted as invalid octal numbers. I've updated the previous comment in
701        configure.ac, and also added a check that gives an error if 08 or 09 are
702        used.
704    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
705        causing partial matching to fail when the end of the subject matched \W
706        in a UTF-8 pattern where \W was quantified with a minimum of 3.
708    15. There were some discrepancies between the declarations in pcre_internal.h
709        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
710        their definitions. The declarations used "const uschar *" and the
711        definitions used USPTR. Even though USPTR is normally defined as "const
712        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
713        reported that: "This difference in casting confuses some C++ compilers, for
714        example, SunCC recognizes above declarations as different functions and
715        generates broken code for hbpcre." I have changed the declarations to use
716        USPTR.
718    16. GNU libtool is named differently on some systems. The autogen.sh script now
719        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
720        (FreeBSD).
722    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
723        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
724        comment: "Figure out how to create a longlong from a string: strtoll and
725        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
726        instance, but it only takes 2 args instead of 3!"
728    18. A subtle bug concerned with back references has been fixed by a change of
729        specification, with a corresponding code fix. A pattern such as
730        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
731        refers, was giving matches when it shouldn't. For example, xa=xaaa would
732        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
733        same bug. Such groups have to be quantified to be useful, or contained
734        inside another quantified group. (If there's no repetition, the reference
735        can never match.) The problem arises because, having left the group and
736        moved on to the rest of the pattern, a later failure that backtracks into
737        the group uses the captured value from the final iteration of the group
738        rather than the correct earlier one. I have fixed this in PCRE by forcing
739        any group that contains a reference to itself to be an atomic group; that
740        is, there cannot be any backtracking into it once it has completed. This is
741        similar to recursive and subroutine calls.
744    Version 8.00 19-Oct-09
745    ----------------------
747    1.  The table for translating pcre_compile() error codes into POSIX error codes
748        was out-of-date, and there was no check on the pcre_compile() error code
749        being within the table. This could lead to an OK return being given in
750        error.
752    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
753        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
754        in a Windows environment.
756    3.  The pcregrep --count option prints the count for each file even when it is
757        zero, as does GNU grep. However, pcregrep was also printing all files when
758        --files-with-matches was added. Now, when both options are given, it prints
759        counts only for those files that have at least one match. (GNU grep just
760        prints the file name in this circumstance, but including the count seems
761        more useful - otherwise, why use --count?) Also ensured that the
762        combination -clh just lists non-zero counts, with no names.
764    4.  The long form of the pcregrep -F option was incorrectly implemented as
765        --fixed_strings instead of --fixed-strings. This is an incompatible change,
766        but it seems right to fix it, and I didn't think it was worth preserving
767        the old behaviour.
769    5.  The command line items --regex=pattern and --regexp=pattern were not
770        recognized by pcregrep, which required --regex pattern or --regexp pattern
771        (with a space rather than an '='). The man page documented the '=' forms,
772        which are compatible with GNU grep; these now work.
774    6.  No libpcreposix.pc file was created for pkg-config; there was just
775        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
777    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
778        when UCP support is not needed, by modifying the Python script that
779        generates it from Unicode data files. This should not matter if the module
780        is correctly used as a library, but I received one complaint about 50K of
781        unwanted data. My guess is that the person linked everything into his
782        program rather than using a library. Anyway, it does no harm.
784    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
785        was a minimum greater than 1 for a wide character in a possessive
786        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
787        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
788        character. Chaos in the form of incorrect output or a compiling loop could
789        result.
791    9.  The restrictions on what a pattern can contain when partial matching is
792        requested for pcre_exec() have been removed. All patterns can now be
793        partially matched by this function. In addition, if there are at least two
794        slots in the offset vector, the offset of the earliest inspected character
795        for the match and the offset of the end of the subject are set in them when
796        PCRE_ERROR_PARTIAL is returned.
798    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
799        synonymous with PCRE_PARTIAL, for backwards compatibility, and
800        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
801        and may be more useful for multi-segment matching.
803    11. Partial matching with pcre_exec() is now more intuitive. A partial match
804        used to be given if ever the end of the subject was reached; now it is
805        given only if matching could not proceed because another character was
806        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
807        string "Z", which now yields "no match" instead of "partial match". In the
808        case of pcre_dfa_exec(), "no match" is given if every matching path for the
809        final character ended with (*FAIL).
811    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
812        if the pattern had a "must contain" character that was already found in the
813        earlier partial match, unless partial matching was again requested. For
814        example, with the pattern /dog.(body)?/, the "must contain" character is
815        "g". If the first part-match was for the string "dog", restarting with
816        "sbody" failed. This bug has been fixed.
818    13. The string returned by pcre_dfa_exec() after a partial match has been
819        changed so that it starts at the first inspected character rather than the
820        first character of the match. This makes a difference only if the pattern
821        starts with a lookbehind assertion or \b or \B (\K is not supported by
822        pcre_dfa_exec()). It's an incompatible change, but it makes the two
823        matching functions compatible, and I think it's the right thing to do.
825    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
826        so that the demonstration program is easily available in environments where
827        PCRE has not been installed from source.
829    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
830        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
831        library.
833    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
834        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
835        is not the first non-POSIX option to be added. Clearly some people find
836        these options useful.
838    17. If a caller to the POSIX matching function regexec() passes a non-zero
839        value for nmatch with a NULL value for pmatch, the value of
840        nmatch is forced to zero.
842    18. RunGrepTest did not have a test for the availability of the -u option of
843        the diff command, as RunTest does. It now checks in the same way as
844        RunTest, and also checks for the -b option.
846    19. If an odd number of negated classes containing just a single character
847        interposed, within parentheses, between a forward reference to a named
848        subpattern and the definition of the subpattern, compilation crashed with
849        an internal error, complaining that it could not find the referenced
850        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
851        [The bug was that it was starting one character too far in when skipping
852        over the character class, thus treating the ] as data rather than
853        terminating the class. This meant it could skip too much.]
855    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
856        /g option in pcretest when the pattern contains \K, which makes it possible
857        to have an empty string match not at the start, even when the pattern is
858        anchored. Updated pcretest and pcredemo to use this option.
860    21. If the maximum number of capturing subpatterns in a recursion was greater
861        than the maximum at the outer level, the higher number was returned, but
862        with unset values at the outer level. The correct (outer level) value is
863        now given.
865    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
866        PCRE did not set those parentheses (unlike Perl). I have now found a way to
867        make it do so. The string so far is captured, making this feature
868        compatible with Perl.
870    23. The tests have been re-organized, adding tests 11 and 12, to make it
871        possible to check the Perl 5.10 features against Perl 5.10.
873    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
874        pattern matches a fixed length string. PCRE did not allow this; now it
875        does. Neither allows recursion.
877    25. I finally figured out how to implement a request to provide the minimum
878        length of subject string that was needed in order to match a given pattern.
879        (It was back references and recursion that I had previously got hung up
880        on.) This code has now been added to pcre_study(); it finds a lower bound
881        to the length of subject needed. It is not necessarily the greatest lower
882        bound, but using it to avoid searching strings that are too short does give
883        some useful speed-ups. The value is available to calling programs via
884        pcre_fullinfo().
886    26. While implementing 25, I discovered to my embarrassment that pcretest had
887        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
888        study optimizations had never been tested with that matching function.
889        Oops. What is worse, even when it was passed study data, there was a bug in
890        pcre_dfa_exec() that meant it never actually used it. Double oops. There
891        were also very few tests of studied patterns with pcre_dfa_exec().
893    27. If (?| is used to create subpatterns with duplicate numbers, they are now
894        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
895        on the other side of the coin, they are no longer allowed to have different
896        names, because these cannot be distinguished in PCRE, and this has caused
897        confusion. (This is a difference from Perl.)
899    28. When duplicate subpattern names are present (necessarily with different
900        numbers, as required by 27 above), and a test is made by name in a
901        conditional pattern, either for a subpattern having been matched, or for
902        recursion in such a pattern, all the associated numbered subpatterns are
903        tested, and the overall condition is true if the condition is true for any
904        one of them. This is the way Perl works, and is also more like the way
905        testing by number works.
908    Version 7.9 11-Apr-09
909    ---------------------
911    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
912        (pcretest), all targets were linked against these libraries. This included
913        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
914        libraries. This caused unwanted dependencies to be created. This problem
915        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
916        pcretest is linked with readline.
918    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
919        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
920        moved outside it again, because FALSE and TRUE are already defined in AIX,
921        but BOOL is not.
923    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
924        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
926    4.  The pcregrep documentation said spaces were inserted as well as colons (or
927        hyphens) following file names and line numbers when outputting matching
928        lines. This is not true; no spaces are inserted. I have also clarified the
929        wording for the --colour (or --color) option.
931    5.  In pcregrep, when --colour was used with -o, the list of matching strings
932        was not coloured; this is different to GNU grep, so I have changed it to be
933        the same.
935    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
936        each matching line was coloured. Now it goes on to look for further matches
937        of any of the test patterns, which is the same behaviour as GNU grep.
939    7.  A pattern that could match an empty string could cause pcregrep to loop; it
940        doesn't make sense to accept an empty string match in pcregrep, so I have
941        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
942        seems to be how GNU grep behaves.
944    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
945        start or after a newline", because the conditional assertion was not being
946        correctly handled. The rule now is that both the assertion and what follows
947        in the first alternative must satisfy the test.
949    9.  If auto-callout was enabled in a pattern with a conditional group whose
950        condition was an assertion, PCRE could crash during matching, both with
951        pcre_exec() and pcre_dfa_exec().
953    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
954        used for matching.
956    11. Unicode property support in character classes was not working for
957        characters (bytes) greater than 127 when not in UTF-8 mode.
959    12. Added the -M command line option to pcretest.
961    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
963    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
965    16. Added comments and documentation about mis-use of no_arg in the C++
966        wrapper.
968    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
969        from Martin Jerabek that uses macro names for all relevant character and
970        string constants.
972    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
973        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
974        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
975        these, but not everybody uses configure.
977    19. A conditional group that had only one branch was not being correctly
978        recognized as an item that could match an empty string. This meant that an
979        enclosing group might also not be so recognized, causing infinite looping
980        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
981        with the subject "ab", where knowledge that the repeated group can match
982        nothing is needed in order to break the loop.
984    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
985        exec(), but without supplying a callout function, matching went wrong.
987    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
988        leak if the size of the offset vector was greater than 30. When the vector
989        is smaller, the saved offsets during recursion go onto a local stack
990        vector, but for larger vectors malloc() is used. It was failing to free
991        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
992        error, in fact).
994    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
995        heapframe that is used only when UTF-8 support is enabled. This caused no
996        problem, but was untidy.
998    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
999        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
1000        included within another project.
