ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 57 by nigel, Sat Feb 24 21:39:50 2007 UTC revision 661 by ph10, Sun Aug 21 09:00:54 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 in
10    a POSIX name.
13    Version 8.13 16-Aug-2011
14    ------------------------
16    1.  The Unicode data tables have been updated to Unicode 6.0.0.
18    2.  Two minor typos in pcre_internal.h have been fixed.
20    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
21        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
22        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
24    4.  There were a number of related bugs in the code for matching backrefences
25        caselessly in UTF-8 mode when codes for the characters concerned were
26        different numbers of bytes. For example, U+023A and U+2C65 are an upper
27        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
28        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
29        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
30        2-byte code at the end of the subject (it thought there wasn't enough data
31        left).
33    5.  Comprehensive information about what went wrong is now returned by
34        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
35        as the output vector has at least 2 elements. The offset of the start of
36        the failing character and a reason code are placed in the vector.
38    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
39        now returned is for the first byte of the failing character, instead of the
40        last byte inspected. This is an incompatible change, but I hope it is small
41        enough not to be a problem. It makes the returned offset consistent with
42        pcre_exec() and pcre_dfa_exec().
44    7.  pcretest now gives a text phrase as well as the error number when
45        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
46        failure, the offset and reason code are output.
48    8.  When \R was used with a maximizing quantifier it failed to skip backwards
49        over a \r\n pair if the subsequent match failed. Instead, it just skipped
50        back over a single character (\n). This seems wrong (because it treated the
51        two characters as a single entity when going forwards), conflicts with the
52        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
53        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
54        has been changed.
56    9.  Some internal refactoring has changed the processing so that the handling
57        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
58        time (the PCRE_DOTALL option was changed this way some time ago: version
59        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
60        which was always a bit of a fudge. It also means that there is one less
61        argument for the match() function, which reduces its stack requirements
62        slightly. This change also fixes an incompatibility with Perl: the pattern
63        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
65    10. More internal refactoring has drastically reduced the number of recursive
66        calls to match() for possessively repeated groups such as (abc)++ when
67        using pcre_exec().
69    11. While implementing 10, a number of bugs in the handling of groups were
70        discovered and fixed:
72        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
73        (a|)*(?1) gave a compile-time internal error.
74        ((a|)+)+  did not notice that the outer group could match an empty string.
75        (^a|^)+   was not marked as anchored.
76        (.*a|.*)+ was not marked as matching at start or after a newline.
78    12. Yet more internal refactoring has removed another argument from the match()
79        function. Special calls to this function are now indicated by setting a
80        value in a variable in the "match data" data block.
82    13. Be more explicit in pcre_study() instead of relying on "default" for
83        opcodes that mean there is no starting character; this means that when new
84        ones are added and accidentally left out of pcre_study(), testing should
85        pick them up.
87    14. The -s option of pcretest has been documented for ages as being an old
88        synonym of -m (show memory usage). I have changed it to mean "force study
89        for every regex", that is, assume /S for every regex. This is similar to -i
90        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
91        using it. It makes it easier to run collections of tests with and without
92        study enabled, and thereby test pcre_study() more easily. All the standard
93        tests are now run with and without -s (but some patterns can be marked as
94        "never study" - see 20 below).
96    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
97        restoration of the capturing data to the outer values was not happening
98        correctly.
100    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
101        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
102        pattern had matched an empty string, and so incorrectly returned a no
103        match.
105    17. There was optimizing code for the last branch of non-capturing parentheses,
106        and also for the obeyed branch of a conditional subexpression, which used
107        tail recursion to cut down on stack usage. Unfortunately, now that there is
108        the possibility of (*THEN) occurring in these branches, tail recursion is
109        no longer possible because the return has to be checked for (*THEN). These
110        two optimizations have therefore been removed.
112    18. If a pattern containing \R was studied, it was assumed that \R always
113        matched two bytes, thus causing the minimum subject length to be
114        incorrectly computed because \R can also match just one byte.
116    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
117        was incorrectly computed.
119    20. If /S is present twice on a test pattern in pcretest input, it now
120        *disables* studying, thereby overriding the use of -s on the command line
121        (see 14 above). This is necessary for one or two tests to keep the output
122        identical in both cases.
124    21. When (*ACCEPT) was used in an assertion that matched an empty string and
125        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
127    22. When an atomic group that contained a capturing parenthesis was
128        successfully matched, but the branch in which it appeared failed, the
129        capturing was not being forgotten if a higher numbered group was later
130        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
131        group 1 to "a", when in fact it should be unset. This applied to multi-
132        branched capturing and non-capturing groups, repeated or not, and also to
133        positive assertions (capturing in negative assertions does not happen
134        in PCRE) and also to nested atomic groups.
136    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
137        subject after a captured substring, to make it easier to tell which of a
138        number of identical substrings has been captured.
140    24. The way atomic groups are processed by pcre_exec() has been changed so that
141        if they are repeated, backtracking one repetition now resets captured
142        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
143        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
144        "aaa". Previously, it would have been "a". As part of this code
145        refactoring, the way recursive calls are handled has also been changed.
147    25. If an assertion condition captured any substrings, they were not passed
148        back unless some other capturing happened later. For example, if
149        (?(?=(a))a) was matched against "a", no capturing was returned.
151    26. When studying a pattern that contained subroutine calls or assertions,
152        the code for finding the minimum length of a possible match was handling
153        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
154        group 1 called group 2 while simultaneously a separate group 2 called group
155        1). A stack overflow occurred in this case. I have fixed this by limiting
156        the recursion depth to 10.
158    27. Updated RunTest.bat in the distribution to the version supplied by Tom
159        Fortmann. This supports explicit test numbers on the command line, and has
160        argument validation and error reporting.
162    28. An instance of \X with an unlimited repeat could fail if at any point the
163        first character it looked at was a mark character.
165    29. Some minor code refactoring concerning Unicode properties and scripts
166        should reduce the stack requirement of match() slightly.
168    30. Added the '=' option to pcretest to check the setting of unused capturing
169        slots at the end of the pattern, which are documented as being -1, but are
170        not included in the return count.
172    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
173        compiled something random. Now it gives a compile-time error (as does
174        Perl).
176    32. A *MARK encountered during the processing of a positive assertion is now
177        recorded and passed back (compatible with Perl).
179    33. If --only-matching or --colour was set on a pcregrep call whose pattern
180        had alternative anchored branches, the search for a second match in a line
181        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
182        matched the line "0102" twice. The same bug affected patterns that started
183        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
184        twice.
186    34. Previously, PCRE did not allow quantification of assertions. However, Perl
187        does, and because of capturing effects, quantifying parenthesized
188        assertions may at times be useful. Quantifiers are now allowed for
189        parenthesized assertions.
191    35. A minor code tidy in pcre_compile() when checking options for \R usage.
193    36. \g was being checked for fancy things in a character class, when it should
194        just be a literal "g".
196    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
197        appearance of a nested POSIX class supersedes an apparent external class.
198        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
199        unescaped square brackets may also appear as part of class names. For
200        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
201        more like Perl. (But see 8.20/1 above.)
203    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
204        was because it thought it was \N{name}, which is not supported).
206    39. Add minix to OS list not supporting the -S option in pcretest.
208    40. PCRE tries to detect cases of infinite recursion at compile time, but it
209        cannot analyze patterns in sufficient detail to catch mutual recursions
210        such as ((?1))((?2)). There is now a runtime test that gives an error if a
211        subgroup is called recursively as a subpattern for a second time at the
212        same position in the subject string. In previous releases this might have
213        been caught by the recursion limit, or it might have run out of stack.
215    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
216        happen only once. PCRE was, however incorrectly giving a compile time error
217        "recursive call could loop indefinitely" because it cannot analyze the
218        pattern in sufficient detail. The compile time test no longer happens when
219        PCRE is compiling a conditional subpattern, but actual runaway loops are
220        now caught at runtime (see 40 above).
222    42. It seems that Perl allows any characters other than a closing parenthesis
223        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
224        has been changed to be the same.
226    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
227        as not to get warnings when autogen.sh is called. Also changed
228        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
230    44. To help people who use pcregrep to scan files containing exceedingly long
231        lines, the following changes have been made:
233        (a) The default value of the buffer size parameter has been increased from
234            8K to 20K. (The actual buffer used is three times this size.)
236        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
237            PCRE is built.
239        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
240            to be set at run time.
242        (d) Numerical values in pcregrep options can be followed by K or M, for
243            example --buffer-size=50K.
245        (e) If a line being scanned overflows pcregrep's buffer, an error is now
246            given and the return code is set to 2.
248    45. Add a pointer to the latest mark to the callout data block.
250    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
251        partial match of an empty string instead of no match. This was specific to
252        the use of ".".
254    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
255        complete match instead of a partial match. This bug was dependent on both
256        the PCRE_UTF8 and PCRE_DOTALL options being set.
258    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
259        starting byte set, because \b was not being ignored.
262    Version 8.12 15-Jan-2011
263    ------------------------
265    1.  Fixed some typos in the markup of the man pages, and wrote a script that
266        checks for such things as part of the documentation building process.
268    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
269        --match-limit and --recursion-limit options (added for 8.11). In
270        particular, this made one of the standard tests fail. (The integer value
271        went into the wrong half of a long int.)
273    3.  If the --colour option was given to pcregrep with -v (invert match), it
274        did strange things, either producing crazy output, or crashing. It should,
275        of course, ignore a request for colour when reporting lines that do not
276        match.
278    4.  Another pcregrep bug caused similar problems if --colour was specified with
279        -M (multiline) and the pattern match finished with a line ending.
281    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
282        matched in multiline mode, the following line was shown as part of the
283        match. This seems wrong, so I have changed it.
285    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
286        the check for further matches in the same line (so they could be coloured)
287        to overrun the end of the current line. If another match was found, it was
288        incorrectly shown (and then shown again when found in the next line).
290    7.  If pcregrep was compiled under Windows, there was a reference to the
291        function pcregrep_exit() before it was defined. I am assuming this was
292        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
293        reported by a user. I've moved the definition above the reference.
296    Version 8.11 10-Dec-2010
297    ------------------------
299    1.  (*THEN) was not working properly if there were untried alternatives prior
300        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
301        backtracked to try for "b" instead of moving to the next alternative branch
302        at the same level (in this case, to look for "c"). The Perl documentation
303        is clear that when (*THEN) is backtracked onto, it goes to the "next
304        alternative in the innermost enclosing group".
306    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
307        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
308        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
309        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
310        (*THEN).
312    3.  If \s appeared in a character class, it removed the VT character from
313        the class, even if it had been included by some previous item, for example
314        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
315        of \s, but is part of the POSIX "space" class.)
317    4.  A partial match never returns an empty string (because you can always
318        match an empty string at the end of the subject); however the checking for
319        an empty string was starting at the "start of match" point. This has been
320        changed to the "earliest inspected character" point, because the returned
321        data for a partial match starts at this character. This means that, for
322        example, /(?<=abc)def/ gives a partial match for the subject "abc"
323        (previously it gave "no match").
325    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
326        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
327        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
328        has an implication that the given string is incomplete (because a partial
329        match is preferred over a full match). For this reason, these items now
330        give a partial match in this situation. [Aside: previously, the one case
331        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
332        match rather than a full match, which was wrong by the old rules, but is
333        now correct.]
335    6.  There was a bug in the handling of #-introduced comments, recognized when
336        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
337        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
338        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
339        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
340        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
341        places in pcre_compile().
343    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
344        comments when looking ahead for named forward references to subpatterns,
345        the only newline sequence it recognized was NL. It now handles newlines
346        according to the set newline convention.
348    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
349        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
350        cater for a lack of strerror(). These oversights have been fixed.
352    9.  Added --match-limit and --recursion-limit to pcregrep.
354    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
356    11. When the -o option was used, pcregrep was setting a return code of 1, even
357        when matches were found, and --line-buffered was not being honoured.
359    12. Added an optional parentheses number to the -o and --only-matching options
360        of pcregrep.
362    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
363        can match an empty string. The code to do it in pcretest and pcredemo
364        needed fixing:
366        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
367            only one byte after an empty string match just before CRLF (this case
368            just got forgotten; "any" and "anycrlf" were OK).
370        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
371            mode when an empty string match preceded an ASCII character followed by
372            a non-ASCII character. (The code for advancing by one character rather
373            than one byte was nonsense.)
375        (c) The pcredemo.c sample program did not have any code at all to handle
376            the cases when CRLF is a valid newline sequence.
378    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
379        as a starting offset was within the subject string. There is now a new
380        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
381        negative or greater than the length of the string. In order to test this,
382        pcretest is extended to allow the setting of negative starting offsets.
384    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
385        starting offset points to the beginning of a UTF-8 character was
386        unnecessarily clumsy. I tidied it up.
388    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
389        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
391    17. Nobody had reported that the --include_dir option, which was added in
392        release 7.7 should have been called --include-dir (hyphen, not underscore)
393        for compatibility with GNU grep. I have changed it to --include-dir, but
394        left --include_dir as an undocumented synonym, and the same for
395        --exclude-dir, though that is not available in GNU grep, at least as of
396        release 2.5.4.
398    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
399        characters from a string of bytes) have been redefined so as not to use
400        loops, in order to improve performance in some environments. At the same
401        time, I abstracted some of the common code into auxiliary macros to save
402        repetition (this should not affect the compiled code).
404    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
405        compile-time error is now given if \c is not followed by an ASCII
406        character, that is, a byte less than 128. (In EBCDIC mode, the code is
407        different, and any byte value is allowed.)
409    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
410        START_OPTIMIZE option, which is now allowed at compile time - but just
411        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
412        to pcregrep and other applications that have no direct access to PCRE
413        options. The new /Y option in pcretest sets this option when calling
414        pcre_compile().
416    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
417        back references. Groups containing recursive back references were forced to
418        be atomic by that change, but in the case of named groups, the amount of
419        memory required was incorrectly computed, leading to "Failed: internal
420        error: code overflow". This has been fixed.
422    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
423        pcretest.c, to avoid build problems in some Borland environments.
426    Version 8.10 25-Jun-2010
427    ------------------------
429    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
430        THEN.
432    2.  (*ACCEPT) was not working when inside an atomic group.
434    3.  Inside a character class, \B is treated as a literal by default, but
435        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
436        causes the error). The code is unchanged, but I tidied the documentation.
438    4.  Inside a character class, PCRE always treated \R and \X as literals,
439        whereas Perl faults them if its -w option is set. I have changed PCRE so
440        that it faults them when PCRE_EXTRA is set.
442    5.  Added support for \N, which always matches any character other than
443        newline. (It is the same as "." when PCRE_DOTALL is not set.)
445    6.  When compiling pcregrep with newer versions of gcc which may have
446        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
447        declared with attribute warn_unused_result" were given. Just casting the
448        result to (void) does not stop the warnings; a more elaborate fudge is
449        needed. I've used a macro to implement this.
451    7.  Minor change to pcretest.c to avoid a compiler warning.
453    8.  Added four artifical Unicode properties to help with an option to make
454        \s etc use properties (see next item). The new properties are: Xan
455        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
457    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
458        use Unicode properties. (*UCP) at the start of a pattern can be used to set
459        this option. Modified pcretest to add /W to test this facility. Added
460        REG_UCP to make it available via the POSIX interface.
462    10. Added --line-buffered to pcregrep.
464    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
465        studied, and the match started with a letter with a code point greater than
466        127 whose first byte was different to the first byte of the other case of
467        the letter, the other case of this starting letter was not recognized
468        (#976).
470    12. If a pattern that was studied started with a repeated Unicode property
471        test, for example, \p{Nd}+, there was the theoretical possibility of
472        setting up an incorrect bitmap of starting bytes, but fortunately it could
473        not have actually happened in practice until change 8 above was made (it
474        added property types that matched character-matching opcodes).
