/[pcre]/code/branches/oldtrunk8/ChangeLog
ViewVC logotype

Diff of /code/branches/oldtrunk8/ChangeLog

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

revision 59 by nigel, Sat Feb 24 21:39:54 2007 UTC revision 632 by ph10, Fri Jul 22 17:47:49 2011 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    Version 8.13 30-Apr-2011
5    ------------------------
6    
7    1.  The Unicode data tables have been updated to Unicode 6.0.0.
8    
9    2.  Two minor typos in pcre_internal.h have been fixed.
10    
11    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
12        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
13        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
14    
15    4.  There were a number of related bugs in the code for matching backrefences
16        caselessly in UTF-8 mode when codes for the characters concerned were
17        different numbers of bytes. For example, U+023A and U+2C65 are an upper
18        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
19        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
20        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
21        2-byte code at the end of the subject (it thought there wasn't enough data
22        left).
23    
24    5.  Comprehensive information about what went wrong is now returned by
25        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
26        as the output vector has at least 2 elements. The offset of the start of
27        the failing character and a reason code are placed in the vector.
28    
29    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
30        now returned is for the first byte of the failing character, instead of the
31        last byte inspected. This is an incompatible change, but I hope it is small
32        enough not to be a problem. It makes the returned offset consistent with
33        pcre_exec() and pcre_dfa_exec().
34    
35    7.  pcretest now gives a text phrase as well as the error number when
36        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
37        failure, the offset and reason code are output.
38    
39    8.  When \R was used with a maximizing quantifier it failed to skip backwards
40        over a \r\n pair if the subsequent match failed. Instead, it just skipped
41        back over a single character (\n). This seems wrong (because it treated the
42        two characters as a single entity when going forwards), conflicts with the
43        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
44        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
45        has been changed.
46    
47    9.  Some internal refactoring has changed the processing so that the handling
48        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
49        time (the PCRE_DOTALL option was changed this way some time ago: version
50        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
51        which was always a bit of a fudge. It also means that there is one less
52        argument for the match() function, which reduces its stack requirements
53        slightly. This change also fixes an incompatibility with Perl: the pattern
54        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
55    
56    10. More internal refactoring has drastically reduced the number of recursive
57        calls to match() for possessively repeated groups such as (abc)++ when
58        using pcre_exec().
59    
60    11. While implementing 10, a number of bugs in the handling of groups were
61        discovered and fixed:
62    
63        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
64        (a|)*(?1) gave a compile-time internal error.
65        ((a|)+)+  did not notice that the outer group could match an empty string.
66        (^a|^)+   was not marked as anchored.
67        (.*a|.*)+ was not marked as matching at start or after a newline.
68    
69    12. Yet more internal refactoring has removed another argument from the match()
70        function. Special calls to this function are now indicated by setting a
71        value in a variable in the "match data" data block.
72    
73    13. Be more explicit in pcre_study() instead of relying on "default" for
74        opcodes that mean there is no starting character; this means that when new
75        ones are added and accidentally left out of pcre_study(), testing should
76        pick them up.
77    
78    14. The -s option of pcretest has been documented for ages as being an old
79        synonym of -m (show memory usage). I have changed it to mean "force study
80        for every regex", that is, assume /S for every regex. This is similar to -i
81        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
82        using it. It makes it easier to run collections of tests with and without
83        study enabled, and thereby test pcre_study() more easily. All the standard
84        tests are now run with and without -s (but some patterns can be marked as
85        "never study" - see 20 below).
86    
87    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
88        restoration of the capturing data to the outer values was not happening
89        correctly.
90    
91    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
92        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
93        pattern had matched an empty string, and so incorrectly returned a no
94        match.
95    
96    17. There was optimizing code for the last branch of non-capturing parentheses,
97        and also for the obeyed branch of a conditional subexpression, which used
98        tail recursion to cut down on stack usage. Unfortunately, not that there is
99        the possibility of (*THEN) occurring in these branches, tail recursion is
100        no longer possible because the return has to be checked for (*THEN). These
101        two optimizations have therefore been removed.
102    
103    18. If a pattern containing \R was studied, it was assumed that \R always
104        matched two bytes, thus causing the minimum subject length to be
105        incorrectly computed because \R can also match just one byte.
106    
107    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
108        was incorrectly computed.
109    
110    20. If /S is present twice on a test pattern in pcretest input, it *disables*
111        studying, thereby overriding the use of -s on the command line. This is
112        necessary for one or two tests to keep the output identical in both cases.
113    
114    21. When (*ACCEPT) was used in an assertion that matched an empty string and
115        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
116    
117    22. When an atomic group that contained a capturing parenthesis was
118        successfully matched, but the branch in which it appeared failed, the
119        capturing was not being forgotten if a higher numbered group was later
120        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
121        group 1 to "a", when in fact it should be unset. This applied to multi-
122        branched capturing and non-capturing groups, repeated or not, and also to
123        positive assertions (capturing in negative assertions is not well defined
124        in PCRE) and also to nested atomic groups.
125    
126    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
127        subject after a captured substring (to make it easier to tell which of a
128        number of identical substrings has been captured).
129    
130    24. The way atomic groups are processed by pcre_exec() has been changed so that
131        if they are repeated, backtracking one repetition now resets captured
132        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
133        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
134        "aaa". Previously, it would have been "a". As part of this code
135        refactoring, the way recursive calls are handled has also been changed.
136    
137    24. If an assertion condition captured any substrings, they were not passed
138        back unless some other capturing happened later. For example, if
139        (?(?=(a))a) was matched against "a", no capturing was returned.
140    
141    25. When studying a pattern that contained subroutine calls or assertions,
142        the code for finding the minimum length of a possible match was handling
143        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
144        group 1 called group 2 while simultaneously a separate group 2 called group
145        1). A stack overflow occurred in this case. I have fixed this by limiting
146        the recursion depth to 10.
147    