1002    24. Steven Van Ingelgem's patches to add more options to the CMake support,
1003        slightly modified by me:
1005          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
1006              not building pcregrep.
1008          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
1009              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
1011    25. Forward references, both numeric and by name, in patterns that made use of
1012        duplicate group numbers, could behave incorrectly or give incorrect errors,
1013        because when scanning forward to find the reference group, PCRE was not
1014        taking into account the duplicate group numbers. A pattern such as
1015        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
1017    26. Changed a few more instances of "const unsigned char *" to USPTR, making
1018        the feature of a custom pointer more persuasive (as requested by a user).
1020    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
1021        pcretest.c, inside #ifndefs, because it seems they are sometimes already
1022        pre-defined.
1024    28. Added support for (*UTF8) at the start of a pattern.
1026    29. Arrange for flags added by the "release type" setting in CMake to be shown
1027        in the configuration summary.
1030    Version 7.8 05-Sep-08
1031    ---------------------
1033    1.  Replaced UCP searching code with optimized version as implemented for Ad
1034        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
1035        stage table and inline lookup instead of a function, giving speed ups of 2
1036        to 5 times on some simple patterns that I tested. Permission was given to
1037        distribute the MultiStage2.py script that generates the tables (it's not in
1038        the tarball, but is in the Subversion repository).
1040    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
1041        scripts.
1043    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
1044        a group with a zero qualifier. The result of the study could be incorrect,
1045        or the function might crash, depending on the pattern.
1047    4.  Caseless matching was not working for non-ASCII characters in back
1048        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
1049        It now works when Unicode Property Support is available.
1051    5.  In pcretest, an escape such as \x{de} in the data was always generating
1052        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
1053        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
1054        truncation.
1056    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
1058    7.  Added two (int) casts to pcregrep when printing the difference of two
1059        pointers, in case they are 64-bit values.
1061    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
1062        test 2 if it fails.
1064    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
1065        and a #define of that name to empty if it is not externally set. This is to
1066        allow users of MSVC to set it if necessary.
1068    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
1069        the convenience functions in the pcre_get.c source file.
1071    11. An option change at the start of a pattern that had top-level alternatives
1072        could cause overwriting and/or a crash. This command provoked a crash in
1073        some environments:
1075          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
1077        This potential security problem was recorded as CVE-2008-2371.
1079    12. For a pattern where the match had to start at the beginning or immediately
1080        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
1081        pcre_dfa_exec() could read past the end of the passed subject if there was
1082        no match. To help with detecting such bugs (e.g. with valgrind), I modified
1083        pcretest so that it places the subject at the end of its malloc-ed buffer.
1085    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
1086        exec() might read past the end of the data buffer in UTF-8 mode.
1088    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
1089        the data contained the byte 0x85 as part of a UTF-8 character within its
1090        first line. This applied both to normal and DFA matching.
1092    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
1093        /^[^d]*?$/8 failed to match "abc".
1095    16. Added a missing copyright notice to pcrecpp_internal.h.
1097    17. Make it more clear in the documentation that values returned from
1098        pcre_exec() in ovector are byte offsets, not character counts.
1100    18. Tidied a few places to stop certain compilers from issuing warnings.
1102    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
1103        supplied by Stefan Weber. I made a further small update for 7.8 because
1104        there is a change of source arrangements: the pcre_searchfuncs.c module is
1105        replaced by pcre_ucd.c.
1108    Version 7.7 07-May-08
1109    ---------------------
1111    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
1112        a string to a long long, pretend we don't even have a long long." This is
1113        done by checking for the strtoq, strtoll, and _strtoi64 functions.
1115    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
1116        pre-7.6 versions, which defined a global no_arg variable instead of putting
1117        it in the RE class. (See also #8 below.)
1119    3.  Remove a line of dead code, identified by coverity and reported by Nuno
1120        Lopes.
1122    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
1124        (1) The include/exclude patterns were being applied to the whole pathnames
1125            of files, instead of just to the final components.
1127        (2) If there was more than one level of directory, the subdirectories were
1128            skipped unless they satisfied the include/exclude conditions. This is
1129            inconsistent with GNU grep (and could even be seen as contrary to the
1130            pcregrep specification - which I improved to make it absolutely clear).
1131            The action now is always to scan all levels of directory, and just
1132            apply the include/exclude patterns to regular files.
1134    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
1135        --exclude_dir in the tests to avoid scanning .svn directories.
1137    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
1138        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
1139        doesn't support NULs in patterns.
1141    7.  Added some missing "const"s to declarations of static tables in
1142        pcre_compile.c and pcre_dfa_exec.c.
1144    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
1145        caused by fix #2  above. (Subsequently also a second patch to fix the
1146        first patch. And a third patch - this was a messy problem.)
1148    9.  Applied Craig's patch to remove the use of push_back().
1150    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
1151        matching function regexec().
1153    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
1154        which, however, unlike Perl's \g{...}, are subroutine calls, not back
1155        references. PCRE supports relative numbers with this syntax (I don't think
1156        Oniguruma does).
1158    12. Previously, a group with a zero repeat such as (...){0} was completely
1159        omitted from the compiled regex. However, this means that if the group
1160        was called as a subroutine from elsewhere in the pattern, things went wrong
1161        (an internal error was given). Such groups are now left in the compiled
1162        pattern, with a new opcode that causes them to be skipped at execution
1163        time.
1165    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
1166        to the way PCRE behaves:
1168        (a) A lone ] character is dis-allowed (Perl treats it as data).
1170        (b) A back reference to an unmatched subpattern matches an empty string
1171            (Perl fails the current match path).
1173        (c) A data ] in a character class must be notated as \] because if the
1174            first data character in a class is ], it defines an empty class. (In
1175            Perl it is not possible to have an empty class.) The empty class []
1176            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
1177            The negative empty class [^] matches any one character, independently
1178            of the DOTALL setting.
1180    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
1181        non-existent subpattern following a character class starting with ']' and
1182        containing () gave an internal compiling error instead of "reference to
1183        non-existent subpattern". Fortunately, when the pattern did exist, the
1184        compiled code was correct. (When scanning forwards to check for the
1185        existencd of the subpattern, it was treating the data ']' as terminating
1186        the class, so got the count wrong. When actually compiling, the reference
1187        was subsequently set up correctly.)
1189    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
1190        it was being rejected as not supported by pcre_dfa_exec(), even though
1191        other assertions are supported. I have made pcre_dfa_exec() support
1192        (*FAIL).
1194    16. The implementation of 13c above involved the invention of a new opcode,
1195        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
1196        cannot be changed at match time, I realized I could make a small
1197        improvement to matching performance by compiling OP_ALLANY instead of
1198        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
1199        on the OP_ANY path.
1201    17. Compiling pcretest on Windows with readline support failed without the
1202        following two fixes: (1) Make the unistd.h include conditional on
1203        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
1205    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
1206        ncurses library to be included for pcretest when ReadLine support is
1207        requested, but also to allow for it to be overridden. This patch came from
1208        Daniel Bergström.
1210    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
1211        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
1212        any errors with the current Unicode tables. Thanks to Peter Kankowski for
1213        spotting this.
1216    Version 7.6 28-Jan-08
1217    ---------------------
1219    1.  A character class containing a very large number of characters with
1220        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
1221        overflow.
1223    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
1224        HAVE_LONG_LONG is not defined.
1226    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
1227        bring them up to date and include new features. This patch includes:
1229        - Fixed PH's badly added libz and libbz2 support.
1230        - Fixed a problem with static linking.
1231        - Added pcredemo. [But later removed - see 7 below.]
1232        - Fixed dftables problem and added an option.
1233        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
1234            HAVE_LONG_LONG.
1235        - Added readline support for pcretest.
1236        - Added an listing of the option settings after cmake has run.
1238    4.  A user submitted a patch to Makefile that makes it easy to create
1239        "pcre.dll" under mingw when using Configure/Make. I added stuff to
1240        Makefile.am that cause it to include this special target, without
1241        affecting anything else. Note that the same mingw target plus all
1242        the other distribution libraries and programs are now supported
1243        when configuring with CMake (see 6 below) instead of with
1244        Configure/Make.
1246    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
1247        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
1248        exported in the Windows port". It has not yet been confirmed that the patch
1249        solves the problem, but it does no harm.
1251    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
1252        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
1253        with CMake, and also correct the comment about stack recursion.
1255    7.  Remove the automatic building of pcredemo from the ./configure system and
1256        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
1257        of a program that users should build themselves after PCRE is installed, so
1258        building it automatically is not really right. What is more, it gave
1259        trouble in some build environments.
1261    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
1264    Version 7.5 10-Jan-08
1265    ---------------------
1267    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
1268        values in parens when parsing an RE using the C++ wrapper."
1270    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
1271        Characters greater than 255 were excluded from the class instead of being
1272        included.
1274    3.  The same bug as (2) above applied to negated POSIX classes such as
1275        [:^space:].
1277    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
1278        defined or documented. It seems to have been a typo for PCRE_STATIC, so
1279        I have changed it.
1281    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
1282        first named subpattern) and a construct such as (?&a) would reference the
1283        first named subpattern whose name started with "a" (in other words, the
1284        length check was missing). Both these problems are fixed. "Subpattern name
1285        expected" is now given for (?&) (a zero-length name), and this patch also
1286        makes it give the same error for \k'' (previously it complained that that
1287        was a reference to a non-existent subpattern).
1289    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
1290        this is right because (?- can be followed by option settings as well as by
1291        digits. I have, however, made the messages clearer.
1293    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
1294        than the number used in the conditional) now cause a compile-time error.
1295        This is actually not compatible with Perl, which accepts such patterns, but
1296        treats the conditional as always being FALSE (as PCRE used to), but it
1297        seems to me that giving a diagnostic is better.
1299    8.  Change "alphameric" to the more common word "alphanumeric" in comments
1300        and messages.
1302    9.  Fix two occurrences of "backslash" in comments that should have been
1303        "backspace".
1305    10. Remove two redundant lines of code that can never be obeyed (their function
1306        was moved elsewhere).
1308    11. The program that makes PCRE's Unicode character property table had a bug
1309        which caused it to generate incorrect table entries for sequences of
1310        characters that have the same character type, but are in different scripts.
1311        It amalgamated them into a single range, with the script of the first of
1312        them. In other words, some characters were in the wrong script. There were
1313        thirteen such cases, affecting characters in the following ranges:
1315          U+002b0 - U+002c1
1316          U+0060c - U+0060d
1317          U+0061e - U+00612
1318          U+0064b - U+0065e
1319          U+0074d - U+0076d
1320          U+01800 - U+01805
1321          U+01d00 - U+01d77
1322          U+01d9b - U+01dbf
1323          U+0200b - U+0200f
1324          U+030fc - U+030fe
1325          U+03260 - U+0327f
1326          U+0fb46 - U+0fbb1
1327          U+10450 - U+1049d
1329    12. The -o option (show only the matching part of a line) for pcregrep was not
1330        compatible with GNU grep in that, if there was more than one match in a
1331        line, it showed only the first of them. It now behaves in the same way as
1332        GNU grep.
1334    13. If the -o and -v options were combined for pcregrep, it printed a blank
1335        line for every non-matching line. GNU grep prints nothing, and pcregrep now
1336        does the same. The return code can be used to tell if there were any
1337        non-matching lines.