476    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
477        possible starting bytes for non-anchored patterns.
479    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
480        \R, and also a number of cases that involve Unicode properties, both
481        explicit and implicit when PCRE_UCP is set.
483    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
484        input, it could crash or give wrong results if characters with values
485        greater than 0xc0 were present in the subject string. (Detail: it assumed
486        UTF-8 input when processing these items.)
488    16. Added a lot of (int) casts to avoid compiler warnings in systems where
489        size_t is 64-bit (#991).
491    17. Added a check for running out of memory when PCRE is compiled with
492        --disable-stack-for-recursion (#990).
494    18. If the last data line in a file for pcretest does not have a newline on
495        the end, a newline was missing in the output.
497    19. The default pcre_chartables.c file recognizes only ASCII characters (values
498        less than 128) in its various bitmaps. However, there is a facility for
499        generating tables according to the current locale when PCRE is compiled. It
500        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
501        characters, are recognized by isspace() and therefore were getting set in
502        these tables, and indeed these tables seem to approximate to ISO 8859. This
503        caused a problem in UTF-8 mode when pcre_study() was used to create a list
504        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
505        which of course cannot start UTF-8 characters. I have changed the code so
506        that only real ASCII characters (less than 128) and the correct starting
507        bytes for UTF-8 encodings are set for characters greater than 127 when in
508        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
509        altogether.)
511    20. Added the /T option to pcretest so as to be able to run tests with non-
512        standard character tables, thus making it possible to include the tests
513        used for 19 above in the standard set of tests.
515    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
516        reference to a subpattern the other side of a comment that contains an
517        opening parenthesis caused either an internal compiling error, or a
518        reference to the wrong subpattern.
521    Version 8.02 19-Mar-2010
522    ------------------------
524    1.  The Unicode data tables have been updated to Unicode 5.2.0.
526    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
527        configured.
529    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
530        original author of that file, following a query about its status.
532    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
533        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
535    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
536        quantifier applied to a forward-referencing subroutine call, could compile
537        incorrect code or give the error "internal error: previously-checked
538        referenced subpattern not found".
540    6.  Both MS Visual Studio and Symbian OS have problems with initializing
541        variables to point to external functions. For these systems, therefore,
542        pcre_malloc etc. are now initialized to local functions that call the
543        relevant global functions.
545    7.  There were two entries missing in the vectors called coptable and poptable
546        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
547        I've fixed the data, and added a kludgy way of testing at compile time that
548        the lengths are correct (equal to the number of opcodes).
550    8.  Following on from 7, I added a similar kludge to check the length of the
551        eint vector in pcreposix.c.
553    9.  Error texts for pcre_compile() are held as one long string to avoid too
554        much relocation at load time. To find a text, the string is searched,
555        counting zeros. There was no check for running off the end of the string,
556        which could happen if a new error number was added without updating the
557        string.
559    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
561    11. \K was not working if it appeared in an atomic group or in a group that
562        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
563        \K is "not well defined" if used in an assertion. PCRE now accepts it if
564        the assertion is positive, but not if it is negative.
566    12. Change 11 fortuitously reduced the size of the stack frame used in the
567        "match()" function of pcre_exec.c by one pointer. Forthcoming
568        implementation of support for (*MARK) will need an extra pointer on the
569        stack; I have reserved it now, so that the stack frame size does not
570        decrease.
572    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
573        item in branch that calls a recursion is a subroutine call - as in the
574        second branch in the above example - was incorrectly given the compile-
575        time error "recursive call could loop indefinitely" because pcre_compile()
576        was not correctly checking the subroutine for matching a non-empty string.
578    14. The checks for overrunning compiling workspace could trigger after an
579        overrun had occurred. This is a "should never occur" error, but it can be
580        triggered by pathological patterns such as hundreds of nested parentheses.
581        The checks now trigger 100 bytes before the end of the workspace.
583    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
586    Version 8.01 19-Jan-2010
587    ------------------------
589    1.  If a pattern contained a conditional subpattern with only one branch (in
590        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
591        computed the wrong minimum data length (which is of course zero for such
592        subpatterns). This could cause incorrect "no match" results.
594    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
595        the pattern is reset in the first branch, pcre_compile() failed with
596        "internal error: code overflow at offset...". This happened only when
597        the reset was to the original external option setting. (An optimization
598        abstracts leading options settings into an external setting, which was the
599        cause of this.)
601    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
602        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
603        assertion pattern did not match (meaning that the assertion was true), it
604        was incorrectly treated as false if the SKIP had been reached during the
605        matching. This also applied to assertions used as conditions.
607    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
608        assertion subpattern, including such a pattern used as a condition,
609        unpredictable results occurred, instead of the error return
612    5.  The C++ GlobalReplace function was not working like Perl for the special
613        situation when an empty string is matched. It now does the fancy magic
614        stuff that is necessary.
616    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
617        removed. (These were left over from very, very early versions of PCRE.)
619    7.  Some cosmetic changes to the code to make life easier when compiling it
620        as part of something else:
622        (a) Change DEBUG to PCRE_DEBUG.
624        (b) In pcre_compile(), rename the member of the "branch_chain" structure
625            called "current" as "current_branch", to prevent a collision with the
626            Linux macro when compiled as a kernel module.
628        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
629            prevent a collision with the Linux macro when compiled as a kernel
630            module.
632    8.  In pcre_compile() there are some checks for integer overflows that used to
633        cast potentially large values to (double). This has been changed to that
634        when building, a check for int64_t is made, and if it is found, it is used
635        instead, thus avoiding the use of floating point arithmetic. (There is no
636        other use of FP in PCRE.) If int64_t is not found, the fallback is to
637        double.
639    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
640        2005 (difference between two addresses compared to an unsigned value).
642    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
643        custom one, because of the following reported problem in Windows:
645          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
646              under Win32.
647          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
648              therefore missing the function definition.
649          - The compiler thus generates a "C" signature for the test function.
650          - The linker fails to find the "C" function.
651          - PCRE fails to configure if asked to do so against libbz2.
653    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
654        messages were output:
656          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
657          rerunning libtoolize, to keep the correct libtool macros in-tree.
658          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
660        I have done both of these things.
662    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
663        most of the time, it *can* run out if it is given a pattern that contains a
664        runaway infinite recursion. I updated the discussion in the pcrestack man
665        page.
667    13. Now that we have gone to the x.xx style of version numbers, the minor
668        version may start with zero. Using 08 or 09 is a bad idea because users
669        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
670        interpreted as invalid octal numbers. I've updated the previous comment in
671        configure.ac, and also added a check that gives an error if 08 or 09 are
672        used.
674    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
675        causing partial matching to fail when the end of the subject matched \W
676        in a UTF-8 pattern where \W was quantified with a minimum of 3.
678    15. There were some discrepancies between the declarations in pcre_internal.h
679        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
680        their definitions. The declarations used "const uschar *" and the
681        definitions used USPTR. Even though USPTR is normally defined as "const
682        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
683        reported that: "This difference in casting confuses some C++ compilers, for
684        example, SunCC recognizes above declarations as different functions and
685        generates broken code for hbpcre." I have changed the declarations to use
686        USPTR.
688    16. GNU libtool is named differently on some systems. The autogen.sh script now
689        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
690        (FreeBSD).
692    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
693        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
694        comment: "Figure out how to create a longlong from a string: strtoll and
695        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
696        instance, but it only takes 2 args instead of 3!"
698    18. A subtle bug concerned with back references has been fixed by a change of
699        specification, with a corresponding code fix. A pattern such as
700        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
701        refers, was giving matches when it shouldn't. For example, xa=xaaa would
702        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
703        same bug. Such groups have to be quantified to be useful, or contained
704        inside another quantified group. (If there's no repetition, the reference
705        can never match.) The problem arises because, having left the group and
706        moved on to the rest of the pattern, a later failure that backtracks into
707        the group uses the captured value from the final iteration of the group
708        rather than the correct earlier one. I have fixed this in PCRE by forcing
709        any group that contains a reference to itself to be an atomic group; that
710        is, there cannot be any backtracking into it once it has completed. This is
711        similar to recursive and subroutine calls.
714    Version 8.00 19-Oct-09
715    ----------------------
717    1.  The table for translating pcre_compile() error codes into POSIX error codes
718        was out-of-date, and there was no check on the pcre_compile() error code
719        being within the table. This could lead to an OK return being given in
720        error.
722    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
723        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
724        in a Windows environment.
726    3.  The pcregrep --count option prints the count for each file even when it is
727        zero, as does GNU grep. However, pcregrep was also printing all files when
728        --files-with-matches was added. Now, when both options are given, it prints
729        counts only for those files that have at least one match. (GNU grep just
730        prints the file name in this circumstance, but including the count seems
731        more useful - otherwise, why use --count?) Also ensured that the
732        combination -clh just lists non-zero counts, with no names.
734    4.  The long form of the pcregrep -F option was incorrectly implemented as
735        --fixed_strings instead of --fixed-strings. This is an incompatible change,
736        but it seems right to fix it, and I didn't think it was worth preserving
737        the old behaviour.
739    5.  The command line items --regex=pattern and --regexp=pattern were not
740        recognized by pcregrep, which required --regex pattern or --regexp pattern
741        (with a space rather than an '='). The man page documented the '=' forms,
742        which are compatible with GNU grep; these now work.
744    6.  No libpcreposix.pc file was created for pkg-config; there was just
745        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
747    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
748        when UCP support is not needed, by modifying the Python script that
749        generates it from Unicode data files. This should not matter if the module
750        is correctly used as a library, but I received one complaint about 50K of
751        unwanted data. My guess is that the person linked everything into his
752        program rather than using a library. Anyway, it does no harm.
754    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
755        was a minimum greater than 1 for a wide character in a possessive
756        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
757        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
758        character. Chaos in the form of incorrect output or a compiling loop could
759        result.
761    9.  The restrictions on what a pattern can contain when partial matching is
762        requested for pcre_exec() have been removed. All patterns can now be
763        partially matched by this function. In addition, if there are at least two
764        slots in the offset vector, the offset of the earliest inspected character
765        for the match and the offset of the end of the subject are set in them when
766        PCRE_ERROR_PARTIAL is returned.
768    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
769        synonymous with PCRE_PARTIAL, for backwards compatibility, and
770        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
771        and may be more useful for multi-segment matching.
773    11. Partial matching with pcre_exec() is now more intuitive. A partial match
774        used to be given if ever the end of the subject was reached; now it is
775        given only if matching could not proceed because another character was
776        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
777        string "Z", which now yields "no match" instead of "partial match". In the
778        case of pcre_dfa_exec(), "no match" is given if every matching path for the
779        final character ended with (*FAIL).
781    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
782        if the pattern had a "must contain" character that was already found in the
783        earlier partial match, unless partial matching was again requested. For
784        example, with the pattern /dog.(body)?/, the "must contain" character is
785        "g". If the first part-match was for the string "dog", restarting with
786        "sbody" failed. This bug has been fixed.
788    13. The string returned by pcre_dfa_exec() after a partial match has been
789        changed so that it starts at the first inspected character rather than the
790        first character of the match. This makes a difference only if the pattern
791        starts with a lookbehind assertion or \b or \B (\K is not supported by
792        pcre_dfa_exec()). It's an incompatible change, but it makes the two
793        matching functions compatible, and I think it's the right thing to do.
795    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
796        so that the demonstration program is easily available in environments where
797        PCRE has not been installed from source.
799    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
800        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
801        library.
803    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
804        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
805        is not the first non-POSIX option to be added. Clearly some people find
806        these options useful.
808    17. If a caller to the POSIX matching function regexec() passes a non-zero
809        value for nmatch with a NULL value for pmatch, the value of
810        nmatch is forced to zero.
812    18. RunGrepTest did not have a test for the availability of the -u option of
813        the diff command, as RunTest does. It now checks in the same way as
814        RunTest, and also checks for the -b option.
816    19. If an odd number of negated classes containing just a single character
817        interposed, within parentheses, between a forward reference to a named
818        subpattern and the definition of the subpattern, compilation crashed with
819        an internal error, complaining that it could not find the referenced
820        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
821        [The bug was that it was starting one character too far in when skipping
822        over the character class, thus treating the ] as data rather than
823        terminating the class. This meant it could skip too much.]
825    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
826        /g option in pcretest when the pattern contains \K, which makes it possible
827        to have an empty string match not at the start, even when the pattern is
828        anchored. Updated pcretest and pcredemo to use this option.
830    21. If the maximum number of capturing subpatterns in a recursion was greater
831        than the maximum at the outer level, the higher number was returned, but
832        with unset values at the outer level. The correct (outer level) value is
833        now given.
835    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
836        PCRE did not set those parentheses (unlike Perl). I have now found a way to
837        make it do so. The string so far is captured, making this feature
838        compatible with Perl.
840    23. The tests have been re-organized, adding tests 11 and 12, to make it
841        possible to check the Perl 5.10 features against Perl 5.10.
843    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
844        pattern matches a fixed length string. PCRE did not allow this; now it
845        does. Neither allows recursion.
847    25. I finally figured out how to implement a request to provide the minimum
848        length of subject string that was needed in order to match a given pattern.
849        (It was back references and recursion that I had previously got hung up
850        on.) This code has now been added to pcre_study(); it finds a lower bound
851        to the length of subject needed. It is not necessarily the greatest lower
852        bound, but using it to avoid searching strings that are too short does give
853        some useful speed-ups. The value is available to calling programs via
854        pcre_fullinfo().
856    26. While implementing 25, I discovered to my embarrassment that pcretest had
857        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
858        study optimizations had never been tested with that matching function.
859        Oops. What is worse, even when it was passed study data, there was a bug in
860        pcre_dfa_exec() that meant it never actually used it. Double oops. There
861        were also very few tests of studied patterns with pcre_dfa_exec().
863    27. If (?| is used to create subpatterns with duplicate numbers, they are now
864        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
865        on the other side of the coin, they are no longer allowed to have different
866        names, because these cannot be distinguished in PCRE, and this has caused
867        confusion. (This is a difference from Perl.)
869    28. When duplicate subpattern names are present (necessarily with different
870        numbers, as required by 27 above), and a test is made by name in a
871        conditional pattern, either for a subpattern having been matched, or for
872        recursion in such a pattern, all the associated numbered subpatterns are
873        tested, and the overall condition is true if the condition is true for any
874        one of them. This is the way Perl works, and is also more like the way
875        testing by number works.
878    Version 7.9 11-Apr-09
879    ---------------------
881    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
882        (pcretest), all targets were linked against these libraries. This included
883        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
884        libraries. This caused unwanted dependencies to be created. This problem
885        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
886        pcretest is linked with readline.
888    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
889        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
890        moved outside it again, because FALSE and TRUE are already defined in AIX,
891        but BOOL is not.
893    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
894        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
896    4.  The pcregrep documentation said spaces were inserted as well as colons (or
897        hyphens) following file names and line numbers when outputting matching
898        lines. This is not true; no spaces are inserted. I have also clarified the
899        wording for the --colour (or --color) option.
901    5.  In pcregrep, when --colour was used with -o, the list of matching strings
902        was not coloured; this is different to GNU grep, so I have changed it to be
903        the same.
905    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
906        each matching line was coloured. Now it goes on to look for further matches
907        of any of the test patterns, which is the same behaviour as GNU grep.
909    7.  A pattern that could match an empty string could cause pcregrep to loop; it
910        doesn't make sense to accept an empty string match in pcregrep, so I have
911        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
912        seems to be how GNU grep behaves.
914    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
915        start or after a newline", because the conditional assertion was not being
916        correctly handled. The rule now is that both the assertion and what follows
917        in the first alternative must satisfy the test.
919    9.  If auto-callout was enabled in a pattern with a conditional group whose
920        condition was an assertion, PCRE could crash during matching, both with
921        pcre_exec() and pcre_dfa_exec().
923    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
924        used for matching.
926    11. Unicode property support in character classes was not working for
927        characters (bytes) greater than 127 when not in UTF-8 mode.
929    12. Added the -M command line option to pcretest.
931    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
933    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
935    16. Added comments and documentation about mis-use of no_arg in the C++
936        wrapper.
938    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
939        from Martin Jerabek that uses macro names for all relevant character and
940        string constants.
942    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
943        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
944        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
945        these, but not everybody uses configure.
947    19. A conditional group that had only one branch was not being correctly
948        recognized as an item that could match an empty string. This meant that an
949        enclosing group might also not be so recognized, causing infinite looping
950        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
951        with the subject "ab", where knowledge that the repeated group can match
952        nothing is needed in order to break the loop.
954    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
955        exec(), but without supplying a callout function, matching went wrong.
957    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
958        leak if the size of the offset vector was greater than 30. When the vector
959        is smaller, the saved offsets during recursion go onto a local stack
960        vector, but for larger vectors malloc() is used. It was failing to free
961        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
962        error, in fact).
964    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
965        heapframe that is used only when UTF-8 support is enabled. This caused no
966        problem, but was untidy.
968    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
969        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
970        included within another project.
972    24. Steven Van Ingelgem's patches to add more options to the CMake support,
973        slightly modified by me:
975          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
976              not building pcregrep.
978          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
979              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
981    25. Forward references, both numeric and by name, in patterns that made use of
982        duplicate group numbers, could behave incorrectly or give incorrect errors,
983        because when scanning forward to find the reference group, PCRE was not
984        taking into account the duplicate group numbers. A pattern such as
985        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
987    26. Changed a few more instances of "const unsigned char *" to USPTR, making
988        the feature of a custom pointer more persuasive (as requested by a user).
990    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
991        pcretest.c, inside #ifndefs, because it seems they are sometimes already
992        pre-defined.
994    28. Added support for (*UTF8) at the start of a pattern.
996    29. Arrange for flags added by the "release type" setting in CMake to be shown
997        in the configuration summary.
1000    Version 7.8 05-Sep-08
1001    ---------------------
1003    1.  Replaced UCP searching code with optimized version as implemented for Ad
1004        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
1005        stage table and inline lookup instead of a function, giving speed ups of 2
1006        to 5 times on some simple patterns that I tested. Permission was given to
1007        distribute the MultiStage2.py script that generates the tables (it's not in
1008        the tarball, but is in the Subversion repository).
1010    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
1011        scripts.
1013    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
1014        a group with a zero qualifier. The result of the study could be incorrect,
1015        or the function might crash, depending on the pattern.
1017    4.  Caseless matching was not working for non-ASCII characters in back
1018        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
1019        It now works when Unicode Property Support is available.
1021    5.  In pcretest, an escape such as \x{de} in the data was always generating
1022        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
1023        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
1024        truncation.
1026    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
1028    7.  Added two (int) casts to pcregrep when printing the difference of two
1029        pointers, in case they are 64-bit values.
1031    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
1032        test 2 if it fails.
1034    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
1035        and a #define of that name to empty if it is not externally set. This is to
1036        allow users of MSVC to set it if necessary.
1038    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
1039        the convenience functions in the pcre_get.c source file.
1041    11. An option change at the start of a pattern that had top-level alternatives
1042        could cause overwriting and/or a crash. This command provoked a crash in
1043        some environments:
1045          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
1047        This potential security problem was recorded as CVE-2008-2371.
1049    12. For a pattern where the match had to start at the beginning or immediately
1050        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
1051        pcre_dfa_exec() could read past the end of the passed subject if there was
1052        no match. To help with detecting such bugs (e.g. with valgrind), I modified
1053        pcretest so that it places the subject at the end of its malloc-ed buffer.
1055    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
1056        exec() might read past the end of the data buffer in UTF-8 mode.
1058    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
1059        the data contained the byte 0x85 as part of a UTF-8 character within its
1060        first line. This applied both to normal and DFA matching.
1062    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
1063        /^[^d]*?$/8 failed to match "abc".
1065    16. Added a missing copyright notice to pcrecpp_internal.h.
1067    17. Make it more clear in the documentation that values returned from
1068        pcre_exec() in ovector are byte offsets, not character counts.
1070    18. Tidied a few places to stop certain compilers from issuing warnings.
1072    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
1073        supplied by Stefan Weber. I made a further small update for 7.8 because
1074        there is a change of source arrangements: the pcre_searchfuncs.c module is
1075        replaced by pcre_ucd.c.
1078    Version 7.7 07-May-08
1079    ---------------------
1081    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
1082        a string to a long long, pretend we don't even have a long long." This is
1083        done by checking for the strtoq, strtoll, and _strtoi64 functions.
1085    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
1086        pre-7.6 versions, which defined a global no_arg variable instead of putting
1087        it in the RE class. (See also #8 below.)
1089    3.  Remove a line of dead code, identified by coverity and reported by Nuno
1090        Lopes.
1092    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
1094        (1) The include/exclude patterns were being applied to the whole pathnames
1095            of files, instead of just to the final components.
1097        (2) If there was more than one level of directory, the subdirectories were
1098            skipped unless they satisfied the include/exclude conditions. This is
1099            inconsistent with GNU grep (and could even be seen as contrary to the
1100            pcregrep specification - which I improved to make it absolutely clear).
1101            The action now is always to scan all levels of directory, and just
1102            apply the include/exclude patterns to regular files.
1104    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
1105        --exclude_dir in the tests to avoid scanning .svn directories.
1107    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
1108        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
1109        doesn't support NULs in patterns.
1111    7.  Added some missing "const"s to declarations of static tables in
1112        pcre_compile.c and pcre_dfa_exec.c.
1114    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
1115        caused by fix #2  above. (Subsequently also a second patch to fix the
1116        first patch. And a third patch - this was a messy problem.)
1118    9.  Applied Craig's patch to remove the use of push_back().
1120    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
1121        matching function regexec().
1123    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
1124        which, however, unlike Perl's \g{...}, are subroutine calls, not back
1125        references. PCRE supports relative numbers with this syntax (I don't think
1126        Oniguruma does).
1128    12. Previously, a group with a zero repeat such as (...){0} was completely
1129        omitted from the compiled regex. However, this means that if the group
1130        was called as a subroutine from elsewhere in the pattern, things went wrong
1131        (an internal error was given). Such groups are now left in the compiled
1132        pattern, with a new opcode that causes them to be skipped at execution
1133        time.
1135    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
1136        to the way PCRE behaves:
1138        (a) A lone ] character is dis-allowed (Perl treats it as data).
1140        (b) A back reference to an unmatched subpattern matches an empty string
1141            (Perl fails the current match path).
1143        (c) A data ] in a character class must be notated as \] because if the
1144            first data character in a class is ], it defines an empty class. (In
1145            Perl it is not possible to have an empty class.) The empty class []
1146            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
1147            The negative empty class [^] matches any one character, independently
1148            of the DOTALL setting.
1150    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
1151        non-existent subpattern following a character class starting with ']' and
1152        containing () gave an internal compiling error instead of "reference to
1153        non-existent subpattern". Fortunately, when the pattern did exist, the
1154        compiled code was correct. (When scanning forwards to check for the
1155        existencd of the subpattern, it was treating the data ']' as terminating
1156        the class, so got the count wrong. When actually compiling, the reference
1157        was subsequently set up correctly.)
1159    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
1160        it was being rejected as not supported by pcre_dfa_exec(), even though
1161        other assertions are supported. I have made pcre_dfa_exec() support
1162        (*FAIL).
1164    16. The implementation of 13c above involved the invention of a new opcode,
1165        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
1166        cannot be changed at match time, I realized I could make a small
1167        improvement to matching performance by compiling OP_ALLANY instead of
1168        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
1169        on the OP_ANY path.
1171    17. Compiling pcretest on Windows with readline support failed without the
1172        following two fixes: (1) Make the unistd.h include conditional on
1173        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
1175    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
1176        ncurses library to be included for pcretest when ReadLine support is
1177        requested, but also to allow for it to be overridden. This patch came from
1178        Daniel Bergström.
1180    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
1181        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
1182        any errors with the current Unicode tables. Thanks to Peter Kankowski for
1183        spotting this.
1186    Version 7.6 28-Jan-08
1187    ---------------------
1189    1.  A character class containing a very large number of characters with
1190        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
1191        overflow.
1193    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
1194        HAVE_LONG_LONG is not defined.
1196    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
1197        bring them up to date and include new features. This patch includes:
1199        - Fixed PH's badly added libz and libbz2 support.
1200        - Fixed a problem with static linking.
1201        - Added pcredemo. [But later removed - see 7 below.]
1202        - Fixed dftables problem and added an option.
1203        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
1204            HAVE_LONG_LONG.
1205        - Added readline support for pcretest.
1206        - Added an listing of the option settings after cmake has run.
1208    4.  A user submitted a patch to Makefile that makes it easy to create
1209        "pcre.dll" under mingw when using Configure/Make. I added stuff to
1210        Makefile.am that cause it to include this special target, without
1211        affecting anything else. Note that the same mingw target plus all
1212        the other distribution libraries and programs are now supported
1213        when configuring with CMake (see 6 below) instead of with
1214        Configure/Make.
1216    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
1217        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
1218        exported in the Windows port". It has not yet been confirmed that the patch
1219        solves the problem, but it does no harm.
1221    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
1222        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
1223        with CMake, and also correct the comment about stack recursion.
1225    7.  Remove the automatic building of pcredemo from the ./configure system and
1226        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
1227        of a program that users should build themselves after PCRE is installed, so
1228        building it automatically is not really right. What is more, it gave
1229        trouble in some build environments.
1231    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
1234    Version 7.5 10-Jan-08
1235    ---------------------
1237    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
1238        values in parens when parsing an RE using the C++ wrapper."
1240    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
1241        Characters greater than 255 were excluded from the class instead of being
1242        included.
1244    3.  The same bug as (2) above applied to negated POSIX classes such as
1245        [:^space:].
1247    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
1248        defined or documented. It seems to have been a typo for PCRE_STATIC, so
1249        I have changed it.
1251    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
1252        first named subpattern) and a construct such as (?&a) would reference the
1253        first named subpattern whose name started with "a" (in other words, the
1254        length check was missing). Both these problems are fixed. "Subpattern name
1255        expected" is now given for (?&) (a zero-length name), and this patch also
1256        makes it give the same error for \k'' (previously it complained that that
1257        was a reference to a non-existent subpattern).
1259    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
1260        this is right because (?- can be followed by option settings as well as by
1261        digits. I have, however, made the messages clearer.
1263    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
1264        than the number used in the conditional) now cause a compile-time error.
1265        This is actually not compatible with Perl, which accepts such patterns, but
1266        treats the conditional as always being FALSE (as PCRE used to), but it
1267        seems to me that giving a diagnostic is better.
1269    8.  Change "alphameric" to the more common word "alphanumeric" in comments
1270        and messages.
1272    9.  Fix two occurrences of "backslash" in comments that should have been
1273        "backspace".
1275    10. Remove two redundant lines of code that can never be obeyed (their function
1276        was moved elsewhere).
1278    11. The program that makes PCRE's Unicode character property table had a bug
1279        which caused it to generate incorrect table entries for sequences of
1280        characters that have the same character type, but are in different scripts.
1281        It amalgamated them into a single range, with the script of the first of
1282        them. In other words, some characters were in the wrong script. There were
1283        thirteen such cases, affecting characters in the following ranges:
1285          U+002b0 - U+002c1
1286          U+0060c - U+0060d
1287          U+0061e - U+00612
1288          U+0064b - U+0065e
1289          U+0074d - U+0076d
1290          U+01800 - U+01805
1291          U+01d00 - U+01d77
1292          U+01d9b - U+01dbf
1293          U+0200b - U+0200f
1294          U+030fc - U+030fe
1295          U+03260 - U+0327f
1296          U+0fb46 - U+0fbb1
1297          U+10450 - U+1049d
1299    12. The -o option (show only the matching part of a line) for pcregrep was not
1300        compatible with GNU grep in that, if there was more than one match in a
1301        line, it showed only the first of them. It now behaves in the same way as
1302        GNU grep.
1304    13. If the -o and -v options were combined for pcregrep, it printed a blank
1305        line for every non-matching line. GNU grep prints nothing, and pcregrep now
1306        does the same. The return code can be used to tell if there were any
1307        non-matching lines.
1309    14. Added --file-offsets and --line-offsets to pcregrep.
1311    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
1312        infinitely looping recursion. The bug was that positive lookaheads were not
1313        being skipped when checking for a possible empty match (negative lookaheads
1314        and both kinds of lookbehind were skipped).
1316    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
1317        inclusion of <windows.h> to before rather than after the definition of
1318        INVALID_FILE_ATTRIBUTES (patch from David Byron).
1320    17. Specifying a possessive quantifier with a specific limit for a Unicode
1321        character property caused pcre_compile() to compile bad code, which led at
1322        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
1323        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
1324        caused the error; without that there was no problem.
1326    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
1328    19. Added --enable-pcretest-libreadline.
1330    20. In pcrecpp.cc, the variable 'count' was incremented twice in
1331        RE::GlobalReplace(). As a result, the number of replacements returned was
1332        double what it should be. I removed one of the increments, but Craig sent a
1333        later patch that removed the other one (the right fix) and added unit tests
1334        that check the return values (which was not done before).
1336    21. Several CMake things:
1338        (1) Arranged that, when cmake is used on Unix, the libraries end up with
1339            the names libpcre and libpcreposix, not just pcre and pcreposix.
1341        (2) The above change means that pcretest and pcregrep are now correctly
1342            linked with the newly-built libraries, not previously installed ones.
1346    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
1347        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
1348        UTF-8 newline character). The key issue is that the pattern starts .*;
1349        this means that the match must be either at the beginning, or after a
1350        newline. The bug was in the code for advancing after a failed match and
1351        checking that the new position followed a newline. It was not taking
1352        account of UTF-8 characters correctly.
1354    23. PCRE was behaving differently from Perl in the way it recognized POSIX
1355        character classes. PCRE was not treating the sequence [:...:] as a
1356        character class unless the ... were all letters. Perl, however, seems to
1357        allow any characters between [: and :], though of course it rejects as
1358        unknown any "names" that contain non-letters, because all the known class
1359        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
1360        for example, whereas PCRE did not - it did not recognize a POSIX character
1361        class. This seemed a bit dangerous, so the code has been changed to be
1362        closer to Perl. The behaviour is not identical to Perl, because PCRE will
1363        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
1364        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
1365        Perl does, and where it didn't before.
1367    24. Rewrite so as to remove the single use of %n from pcregrep because in some
1368        Windows environments %n is disabled by default.
1371    Version 7.4 21-Sep-07
1372    ---------------------
1374    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
1375        means that a class such as [\s] counted as "explicit reference to CR or
1376        LF". That isn't really right - the whole point of the change was to try to
1377        help when there was an actual mention of one of the two characters. So now
1378        the change happens only if \r or \n (or a literal CR or LF) character is
1379        encountered.
1381    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
1382        of both had grown to the point where there were only 3 bits left.
1383        Fortunately, there was spare space in the data structure, and so I have
1384        moved the internal flags into a new 16-bit field to free up more option
1385        bits.
1387    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
1388        but did not set the internal JCHANGED flag - either of these is enough to
1389        control the way the "get" function works - but the PCRE_INFO_JCHANGED
1390        facility is supposed to tell if (?J) was ever used, so now (?J) at the
1391        start sets both bits.