1339    14. Added --file-offsets and --line-offsets to pcregrep.
1341    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
1342        infinitely looping recursion. The bug was that positive lookaheads were not
1343        being skipped when checking for a possible empty match (negative lookaheads
1344        and both kinds of lookbehind were skipped).
1346    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
1347        inclusion of <windows.h> to before rather than after the definition of
1348        INVALID_FILE_ATTRIBUTES (patch from David Byron).
1350    17. Specifying a possessive quantifier with a specific limit for a Unicode
1351        character property caused pcre_compile() to compile bad code, which led at
1352        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
1353        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
1354        caused the error; without that there was no problem.
1356    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
1358    19. Added --enable-pcretest-libreadline.
1360    20. In pcrecpp.cc, the variable 'count' was incremented twice in
1361        RE::GlobalReplace(). As a result, the number of replacements returned was
1362        double what it should be. I removed one of the increments, but Craig sent a
1363        later patch that removed the other one (the right fix) and added unit tests
1364        that check the return values (which was not done before).
1366    21. Several CMake things:
1368        (1) Arranged that, when cmake is used on Unix, the libraries end up with
1369            the names libpcre and libpcreposix, not just pcre and pcreposix.
1371        (2) The above change means that pcretest and pcregrep are now correctly
1372            linked with the newly-built libraries, not previously installed ones.
1376    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
1377        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
1378        UTF-8 newline character). The key issue is that the pattern starts .*;
1379        this means that the match must be either at the beginning, or after a
1380        newline. The bug was in the code for advancing after a failed match and
1381        checking that the new position followed a newline. It was not taking
1382        account of UTF-8 characters correctly.
1384    23. PCRE was behaving differently from Perl in the way it recognized POSIX
1385        character classes. PCRE was not treating the sequence [:...:] as a
1386        character class unless the ... were all letters. Perl, however, seems to
1387        allow any characters between [: and :], though of course it rejects as
1388        unknown any "names" that contain non-letters, because all the known class
1389        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
1390        for example, whereas PCRE did not - it did not recognize a POSIX character
1391        class. This seemed a bit dangerous, so the code has been changed to be
1392        closer to Perl. The behaviour is not identical to Perl, because PCRE will
1393        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
1394        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
1395        Perl does, and where it didn't before.
1397    24. Rewrite so as to remove the single use of %n from pcregrep because in some
1398        Windows environments %n is disabled by default.
1401    Version 7.4 21-Sep-07
1402    ---------------------
1404    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
1405        means that a class such as [\s] counted as "explicit reference to CR or
1406        LF". That isn't really right - the whole point of the change was to try to
1407        help when there was an actual mention of one of the two characters. So now
1408        the change happens only if \r or \n (or a literal CR or LF) character is
1409        encountered.
1411    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
1412        of both had grown to the point where there were only 3 bits left.
1413        Fortunately, there was spare space in the data structure, and so I have
1414        moved the internal flags into a new 16-bit field to free up more option
1415        bits.
1417    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
1418        but did not set the internal JCHANGED flag - either of these is enough to
1419        control the way the "get" function works - but the PCRE_INFO_JCHANGED
1420        facility is supposed to tell if (?J) was ever used, so now (?J) at the
1421        start sets both bits.
1423    4.  Added options (at build time, compile time, exec time) to change \R from
1424        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
1426    5.  doc/pcresyntax.html was missing from the distribution.
1428    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
1429        compatibility, even though it is no longer used.
1431    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
1432        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
1433        windows.h file is present (where different names are used). [This was
1434        reversed later after testing - see 16 below.]
1436    8.  Changed all #include <config.h> to #include "config.h". There were also
1437        some further <pcre.h> cases that I changed to "pcre.h".
1439    9.  When pcregrep was used with the --colour option, it missed the line ending
1440        sequence off the lines that it output.
1442    10. It was pointed out to me that arrays of string pointers cause lots of
1443        relocations when a shared library is dynamically loaded. A technique of
1444        using a single long string with a table of offsets can drastically reduce
1445        these. I have refactored PCRE in four places to do this. The result is
1446        dramatic:
1448          Originally:                          290
1449          After changing UCP table:            187
1450          After changing error message table:   43
1451          After changing table of "verbs"       36
1452          After changing table of Posix names   22
1454        Thanks to the folks working on Gregex for glib for this insight.
1456    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
1457        unicode-properties was also set.
1459    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
1461    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
1462        checked only for CRLF.
1464    14. Added casts to pcretest.c to avoid compiler warnings.
1466    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
1468    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
1469        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
1470        entirely. This removes changes made in 7 above.
1472    17. The CMake files have been updated, and there is now more information about
1473        building with CMake in the NON-UNIX-USE document.
1476    Version 7.3 28-Aug-07
1477    ---------------------
1479     1. In the rejigging of the build system that eventually resulted in 7.1, the
1480        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
1481        brackets there is not right, since it causes compilers to look for an
1482        installed pcre.h, not the version that is in the source that is being
1483        compiled (which of course may be different). I have changed it back to:
1485          #include "pcre.h"
1487        I have a vague recollection that the change was concerned with compiling in
1488        different directories, but in the new build system, that is taken care of
1489        by the VPATH setting the Makefile.
1491     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
1492        when the subject happened to end in the byte 0x85 (e.g. if the last
1493        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
1494        characters but of course it shouldn't be taken as a newline when it is part
1495        of another character. The bug was that, for an unlimited repeat of . in
1496        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
1497        characters when looking for a newline.
1499     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
1501     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
1502        in debug output.
1504     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
1505        long printing in the pcrecpp unittest when running under MinGW.
1507     6. ESC_K was left out of the EBCDIC table.
1509     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
1510        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
1511        limit also applies to "virtual nesting" when a pattern is recursive, and in
1512        this case 1000 isn't so big. I have been able to remove this limit at the
1513        expense of backing off one optimization in certain circumstances. Normally,
1514        when pcre_exec() would call its internal match() function recursively and
1515        immediately return the result unconditionally, it uses a "tail recursion"
1516        feature to save stack. However, when a subpattern that can match an empty
1517        string has an unlimited repetition quantifier, it no longer makes this
1518        optimization. That gives it a stack frame in which to save the data for
1519        checking that an empty string has been matched. Previously this was taken
1520        from the 1000-entry workspace that had been reserved. So now there is no
1521        explicit limit, but more stack is used.
1523     8. Applied Daniel's patches to solve problems with the import/export magic
1524        syntax that is required for Windows, and which was going wrong for the
1525        pcreposix and pcrecpp parts of the library. These were overlooked when this
1526        problem was solved for the main library.
1528     9. There were some crude static tests to avoid integer overflow when computing
1529        the size of patterns that contain repeated groups with explicit upper
1530        limits. As the maximum quantifier is 65535, the maximum group length was
1531        set at 30,000 so that the product of these two numbers did not overflow a
1532        32-bit integer. However, it turns out that people want to use groups that
1533        are longer than 30,000 bytes (though not repeat them that many times).
1534        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1535        made it possible to implement the integer overflow checks in a much more
1536        dynamic way, which I have now done. The artificial limitation on group
1537        length has been removed - we now have only the limit on the total length of
1538        the compiled pattern, which depends on the LINK_SIZE setting.
1540    10. Fixed a bug in the documentation for get/copy named substring when
1541        duplicate names are permitted. If none of the named substrings are set, the
1542        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1543        empty string.
1545    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1546        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1547        because the ] is interpreted as the first data character and the
1548        terminating ] is not found. PCRE has been made compatible with Perl in this
1549        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1550        cause memory overwriting.
1552    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1553        string has been matched (to stop an infinite loop). It was not recognizing
1554        a conditional subpattern that could match an empty string if that
1555        subpattern was within another subpattern. For example, it looped when
1556        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
1557        condition was not nested. This bug has been fixed.
1559    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1560        past the start of the subject in the presence of bytes with the top bit
1561        set, for example "\x8aBCD".
1563    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1564        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1566    14. Optimized (?!) to (*FAIL).
1568    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1569        This restricts code points to be within the range 0 to 0x10FFFF, excluding
1570        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1571        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1572        does: it's just the validity check that is more restrictive.
1574    16. Inserted checks for integer overflows during escape sequence (backslash)
1575        processing, and also fixed erroneous offset values for syntax errors during
1576        backslash processing.
1578    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1579        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1581    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1582        caused an overrun.
1584    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1585        something other than just ASCII characters) inside a group that had an
1586        unlimited repeat caused a loop at compile time (while checking to see
1587        whether the group could match an empty string).
1589    20. Debugging a pattern containing \p or \P could cause a crash. For example,
1590        [\P{Any}] did so. (Error in the code for printing property names.)
1592    21. An orphan \E inside a character class could cause a crash.
1594    22. A repeated capturing bracket such as (A)? could cause a wild memory
1595        reference during compilation.
1597    23. There are several functions in pcre_compile() that scan along a compiled
1598        expression for various reasons (e.g. to see if it's fixed length for look
1599        behind). There were bugs in these functions when a repeated \p or \P was
1600        present in the pattern. These operators have additional parameters compared
1601        with \d, etc, and these were not being taken into account when moving along
1602        the compiled data. Specifically:
1604        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1605            length.
1607        (b) An item such as \pL+ within a repeated group could cause crashes or
1608            loops.
1610        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1611            "reference to non-existent subpattern" error.
1613        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1615    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1616        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1618    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1620    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1621        character were causing crashes (broken optimization).
1623    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1624        \p or \P) caused a compile-time loop.
1626    28. More problems have arisen in unanchored patterns when CRLF is a valid line
1627        break. For example, the unstudied pattern [\r\n]A does not match the string
1628        "\r\nA" because change 7.0/46 below moves the current point on by two
1629        characters after failing to match at the start. However, the pattern \nA
1630        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
1631        the same is true. There doesn't seem any very clean way out of this, but
1632        what I have chosen to do makes the common cases work: PCRE now takes note
1633        of whether there can be an explicit match for \r or \n anywhere in the
1634        pattern, and if so, 7.0/46 no longer applies. As part of this change,
1635        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1636        pattern has explicit CR or LF references.
1638    29. Added (*CR) etc for changing newline setting at start of pattern.
1641    Version 7.2 19-Jun-07
1642    ---------------------
1644     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
1645        which is apparently normally available under Windows.