1393    4.  Added options (at build time, compile time, exec time) to change \R from
1394        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
1396    5.  doc/pcresyntax.html was missing from the distribution.
1398    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
1399        compatibility, even though it is no longer used.
1401    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
1402        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
1403        windows.h file is present (where different names are used). [This was
1404        reversed later after testing - see 16 below.]
1406    8.  Changed all #include <config.h> to #include "config.h". There were also
1407        some further <pcre.h> cases that I changed to "pcre.h".
1409    9.  When pcregrep was used with the --colour option, it missed the line ending
1410        sequence off the lines that it output.
1412    10. It was pointed out to me that arrays of string pointers cause lots of
1413        relocations when a shared library is dynamically loaded. A technique of
1414        using a single long string with a table of offsets can drastically reduce
1415        these. I have refactored PCRE in four places to do this. The result is
1416        dramatic:
1418          Originally:                          290
1419          After changing UCP table:            187
1420          After changing error message table:   43
1421          After changing table of "verbs"       36
1422          After changing table of Posix names   22
1424        Thanks to the folks working on Gregex for glib for this insight.
1426    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
1427        unicode-properties was also set.
1429    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
1431    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
1432        checked only for CRLF.
1434    14. Added casts to pcretest.c to avoid compiler warnings.
1436    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
1438    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
1439        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
1440        entirely. This removes changes made in 7 above.
1442    17. The CMake files have been updated, and there is now more information about
1443        building with CMake in the NON-UNIX-USE document.
1446    Version 7.3 28-Aug-07
1447    ---------------------
1449     1. In the rejigging of the build system that eventually resulted in 7.1, the
1450        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
1451        brackets there is not right, since it causes compilers to look for an
1452        installed pcre.h, not the version that is in the source that is being
1453        compiled (which of course may be different). I have changed it back to:
1455          #include "pcre.h"
1457        I have a vague recollection that the change was concerned with compiling in
1458        different directories, but in the new build system, that is taken care of
1459        by the VPATH setting the Makefile.
1461     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
1462        when the subject happened to end in the byte 0x85 (e.g. if the last
1463        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
1464        characters but of course it shouldn't be taken as a newline when it is part
1465        of another character. The bug was that, for an unlimited repeat of . in
1466        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
1467        characters when looking for a newline.
1469     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
1471     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
1472        in debug output.
1474     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
1475        long printing in the pcrecpp unittest when running under MinGW.
1477     6. ESC_K was left out of the EBCDIC table.
1479     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
1480        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
1481        limit also applies to "virtual nesting" when a pattern is recursive, and in
1482        this case 1000 isn't so big. I have been able to remove this limit at the
1483        expense of backing off one optimization in certain circumstances. Normally,
1484        when pcre_exec() would call its internal match() function recursively and
1485        immediately return the result unconditionally, it uses a "tail recursion"
1486        feature to save stack. However, when a subpattern that can match an empty
1487        string has an unlimited repetition quantifier, it no longer makes this
1488        optimization. That gives it a stack frame in which to save the data for
1489        checking that an empty string has been matched. Previously this was taken
1490        from the 1000-entry workspace that had been reserved. So now there is no
1491        explicit limit, but more stack is used.
1493     8. Applied Daniel's patches to solve problems with the import/export magic
1494        syntax that is required for Windows, and which was going wrong for the
1495        pcreposix and pcrecpp parts of the library. These were overlooked when this
1496        problem was solved for the main library.
1498     9. There were some crude static tests to avoid integer overflow when computing
1499        the size of patterns that contain repeated groups with explicit upper
1500        limits. As the maximum quantifier is 65535, the maximum group length was
1501        set at 30,000 so that the product of these two numbers did not overflow a
1502        32-bit integer. However, it turns out that people want to use groups that
1503        are longer than 30,000 bytes (though not repeat them that many times).
1504        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1505        made it possible to implement the integer overflow checks in a much more
1506        dynamic way, which I have now done. The artificial limitation on group
1507        length has been removed - we now have only the limit on the total length of
1508        the compiled pattern, which depends on the LINK_SIZE setting.
1510    10. Fixed a bug in the documentation for get/copy named substring when
1511        duplicate names are permitted. If none of the named substrings are set, the
1512        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1513        empty string.
1515    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1516        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1517        because the ] is interpreted as the first data character and the
1518        terminating ] is not found. PCRE has been made compatible with Perl in this
1519        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1520        cause memory overwriting.
1522    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1523        string has been matched (to stop an infinite loop). It was not recognizing
1524        a conditional subpattern that could match an empty string if that
1525        subpattern was within another subpattern. For example, it looped when
1526        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
1527        condition was not nested. This bug has been fixed.
1529    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1530        past the start of the subject in the presence of bytes with the top bit
1531        set, for example "\x8aBCD".
1533    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1534        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1536    14. Optimized (?!) to (*FAIL).
1538    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1539        This restricts code points to be within the range 0 to 0x10FFFF, excluding
1540        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1541        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1542        does: it's just the validity check that is more restrictive.
1544    16. Inserted checks for integer overflows during escape sequence (backslash)
1545        processing, and also fixed erroneous offset values for syntax errors during
1546        backslash processing.
1548    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1549        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1551    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1552        caused an overrun.
1554    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1555        something other than just ASCII characters) inside a group that had an
1556        unlimited repeat caused a loop at compile time (while checking to see
1557        whether the group could match an empty string).
1559    20. Debugging a pattern containing \p or \P could cause a crash. For example,
1560        [\P{Any}] did so. (Error in the code for printing property names.)
1562    21. An orphan \E inside a character class could cause a crash.
1564    22. A repeated capturing bracket such as (A)? could cause a wild memory
1565        reference during compilation.
1567    23. There are several functions in pcre_compile() that scan along a compiled
1568        expression for various reasons (e.g. to see if it's fixed length for look
1569        behind). There were bugs in these functions when a repeated \p or \P was
1570        present in the pattern. These operators have additional parameters compared
1571        with \d, etc, and these were not being taken into account when moving along
1572        the compiled data. Specifically:
1574        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1575            length.
1577        (b) An item such as \pL+ within a repeated group could cause crashes or
1578            loops.
1580        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1581            "reference to non-existent subpattern" error.
1583        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1585    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1586        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1588    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1590    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1591        character were causing crashes (broken optimization).
1593    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1594        \p or \P) caused a compile-time loop.
1596    28. More problems have arisen in unanchored patterns when CRLF is a valid line
1597        break. For example, the unstudied pattern [\r\n]A does not match the string
1598        "\r\nA" because change 7.0/46 below moves the current point on by two
1599        characters after failing to match at the start. However, the pattern \nA
1600        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
1601        the same is true. There doesn't seem any very clean way out of this, but
1602        what I have chosen to do makes the common cases work: PCRE now takes note
1603        of whether there can be an explicit match for \r or \n anywhere in the
1604        pattern, and if so, 7.0/46 no longer applies. As part of this change,
1605        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1606        pattern has explicit CR or LF references.
1608    29. Added (*CR) etc for changing newline setting at start of pattern.
1611    Version 7.2 19-Jun-07
1612    ---------------------
1614     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
1615        which is apparently normally available under Windows.
1617     2. Re-jig the pcregrep tests with different newline settings in an attempt
1618        to make them independent of the local environment's newline setting.
1620     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
1622     4. Some of the "internals" tests were previously cut out when the link size
1623        was not 2, because the output contained actual offsets. The recent new
1624        "Z" feature of pcretest means that these can be cut out, making the tests
1625        usable with all link sizes.
1627     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
1628        stack recursion. This gives a massive performance boost under BSD, but just
1629        a small improvement under Linux. However, it saves one field in the frame
1630        in all cases.
1632     6. Added more features from the forthcoming Perl 5.10:
1634        (a) (?-n) (where n is a string of digits) is a relative subroutine or
1635            recursion call. It refers to the nth most recently opened parentheses.
1637        (b) (?+n) is also a relative subroutine call; it refers to the nth next
1638            to be opened parentheses.
1640        (c) Conditions that refer to capturing parentheses can be specified
1641            relatively, for example, (?(-2)... or (?(+3)...
1643        (d) \K resets the start of the current match so that everything before
1644            is not part of it.
1646        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1648        (f) \g{name} is another synonym - part of Perl 5.10's unification of
1649            reference syntax.
1651        (g) (?| introduces a group in which the numbering of parentheses in each
1652            alternative starts with the same number.
1654        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
1656     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
1659     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
1660        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
1661        for detecting groups that can match an empty string.
1663     9. A pattern with a very large number of alternatives (more than several
1664        hundred) was running out of internal workspace during the pre-compile
1665        phase, where pcre_compile() figures out how much memory will be needed. A
1666        bit of new cunning has reduced the workspace needed for groups with
1667        alternatives. The 1000-alternative test pattern now uses 12 bytes of
1668        workspace instead of running out of the 4096 that are available.
1670    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
1672    11. Applied patch from Google to remove an optimization that didn't quite work.
1673        The report of the bug said:
1675          pcrecpp::RE("a*").FullMatch("aaa") matches, while
1676          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
1677          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
1679    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
1680        it matched the wrong number of bytes.
1683    Version 7.1 24-Apr-07
1684    ---------------------
1686     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
1687        that is more "standard", making use of automake and other Autotools. There
1688        is some re-arrangement of the files and adjustment of comments consequent
1689        on this.
1691     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
1692        for recursive directory scanning broke on some systems because the files
1693        are not scanned in any specific order and on different systems the order
1694        was different. A call to "sort" has been inserted into RunGrepTest for the
1695        approprate test as a short-term fix. In the longer term there may be an
1696        alternative.
1698     3. I had an email from Eric Raymond about problems translating some of PCRE's
1699        man pages to HTML (despite the fact that I distribute HTML pages, some
1700        people do their own conversions for various reasons). The problems
1701        concerned the use of low-level troff macros .br and .in. I have therefore
1702        removed all such uses from the man pages (some were redundant, some could
1703        be replaced by .nf/.fi pairs). The 132html script that I use to generate
1704        HTML has been updated to handle .nf/.fi and to complain if it encounters
1705        .br or .in.
1707     4. Updated comments in configure.ac that get placed in config.h.in and also
1708        arranged for config.h to be included in the distribution, with the name
1709        config.h.generic, for the benefit of those who have to compile without
1710        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
1712     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
1713        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
1714        makevp.bat for the current PCRE, using the additional files
1715        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
1717     6. A Windows user reported a minor discrepancy with test 2, which turned out
1718        to be caused by a trailing space on an input line that had got lost in his
1719        copy. The trailing space was an accident, so I've just removed it.
1721     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
1722        that is needed.
1724     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
1725        as "const" (a) because they are and (b) because it helps the PHP
1726        maintainers who have recently made a script to detect big data structures
1727        in the php code that should be moved to the .rodata section. I remembered
1728        to update Builducptable as well, so it won't revert if ucptable.h is ever
1729        re-created.
1731     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
1732        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
1733        order to be able to cut out the UTF-8 tables in the latter when UTF-8
1734        support is not required. This saves 1.5-2K of code, which is important in
1735        some applications.
1737        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
1738        so as not to refer to the tables, even though these functions will never be
1739        called when UTF-8 support is disabled. Otherwise there are problems with a
1740        shared library.
1742    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
1744        (a) It was defining its arguments as char * instead of void *.
1746        (b) It was assuming that all moves were upwards in memory; this was true
1747            a long time ago when I wrote it, but is no longer the case.
1749        The emulated memove() is provided for those environments that have neither
1750        memmove() nor bcopy(). I didn't think anyone used it these days, but that
1751        is clearly not the case, as these two bugs were recently reported.
1753    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
1754        and Detrail to create the HTML documentation, the .txt form of the man
1755        pages, and it removes trailing spaces from listed files. It also creates
1756        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
1757        case, it wraps all the #defines with #ifndefs. This script should be run
1758        before "make dist".
1760    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1761        with Unicode property support.
1763        (a) For a maximizing quantifier, if the two different cases of the
1764            character were of different lengths in their UTF-8 codings (there are
1765            some cases like this - I found 11), and the matching function had to
1766            back up over a mixture of the two cases, it incorrectly assumed they
1767            were both the same length.
1769        (b) When PCRE was configured to use the heap rather than the stack for
1770            recursion during matching, it was not correctly preserving the data for
1771            the other case of a UTF-8 character when checking ahead for a match
1772            while processing a minimizing repeat. If the check also involved
1773            matching a wide character, but failed, corruption could cause an
1774            erroneous result when trying to check for a repeat of the original
1775            character.
1777    13. Some tidying changes to the testing mechanism:
1779        (a) The RunTest script now detects the internal link size and whether there
1780            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1781            values substituted by "configure". (The RunGrepTest script already did
1782            this for UTF-8.) The configure.ac script no longer substitutes the
1783            relevant variables.
1785        (b) The debugging options /B and /D in pcretest show the compiled bytecode
1786            with length and offset values. This means that the output is different
1787            for different internal link sizes. Test 2 is skipped for link sizes
1788            other than 2 because of this, bypassing the problem. Unfortunately,
1789            there was also a test in test 3 (the locale tests) that used /B and
1790            failed for link sizes other than 2. Rather than cut the whole test out,
1791            I have added a new /Z option to pcretest that replaces the length and
1792            offset values with spaces. This is now used to make test 3 independent
1793            of link size. (Test 2 will be tidied up later.)
1795    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1796        segmentation fault instead of returning the appropriate error message.
1798    15. In multiline mode when the newline sequence was set to "any", the pattern
1799        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1800        This doesn't seem right; it now treats the CRLF combination as the line
1801        ending, and so does not match in that case. It's only a pattern such as ^$
1802        that would hit this one: something like ^ABC$ would have failed after \r
1803        and then tried again after \r\n.
1805    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1806        in an attempt to make files that differ only in their line terminators
1807        compare equal. This works on Linux.
1809    17. Under certain error circumstances pcregrep might try to free random memory
1810        as it exited. This is now fixed, thanks to valgrind.
1812    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1813        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1814        was because its rules for how to advance for /g after matching an empty
1815        string at the end of a line did not allow for this case. They now check for
1816        it specially.
1818    20. pcretest is supposed to handle patterns and data of any length, by
1819        extending its buffers when necessary. It was getting this wrong when the
1820        buffer for a data line had to be extended.
1822    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1823        CRLF as a newline sequence.
1825    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1826        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1827        I have nevertheless tidied it up.
1829    23. Added some casts to kill warnings from HP-UX ia64 compiler.
1831    24. Added a man page for pcre-config.
1834    Version 7.0 19-Dec-06
1835    ---------------------
1837     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1838        moving to gcc 4.1.1.
1840     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1841        sys/time.h, which is documented as needed for this function. It doesn't
1842        seem to matter on Linux, but it showed up on some releases of OS X.
1844     3. It seems that there are systems where bytes whose values are greater than
1845        127 match isprint() in the "C" locale. The "C" locale should be the
1846        default when a C program starts up. In most systems, only ASCII printing
1847        characters match isprint(). This difference caused the output from pcretest
1848        to vary, making some of the tests fail. I have changed pcretest so that:
1850        (a) When it is outputting text in the compiled version of a pattern, bytes
1851            other than 32-126 are always shown as hex escapes.
1853        (b) When it is outputting text that is a matched part of a subject string,
1854            it does the same, unless a different locale has been set for the match
1855            (using the /L modifier). In this case, it uses isprint() to decide.