1647     2. Re-jig the pcregrep tests with different newline settings in an attempt
1648        to make them independent of the local environment's newline setting.
1650     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
1652     4. Some of the "internals" tests were previously cut out when the link size
1653        was not 2, because the output contained actual offsets. The recent new
1654        "Z" feature of pcretest means that these can be cut out, making the tests
1655        usable with all link sizes.
1657     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
1658        stack recursion. This gives a massive performance boost under BSD, but just
1659        a small improvement under Linux. However, it saves one field in the frame
1660        in all cases.
1662     6. Added more features from the forthcoming Perl 5.10:
1664        (a) (?-n) (where n is a string of digits) is a relative subroutine or
1665            recursion call. It refers to the nth most recently opened parentheses.
1667        (b) (?+n) is also a relative subroutine call; it refers to the nth next
1668            to be opened parentheses.
1670        (c) Conditions that refer to capturing parentheses can be specified
1671            relatively, for example, (?(-2)... or (?(+3)...
1673        (d) \K resets the start of the current match so that everything before
1674            is not part of it.
1676        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1678        (f) \g{name} is another synonym - part of Perl 5.10's unification of
1679            reference syntax.
1681        (g) (?| introduces a group in which the numbering of parentheses in each
1682            alternative starts with the same number.
1684        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
1686     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
1689     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
1690        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
1691        for detecting groups that can match an empty string.
1693     9. A pattern with a very large number of alternatives (more than several
1694        hundred) was running out of internal workspace during the pre-compile
1695        phase, where pcre_compile() figures out how much memory will be needed. A
1696        bit of new cunning has reduced the workspace needed for groups with
1697        alternatives. The 1000-alternative test pattern now uses 12 bytes of
1698        workspace instead of running out of the 4096 that are available.
1700    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
1702    11. Applied patch from Google to remove an optimization that didn't quite work.
1703        The report of the bug said:
1705          pcrecpp::RE("a*").FullMatch("aaa") matches, while
1706          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
1707          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
1709    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
1710        it matched the wrong number of bytes.
1713    Version 7.1 24-Apr-07
1714    ---------------------
1716     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
1717        that is more "standard", making use of automake and other Autotools. There
1718        is some re-arrangement of the files and adjustment of comments consequent
1719        on this.
1721     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
1722        for recursive directory scanning broke on some systems because the files
1723        are not scanned in any specific order and on different systems the order
1724        was different. A call to "sort" has been inserted into RunGrepTest for the
1725        approprate test as a short-term fix. In the longer term there may be an
1726        alternative.
1728     3. I had an email from Eric Raymond about problems translating some of PCRE's
1729        man pages to HTML (despite the fact that I distribute HTML pages, some
1730        people do their own conversions for various reasons). The problems
1731        concerned the use of low-level troff macros .br and .in. I have therefore
1732        removed all such uses from the man pages (some were redundant, some could
1733        be replaced by .nf/.fi pairs). The 132html script that I use to generate
1734        HTML has been updated to handle .nf/.fi and to complain if it encounters
1735        .br or .in.
1737     4. Updated comments in configure.ac that get placed in config.h.in and also
1738        arranged for config.h to be included in the distribution, with the name
1739        config.h.generic, for the benefit of those who have to compile without
1740        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
1742     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
1743        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
1744        makevp.bat for the current PCRE, using the additional files
1745        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
1747     6. A Windows user reported a minor discrepancy with test 2, which turned out
1748        to be caused by a trailing space on an input line that had got lost in his
1749        copy. The trailing space was an accident, so I've just removed it.
1751     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
1752        that is needed.
1754     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
1755        as "const" (a) because they are and (b) because it helps the PHP
1756        maintainers who have recently made a script to detect big data structures
1757        in the php code that should be moved to the .rodata section. I remembered
1758        to update Builducptable as well, so it won't revert if ucptable.h is ever
1759        re-created.
1761     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
1762        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
1763        order to be able to cut out the UTF-8 tables in the latter when UTF-8
1764        support is not required. This saves 1.5-2K of code, which is important in
1765        some applications.
1767        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
1768        so as not to refer to the tables, even though these functions will never be
1769        called when UTF-8 support is disabled. Otherwise there are problems with a
1770        shared library.
1772    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
1774        (a) It was defining its arguments as char * instead of void *.
1776        (b) It was assuming that all moves were upwards in memory; this was true
1777            a long time ago when I wrote it, but is no longer the case.
1779        The emulated memove() is provided for those environments that have neither
1780        memmove() nor bcopy(). I didn't think anyone used it these days, but that
1781        is clearly not the case, as these two bugs were recently reported.
1783    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
1784        and Detrail to create the HTML documentation, the .txt form of the man
1785        pages, and it removes trailing spaces from listed files. It also creates
1786        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
1787        case, it wraps all the #defines with #ifndefs. This script should be run
1788        before "make dist".
1790    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1791        with Unicode property support.
1793        (a) For a maximizing quantifier, if the two different cases of the
1794            character were of different lengths in their UTF-8 codings (there are
1795            some cases like this - I found 11), and the matching function had to
1796            back up over a mixture of the two cases, it incorrectly assumed they
1797            were both the same length.
1799        (b) When PCRE was configured to use the heap rather than the stack for
1800            recursion during matching, it was not correctly preserving the data for
1801            the other case of a UTF-8 character when checking ahead for a match
1802            while processing a minimizing repeat. If the check also involved
1803            matching a wide character, but failed, corruption could cause an
1804            erroneous result when trying to check for a repeat of the original
1805            character.
1807    13. Some tidying changes to the testing mechanism:
1809        (a) The RunTest script now detects the internal link size and whether there
1810            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1811            values substituted by "configure". (The RunGrepTest script already did
1812            this for UTF-8.) The configure.ac script no longer substitutes the
1813            relevant variables.
1815        (b) The debugging options /B and /D in pcretest show the compiled bytecode
1816            with length and offset values. This means that the output is different
1817            for different internal link sizes. Test 2 is skipped for link sizes
1818            other than 2 because of this, bypassing the problem. Unfortunately,
1819            there was also a test in test 3 (the locale tests) that used /B and
1820            failed for link sizes other than 2. Rather than cut the whole test out,
1821            I have added a new /Z option to pcretest that replaces the length and
1822            offset values with spaces. This is now used to make test 3 independent
1823            of link size. (Test 2 will be tidied up later.)
1825    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1826        segmentation fault instead of returning the appropriate error message.
1828    15. In multiline mode when the newline sequence was set to "any", the pattern
1829        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1830        This doesn't seem right; it now treats the CRLF combination as the line
1831        ending, and so does not match in that case. It's only a pattern such as ^$
1832        that would hit this one: something like ^ABC$ would have failed after \r
1833        and then tried again after \r\n.
1835    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1836        in an attempt to make files that differ only in their line terminators
1837        compare equal. This works on Linux.
1839    17. Under certain error circumstances pcregrep might try to free random memory
1840        as it exited. This is now fixed, thanks to valgrind.
1842    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1843        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1844        was because its rules for how to advance for /g after matching an empty
1845        string at the end of a line did not allow for this case. They now check for
1846        it specially.
1848    20. pcretest is supposed to handle patterns and data of any length, by
1849        extending its buffers when necessary. It was getting this wrong when the
1850        buffer for a data line had to be extended.
1852    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1853        CRLF as a newline sequence.
1855    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1856        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1857        I have nevertheless tidied it up.
1859    23. Added some casts to kill warnings from HP-UX ia64 compiler.
1861    24. Added a man page for pcre-config.
1864    Version 7.0 19-Dec-06
1865    ---------------------
1867     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1868        moving to gcc 4.1.1.
1870     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1871        sys/time.h, which is documented as needed for this function. It doesn't
1872        seem to matter on Linux, but it showed up on some releases of OS X.
1874     3. It seems that there are systems where bytes whose values are greater than
1875        127 match isprint() in the "C" locale. The "C" locale should be the
1876        default when a C program starts up. In most systems, only ASCII printing
1877        characters match isprint(). This difference caused the output from pcretest
1878        to vary, making some of the tests fail. I have changed pcretest so that:
1880        (a) When it is outputting text in the compiled version of a pattern, bytes
1881            other than 32-126 are always shown as hex escapes.
1883        (b) When it is outputting text that is a matched part of a subject string,
1884            it does the same, unless a different locale has been set for the match
1885            (using the /L modifier). In this case, it uses isprint() to decide.
1887     4. Fixed a major bug that caused incorrect computation of the amount of memory
1888        required for a compiled pattern when options that changed within the
1889        pattern affected the logic of the preliminary scan that determines the
1890        length. The relevant options are -x, and -i in UTF-8 mode. The result was
1891        that the computed length was too small. The symptoms of this bug were
1892        either the PCRE error "internal error: code overflow" from pcre_compile(),
1893        or a glibc crash with a message such as "pcretest: free(): invalid next
1894        size (fast)". Examples of patterns that provoked this bug (shown in
1895        pcretest format) are:
1897          /(?-x: )/x
1898          /(?x)(?-x: \s*#\s*)/
1899          /((?i)[\x{c0}])/8
1900          /(?i:[\x{c0}])/8
1902        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1903        is now done differently.
1905     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1906        wrapper classes; (b) implement a new function in the C++ scanner that is
1907        more efficient than the old way of doing things because it avoids levels of
1908        recursion in the regex matching; (c) add a paragraph to the documentation
1909        for the FullMatch() function.
1911     6. The escape sequence \n was being treated as whatever was defined as
1912        "newline". Not only was this contrary to the documentation, which states
1913        that \n is character 10 (hex 0A), but it also went horribly wrong when
1914        "newline" was defined as CRLF. This has been fixed.
1916     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1917        was being set to -1 for the "end of line" case (supposedly a value that no
1918        character can have). Though this value is never used (the check for end of
1919        line is "zero bytes in current character"), it caused compiler complaints.
1920        I've changed it to 0xffffffff.
1922     8. In pcre_version.c, the version string was being built by a sequence of
1923        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1924        string (as it is for production releases) called a macro with an empty
1925        argument. The C standard says the result of this is undefined. The gcc
1926        compiler treats it as an empty string (which was what was wanted) but it is
1927        reported that Visual C gives an error. The source has been hacked around to
1928        avoid this problem.
1930     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1931        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1932        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1933        of them did).
1935    10. Originally, pcretest opened its input and output without "b"; then I was
1936        told that "b" was needed in some environments, so it was added for release
1937        5.0 to both the input and output. (It makes no difference on Unix-like
1938        systems.) Later I was told that it is wrong for the input on Windows. I've
1939        now abstracted the modes into two macros, to make it easier to fiddle with
1940        them, and removed "b" from the input mode under Windows.
1942    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1944    12. Added -help and --help to pcretest as an official way of being reminded
1945        of the options.