1857     4. Fixed a major bug that caused incorrect computation of the amount of memory
1858        required for a compiled pattern when options that changed within the
1859        pattern affected the logic of the preliminary scan that determines the
1860        length. The relevant options are -x, and -i in UTF-8 mode. The result was
1861        that the computed length was too small. The symptoms of this bug were
1862        either the PCRE error "internal error: code overflow" from pcre_compile(),
1863        or a glibc crash with a message such as "pcretest: free(): invalid next
1864        size (fast)". Examples of patterns that provoked this bug (shown in
1865        pcretest format) are:
1867          /(?-x: )/x
1868          /(?x)(?-x: \s*#\s*)/
1869          /((?i)[\x{c0}])/8
1870          /(?i:[\x{c0}])/8
1872        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1873        is now done differently.
1875     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1876        wrapper classes; (b) implement a new function in the C++ scanner that is
1877        more efficient than the old way of doing things because it avoids levels of
1878        recursion in the regex matching; (c) add a paragraph to the documentation
1879        for the FullMatch() function.
1881     6. The escape sequence \n was being treated as whatever was defined as
1882        "newline". Not only was this contrary to the documentation, which states
1883        that \n is character 10 (hex 0A), but it also went horribly wrong when
1884        "newline" was defined as CRLF. This has been fixed.
1886     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1887        was being set to -1 for the "end of line" case (supposedly a value that no
1888        character can have). Though this value is never used (the check for end of
1889        line is "zero bytes in current character"), it caused compiler complaints.
1890        I've changed it to 0xffffffff.
1892     8. In pcre_version.c, the version string was being built by a sequence of
1893        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1894        string (as it is for production releases) called a macro with an empty
1895        argument. The C standard says the result of this is undefined. The gcc
1896        compiler treats it as an empty string (which was what was wanted) but it is
1897        reported that Visual C gives an error. The source has been hacked around to
1898        avoid this problem.
1900     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1901        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1902        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1903        of them did).
1905    10. Originally, pcretest opened its input and output without "b"; then I was
1906        told that "b" was needed in some environments, so it was added for release
1907        5.0 to both the input and output. (It makes no difference on Unix-like
1908        systems.) Later I was told that it is wrong for the input on Windows. I've
1909        now abstracted the modes into two macros, to make it easier to fiddle with
1910        them, and removed "b" from the input mode under Windows.
1912    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1914    12. Added -help and --help to pcretest as an official way of being reminded
1915        of the options.
1917    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1918        and pcrecpp.cc because they annoy compilers at high warning levels.
1920    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1922    15. Fixed an occurrence of == in configure.ac that should have been = (shell
1923        scripts are not C programs :-) and which was not noticed because it works
1924        on Linux.
1926    16. pcretest is supposed to handle any length of pattern and data line (as one
1927        line or as a continued sequence of lines) by extending its input buffer if
1928        necessary. This feature was broken for very long pattern lines, leading to
1929        a string of junk being passed to pcre_compile() if the pattern was longer
1930        than about 50K.
1932    17. I have done a major re-factoring of the way pcre_compile() computes the
1933        amount of memory needed for a compiled pattern. Previously, there was code
1934        that made a preliminary scan of the pattern in order to do this. That was
1935        OK when PCRE was new, but as the facilities have expanded, it has become
1936        harder and harder to keep it in step with the real compile phase, and there
1937        have been a number of bugs (see for example, 4 above). I have now found a
1938        cunning way of running the real compile function in a "fake" mode that
1939        enables it to compute how much memory it would need, while actually only
1940        ever using a few hundred bytes of working memory and without too many
1941        tests of the mode. This should make future maintenance and development
1942        easier. A side effect of this work is that the limit of 200 on the nesting
1943        depth of parentheses has been removed (though this was never a serious
1944        limitation, I suspect). However, there is a downside: pcre_compile() now
1945        runs more slowly than before (30% or more, depending on the pattern). I
1946        hope this isn't a big issue. There is no effect on runtime performance.
1948    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1949        newline (only possible for the last line of a file) and it was a
1950        pattern that set a locale (followed by /Lsomething), pcretest crashed.
1952    19. Added additional timing features to pcretest. (1) The -tm option now times
1953        matching only, not compiling. (2) Both -t and -tm can be followed, as a
1954        separate command line item, by a number that specifies the number of
1955        repeats to use when timing. The default is 50000; this gives better
1956        precision, but takes uncomfortably long for very large patterns.
1958    20. Extended pcre_study() to be more clever in cases where a branch of a
1959        subpattern has no definite first character. For example, (a*|b*)[cd] would
1960        previously give no result from pcre_study(). Now it recognizes that the
1961        first character must be a, b, c, or d.
1963    21. There was an incorrect error "recursive call could loop indefinitely" if
1964        a subpattern (or the entire pattern) that was being tested for matching an
1965        empty string contained only one non-empty item after a nested subpattern.
1966        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1967        incorrectly, because the \d was being skipped in the check.
1969    22. The pcretest program now has a new pattern option /B and a command line
1970        option -b, which is equivalent to adding /B to every pattern. This causes
1971        it to show the compiled bytecode, without the additional information that
1972        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1973        is the same as /B/I).
1975    23. A new optimization is now able automatically to treat some sequences such
1976        as a*b as a*+b. More specifically, if something simple (such as a character
1977        or a simple class like \d) has an unlimited quantifier, and is followed by
1978        something that cannot possibly match the quantified thing, the quantifier
1979        is automatically "possessified".
1981    24. A recursive reference to a subpattern whose number was greater than 39
1982        went wrong under certain circumstances in UTF-8 mode. This bug could also
1983        have affected the operation of pcre_study().
1985    25. Realized that a little bit of performance could be had by replacing
1986        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1988    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1990    27. Possessive quantifiers such as a++ were previously implemented by turning
1991        them into atomic groups such as ($>a+). Now they have their own opcodes,
1992        which improves performance. This includes the automatically created ones
1993        from 23 above.
1995    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1996        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1997        the first matched character to be a colon. This applied both to named and
1998        numbered groups.
2000    29. The ucpinternal.h header file was missing its idempotency #ifdef.
2002    30. I was sent a "project" file called libpcre.a.dev which I understand makes
2003        building PCRE on Windows easier, so I have included it in the distribution.
2005    31. There is now a check in pcretest against a ridiculously large number being
2006        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
2007        loop, the loop is abandoned.
2009    32. Forward references to subpatterns in conditions such as (?(2)...) where
2010        subpattern 2 is defined later cause pcre_compile() to search forwards in
2011        the pattern for the relevant set of parentheses. This search went wrong
2012        when there were unescaped parentheses in a character class, parentheses
2013        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
2015    33. "Subroutine" calls and backreferences were previously restricted to
2016        referencing subpatterns earlier in the regex. This restriction has now
2017        been removed.
2019    34. Added a number of extra features that are going to be in Perl 5.10. On the
2020        whole, these are just syntactic alternatives for features that PCRE had
2021        previously implemented using the Python syntax or my own invention. The
2022        other formats are all retained for compatibility.
2024        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
2025            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
2026            also .NET compatible.
2028        (b) A recursion or subroutine call to a named group can now be defined as
2029            (?&name) as well as (?P>name).
2031        (c) A backreference to a named group can now be defined as \k<name> or
2032            \k'name' as well as (?P=name). The new forms, as well as being in Perl
2033            5.10, are also .NET compatible.
2035        (d) A conditional reference to a named group can now use the syntax
2036            (?(<name>) or (?('name') as well as (?(name).
2038        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
2039            groups (named and numbered) that are never evaluated inline, but can be
2040            called as "subroutines" from elsewhere. In effect, the DEFINE condition
2041            is always false. There may be only one alternative in such a group.
2043        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
2044            as the simple (?(R). The condition is true only if the most recent
2045            recursion is that of the given number or name. It does not search out
2046            through the entire recursion stack.
2048        (g) The escape \gN or \g{N} has been added, where N is a positive or
2049            negative number, specifying an absolute or relative reference.
2051    35. Tidied to get rid of some further signed/unsigned compiler warnings and
2052        some "unreachable code" warnings.
2054    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
2055        things, this adds five new scripts.
2057    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
2058        There were also incompatibilities regarding the handling of \Q..\E inside
2059        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
2060        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
2062    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
2063        matches an empty string, and forcibly breaks the loop. There were bugs in
2064        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
2065        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
2066        separate and independent bugs (that affected different cases) have been
2067        fixed.
2069    39. Refactored the code to abolish the use of different opcodes for small
2070        capturing bracket numbers. This is a tidy that I avoided doing when I
2071        removed the limit on the number of capturing brackets for 3.5 back in 2001.
2072        The new approach is not only tidier, it makes it possible to reduce the
2073        memory needed to fix the previous bug (38).
2075    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
2076        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
2077        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
2078        mode.
2080    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
2081        report.
2083    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
2084        copy construction and assignment in the C++ wrapper.
2086    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
2087        couple of bugs that could have given wrong results in the "--newline=crlf"
2088        case.
2090    44. Added a number of casts and did some reorganization of signed/unsigned int
2091        variables following suggestions from Dair Grant. Also renamed the variable
2092        "this" as "item" because it is a C++ keyword.
2094    45. Arranged for dftables to add
2096          #include "pcre_internal.h"
2098        to pcre_chartables.c because without it, gcc 4.x may remove the array
2099        definition from the final binary if PCRE is built into a static library and
2100        dead code stripping is activated.
2102    46. For an unanchored pattern, if a match attempt fails at the start of a
2103        newline sequence, and the newline setting is CRLF or ANY, and the next two
2104        characters are CRLF, advance by two characters instead of one.
2107    Version 6.7 04-Jul-06
2108    ---------------------
2110     1. In order to handle tests when input lines are enormously long, pcretest has
2111        been re-factored so that it automatically extends its buffers when
2112        necessary. The code is crude, but this _is_ just a test program. The
2113        default size has been increased from 32K to 50K.
2115     2. The code in pcre_study() was using the value of the re argument before
2116        testing it for NULL. (Of course, in any sensible call of the function, it
2117        won't be NULL.)
2119     3. The memmove() emulation function in pcre_internal.h, which is used on
2120        systems that lack both memmove() and bcopy() - that is, hardly ever -
2121        was missing a "static" storage class specifier.
2123     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
2124        containing an extended class (one that cannot be represented by a bitmap
2125        because it contains high-valued characters or Unicode property items, e.g.
2126        [\pZ]). Almost always one would set UTF-8 mode when processing such a
2127        pattern, but PCRE should not loop if you do not (it no longer does).
2128        [Detail: two cases were found: (a) a repeated subpattern containing an
2129        extended class; (b) a recursive reference to a subpattern that followed a
2130        previous extended class. It wasn't skipping over the extended class
2131        correctly when UTF-8 mode was not set.]
2133     5. A negated single-character class was not being recognized as fixed-length
2134        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
2135        compile error "lookbehind assertion is not fixed length".
2137     6. The RunPerlTest auxiliary script was showing an unexpected difference
2138        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
2139        write a Perl script that can interpret lines of an input file either as
2140        byte characters or as UTF-8, which is what "perltest" was being required to
2141        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
2142        can't do is switch easily at run time between having the "use utf8;" pragma
2143        or not. In the end, I fudged it by using the RunPerlTest script to insert
2144        "use utf8;" explicitly for the UTF-8 tests.
2146     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
2147        the end of the subject string, contrary to the documentation and to what
2148        Perl does. This was true of both matching functions. Now it matches only at
2149        the start of the subject and immediately after *internal* newlines.
2151     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
2152        a pointer to an int instead of a pointer to an unsigned long int. This
2153        caused problems on 64-bit systems.
2155     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
2156        instance of the 'standard' template library not being so standard".
2158    10. There was no check on the number of named subpatterns nor the maximum
2159        length of a subpattern name. The product of these values is used to compute
2160        the size of the memory block for a compiled pattern. By supplying a very
2161        long subpattern name and a large number of named subpatterns, the size
2162        computation could be caused to overflow. This is now prevented by limiting
2163        the length of names to 32 characters, and the number of named subpatterns
2164        to 10,000.
2166    11. Subpatterns that are repeated with specific counts have to be replicated in
2167        the compiled pattern. The size of memory for this was computed from the
2168        length of the subpattern and the repeat count. The latter is limited to
2169        65535, but there was no limit on the former, meaning that integer overflow
2170        could in principle occur. The compiled length of a repeated subpattern is
2171        now limited to 30,000 bytes in order to prevent this.
2173    12. Added the optional facility to have named substrings with the same name.
2175    13. Added the ability to use a named substring as a condition, using the
2176        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
2177        are numbers (not recommended). Forward references are permitted.
2179    14. Added forward references in named backreferences (if you see what I mean).
2181    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
2182        pattern could run off the end of the subject. For example, the pattern
2183        "(?s)(.{1,5})"8 did this with the subject "ab".
2185    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
2186        PCRE_CASELESS was set when matching characters that were quantified with ?
2187        or *.
2189    17. A character class other than a single negated character that had a minimum
2190        but no maximum quantifier - for example [ab]{6,} - was not handled
2191        correctly by pce_dfa_exec(). It would match only one character.
2193    18. A valid (though odd) pattern that looked like a POSIX character
2194        class but used an invalid character after [ (for example [[,abc,]]) caused
2195        pcre_compile() to give the error "Failed: internal error: code overflow" or
2196        in some cases to crash with a glibc free() error. This could even happen if
2197        the pattern terminated after [[ but there just happened to be a sequence of
2198        letters, a binary zero, and a closing ] in the memory that followed.
2200    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
2201        over the years. Originally (before any Unicode support), just the bottom 8
2202        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
2203        output from "man perlunicode" includes this:
2205          The regular expression compiler produces polymorphic opcodes.  That
2206          is, the pattern adapts to the data and automatically switches to
2207          the Unicode character scheme when presented with Unicode data--or
2208          instead uses a traditional byte scheme when presented with byte
2209          data.
2211        Sadly, a wide octal escape does not cause a switch, and in a string with
2212        no other multibyte characters, these octal escapes are treated as before.
2213        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
2214        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
2215        Unicode string.
2217        I have not perpetrated such confusion in PCRE. Up till now, it took just
2218        the bottom 8 bits, as in old Perl. I have now made octal escapes with
2219        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
2220        translate to the appropriate multibyte character.
2222    29. Applied some refactoring to reduce the number of warnings from Microsoft
2223        and Borland compilers. This has included removing the fudge introduced
2224        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
2225        a warning about an unused variable.
2227    21. PCRE has not included VT (character 0x0b) in the set of whitespace
2228        characters since release 4.0, because Perl (from release 5.004) does not.
2229        [Or at least, is documented not to: some releases seem to be in conflict
2230        with the documentation.] However, when a pattern was studied with
2231        pcre_study() and all its branches started with \s, PCRE still included VT
2232        as a possible starting character. Of course, this did no harm; it just
2233        caused an unnecessary match attempt.
2235    22. Removed a now-redundant internal flag bit that recorded the fact that case
2236        dependency changed within the pattern. This was once needed for "required
2237        byte" processing, but is no longer used. This recovers a now-scarce options
2238        bit. Also moved the least significant internal flag bit to the most-
2239        significant bit of the word, which was not previously used (hangover from
2240        the days when it was an int rather than a uint) to free up another bit for
2241        the future.
2243    23. Added support for CRLF line endings as well as CR and LF. As well as the
2244        default being selectable at build time, it can now be changed at runtime
2245        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
2246        specify that it is scanning data with non-default line endings.
2248    24. Changed the definition of CXXLINK to make it agree with the definition of
2249        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
2251    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
2252        recursions. This makes a big different to stack usage for some patterns.
2254    26. If a subpattern containing a named recursion or subroutine reference such
2255        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
2256        the space required for the compiled pattern went wrong and gave too small a
2257        value. Depending on the environment, this could lead to "Failed: internal
2258        error: code overflow at offset 49" or "glibc detected double free or
2259        corruption" errors.