1947    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1948        and pcrecpp.cc because they annoy compilers at high warning levels.
1950    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1952    15. Fixed an occurrence of == in configure.ac that should have been = (shell
1953        scripts are not C programs :-) and which was not noticed because it works
1954        on Linux.
1956    16. pcretest is supposed to handle any length of pattern and data line (as one
1957        line or as a continued sequence of lines) by extending its input buffer if
1958        necessary. This feature was broken for very long pattern lines, leading to
1959        a string of junk being passed to pcre_compile() if the pattern was longer
1960        than about 50K.
1962    17. I have done a major re-factoring of the way pcre_compile() computes the
1963        amount of memory needed for a compiled pattern. Previously, there was code
1964        that made a preliminary scan of the pattern in order to do this. That was
1965        OK when PCRE was new, but as the facilities have expanded, it has become
1966        harder and harder to keep it in step with the real compile phase, and there
1967        have been a number of bugs (see for example, 4 above). I have now found a
1968        cunning way of running the real compile function in a "fake" mode that
1969        enables it to compute how much memory it would need, while actually only
1970        ever using a few hundred bytes of working memory and without too many
1971        tests of the mode. This should make future maintenance and development
1972        easier. A side effect of this work is that the limit of 200 on the nesting
1973        depth of parentheses has been removed (though this was never a serious
1974        limitation, I suspect). However, there is a downside: pcre_compile() now
1975        runs more slowly than before (30% or more, depending on the pattern). I
1976        hope this isn't a big issue. There is no effect on runtime performance.
1978    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1979        newline (only possible for the last line of a file) and it was a
1980        pattern that set a locale (followed by /Lsomething), pcretest crashed.
1982    19. Added additional timing features to pcretest. (1) The -tm option now times
1983        matching only, not compiling. (2) Both -t and -tm can be followed, as a
1984        separate command line item, by a number that specifies the number of
1985        repeats to use when timing. The default is 50000; this gives better
1986        precision, but takes uncomfortably long for very large patterns.
1988    20. Extended pcre_study() to be more clever in cases where a branch of a
1989        subpattern has no definite first character. For example, (a*|b*)[cd] would
1990        previously give no result from pcre_study(). Now it recognizes that the
1991        first character must be a, b, c, or d.
1993    21. There was an incorrect error "recursive call could loop indefinitely" if
1994        a subpattern (or the entire pattern) that was being tested for matching an
1995        empty string contained only one non-empty item after a nested subpattern.
1996        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1997        incorrectly, because the \d was being skipped in the check.
1999    22. The pcretest program now has a new pattern option /B and a command line
2000        option -b, which is equivalent to adding /B to every pattern. This causes
2001        it to show the compiled bytecode, without the additional information that
2002        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
2003        is the same as /B/I).
2005    23. A new optimization is now able automatically to treat some sequences such
2006        as a*b as a*+b. More specifically, if something simple (such as a character
2007        or a simple class like \d) has an unlimited quantifier, and is followed by
2008        something that cannot possibly match the quantified thing, the quantifier
2009        is automatically "possessified".
2011    24. A recursive reference to a subpattern whose number was greater than 39
2012        went wrong under certain circumstances in UTF-8 mode. This bug could also
2013        have affected the operation of pcre_study().
2015    25. Realized that a little bit of performance could be had by replacing
2016        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
2018    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
2020    27. Possessive quantifiers such as a++ were previously implemented by turning
2021        them into atomic groups such as ($>a+). Now they have their own opcodes,
2022        which improves performance. This includes the automatically created ones
2023        from 23 above.
2025    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
2026        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
2027        the first matched character to be a colon. This applied both to named and
2028        numbered groups.
2030    29. The ucpinternal.h header file was missing its idempotency #ifdef.
2032    30. I was sent a "project" file called libpcre.a.dev which I understand makes
2033        building PCRE on Windows easier, so I have included it in the distribution.
2035    31. There is now a check in pcretest against a ridiculously large number being
2036        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
2037        loop, the loop is abandoned.
2039    32. Forward references to subpatterns in conditions such as (?(2)...) where
2040        subpattern 2 is defined later cause pcre_compile() to search forwards in
2041        the pattern for the relevant set of parentheses. This search went wrong
2042        when there were unescaped parentheses in a character class, parentheses
2043        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
2045    33. "Subroutine" calls and backreferences were previously restricted to
2046        referencing subpatterns earlier in the regex. This restriction has now
2047        been removed.
2049    34. Added a number of extra features that are going to be in Perl 5.10. On the
2050        whole, these are just syntactic alternatives for features that PCRE had
2051        previously implemented using the Python syntax or my own invention. The
2052        other formats are all retained for compatibility.
2054        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
2055            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
2056            also .NET compatible.
2058        (b) A recursion or subroutine call to a named group can now be defined as
2059            (?&name) as well as (?P>name).
2061        (c) A backreference to a named group can now be defined as \k<name> or
2062            \k'name' as well as (?P=name). The new forms, as well as being in Perl
2063            5.10, are also .NET compatible.
2065        (d) A conditional reference to a named group can now use the syntax
2066            (?(<name>) or (?('name') as well as (?(name).
2068        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
2069            groups (named and numbered) that are never evaluated inline, but can be
2070            called as "subroutines" from elsewhere. In effect, the DEFINE condition
2071            is always false. There may be only one alternative in such a group.
2073        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
2074            as the simple (?(R). The condition is true only if the most recent
2075            recursion is that of the given number or name. It does not search out
2076            through the entire recursion stack.
2078        (g) The escape \gN or \g{N} has been added, where N is a positive or
2079            negative number, specifying an absolute or relative reference.
2081    35. Tidied to get rid of some further signed/unsigned compiler warnings and
2082        some "unreachable code" warnings.
2084    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
2085        things, this adds five new scripts.
2087    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
2088        There were also incompatibilities regarding the handling of \Q..\E inside
2089        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
2090        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
2092    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
2093        matches an empty string, and forcibly breaks the loop. There were bugs in
2094        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
2095        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
2096        separate and independent bugs (that affected different cases) have been
2097        fixed.
2099    39. Refactored the code to abolish the use of different opcodes for small
2100        capturing bracket numbers. This is a tidy that I avoided doing when I
2101        removed the limit on the number of capturing brackets for 3.5 back in 2001.
2102        The new approach is not only tidier, it makes it possible to reduce the
2103        memory needed to fix the previous bug (38).
2105    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
2106        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
2107        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
2108        mode.
2110    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
2111        report.
2113    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
2114        copy construction and assignment in the C++ wrapper.
2116    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
2117        couple of bugs that could have given wrong results in the "--newline=crlf"
2118        case.
2120    44. Added a number of casts and did some reorganization of signed/unsigned int
2121        variables following suggestions from Dair Grant. Also renamed the variable
2122        "this" as "item" because it is a C++ keyword.
2124    45. Arranged for dftables to add
2126          #include "pcre_internal.h"
2128        to pcre_chartables.c because without it, gcc 4.x may remove the array
2129        definition from the final binary if PCRE is built into a static library and
2130        dead code stripping is activated.
2132    46. For an unanchored pattern, if a match attempt fails at the start of a
2133        newline sequence, and the newline setting is CRLF or ANY, and the next two
2134        characters are CRLF, advance by two characters instead of one.
2137    Version 6.7 04-Jul-06
2138    ---------------------
2140     1. In order to handle tests when input lines are enormously long, pcretest has
2141        been re-factored so that it automatically extends its buffers when
2142        necessary. The code is crude, but this _is_ just a test program. The
2143        default size has been increased from 32K to 50K.
2145     2. The code in pcre_study() was using the value of the re argument before
2146        testing it for NULL. (Of course, in any sensible call of the function, it
2147        won't be NULL.)
2149     3. The memmove() emulation function in pcre_internal.h, which is used on
2150        systems that lack both memmove() and bcopy() - that is, hardly ever -
2151        was missing a "static" storage class specifier.
2153     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
2154        containing an extended class (one that cannot be represented by a bitmap
2155        because it contains high-valued characters or Unicode property items, e.g.
2156        [\pZ]). Almost always one would set UTF-8 mode when processing such a
2157        pattern, but PCRE should not loop if you do not (it no longer does).
2158        [Detail: two cases were found: (a) a repeated subpattern containing an
2159        extended class; (b) a recursive reference to a subpattern that followed a
2160        previous extended class. It wasn't skipping over the extended class
2161        correctly when UTF-8 mode was not set.]
2163     5. A negated single-character class was not being recognized as fixed-length
2164        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
2165        compile error "lookbehind assertion is not fixed length".
2167     6. The RunPerlTest auxiliary script was showing an unexpected difference
2168        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
2169        write a Perl script that can interpret lines of an input file either as
2170        byte characters or as UTF-8, which is what "perltest" was being required to
2171        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
2172        can't do is switch easily at run time between having the "use utf8;" pragma
2173        or not. In the end, I fudged it by using the RunPerlTest script to insert
2174        "use utf8;" explicitly for the UTF-8 tests.
2176     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
2177        the end of the subject string, contrary to the documentation and to what
2178        Perl does. This was true of both matching functions. Now it matches only at
2179        the start of the subject and immediately after *internal* newlines.
2181     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
2182        a pointer to an int instead of a pointer to an unsigned long int. This
2183        caused problems on 64-bit systems.
2185     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
2186        instance of the 'standard' template library not being so standard".
2188    10. There was no check on the number of named subpatterns nor the maximum
2189        length of a subpattern name. The product of these values is used to compute
2190        the size of the memory block for a compiled pattern. By supplying a very
2191        long subpattern name and a large number of named subpatterns, the size
2192        computation could be caused to overflow. This is now prevented by limiting
2193        the length of names to 32 characters, and the number of named subpatterns
2194        to 10,000.
2196    11. Subpatterns that are repeated with specific counts have to be replicated in
2197        the compiled pattern. The size of memory for this was computed from the
2198        length of the subpattern and the repeat count. The latter is limited to
2199        65535, but there was no limit on the former, meaning that integer overflow
2200        could in principle occur. The compiled length of a repeated subpattern is
2201        now limited to 30,000 bytes in order to prevent this.
2203    12. Added the optional facility to have named substrings with the same name.
2205    13. Added the ability to use a named substring as a condition, using the
2206        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
2207        are numbers (not recommended). Forward references are permitted.
2209    14. Added forward references in named backreferences (if you see what I mean).
2211    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
2212        pattern could run off the end of the subject. For example, the pattern
2213        "(?s)(.{1,5})"8 did this with the subject "ab".
2215    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
2216        PCRE_CASELESS was set when matching characters that were quantified with ?
2217        or *.
2219    17. A character class other than a single negated character that had a minimum
2220        but no maximum quantifier - for example [ab]{6,} - was not handled
2221        correctly by pce_dfa_exec(). It would match only one character.
2223    18. A valid (though odd) pattern that looked like a POSIX character
2224        class but used an invalid character after [ (for example [[,abc,]]) caused
2225        pcre_compile() to give the error "Failed: internal error: code overflow" or
2226        in some cases to crash with a glibc free() error. This could even happen if
2227        the pattern terminated after [[ but there just happened to be a sequence of
2228        letters, a binary zero, and a closing ] in the memory that followed.
2230    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
2231        over the years. Originally (before any Unicode support), just the bottom 8
2232        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
2233        output from "man perlunicode" includes this:
2235          The regular expression compiler produces polymorphic opcodes.  That
2236          is, the pattern adapts to the data and automatically switches to
2237          the Unicode character scheme when presented with Unicode data--or
2238          instead uses a traditional byte scheme when presented with byte
2239          data.
2241        Sadly, a wide octal escape does not cause a switch, and in a string with
2242        no other multibyte characters, these octal escapes are treated as before.
2243        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
2244        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
2245        Unicode string.
2247        I have not perpetrated such confusion in PCRE. Up till now, it took just
2248        the bottom 8 bits, as in old Perl. I have now made octal escapes with
2249        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
2250        translate to the appropriate multibyte character.
2252    29. Applied some refactoring to reduce the number of warnings from Microsoft
2253        and Borland compilers. This has included removing the fudge introduced
2254        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
2255        a warning about an unused variable.
2257    21. PCRE has not included VT (character 0x0b) in the set of whitespace
2258        characters since release 4.0, because Perl (from release 5.004) does not.
2259        [Or at least, is documented not to: some releases seem to be in conflict
2260        with the documentation.] However, when a pattern was studied with
2261        pcre_study() and all its branches started with \s, PCRE still included VT
2262        as a possible starting character. Of course, this did no harm; it just
2263        caused an unnecessary match attempt.
2265    22. Removed a now-redundant internal flag bit that recorded the fact that case
2266        dependency changed within the pattern. This was once needed for "required
2267        byte" processing, but is no longer used. This recovers a now-scarce options
2268        bit. Also moved the least significant internal flag bit to the most-
2269        significant bit of the word, which was not previously used (hangover from
2270        the days when it was an int rather than a uint) to free up another bit for
2271        the future.
2273    23. Added support for CRLF line endings as well as CR and LF. As well as the
2274        default being selectable at build time, it can now be changed at runtime
2275        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
2276        specify that it is scanning data with non-default line endings.
2278    24. Changed the definition of CXXLINK to make it agree with the definition of
2279        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
2281    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
2282        recursions. This makes a big different to stack usage for some patterns.
2284    26. If a subpattern containing a named recursion or subroutine reference such
2285        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
2286        the space required for the compiled pattern went wrong and gave too small a
2287        value. Depending on the environment, this could lead to "Failed: internal
2288        error: code overflow at offset 49" or "glibc detected double free or
2289        corruption" errors.
2291    27. Applied patches from Google (a) to support the new newline modes and (b) to
2292        advance over multibyte UTF-8 characters in GlobalReplace.
2294    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
2295        difference for some implementation of PCRE in some Windows version.
2297    29. Added some extra testing facilities to pcretest:
2299        \q<number>   in a data line sets the "match limit" value
2300        \Q<number>   in a data line sets the "match recursion limt" value
2301        -S <number>  sets the stack size, where <number> is in megabytes
2303        The -S option isn't available for Windows.
2306    Version 6.6 06-Feb-06
2307    ---------------------
2309     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
2310        in pcreposix.h. I have copied the definition from pcre.h.
2312     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
2313        because pcre.h is no longer a built file.
2315     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
2316        not normally included in the compiled code.
2319    Version 6.5 01-Feb-06
2320    ---------------------
2322     1. When using the partial match feature with pcre_dfa_exec(), it was not
2323        anchoring the second and subsequent partial matches at the new starting
2324        point. This could lead to incorrect results. For example, with the pattern
2325        /1234/, partially matching against "123" and then "a4" gave a match.
2327     2. Changes to pcregrep:
2329        (a) All non-match returns from pcre_exec() were being treated as failures
2330            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
2331            error message is output. Some extra information is given for the
2332            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
2333            probably the only errors that are likely to be caused by users (by
2334            specifying a regex that has nested indefinite repeats, for instance).
2335            If there are more than 20 of these errors, pcregrep is abandoned.
2337        (b) A binary zero was treated as data while matching, but terminated the
2338            output line if it was written out. This has been fixed: binary zeroes
2339            are now no different to any other data bytes.
2341        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
2342            used to set a locale for matching. The --locale=xxxx long option has
2343            been added (no short equivalent) to specify a locale explicitly on the
2344            pcregrep command, overriding the environment variables.
2346        (d) When -B was used with -n, some line numbers in the output were one less
2347            than they should have been.
2349        (e) Added the -o (--only-matching) option.
2351        (f) If -A or -C was used with -c (count only), some lines of context were
2352            accidentally printed for the final match.
2354        (g) Added the -H (--with-filename) option.
2356        (h) The combination of options -rh failed to suppress file names for files
2357            that were found from directory arguments.
2359        (i) Added the -D (--devices) and -d (--directories) options.
2361        (j) Added the -F (--fixed-strings) option.
2363        (k) Allow "-" to be used as a file name for -f as well as for a data file.
2365        (l) Added the --colo(u)r option.
2367        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
2368            is not present by default.
2370     3. A nasty bug was discovered in the handling of recursive patterns, that is,
2371        items such as (?R) or (?1), when the recursion could match a number of
2372        alternatives. If it matched one of the alternatives, but subsequently,
2373        outside the recursion, there was a failure, the code tried to back up into
2374        the recursion. However, because of the way PCRE is implemented, this is not
2375        possible, and the result was an incorrect result from the match.
2377        In order to prevent this happening, the specification of recursion has
2378        been changed so that all such subpatterns are automatically treated as
2379        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
2381     4. I had overlooked the fact that, in some locales, there are characters for
2382        which isalpha() is true but neither isupper() nor islower() are true. In
2383        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
2384        and ordfeminine) are like this. This affected the treatment of \w and \W
2385        when they appeared in character classes, but not when they appeared outside
2386        a character class. The bit map for "word" characters is now created
2387        separately from the results of isalnum() instead of just taking it from the
2388        upper, lower, and digit maps. (Plus the underscore character, of course.)
2390     5. The above bug also affected the handling of POSIX character classes such as
2391        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
2392        permanent tables. Instead, the bit maps for such a class were previously
2393        created as the appropriate unions of the upper, lower, and digit bitmaps.
2394        Now they are created by subtraction from the [[:word:]] class, which has
2395        its own bitmap.
2397     6. The [[:blank:]] character class matches horizontal, but not vertical space.
2398        It is created by subtracting the vertical space characters (\x09, \x0a,
2399        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
2400        subtraction was done in the overall bitmap for a character class, meaning
2401        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
2402        be recognized. This bug has been fixed.
2404     7. Patches from the folks at Google:
2406          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
2407          real life, but is still worth protecting against".
2409          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
2410          regular expressions".
2412          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
2413          have it.
2415          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
2416          "configure" and the latter not, in order to fix a problem somebody had
2417          with compiling the Arg class on HP-UX.
2419          (e) Improve the error-handling of the C++ wrapper a little bit.
2421          (f) New tests for checking recursion limiting.
2423     8. The pcre_memmove() function, which is used only if the environment does not
2424        have a standard memmove() function (and is therefore rarely compiled),
2425        contained two bugs: (a) use of int instead of size_t, and (b) it was not
2426        returning a result (though PCRE never actually uses the result).
2428     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
2429        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
2430        returned instead of calling malloc() with an overflowing number that would
2431        most likely cause subsequent chaos.
2433    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
2435    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
2436        with this option is matched, the nmatch and pmatch options of regexec() are
2437        ignored.
2439    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
2440        provided in case anyone wants to the the POSIX interface with UTF-8
2441        strings.
2443    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
2444        C++ linking (needed for some HP-UX environments).
2446    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
2447        (unused parameter) and in the pcre_printint() function (omitted "default"
2448        switch label when the default is to do nothing).
2450    15. Added some code to make it possible, when PCRE is compiled as a C++
2451        library, to replace subject pointers for pcre_exec() with a smart pointer
2452        class, thus making it possible to process discontinuous strings.
2454    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
2455        much the same function. They were added by different people who were trying
2456        to make PCRE easy to compile on non-Unix systems. It has been suggested
2457        that PCRE_EXPORT be abolished now that there is more automatic apparatus
2458        for compiling on Windows systems. I have therefore replaced it with
2459        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
2460        defaults to "extern" for C or "extern C" for C++, which works fine on
2461        Unix-like systems. It is now possible to override the value of PCRE_DATA_
2462        SCOPE with something explicit in config.h. In addition:
2464        (a) pcreposix.h still had just "extern" instead of either of these macros;
2465            I have replaced it with PCRE_DATA_SCOPE.
2467        (b) Functions such as _pcre_xclass(), which are internal to the library,
2468            but external in the C sense, all had PCRE_EXPORT in their definitions.
2469            This is apparently wrong for the Windows case, so I have removed it.
2470            (It makes no difference on Unix-like systems.)
2472    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
2473        of recursive calls to match(). This is different to MATCH_LIMIT because
2474        that limits the total number of calls to match(), not all of which increase
2475        the depth of recursion. Limiting the recursion depth limits the amount of
2476        stack (or heap if NO_RECURSE is set) that is used. The default can be set
2477        when PCRE is compiled, and changed at run time. A patch from Google adds
2478        this functionality to the C++ interface.
2480    18. Changes to the handling of Unicode character properties:
2482        (a) Updated the table to Unicode 4.1.0.
2484        (b) Recognize characters that are not in the table as "Cn" (undefined).
2486        (c) I revised the way the table is implemented to a much improved format
2487            which includes recognition of ranges. It now supports the ranges that
2488            are defined in UnicodeData.txt, and it also amalgamates other
2489            characters into ranges. This has reduced the number of entries in the
2490            table from around 16,000 to around 3,000, thus reducing its size
2491            considerably. I realized I did not need to use a tree structure after
2492            all - a binary chop search is just as efficient. Having reduced the
2493            number of entries, I extended their size from 6 bytes to 8 bytes to
2494            allow for more data.
2496        (d) Added support for Unicode script names via properties such as \p{Han}.
2498    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
2499        matching that character.
2501    20. When matching a repeated Unicode property with a minimum greater than zero,
2502        (for example \pL{2,}), PCRE could look past the end of the subject if it
2503        reached it while seeking the minimum number of characters. This could
2504        happen only if some of the characters were more than one byte long, because
2505        there is a check for at least the minimum number of bytes.