2261    27. Applied patches from Google (a) to support the new newline modes and (b) to
2262        advance over multibyte UTF-8 characters in GlobalReplace.
2264    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
2265        difference for some implementation of PCRE in some Windows version.
2267    29. Added some extra testing facilities to pcretest:
2269        \q<number>   in a data line sets the "match limit" value
2270        \Q<number>   in a data line sets the "match recursion limt" value
2271        -S <number>  sets the stack size, where <number> is in megabytes
2273        The -S option isn't available for Windows.
2276    Version 6.6 06-Feb-06
2277    ---------------------
2279     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
2280        in pcreposix.h. I have copied the definition from pcre.h.
2282     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
2283        because pcre.h is no longer a built file.
2285     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
2286        not normally included in the compiled code.
2289    Version 6.5 01-Feb-06
2290    ---------------------
2292     1. When using the partial match feature with pcre_dfa_exec(), it was not
2293        anchoring the second and subsequent partial matches at the new starting
2294        point. This could lead to incorrect results. For example, with the pattern
2295        /1234/, partially matching against "123" and then "a4" gave a match.
2297     2. Changes to pcregrep:
2299        (a) All non-match returns from pcre_exec() were being treated as failures
2300            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
2301            error message is output. Some extra information is given for the
2302            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
2303            probably the only errors that are likely to be caused by users (by
2304            specifying a regex that has nested indefinite repeats, for instance).
2305            If there are more than 20 of these errors, pcregrep is abandoned.
2307        (b) A binary zero was treated as data while matching, but terminated the
2308            output line if it was written out. This has been fixed: binary zeroes
2309            are now no different to any other data bytes.
2311        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
2312            used to set a locale for matching. The --locale=xxxx long option has
2313            been added (no short equivalent) to specify a locale explicitly on the
2314            pcregrep command, overriding the environment variables.
2316        (d) When -B was used with -n, some line numbers in the output were one less
2317            than they should have been.
2319        (e) Added the -o (--only-matching) option.
2321        (f) If -A or -C was used with -c (count only), some lines of context were
2322            accidentally printed for the final match.
2324        (g) Added the -H (--with-filename) option.
2326        (h) The combination of options -rh failed to suppress file names for files
2327            that were found from directory arguments.
2329        (i) Added the -D (--devices) and -d (--directories) options.
2331        (j) Added the -F (--fixed-strings) option.
2333        (k) Allow "-" to be used as a file name for -f as well as for a data file.
2335        (l) Added the --colo(u)r option.
2337        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
2338            is not present by default.
2340     3. A nasty bug was discovered in the handling of recursive patterns, that is,
2341        items such as (?R) or (?1), when the recursion could match a number of
2342        alternatives. If it matched one of the alternatives, but subsequently,
2343        outside the recursion, there was a failure, the code tried to back up into
2344        the recursion. However, because of the way PCRE is implemented, this is not
2345        possible, and the result was an incorrect result from the match.
2347        In order to prevent this happening, the specification of recursion has
2348        been changed so that all such subpatterns are automatically treated as
2349        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
2351     4. I had overlooked the fact that, in some locales, there are characters for
2352        which isalpha() is true but neither isupper() nor islower() are true. In
2353        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
2354        and ordfeminine) are like this. This affected the treatment of \w and \W
2355        when they appeared in character classes, but not when they appeared outside
2356        a character class. The bit map for "word" characters is now created
2357        separately from the results of isalnum() instead of just taking it from the
2358        upper, lower, and digit maps. (Plus the underscore character, of course.)
2360     5. The above bug also affected the handling of POSIX character classes such as
2361        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
2362        permanent tables. Instead, the bit maps for such a class were previously
2363        created as the appropriate unions of the upper, lower, and digit bitmaps.
2364        Now they are created by subtraction from the [[:word:]] class, which has
2365        its own bitmap.
2367     6. The [[:blank:]] character class matches horizontal, but not vertical space.
2368        It is created by subtracting the vertical space characters (\x09, \x0a,
2369        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
2370        subtraction was done in the overall bitmap for a character class, meaning
2371        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
2372        be recognized. This bug has been fixed.
2374     7. Patches from the folks at Google:
2376          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
2377          real life, but is still worth protecting against".
2379          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
2380          regular expressions".
2382          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
2383          have it.
2385          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
2386          "configure" and the latter not, in order to fix a problem somebody had
2387          with compiling the Arg class on HP-UX.
2389          (e) Improve the error-handling of the C++ wrapper a little bit.
2391          (f) New tests for checking recursion limiting.
2393     8. The pcre_memmove() function, which is used only if the environment does not
2394        have a standard memmove() function (and is therefore rarely compiled),
2395        contained two bugs: (a) use of int instead of size_t, and (b) it was not
2396        returning a result (though PCRE never actually uses the result).
2398     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
2399        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
2400        returned instead of calling malloc() with an overflowing number that would
2401        most likely cause subsequent chaos.
2403    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
2405    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
2406        with this option is matched, the nmatch and pmatch options of regexec() are
2407        ignored.
2409    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
2410        provided in case anyone wants to the the POSIX interface with UTF-8
2411        strings.
2413    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
2414        C++ linking (needed for some HP-UX environments).
2416    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
2417        (unused parameter) and in the pcre_printint() function (omitted "default"
2418        switch label when the default is to do nothing).
2420    15. Added some code to make it possible, when PCRE is compiled as a C++
2421        library, to replace subject pointers for pcre_exec() with a smart pointer
2422        class, thus making it possible to process discontinuous strings.
2424    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
2425        much the same function. They were added by different people who were trying
2426        to make PCRE easy to compile on non-Unix systems. It has been suggested
2427        that PCRE_EXPORT be abolished now that there is more automatic apparatus
2428        for compiling on Windows systems. I have therefore replaced it with
2429        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
2430        defaults to "extern" for C or "extern C" for C++, which works fine on
2431        Unix-like systems. It is now possible to override the value of PCRE_DATA_
2432        SCOPE with something explicit in config.h. In addition:
2434        (a) pcreposix.h still had just "extern" instead of either of these macros;
2435            I have replaced it with PCRE_DATA_SCOPE.
2437        (b) Functions such as _pcre_xclass(), which are internal to the library,
2438            but external in the C sense, all had PCRE_EXPORT in their definitions.
2439            This is apparently wrong for the Windows case, so I have removed it.
2440            (It makes no difference on Unix-like systems.)
2442    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
2443        of recursive calls to match(). This is different to MATCH_LIMIT because
2444        that limits the total number of calls to match(), not all of which increase
2445        the depth of recursion. Limiting the recursion depth limits the amount of
2446        stack (or heap if NO_RECURSE is set) that is used. The default can be set
2447        when PCRE is compiled, and changed at run time. A patch from Google adds
2448        this functionality to the C++ interface.
2450    18. Changes to the handling of Unicode character properties:
2452        (a) Updated the table to Unicode 4.1.0.
2454        (b) Recognize characters that are not in the table as "Cn" (undefined).
2456        (c) I revised the way the table is implemented to a much improved format
2457            which includes recognition of ranges. It now supports the ranges that
2458            are defined in UnicodeData.txt, and it also amalgamates other
2459            characters into ranges. This has reduced the number of entries in the
2460            table from around 16,000 to around 3,000, thus reducing its size
2461            considerably. I realized I did not need to use a tree structure after
2462            all - a binary chop search is just as efficient. Having reduced the
2463            number of entries, I extended their size from 6 bytes to 8 bytes to
2464            allow for more data.
2466        (d) Added support for Unicode script names via properties such as \p{Han}.
2468    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
2469        matching that character.
2471    20. When matching a repeated Unicode property with a minimum greater than zero,
2472        (for example \pL{2,}), PCRE could look past the end of the subject if it
2473        reached it while seeking the minimum number of characters. This could
2474        happen only if some of the characters were more than one byte long, because
2475        there is a check for at least the minimum number of bytes.
2477    21. Refactored the implementation of \p and \P so as to be more general, to
2478        allow for more different types of property in future. This has changed the
2479        compiled form incompatibly. Anybody with saved compiled patterns that use
2480        \p or \P will have to recompile them.
2482    22. Added "Any" and "L&" to the supported property types.
2484    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
2485        but give a compile time error if the value is greater than 0xff.
2487    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
2488        accidentally not being installed or uninstalled.
2490    25. The pcre.h file was built from pcre.h.in, but the only changes that were
2491        made were to insert the current release number. This seemed silly, because
2492        it made things harder for people building PCRE on systems that don't run
2493        "configure". I have turned pcre.h into a distributed file, no longer built
2494        by "configure", with the version identification directly included. There is
2495        no longer a pcre.h.in file.
2497        However, this change necessitated a change to the pcre-config script as
2498        well. It is built from pcre-config.in, and one of the substitutions was the
2499        release number. I have updated configure.ac so that ./configure now finds
2500        the release number by grepping pcre.h.
2502    26. Added the ability to run the tests under valgrind.
2505    Version 6.4 05-Sep-05
2506    ---------------------
2508     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
2509        "--" to be printed when multiple files were scanned, even when none of the
2510        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
2511        consider it to be a bug, and have restored the previous behaviour.
2513     2. A couple of code tidies to get rid of compiler warnings.
2515     3. The pcretest program used to cheat by referring to symbols in the library
2516        whose names begin with _pcre_. These are internal symbols that are not
2517        really supposed to be visible externally, and in some environments it is
2518        possible to suppress them. The cheating is now confined to including
2519        certain files from the library's source, which is a bit cleaner.
2521     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
2522        file's purpose clearer.
2524     5. Reorganized pcre_ucp_findchar().
2527    Version 6.3 15-Aug-05
2528    ---------------------
2530     1. The file libpcre.pc.in did not have general read permission in the tarball.
2532     2. There were some problems when building without C++ support:
2534        (a) If C++ support was not built, "make install" and "make test" still
2535            tried to test it.
2537        (b) There were problems when the value of CXX was explicitly set. Some
2538            changes have been made to try to fix these, and ...
2540        (c) --disable-cpp can now be used to explicitly disable C++ support.
2542        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
2543            backslash in a target when C++ was disabled. This confuses some
2544            versions of "make", apparently. Using an intermediate variable solves
2545            this. (Same for CPP_LOBJ.)
2547     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
2548        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
2549        necessary on certain architectures.
2551     4. Added a setting of -export-symbols-regex to the link command to remove
2552        those symbols that are exported in the C sense, but actually are local
2553        within the library, and not documented. Their names all begin with
2554        "_pcre_". This is not a perfect job, because (a) we have to except some
2555        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
2556        available (and never for static libraries). I have made a note to try to
2557        find a way round (a) in the future.
2560    Version 6.2 01-Aug-05
2561    ---------------------
2563     1. There was no test for integer overflow of quantifier values. A construction
2564        such as {1111111111111111} would give undefined results. What is worse, if
2565        a minimum quantifier for a parenthesized subpattern overflowed and became
2566        negative, the calculation of the memory size went wrong. This could have
2567        led to memory overwriting.
2569     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
2571     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
2572        operating environments where this matters.
2574     4. Applied Giuseppe Maxia's patch to add additional features for controlling
2575        PCRE options from within the C++ wrapper.
2577     5. Named capturing subpatterns were not being correctly counted when a pattern
2578        was compiled. This caused two problems: (a) If there were more than 100
2579        such subpatterns, the calculation of the memory needed for the whole
2580        compiled pattern went wrong, leading to an overflow error. (b) Numerical
2581        back references of the form \12, where the number was greater than 9, were
2582        not recognized as back references, even though there were sufficient
2583        previous subpatterns.
2585     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
2586        versions of gcc, e.g. 2.95.4.
2589    Version 6.1 21-Jun-05
2590    ---------------------
2592     1. There was one reference to the variable "posix" in pcretest.c that was not
2593        surrounded by "#if !defined NOPOSIX".
2595     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
2596        the cross-check on the old pcre_info() function, for the benefit of the
2597        cut-down version of PCRE that is currently imported into Exim.
2599     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
2600        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
2601        patterns that start (?i) but I don't think that matters. The use of (?i) is
2602        just an example; this all applies to the other options as well.
2604     4. Since libtool seems to echo the compile commands it is issuing, the output
2605        from "make" can be reduced a bit by putting "@" in front of each libtool
2606        compile command.
2608     5. Patch from the folks at Google for configure.in to be a bit more thorough
2609        in checking for a suitable C++ installation before trying to compile the
2610        C++ stuff. This should fix a reported problem when a compiler was present,
2611        but no suitable headers.
2613     6. The man pages all had just "PCRE" as their title. I have changed them to
2614        be the relevant file name. I have also arranged that these names are
2615        retained in the file doc/pcre.txt, which is a concatenation in text format
2616        of all the man pages except the little individual ones for each function.
2618     7. The NON-UNIX-USE file had not been updated for the different set of source
2619        files that come with release 6. I also added a few comments about the C++
2620        wrapper.
2623    Version 6.0 07-Jun-05
2624    ---------------------
2626     1. Some minor internal re-organization to help with my DFA experiments.
2628     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
2629        didn't matter for the library itself when fully configured, but did matter
2630        when compiling without UCP support, or within Exim, where the ucp files are
2631        not imported.
2633     3. Refactoring of the library code to split up the various functions into
2634        different source modules. The addition of the new DFA matching code (see
2635        below) to a single monolithic source would have made it really too
2636        unwieldy, quite apart from causing all the code to be include in a
2637        statically linked application, when only some functions are used. This is
2638        relevant even without the DFA addition now that patterns can be compiled in
2639        one application and matched in another.
2641        The downside of splitting up is that there have to be some external
2642        functions and data tables that are used internally in different modules of
2643        the library but which are not part of the API. These have all had their
2644        names changed to start with "_pcre_" so that they are unlikely to clash
2645        with other external names.
2647     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
2648        a different (DFA) algorithm. Although it is slower than the original
2649        function, it does have some advantages for certain types of matching
2650        problem.
2652     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
2653        including restarting after a partial match.
2655     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
2656        defined when compiling for Windows was sent to me. I have put it into the
2657        code, though I have no means of testing or verifying it.
2659     7. Added the pcre_refcount() auxiliary function.
2661     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
2662        match before or at the first newline in the subject string. In pcretest,
2663        the /f option on a pattern can be used to set this.
2665     9. A repeated \w when used in UTF-8 mode with characters greater than 256
2666        would behave wrongly. This has been present in PCRE since release 4.0.
2668    10. A number of changes to the pcregrep command:
2670        (a) Refactored how -x works; insert ^(...)$ instead of setting
2671            PCRE_ANCHORED and checking the length, in preparation for adding
2672            something similar for -w.
2674        (b) Added the -w (match as a word) option.
2676        (c) Refactored the way lines are read and buffered so as to have more
2677            than one at a time available.
2679        (d) Implemented a pcregrep test script.
2681        (e) Added the -M (multiline match) option. This allows patterns to match
2682            over several lines of the subject. The buffering ensures that at least
2683            8K, or the rest of the document (whichever is the shorter) is available
2684            for matching (and similarly the previous 8K for lookbehind assertions).
2686        (f) Changed the --help output so that it now says
2688              -w, --word-regex(p)
2690            instead of two lines, one with "regex" and the other with "regexp"
2691            because that confused at least one person since the short forms are the
2692            same. (This required a bit of code, as the output is generated
2693            automatically from a table. It wasn't just a text change.)
2695        (g) -- can be used to terminate pcregrep options if the next thing isn't an
2696            option but starts with a hyphen. Could be a pattern or a path name
2697            starting with a hyphen, for instance.
2699        (h) "-" can be given as a file name to represent stdin.
2701        (i) When file names are being printed, "(standard input)" is used for
2702            the standard input, for compatibility with GNU grep. Previously
2703            "<stdin>" was used.