2507    21. Refactored the implementation of \p and \P so as to be more general, to
2508        allow for more different types of property in future. This has changed the
2509        compiled form incompatibly. Anybody with saved compiled patterns that use
2510        \p or \P will have to recompile them.
2512    22. Added "Any" and "L&" to the supported property types.
2514    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
2515        but give a compile time error if the value is greater than 0xff.
2517    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
2518        accidentally not being installed or uninstalled.
2520    25. The pcre.h file was built from pcre.h.in, but the only changes that were
2521        made were to insert the current release number. This seemed silly, because
2522        it made things harder for people building PCRE on systems that don't run
2523        "configure". I have turned pcre.h into a distributed file, no longer built
2524        by "configure", with the version identification directly included. There is
2525        no longer a pcre.h.in file.
2527        However, this change necessitated a change to the pcre-config script as
2528        well. It is built from pcre-config.in, and one of the substitutions was the
2529        release number. I have updated configure.ac so that ./configure now finds
2530        the release number by grepping pcre.h.
2532    26. Added the ability to run the tests under valgrind.
2535    Version 6.4 05-Sep-05
2536    ---------------------
2538     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
2539        "--" to be printed when multiple files were scanned, even when none of the
2540        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
2541        consider it to be a bug, and have restored the previous behaviour.
2543     2. A couple of code tidies to get rid of compiler warnings.
2545     3. The pcretest program used to cheat by referring to symbols in the library
2546        whose names begin with _pcre_. These are internal symbols that are not
2547        really supposed to be visible externally, and in some environments it is
2548        possible to suppress them. The cheating is now confined to including
2549        certain files from the library's source, which is a bit cleaner.
2551     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
2552        file's purpose clearer.
2554     5. Reorganized pcre_ucp_findchar().
2557    Version 6.3 15-Aug-05
2558    ---------------------
2560     1. The file libpcre.pc.in did not have general read permission in the tarball.
2562     2. There were some problems when building without C++ support:
2564        (a) If C++ support was not built, "make install" and "make test" still
2565            tried to test it.
2567        (b) There were problems when the value of CXX was explicitly set. Some
2568            changes have been made to try to fix these, and ...
2570        (c) --disable-cpp can now be used to explicitly disable C++ support.
2572        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
2573            backslash in a target when C++ was disabled. This confuses some
2574            versions of "make", apparently. Using an intermediate variable solves
2575            this. (Same for CPP_LOBJ.)
2577     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
2578        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
2579        necessary on certain architectures.
2581     4. Added a setting of -export-symbols-regex to the link command to remove
2582        those symbols that are exported in the C sense, but actually are local
2583        within the library, and not documented. Their names all begin with
2584        "_pcre_". This is not a perfect job, because (a) we have to except some
2585        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
2586        available (and never for static libraries). I have made a note to try to
2587        find a way round (a) in the future.
2590    Version 6.2 01-Aug-05
2591    ---------------------
2593     1. There was no test for integer overflow of quantifier values. A construction
2594        such as {1111111111111111} would give undefined results. What is worse, if
2595        a minimum quantifier for a parenthesized subpattern overflowed and became
2596        negative, the calculation of the memory size went wrong. This could have
2597        led to memory overwriting.
2599     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
2601     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
2602        operating environments where this matters.
2604     4. Applied Giuseppe Maxia's patch to add additional features for controlling
2605        PCRE options from within the C++ wrapper.
2607     5. Named capturing subpatterns were not being correctly counted when a pattern
2608        was compiled. This caused two problems: (a) If there were more than 100
2609        such subpatterns, the calculation of the memory needed for the whole
2610        compiled pattern went wrong, leading to an overflow error. (b) Numerical
2611        back references of the form \12, where the number was greater than 9, were
2612        not recognized as back references, even though there were sufficient
2613        previous subpatterns.
2615     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
2616        versions of gcc, e.g. 2.95.4.
2619    Version 6.1 21-Jun-05
2620    ---------------------
2622     1. There was one reference to the variable "posix" in pcretest.c that was not
2623        surrounded by "#if !defined NOPOSIX".
2625     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
2626        the cross-check on the old pcre_info() function, for the benefit of the
2627        cut-down version of PCRE that is currently imported into Exim.
2629     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
2630        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
2631        patterns that start (?i) but I don't think that matters. The use of (?i) is
2632        just an example; this all applies to the other options as well.
2634     4. Since libtool seems to echo the compile commands it is issuing, the output
2635        from "make" can be reduced a bit by putting "@" in front of each libtool
2636        compile command.
2638     5. Patch from the folks at Google for configure.in to be a bit more thorough
2639        in checking for a suitable C++ installation before trying to compile the
2640        C++ stuff. This should fix a reported problem when a compiler was present,
2641        but no suitable headers.
2643     6. The man pages all had just "PCRE" as their title. I have changed them to
2644        be the relevant file name. I have also arranged that these names are
2645        retained in the file doc/pcre.txt, which is a concatenation in text format
2646        of all the man pages except the little individual ones for each function.
2648     7. The NON-UNIX-USE file had not been updated for the different set of source
2649        files that come with release 6. I also added a few comments about the C++
2650        wrapper.
2653    Version 6.0 07-Jun-05
2654    ---------------------
2656     1. Some minor internal re-organization to help with my DFA experiments.
2658     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
2659        didn't matter for the library itself when fully configured, but did matter
2660        when compiling without UCP support, or within Exim, where the ucp files are
2661        not imported.
2663     3. Refactoring of the library code to split up the various functions into
2664        different source modules. The addition of the new DFA matching code (see
2665        below) to a single monolithic source would have made it really too
2666        unwieldy, quite apart from causing all the code to be include in a
2667        statically linked application, when only some functions are used. This is
2668        relevant even without the DFA addition now that patterns can be compiled in
2669        one application and matched in another.
2671        The downside of splitting up is that there have to be some external
2672        functions and data tables that are used internally in different modules of
2673        the library but which are not part of the API. These have all had their
2674        names changed to start with "_pcre_" so that they are unlikely to clash
2675        with other external names.
2677     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
2678        a different (DFA) algorithm. Although it is slower than the original
2679        function, it does have some advantages for certain types of matching
2680        problem.
2682     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
2683        including restarting after a partial match.
2685     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
2686        defined when compiling for Windows was sent to me. I have put it into the
2687        code, though I have no means of testing or verifying it.
2689     7. Added the pcre_refcount() auxiliary function.
2691     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
2692        match before or at the first newline in the subject string. In pcretest,
2693        the /f option on a pattern can be used to set this.
2695     9. A repeated \w when used in UTF-8 mode with characters greater than 256
2696        would behave wrongly. This has been present in PCRE since release 4.0.
2698    10. A number of changes to the pcregrep command:
2700        (a) Refactored how -x works; insert ^(...)$ instead of setting
2701            PCRE_ANCHORED and checking the length, in preparation for adding
2702            something similar for -w.
2704        (b) Added the -w (match as a word) option.
2706        (c) Refactored the way lines are read and buffered so as to have more
2707            than one at a time available.
2709        (d) Implemented a pcregrep test script.
2711        (e) Added the -M (multiline match) option. This allows patterns to match
2712            over several lines of the subject. The buffering ensures that at least
2713            8K, or the rest of the document (whichever is the shorter) is available
2714            for matching (and similarly the previous 8K for lookbehind assertions).
2716        (f) Changed the --help output so that it now says
2718              -w, --word-regex(p)
2720            instead of two lines, one with "regex" and the other with "regexp"
2721            because that confused at least one person since the short forms are the
2722            same. (This required a bit of code, as the output is generated
2723            automatically from a table. It wasn't just a text change.)
2725        (g) -- can be used to terminate pcregrep options if the next thing isn't an
2726            option but starts with a hyphen. Could be a pattern or a path name
2727            starting with a hyphen, for instance.
2729        (h) "-" can be given as a file name to represent stdin.
2731        (i) When file names are being printed, "(standard input)" is used for
2732            the standard input, for compatibility with GNU grep. Previously
2733            "<stdin>" was used.
2735        (j) The option --label=xxx can be used to supply a name to be used for
2736            stdin when file names are being printed. There is no short form.
2738        (k) Re-factored the options decoding logic because we are going to add
2739            two more options that take data. Such options can now be given in four
2740            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
2742        (l) Added the -A, -B, and -C options for requesting that lines of context
2743            around matches be printed.
2745        (m) Added the -L option to print the names of files that do not contain
2746            any matching lines, that is, the complement of -l.
2748        (n) The return code is 2 if any file cannot be opened, but pcregrep does
2749            continue to scan other files.
2751        (o) The -s option was incorrectly implemented. For compatibility with other
2752            greps, it now suppresses the error message for a non-existent or non-
2753            accessible file (but not the return code). There is a new option called
2754            -q that suppresses the output of matching lines, which was what -s was
2755            previously doing.
2757        (p) Added --include and --exclude options to specify files for inclusion
2758            and exclusion when recursing.
2760    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
2761        Hopefully, it now does.
2763    12. Missing cast in pcre_study().
2765    13. Added an "uninstall" target to the makefile.
2767    14. Replaced "extern" in the function prototypes in Makefile.in with
2768        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
2769        world, but is set differently for Windows.
2771    15. Added a second compiling function called pcre_compile2(). The only
2772        difference is that it has an extra argument, which is a pointer to an
2773        integer error code. When there is a compile-time failure, this is set
2774        non-zero, in addition to the error test pointer being set to point to an
2775        error message. The new argument may be NULL if no error number is required
2776        (but then you may as well call pcre_compile(), which is now just a
2777        wrapper). This facility is provided because some applications need a
2778        numeric error indication, but it has also enabled me to tidy up the way
2779        compile-time errors are handled in the POSIX wrapper.
2781    16. Added VPATH=.libs to the makefile; this should help when building with one
2782        prefix path and installing with another. (Or so I'm told by someone who
2783        knows more about this stuff than I do.)
2785    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
2786        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
2787        match everything, including newlines. This is not POSIX-compatible, but
2788        somebody wanted the feature. From pcretest it can be activated by using
2789        both the P and the s flags.
2791    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
2793    19. libpcre.pc was being incorrectly installed as executable.
2795    20. A couple of places in pcretest check for end-of-line by looking for '\n';
2796        it now also looks for '\r' so that it will work unmodified on Windows.
2798    21. Added Google's contributed C++ wrapper to the distribution.
2800    22. Added some untidy missing memory free() calls in pcretest, to keep
2801        Electric Fence happy when testing.
2805    Version 5.0 13-Sep-04
2806    ---------------------
2808     1. Internal change: literal characters are no longer packed up into items
2809        containing multiple characters in a single byte-string. Each character
2810        is now matched using a separate opcode. However, there may be more than one
2811        byte in the character in UTF-8 mode.
2813     2. The pcre_callout_block structure has two new fields: pattern_position and
2814        next_item_length. These contain the offset in the pattern to the next match
2815        item, and its length, respectively.