2705        (j) The option --label=xxx can be used to supply a name to be used for
2706            stdin when file names are being printed. There is no short form.
2708        (k) Re-factored the options decoding logic because we are going to add
2709            two more options that take data. Such options can now be given in four
2710            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
2712        (l) Added the -A, -B, and -C options for requesting that lines of context
2713            around matches be printed.
2715        (m) Added the -L option to print the names of files that do not contain
2716            any matching lines, that is, the complement of -l.
2718        (n) The return code is 2 if any file cannot be opened, but pcregrep does
2719            continue to scan other files.
2721        (o) The -s option was incorrectly implemented. For compatibility with other
2722            greps, it now suppresses the error message for a non-existent or non-
2723            accessible file (but not the return code). There is a new option called
2724            -q that suppresses the output of matching lines, which was what -s was
2725            previously doing.
2727        (p) Added --include and --exclude options to specify files for inclusion
2728            and exclusion when recursing.
2730    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
2731        Hopefully, it now does.
2733    12. Missing cast in pcre_study().
2735    13. Added an "uninstall" target to the makefile.
2737    14. Replaced "extern" in the function prototypes in Makefile.in with
2738        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
2739        world, but is set differently for Windows.
2741    15. Added a second compiling function called pcre_compile2(). The only
2742        difference is that it has an extra argument, which is a pointer to an
2743        integer error code. When there is a compile-time failure, this is set
2744        non-zero, in addition to the error test pointer being set to point to an
2745        error message. The new argument may be NULL if no error number is required
2746        (but then you may as well call pcre_compile(), which is now just a
2747        wrapper). This facility is provided because some applications need a
2748        numeric error indication, but it has also enabled me to tidy up the way
2749        compile-time errors are handled in the POSIX wrapper.
2751    16. Added VPATH=.libs to the makefile; this should help when building with one
2752        prefix path and installing with another. (Or so I'm told by someone who
2753        knows more about this stuff than I do.)
2755    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
2756        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
2757        match everything, including newlines. This is not POSIX-compatible, but
2758        somebody wanted the feature. From pcretest it can be activated by using
2759        both the P and the s flags.
2761    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
2763    19. libpcre.pc was being incorrectly installed as executable.
2765    20. A couple of places in pcretest check for end-of-line by looking for '\n';
2766        it now also looks for '\r' so that it will work unmodified on Windows.
2768    21. Added Google's contributed C++ wrapper to the distribution.
2770    22. Added some untidy missing memory free() calls in pcretest, to keep
2771        Electric Fence happy when testing.
2775    Version 5.0 13-Sep-04
2776    ---------------------
2778     1. Internal change: literal characters are no longer packed up into items
2779        containing multiple characters in a single byte-string. Each character
2780        is now matched using a separate opcode. However, there may be more than one
2781        byte in the character in UTF-8 mode.
2783     2. The pcre_callout_block structure has two new fields: pattern_position and
2784        next_item_length. These contain the offset in the pattern to the next match
2785        item, and its length, respectively.
2787     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
2788        insertion of callouts before each pattern item. Added the /C option to
2789        pcretest to make use of this.
2791     4. On the advice of a Windows user, the lines
2793          #if defined(_WIN32) || defined(WIN32)
2794          _setmode( _fileno( stdout ), 0x8000 );
2795          #endif  /* defined(_WIN32) || defined(WIN32) */
2797        have been added to the source of pcretest. This apparently does useful
2798        magic in relation to line terminators.
2800     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
2801        for the benefit of those environments where the "b" makes a difference.
2803     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
2804        to know about it. I have put a hack into configure.in that adds in code
2805        to set GCC=yes if CC=icc. This seems to end up at a point in the
2806        generated configure script that is early enough to affect the setting of
2807        compiler options, which is what is needed, but I have no means of testing
2808        whether it really works. (The user who reported this had patched the
2809        generated configure script, which of course I cannot do.)
2811        LATER: After change 22 below (new libtool files), the configure script
2812        seems to know about icc (and also ecc). Therefore, I have commented out
2813        this hack in configure.in.
2815     7. Added support for pkg-config (2 patches were sent in).
2817     8. Negated POSIX character classes that used a combination of internal tables
2818        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2819        [[:^ascii]]. Typically, they would match almost any characters. The other
2820        POSIX classes were not broken in this way.
2822     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2823        to find the match, as PCRE was deluded into thinking that the match had to
2824        start at the start point or following a newline. The same bug applied to
2825        patterns with negative forward assertions or any backward assertions
2826        preceding ".*" at the start, unless the pattern required a fixed first
2827        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2829    10. In UTF-8 mode, when moving forwards in the subject after a failed match
2830        starting at the last subject character, bytes beyond the end of the subject
2831        string were read.
2833    11. Renamed the variable "class" as "classbits" to make life easier for C++
2834        users. (Previously there was a macro definition, but it apparently wasn't
2835        enough.)
2837    12. Added the new field "tables" to the extra data so that tables can be passed
2838        in at exec time, or the internal tables can be re-selected. This allows
2839        a compiled regex to be saved and re-used at a later time by a different
2840        program that might have everything at different addresses.
2842    13. Modified the pcre-config script so that, when run on Solaris, it shows a
2843        -R library as well as a -L library.
2845    14. The debugging options of pcretest (-d on the command line or D on a
2846        pattern) showed incorrect output for anything following an extended class
2847        that contained multibyte characters and which was followed by a quantifier.
2849    15. Added optional support for general category Unicode character properties
2850        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2851        support. It adds about 90K to the size of the library. The meanings of the
2852        inbuilt class escapes such as \d and \s have NOT been changed.
2854    16. Updated pcredemo.c to include calls to free() to release the memory for the
2855        compiled pattern.
2857    17. The generated file chartables.c was being created in the source directory
2858        instead of in the building directory. This caused the build to fail if the
2859        source directory was different from the building directory, and was
2860        read-only.
2862    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2863        file. No doubt somebody will tell me if they don't make sense... Also added
2864        Dan Mooney's comments about building on OpenVMS.
2866    19. Added support for partial matching via the PCRE_PARTIAL option for
2867        pcre_exec() and the \P data escape in pcretest.
2869    20. Extended pcretest with 3 new pattern features:
2871        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
2872              write the compiled pattern to the file whose name is "rest-of-line".
2873              This is a straight binary dump of the data, with the saved pointer to
2874              the character tables forced to be NULL. The study data, if any, is
2875              written too. After writing, pcretest reads a new pattern.
2877        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2878              compiled pattern from the given file. There must not be any
2879              occurrences of "<" in the file name (pretty unlikely); if there are,
2880              pcretest will instead treat the initial "<" as a pattern delimiter.
2881              After reading in the pattern, pcretest goes on to read data lines as
2882              usual.
2884        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2885              and 16-bit fields in a compiled pattern, to simulate a pattern that
2886              was compiled on a host of opposite endianness.
2888    21. The pcre-exec() function can now cope with patterns that were compiled on
2889        hosts of opposite endianness, with this restriction:
2891          As for any compiled expression that is saved and used later, the tables
2892          pointer field cannot be preserved; the extra_data field in the arguments
2893          to pcre_exec() should be used to pass in a tables address if a value
2894          other than the default internal tables were used at compile time.
2896    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2897        now diagnosed as an error. Previously, most of the time, a negative number
2898        would have been treated as zero, but if in addition "ovector" was passed as
2899        NULL, a crash could occur.
2901    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2902        new versions from the libtool 1.5 distribution (the last one is a copy of
2903        a file called libtool.m4). This seems to have fixed the need to patch
2904        "configure" to support Darwin 1.3 (which I used to do). However, I still
2905        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2906        workstation).
2908    24. Changed the PCRE licence to be the more standard "BSD" licence.
2911    Version 4.5 01-Dec-03
2912    ---------------------
2914     1. There has been some re-arrangement of the code for the match() function so
2915        that it can be compiled in a version that does not call itself recursively.
2916        Instead, it keeps those local variables that need separate instances for
2917        each "recursion" in a frame on the heap, and gets/frees frames whenever it
2918        needs to "recurse". Keeping track of where control must go is done by means
2919        of setjmp/longjmp. The whole thing is implemented by a set of macros that
2920        hide most of the details from the main code, and operates only if
2921        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2922        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2923        operating.
2925        To make it easier for callers to provide specially tailored get/free
2926        functions for this usage, two new functions, pcre_stack_malloc, and
2927        pcre_stack_free, are used. They are always called in strict stacking order,
2928        and the size of block requested is always the same.
2930        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2931        PCRE has been compiled to use the stack or the heap for recursion. The
2932        -C option of pcretest uses this to show which version is compiled.
2934        A new data escape \S, is added to pcretest; it causes the amounts of store
2935        obtained and freed by both kinds of malloc/free at match time to be added
2936        to the output.
2938     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2939        what's available on my current Linux desktop machine.
2941     3. When matching a UTF-8 string, the test for a valid string at the start has
2942        been extended. If start_offset is not zero, PCRE now checks that it points
2943        to a byte that is the start of a UTF-8 character. If not, it returns
2944        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2945        this is necessary because there may be backward assertions in the pattern.
2946        When matching the same subject several times, it may save resources to use
2947        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2949     4. The code for checking the validity of UTF-8 strings has been tightened so
2950        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2951        containing "overlong sequences".
2953     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2954        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2955        should have been "&", but it just so happened that all the cases this let
2956        through by mistake were picked up later in the function.
2958     6. I had used a variable called "isblank" - this is a C99 function, causing
2959        some compilers to warn. To avoid this, I renamed it (as "blankclass").
2961     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2962        prompting; (b) run "./pcretest /dev/null" at the start of the test script
2963        so the version is shown; (c) stop "make test" echoing "./RunTest".
2965     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2967     9. The prototype for memmove() for systems that don't have it was using
2968        size_t, but the inclusion of the header that defines size_t was later. I've
2969        moved the #includes for the C headers earlier to avoid this.
2971    10. Added some adjustments to the code to make it easier to compiler on certain
2972        special systems:
2974          (a) Some "const" qualifiers were missing.
2975          (b) Added the macro EXPORT before all exported functions; by default this
2976              is defined to be empty.
2977          (c) Changed the dftables auxiliary program (that builds chartables.c) so
2978              that it reads its output file name as an argument instead of writing
2979              to the standard output and assuming this can be redirected.
2981    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2982        class containing characters with values greater than 255, PCRE compilation
2983        went into a loop.
2985    12. A recursive reference to a subpattern that was within another subpattern
2986        that had a minimum quantifier of zero caused PCRE to crash. For example,
2987        (x(y(?2))z)? provoked this bug with a subject that got as far as the
2988        recursion. If the recursively-called subpattern itself had a zero repeat,
2989        that was OK.
2991    13. In pcretest, the buffer for reading a data line was set at 30K, but the
2992        buffer into which it was copied (for escape processing) was still set at
2993        1024, so long lines caused crashes.
2995    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2996        "internal error: code overflow...". This applied to any character class
2997        that was followed by a possessive quantifier.
2999    15. Modified the Makefile to add libpcre.la as a prerequisite for
3000        libpcreposix.la because I was told this is needed for a parallel build to
3001        work.
3003    16. If a pattern that contained .* following optional items at the start was
3004        studied, the wrong optimizing data was generated, leading to matching
3005        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
3006        matching string must start with a or b or c. The correct conclusion for
3007        this pattern is that a match can start with any character.
3010    Version 4.4 13-Aug-03
3011    ---------------------
3013     1. In UTF-8 mode, a character class containing characters with values between
3014        127 and 255 was not handled correctly if the compiled pattern was studied.
3015        In fixing this, I have also improved the studying algorithm for such
3016        classes (slightly).
3018     2. Three internal functions had redundant arguments passed to them. Removal
3019        might give a very teeny performance improvement.
3021     3. Documentation bug: the value of the capture_top field in a callout is *one
3022        more than* the number of the hightest numbered captured substring.
3024     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
3025        in incorrectly linking with a previously installed version. They now link
3026        explicitly with libpcre.la.
3028     5. configure.in no longer needs to recognize Cygwin specially.
3030     6. A problem in pcre.in for Windows platforms is fixed.
3032     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
3033        pcretest, it used to include the size of the study block as part of its
3034        output. Unfortunately, the structure contains a field that has a different
3035        size on different hardware architectures. This meant that the tests that
3036        showed this size failed. As the block is currently always of a fixed size,
3037        this information isn't actually particularly useful in pcretest output, so
3038        I have just removed it.
3040     8. Three pre-processor statements accidentally did not start in column 1.
3041        Sadly, there are *still* compilers around that complain, even though
3042        standard C has not required this for well over a decade. Sigh.
3044     9. In pcretest, the code for checking callouts passed small integers in the
3045        callout_data field, which is a void * field. However, some picky compilers
3046        complained about the casts involved for this on 64-bit systems. Now
3047        pcretest passes the address of the small integer instead, which should get
3048        rid of the warnings.
3050    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
3051        both compile and run time, and gives an error if an invalid UTF-8 sequence
3052        is found. There is a option for disabling this check in cases where the
3053        string is known to be correct and/or the maximum performance is wanted.
3055    11. In response to a bug report, I changed one line in Makefile.in from
3057            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
3058        to
3059            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
3061        to look similar to other lines, but I have no way of telling whether this
3062        is the right thing to do, as I do not use Windows. No doubt I'll get told
3063        if it's wrong...
3066    Version 4.3 21-May-03
3067    ---------------------
3069    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
3070       Makefile.
3072    2. Some refactoring to improve the quality of the code:
3074       (i)   The utf8_table... variables are now declared "const".
3076       (ii)  The code for \cx, which used the "case flipping" table to upper case
3077             lower case letters, now just substracts 32. This is ASCII-specific,
3078             but the whole concept of \cx is ASCII-specific, so it seems
3079             reasonable.
3081       (iii) PCRE was using its character types table to recognize decimal and
3082             hexadecimal digits in the pattern. This is silly, because it handles
3083             only 0-9, a-f, and A-F, but the character types table is locale-
3084             specific, which means strange things might happen. A private
3085             table is now used for this - though it costs 256 bytes, a table is
3086             much faster than multiple explicit tests. Of course, the standard
3087             character types table is still used for matching digits in subject
3088             strings against \d.
3090       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
3091             ending in _t are). So I've renamed it as ESC_tee.
3093    3. The first argument for regexec() in the POSIX wrapper should have been
3094       defined as "const".
3096    4. Changed pcretest to use malloc() for its buffers so that they can be
3097       Electric Fenced for debugging.
3099    5. There were several places in the code where, in UTF-8 mode, PCRE would try
3100       to read one or more bytes before the start of the subject string. Often this
3101       had no effect on PCRE's behaviour, but in some circumstances it could
3102       provoke a segmentation fault.
3104    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
3105       to try to read one or more bytes before the start of the subject string.
3107    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
3108       UTF-8 support could misbehave in various ways if the subject string
3109       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
3110       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
3111       back over UTF-8 characters.)
3114    Version 4.2 14-Apr-03
3115    ---------------------
3117    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
3119    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
3120         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
3121         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
3122         [WIN_PREFIX]: new variable, "cyg" for Cygwin
3123         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
3124           and BUILD_EXEEXT
3125         Note: automatic setting of the BUILD variables is not yet working
3126         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
3127           compile-time but not at link-time
3128         [LINK]: use for linking executables only
3129         make different versions for Windows and non-Windows
3130         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
3131           libraries
3132         [LINK_FOR_BUILD]: new variable
3133         [OBJEXT]: use throughout
3134         [EXEEXT]: use throughout
3135         <winshared>: new target
3136         <wininstall>: new target
3137         <dftables.o>: use native compiler
3138         <dftables>: use native linker
3139         <install>: handle Windows platform correctly
3140         <clean>: ditto
3141         <check>: ditto
3142         copy DLL to top builddir before testing
3144       As part of these changes, -no-undefined was removed again. This was reported
3145       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
3146       in any case.