2817     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
2818        insertion of callouts before each pattern item. Added the /C option to
2819        pcretest to make use of this.
2821     4. On the advice of a Windows user, the lines
2823          #if defined(_WIN32) || defined(WIN32)
2824          _setmode( _fileno( stdout ), 0x8000 );
2825          #endif  /* defined(_WIN32) || defined(WIN32) */
2827        have been added to the source of pcretest. This apparently does useful
2828        magic in relation to line terminators.
2830     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
2831        for the benefit of those environments where the "b" makes a difference.
2833     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
2834        to know about it. I have put a hack into configure.in that adds in code
2835        to set GCC=yes if CC=icc. This seems to end up at a point in the
2836        generated configure script that is early enough to affect the setting of
2837        compiler options, which is what is needed, but I have no means of testing
2838        whether it really works. (The user who reported this had patched the
2839        generated configure script, which of course I cannot do.)
2841        LATER: After change 22 below (new libtool files), the configure script
2842        seems to know about icc (and also ecc). Therefore, I have commented out
2843        this hack in configure.in.
2845     7. Added support for pkg-config (2 patches were sent in).
2847     8. Negated POSIX character classes that used a combination of internal tables
2848        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2849        [[:^ascii]]. Typically, they would match almost any characters. The other
2850        POSIX classes were not broken in this way.
2852     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2853        to find the match, as PCRE was deluded into thinking that the match had to
2854        start at the start point or following a newline. The same bug applied to
2855        patterns with negative forward assertions or any backward assertions
2856        preceding ".*" at the start, unless the pattern required a fixed first
2857        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2859    10. In UTF-8 mode, when moving forwards in the subject after a failed match
2860        starting at the last subject character, bytes beyond the end of the subject
2861        string were read.
2863    11. Renamed the variable "class" as "classbits" to make life easier for C++
2864        users. (Previously there was a macro definition, but it apparently wasn't
2865        enough.)
2867    12. Added the new field "tables" to the extra data so that tables can be passed
2868        in at exec time, or the internal tables can be re-selected. This allows
2869        a compiled regex to be saved and re-used at a later time by a different
2870        program that might have everything at different addresses.
2872    13. Modified the pcre-config script so that, when run on Solaris, it shows a
2873        -R library as well as a -L library.
2875    14. The debugging options of pcretest (-d on the command line or D on a
2876        pattern) showed incorrect output for anything following an extended class
2877        that contained multibyte characters and which was followed by a quantifier.
2879    15. Added optional support for general category Unicode character properties
2880        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2881        support. It adds about 90K to the size of the library. The meanings of the
2882        inbuilt class escapes such as \d and \s have NOT been changed.
2884    16. Updated pcredemo.c to include calls to free() to release the memory for the
2885        compiled pattern.
2887    17. The generated file chartables.c was being created in the source directory
2888        instead of in the building directory. This caused the build to fail if the
2889        source directory was different from the building directory, and was
2890        read-only.
2892    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2893        file. No doubt somebody will tell me if they don't make sense... Also added
2894        Dan Mooney's comments about building on OpenVMS.
2896    19. Added support for partial matching via the PCRE_PARTIAL option for
2897        pcre_exec() and the \P data escape in pcretest.
2899    20. Extended pcretest with 3 new pattern features:
2901        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
2902              write the compiled pattern to the file whose name is "rest-of-line".
2903              This is a straight binary dump of the data, with the saved pointer to
2904              the character tables forced to be NULL. The study data, if any, is
2905              written too. After writing, pcretest reads a new pattern.
2907        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2908              compiled pattern from the given file. There must not be any
2909              occurrences of "<" in the file name (pretty unlikely); if there are,
2910              pcretest will instead treat the initial "<" as a pattern delimiter.
2911              After reading in the pattern, pcretest goes on to read data lines as
2912              usual.
2914        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2915              and 16-bit fields in a compiled pattern, to simulate a pattern that
2916              was compiled on a host of opposite endianness.
2918    21. The pcre-exec() function can now cope with patterns that were compiled on
2919        hosts of opposite endianness, with this restriction:
2921          As for any compiled expression that is saved and used later, the tables
2922          pointer field cannot be preserved; the extra_data field in the arguments
2923          to pcre_exec() should be used to pass in a tables address if a value
2924          other than the default internal tables were used at compile time.
2926    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2927        now diagnosed as an error. Previously, most of the time, a negative number
2928        would have been treated as zero, but if in addition "ovector" was passed as
2929        NULL, a crash could occur.
2931    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2932        new versions from the libtool 1.5 distribution (the last one is a copy of
2933        a file called libtool.m4). This seems to have fixed the need to patch
2934        "configure" to support Darwin 1.3 (which I used to do). However, I still
2935        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2936        workstation).
2938    24. Changed the PCRE licence to be the more standard "BSD" licence.
2941    Version 4.5 01-Dec-03
2942    ---------------------
2944     1. There has been some re-arrangement of the code for the match() function so
2945        that it can be compiled in a version that does not call itself recursively.
2946        Instead, it keeps those local variables that need separate instances for
2947        each "recursion" in a frame on the heap, and gets/frees frames whenever it
2948        needs to "recurse". Keeping track of where control must go is done by means
2949        of setjmp/longjmp. The whole thing is implemented by a set of macros that
2950        hide most of the details from the main code, and operates only if
2951        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2952        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2953        operating.
2955        To make it easier for callers to provide specially tailored get/free
2956        functions for this usage, two new functions, pcre_stack_malloc, and
2957        pcre_stack_free, are used. They are always called in strict stacking order,
2958        and the size of block requested is always the same.
2960        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2961        PCRE has been compiled to use the stack or the heap for recursion. The
2962        -C option of pcretest uses this to show which version is compiled.
2964        A new data escape \S, is added to pcretest; it causes the amounts of store
2965        obtained and freed by both kinds of malloc/free at match time to be added
2966        to the output.
2968     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2969        what's available on my current Linux desktop machine.
2971     3. When matching a UTF-8 string, the test for a valid string at the start has
2972        been extended. If start_offset is not zero, PCRE now checks that it points
2973        to a byte that is the start of a UTF-8 character. If not, it returns
2974        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2975        this is necessary because there may be backward assertions in the pattern.
2976        When matching the same subject several times, it may save resources to use
2977        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2979     4. The code for checking the validity of UTF-8 strings has been tightened so
2980        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2981        containing "overlong sequences".
2983     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2984        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2985        should have been "&", but it just so happened that all the cases this let
2986        through by mistake were picked up later in the function.
2988     6. I had used a variable called "isblank" - this is a C99 function, causing
2989        some compilers to warn. To avoid this, I renamed it (as "blankclass").
2991     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2992        prompting; (b) run "./pcretest /dev/null" at the start of the test script
2993        so the version is shown; (c) stop "make test" echoing "./RunTest".
2995     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2997     9. The prototype for memmove() for systems that don't have it was using
2998        size_t, but the inclusion of the header that defines size_t was later. I've
2999        moved the #includes for the C headers earlier to avoid this.
3001    10. Added some adjustments to the code to make it easier to compiler on certain
3002        special systems:
3004          (a) Some "const" qualifiers were missing.
3005          (b) Added the macro EXPORT before all exported functions; by default this
3006              is defined to be empty.
3007          (c) Changed the dftables auxiliary program (that builds chartables.c) so
3008              that it reads its output file name as an argument instead of writing
3009              to the standard output and assuming this can be redirected.
3011    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
3012        class containing characters with values greater than 255, PCRE compilation
3013        went into a loop.
3015    12. A recursive reference to a subpattern that was within another subpattern
3016        that had a minimum quantifier of zero caused PCRE to crash. For example,
3017        (x(y(?2))z)? provoked this bug with a subject that got as far as the
3018        recursion. If the recursively-called subpattern itself had a zero repeat,
3019        that was OK.
3021    13. In pcretest, the buffer for reading a data line was set at 30K, but the
3022        buffer into which it was copied (for escape processing) was still set at
3023        1024, so long lines caused crashes.
3025    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
3026        "internal error: code overflow...". This applied to any character class
3027        that was followed by a possessive quantifier.
3029    15. Modified the Makefile to add libpcre.la as a prerequisite for
3030        libpcreposix.la because I was told this is needed for a parallel build to
3031        work.
3033    16. If a pattern that contained .* following optional items at the start was
3034        studied, the wrong optimizing data was generated, leading to matching
3035        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
3036        matching string must start with a or b or c. The correct conclusion for
3037        this pattern is that a match can start with any character.
3040    Version 4.4 13-Aug-03
3041    ---------------------
3043     1. In UTF-8 mode, a character class containing characters with values between
3044        127 and 255 was not handled correctly if the compiled pattern was studied.
3045        In fixing this, I have also improved the studying algorithm for such
3046        classes (slightly).
3048     2. Three internal functions had redundant arguments passed to them. Removal
3049        might give a very teeny performance improvement.
3051     3. Documentation bug: the value of the capture_top field in a callout is *one
3052        more than* the number of the hightest numbered captured substring.
3054     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
3055        in incorrectly linking with a previously installed version. They now link
3056        explicitly with libpcre.la.
3058     5. configure.in no longer needs to recognize Cygwin specially.
3060     6. A problem in pcre.in for Windows platforms is fixed.
3062     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
3063        pcretest, it used to include the size of the study block as part of its
3064        output. Unfortunately, the structure contains a field that has a different
3065        size on different hardware architectures. This meant that the tests that
3066        showed this size failed. As the block is currently always of a fixed size,
3067        this information isn't actually particularly useful in pcretest output, so
3068        I have just removed it.
3070     8. Three pre-processor statements accidentally did not start in column 1.
3071        Sadly, there are *still* compilers around that complain, even though
3072        standard C has not required this for well over a decade. Sigh.
3074     9. In pcretest, the code for checking callouts passed small integers in the
3075        callout_data field, which is a void * field. However, some picky compilers
3076        complained about the casts involved for this on 64-bit systems. Now
3077        pcretest passes the address of the small integer instead, which should get
3078        rid of the warnings.
3080    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
3081        both compile and run time, and gives an error if an invalid UTF-8 sequence
3082        is found. There is a option for disabling this check in cases where the
3083        string is known to be correct and/or the maximum performance is wanted.
3085    11. In response to a bug report, I changed one line in Makefile.in from
3087            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
3088        to
3089            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
3091        to look similar to other lines, but I have no way of telling whether this
3092        is the right thing to do, as I do not use Windows. No doubt I'll get told
3093        if it's wrong...
3096  Version 4.3 21-May-03  Version 4.3 21-May-03
3097  ---------------------  ---------------------

Removed from v.69  
changed lines
  Added in v.681

  ViewVC Help
Powered by ViewVC 1.1.5