3148    3. Some tidies to get rid of compiler warnings:
3150       . In the match_data structure, match_limit was an unsigned long int, whereas
3151         match_call_count was an int. I've made them both unsigned long ints.
3153       . In pcretest the fact that a const uschar * doesn't automatically cast to
3154         a void * provoked a warning.
3156       . Turning on some more compiler warnings threw up some "shadow" variables
3157         and a few more missing casts.
3159    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
3160       option, a class that contained a single character with a value between 128
3161       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
3163    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
3164       option, a class that contained several characters, but with at least one
3165       whose value was between 128 and 255 caused PCRE to crash.
3168    Version 4.1 12-Mar-03
3169    ---------------------
3171    1. Compiling with gcc -pedantic found a couple of places where casts were
3172    needed, and a string in dftables.c that was longer than standard compilers are
3173    required to support.
3175    2. Compiling with Sun's compiler found a few more places where the code could
3176    be tidied up in order to avoid warnings.
3178    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
3179    first of these names is deprecated in the latest Autoconf in favour of the name
3180    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
3181    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
3182    analogy I have changed it to CFLAGS_FOR_BUILD.
3184    4. Added -no-undefined to the linking command in the Makefile, because this is
3185    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
3186    linking step for the pcreposix library.
3188    5. PCRE was failing to diagnose the case of two named groups with the same
3189    name.
3191    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
3192    literal character that is needed in the subject for a match, and scans along to
3193    ensure that it is present before embarking on the full matching process. This
3194    saves time in cases of nested unlimited repeats that are never going to match.
3195    Problem: the scan can take a lot of time if the subject is very long (e.g.
3196    megabytes), thus penalizing straightforward matches. It is now done only if the
3197    amount of subject to be scanned is less than 1000 bytes.
3199    7. A lesser problem with the same optimization is that it was recording the
3200    first character of an anchored pattern as "needed", thus provoking a search
3201    right along the subject, even when the first match of the pattern was going to
3202    fail. The "needed" character is now not set for anchored patterns, unless it
3203    follows something in the pattern that is of non-fixed length. Thus, it still
3204    fulfils its original purpose of finding quick non-matches in cases of nested
3205    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
3208    Version 4.0 17-Feb-03
3209    ---------------------
3211    1. If a comment in an extended regex that started immediately after a meta-item
3212    extended to the end of string, PCRE compiled incorrect data. This could lead to
3213    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
3215    2. Moved to autoconf 2.53 and libtool 1.4.2.
3217    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
3218    the special perltest8 script is no longer needed - all the tests can be run
3219    from a single perltest script.
3221    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
3222    by \s. It has now been removed in PCRE. This means it isn't recognized as
3223    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
3224    class [:space:] *does* include VT, thereby creating a mess.
3226    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
3227    space and tab.
3229    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
3230    its new features into the main test script, reducing the number of scripts.
3232    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
3233    were backward compatible, and made the (?i) apply to the whole pattern, as if
3234    /i were given. Now it behaves more logically, and applies the option setting
3235    only to what follows. PCRE has been changed to follow suit. However, if it
3236    finds options settings right at the start of the pattern, it extracts them into
3237    the global options, as before. Thus, they show up in the info data.
3239    8. Added support for the \Q...\E escape sequence. Characters in between are
3240    treated as literals. This is slightly different from Perl in that $ and @ are
3241    also handled as literals inside the quotes. In Perl, they will cause variable
3242    interpolation. Note the following examples:
3244        Pattern            PCRE matches      Perl matches
3246        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
3247        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
3248        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
3250    For compatibility with Perl, \Q...\E sequences are recognized inside character
3251    classes as well as outside them.
3253    9. Re-organized 3 code statements in pcretest to avoid "overflow in
3254    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
3255    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
3256    signed/unsigned warnings.
3258    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
3259    option for pcretest, so I've replaced it by a simple function that does just
3260    that job.
3262    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
3263    "pcregrep -".
3265    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
3266    Java package. This provides some syntactic sugar for simple cases of what my
3267    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
3268    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
3269    item, you can use this simplified notation. Note that only makes sense with
3270    greedy quantifiers. Consequently, the use of the possessive quantifier forces
3271    greediness, whatever the setting of the PCRE_UNGREEDY option.
3273    13. A change of greediness default within a pattern was not taking effect at
3274    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
3275    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
3276    was abstracted outside.
3278    14. PCRE now supports the \G assertion. It is true when the current matching
3279    position is at the start point of the match. This differs from \A when the
3280    starting offset is non-zero. Used with the /g option of pcretest (or similar
3281    code), it works in the same way as it does for Perl's /g option. If all
3282    alternatives of a regex begin with \G, the expression is anchored to the start
3283    match position, and the "anchored" flag is set in the compiled expression.
3285    15. Some bugs concerning the handling of certain option changes within patterns
3286    have been fixed. These applied to options other than (?ims). For example,
3287    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
3288    been the other way round. Some of this was related to change 7 above.
3290    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
3291    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
3292    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
3293    POSIX classes only within a class (e.g. /[[:alpha:]]/).
3295    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
3296    mode. Unlike ".", it always matches newline, whatever the setting of
3297    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
3298    assertions. Perl allows it, but it doesn't (in general) work because it can't
3299    calculate the length of the lookbehind. At least, that's the case for Perl
3300    5.8.0 - I've been told they are going to document that it doesn't work in
3301    future.
3303    18. Added an error diagnosis for escapes that PCRE does not support: these are
3304    \L, \l, \N, \P, \p, \U, \u, and \X.
3306    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
3307    reading past the end of the pattern in cases such as /[abcd/.
3309    20. PCRE was getting more memory than necessary for patterns with classes that
3310    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
3312    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
3313    compiling PCRE for use with Virtual Pascal.
3315    22. Small fix to the Makefile to make it work properly if the build is done
3316    outside the source tree.
3318    23. Added a new extension: a condition to go with recursion. If a conditional
3319    subpattern starts with (?(R) the "true" branch is used if recursion has
3320    happened, whereas the "false" branch is used only at the top level.
3322    24. When there was a very long string of literal characters (over 255 bytes
3323    without UTF support, over 250 bytes with UTF support), the computation of how
3324    much memory was required could be incorrect, leading to segfaults or other
3325    strange effects.
3327    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
3328    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
3329    there was a subsequent back reference to those brackets. This meant that, for
3330    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
3331    possible to check for precisely this case. All we can do is abandon the
3332    optimization if .* occurs inside capturing brackets when there are any back
3333    references whatsoever. (See below for a better fix that came later.)
3335    26. The handling of the optimization for finding the first character of a
3336    non-anchored pattern, and for finding a character that is required later in the
3337    match were failing in some cases. This didn't break the matching; it just
3338    failed to optimize when it could. The way this is done has been re-implemented.
3340    27. Fixed typo in error message for invalid (?R item (it said "(?p").
3342    28. Added a new feature that provides some of the functionality that Perl
3343    provides with (?{...}). The facility is termed a "callout". The way it is done
3344    in PCRE is for the caller to provide an optional function, by setting
3345    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
3346    global variable. By default it is unset, which disables all calling out. To get
3347    the function called, the regex must include (?C) at appropriate points. This
3348    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
3349    This provides a means of identifying different callout points. When PCRE
3350    reaches such a point in the regex, if pcre_callout has been set, the external
3351    function is called. It is provided with data in a structure called
3352    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
3353    matching continues; if it returns a non-zero value, the match at the current
3354    point fails. However, backtracking will occur if possible. [This was changed
3355    later and other features added - see item 49 below.]
3357    29. pcretest is upgraded to test the callout functionality. It provides a
3358    callout function that displays information. By default, it shows the start of
3359    the match and the current position in the text. There are some new data escapes
3360    to vary what happens:
3362        \C+         in addition, show current contents of captured substrings
3363        \C-         do not supply a callout function
3364        \C!n        return 1 when callout number n is reached
3365        \C!n!m      return 1 when callout number n is reached for the mth time
3367    30. If pcregrep was called with the -l option and just a single file name, it
3368    output "<stdin>" if a match was found, instead of the file name.
3370    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
3371    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
3372    pcre_exec(). This saves a malloc/free per call. The default value of
3373    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
3374    when configuring.
3376    32. The default maximum size of a compiled pattern is 64K. There have been a
3377    few cases of people hitting this limit. The code now uses macros to handle the
3378    storing of links as offsets within the compiled pattern. It defaults to 2-byte
3379    links, but this can be changed to 3 or 4 bytes by --with-link-size when
3380    configuring. Tests 2 and 5 work only with 2-byte links because they output
3381    debugging information about compiled patterns.
3383    33. Internal code re-arrangements:
3385    (a) Moved the debugging function for printing out a compiled regex into
3386        its own source file (printint.c) and used #include to pull it into
3387        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
3388        separate copies.
3390    (b) Defined the list of op-code names for debugging as a macro in
3391        internal.h so that it is next to the definition of the opcodes.
3393    (c) Defined a table of op-code lengths for simpler skipping along compiled
3394        code. This is again a macro in internal.h so that it is next to the
3395        definition of the opcodes.
3397    34. Added support for recursive calls to individual subpatterns, along the
3398    lines of Robin Houston's patch (but implemented somewhat differently).
3400    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
3401    allow it to read and process whole directories in Win32. This code was
3402    contributed by Lionel Fourquaux; it has not been tested by me.
3404    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
3405    used to name a group. Names consist of alphanumerics and underscores, and must
3406    be unique. Back references use the syntax (?P=name) and recursive calls use
3407    (?P>name) which is a PCRE extension to the Python extension. Groups still have
3408    numbers. The function pcre_fullinfo() can be used after compilation to extract
3409    a name/number map. There are three relevant calls:
3411      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
3412      PCRE_INFO_NAMECOUNT            yields the number of entries
3413      PCRE_INFO_NAMETABLE            yields a pointer to the map.
3415    The map is a vector of fixed-size entries. The size of each entry depends on
3416    the length of the longest name used. The first two bytes of each entry are the
3417    group number, most significant byte first. There follows the corresponding
3418    name, zero terminated. The names are in alphabetical order.
3420    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
3421    case instead of 255. Making it the same both with and without UTF-8 support
3422    means that the same test output works with both.
3424    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
3425    calling malloc() with a zero argument.
3427    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
3428    optimization. I've improved things by keeping a bitmap of backreferences with
3429    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
3430    fact referenced, the optimization can be applied. It is unlikely that a
3431    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
3432    the match to follow \n) will appear inside brackets with a number greater than
3433    31, but if it does, any back reference > 31 suppresses the optimization.
3435    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
3436    of disabling numbered capturing parentheses. Any opening parenthesis that is
3437    not followed by ? behaves as if it were followed by ?: but named parentheses
3438    can still be used for capturing (and they will acquire numbers in the usual
3439    way).
3441    41. Redesigned the return codes from the match() function into yes/no/error so
3442    that errors can be passed back from deep inside the nested calls. A malloc
3443    failure while inside a recursive subpattern call now causes the
3444    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
3446    42. It is now possible to set a limit on the number of times the match()
3447    function is called in a call to pcre_exec(). This facility makes it possible to
3448    limit the amount of recursion and backtracking, though not in a directly
3449    obvious way, because the match() function is used in a number of different
3450    circumstances. The count starts from zero for each position in the subject
3451    string (for non-anchored patterns). The default limit is, for compatibility, a
3452    large number, namely 10 000 000. You can change this in two ways:
3454    (a) When configuring PCRE before making, you can use --with-match-limit=n
3455        to set a default value for the compiled library.
3457    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
3458        a different value is set. See 45 below.
3460    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
3462    43. Added a new function pcre_config(int, void *) to enable run-time extraction
3463    of things that can be changed at compile time. The first argument specifies
3464    what is wanted and the second points to where the information is to be placed.
3465    The current list of available information is:
3467      PCRE_CONFIG_UTF8
3469    The output is an integer that is set to one if UTF-8 support is available;
3470    otherwise it is set to zero.
3474    The output is an integer that it set to the value of the code that is used for
3475    newline. It is either LF (10) or CR (13).
3479    The output is an integer that contains the number of bytes used for internal
3480    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
3484    The output is an integer that contains the threshold above which the POSIX
3485    interface uses malloc() for output vectors. See item 31 above.
3489    The output is an unsigned integer that contains the default limit of the number
3490    of match() calls in a pcre_exec() execution. See 42 above.
3492    44. pcretest has been upgraded by the addition of the -C option. This causes it
3493    to extract all the available output from the new pcre_config() function, and to
3494    output it. The program then exits immediately.
3496    45. A need has arisen to pass over additional data with calls to pcre_exec() in
3497    order to support additional features. One way would have been to define
3498    pcre_exec2() (for example) with extra arguments, but this would not have been
3499    extensible, and would also have required all calls to the original function to
3500    be mapped to the new one. Instead, I have chosen to extend the mechanism that
3501    is used for passing in "extra" data from pcre_study().
3503    The pcre_extra structure is now exposed and defined in pcre.h. It currently
3504    contains the following fields:
3506      flags         a bitmap indicating which of the following fields are set
3507      study_data    opaque data from pcre_study()
3508      match_limit   a way of specifying a limit on match() calls for a specific
3509                      call to pcre_exec()
3510      callout_data  data for callouts (see 49 below)
3512    The flag bits are also defined in pcre.h, and are
3518    The pcre_study() function now returns one of these new pcre_extra blocks, with
3519    the actual study data pointed to by the study_data field, and the
3520    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
3521    before. That is, this change is entirely upwards-compatible and requires no
3522    change to existing code.
3524    If you want to pass in additional data to pcre_exec(), you can either place it
3525    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
3526    block.
3528    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
3529    data string contains the escape sequence \M, pcretest calls pcre_exec() several
3530    times with different match limits, until it finds the minimum value needed for
3531    pcre_exec() to complete. The value is then output. This can be instructive; for
3532    most simple matches the number is quite small, but for pathological cases it
3533    gets very large very quickly.
3535    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
3536    returns the size of the data block pointed to by the study_data field in a
3537    pcre_extra block, that is, the value that was passed as the argument to
3538    pcre_malloc() when PCRE was getting memory in which to place the information
3539    created by pcre_study(). The fourth argument should point to a size_t variable.
3540    pcretest has been extended so that this information is shown after a successful
3541    pcre_study() call when information about the compiled regex is being displayed.
3543    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
3544    because what follows is always an absolute path. (Later: it turns out that this
3545    is more than cosmetic for MinGW, because it doesn't like empty path
3546    components.)
3548    49. Some changes have been made to the callout feature (see 28 above):
3550    (i)  A callout function now has three choices for what it returns:
3552           0  =>  success, carry on matching
3553         > 0  =>  failure at this point, but backtrack if possible
3554         < 0  =>  serious error, return this value from pcre_exec()
3556         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
3557         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
3558         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
3559         use by callout functions. It will never be used by PCRE itself.
3561    (ii) The pcre_extra structure (see 45 above) has a void * field called
3562         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
3563         pcre_callout_block structure has a field of the same name. The contents of
3564         the field passed in the pcre_extra structure are passed to the callout
3565         function in the corresponding field in the callout block. This makes it
3566         easier to use the same callout-containing regex from multiple threads. For
3567         testing, the pcretest program has a new data escape
3569           \C*n        pass the number n (may be negative) as callout_data
3571         If the callout function in pcretest receives a non-zero value as
3572         callout_data, it returns that value.
3574    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
3575    there were some redundant $(CFLAGS) in commands that are now specified as
3576    $(LINK), which already includes $(CFLAGS).