148    26. Updated RunTest.bat in the distribution to the version supplied by Tom
149        Fortmann. This supports explicit test numbers on the command line, and has
150        argument validation and error reporting.
151    
152    27. An instance of \X with an unlimited repeat could fail if at any point the
153        first character it looked at was a mark character.
154    
155    28. Some minor code refactoring concerning Unicode properties and scripts
156        should reduce the stack requirement of match() slightly.
157    
158    29. Added the '=' option to pcretest to check the setting of unused capturing
159        slots at the end of the pattern, which are documented as being -1, but are
160        not included in the return count.
161    
162    30. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
163        compiled something random. Now it gives a compile-time error (as does
164        Perl).
165    
166    31. A *MARK encountered during the processing of a positive assertion is now
167        recorded and passed back (compatible with Perl).
168    
169    32. If --only-matching or --colour was set on a pcregrep call whose pattern
170        had alternative anchored branches, the search for a second match in a line
171        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
172        matched the line "0102" twice. The same bug affected patterns that started
173        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
174        twice.
175    
176    
177    Version 8.12 15-Jan-2011
178    ------------------------
179    
180    1.  Fixed some typos in the markup of the man pages, and wrote a script that
181        checks for such things as part of the documentation building process.
182    
183    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
184        --match-limit and --recursion-limit options (added for 8.11). In
185        particular, this made one of the standard tests fail. (The integer value
186        went into the wrong half of a long int.)
187    
188    3.  If the --colour option was given to pcregrep with -v (invert match), it
189        did strange things, either producing crazy output, or crashing. It should,
190        of course, ignore a request for colour when reporting lines that do not
191        match.
192    
193    4.  Another pcregrep bug caused similar problems if --colour was specified with
194        -M (multiline) and the pattern match finished with a line ending.
195    
196    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
197        matched in multiline mode, the following line was shown as part of the
198        match. This seems wrong, so I have changed it.
199    
200    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
201        the check for further matches in the same line (so they could be coloured)
202        to overrun the end of the current line. If another match was found, it was
203        incorrectly shown (and then shown again when found in the next line).
204    
205    7.  If pcregrep was compiled under Windows, there was a reference to the
206        function pcregrep_exit() before it was defined. I am assuming this was
207        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
208        reported by a user. I've moved the definition above the reference.
209    
210    
211    Version 8.11 10-Dec-2010
212    ------------------------
213    
214    1.  (*THEN) was not working properly if there were untried alternatives prior
215        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
216        backtracked to try for "b" instead of moving to the next alternative branch
217        at the same level (in this case, to look for "c"). The Perl documentation
218        is clear that when (*THEN) is backtracked onto, it goes to the "next
219        alternative in the innermost enclosing group".
220    
221    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
222        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
223        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
224        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
225        (*THEN).
226    
227    3.  If \s appeared in a character class, it removed the VT character from
228        the class, even if it had been included by some previous item, for example
229        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
230        of \s, but is part of the POSIX "space" class.)
231    
232    4.  A partial match never returns an empty string (because you can always
233        match an empty string at the end of the subject); however the checking for
234        an empty string was starting at the "start of match" point. This has been
235        changed to the "earliest inspected character" point, because the returned
236        data for a partial match starts at this character. This means that, for
237        example, /(?<=abc)def/ gives a partial match for the subject "abc"
238        (previously it gave "no match").
239    
240    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
241        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
242        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
243        has an implication that the given string is incomplete (because a partial
244        match is preferred over a full match). For this reason, these items now
245        give a partial match in this situation. [Aside: previously, the one case
246        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
247        match rather than a full match, which was wrong by the old rules, but is
248        now correct.]
249    
250    6.  There was a bug in the handling of #-introduced comments, recognized when
251        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
252        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
253        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
254        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
255        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
256        places in pcre_compile().
257    
258    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
259        comments when looking ahead for named forward references to subpatterns,
260        the only newline sequence it recognized was NL. It now handles newlines
261        according to the set newline convention.
262    
263    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
264        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
265        cater for a lack of strerror(). These oversights have been fixed.
266    
267    9.  Added --match-limit and --recursion-limit to pcregrep.
268    
269    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
270    
271    11. When the -o option was used, pcregrep was setting a return code of 1, even
272        when matches were found, and --line-buffered was not being honoured.
273    
274    12. Added an optional parentheses number to the -o and --only-matching options
275        of pcregrep.
276    
277    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
278        can match an empty string. The code to do it in pcretest and pcredemo
279        needed fixing:
280    
281        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
282            only one byte after an empty string match just before CRLF (this case
283            just got forgotten; "any" and "anycrlf" were OK).
284    
285        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
286            mode when an empty string match preceded an ASCII character followed by
287            a non-ASCII character. (The code for advancing by one character rather
288            than one byte was nonsense.)
289    
290        (c) The pcredemo.c sample program did not have any code at all to handle
291            the cases when CRLF is a valid newline sequence.
292    
293    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
294        as a starting offset was within the subject string. There is now a new
295        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
296        negative or greater than the length of the string. In order to test this,
297        pcretest is extended to allow the setting of negative starting offsets.
298    
299    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
300        starting offset points to the beginning of a UTF-8 character was
301        unnecessarily clumsy. I tidied it up.
302    
303    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
304        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
305    
306    17. Nobody had reported that the --include_dir option, which was added in
307        release 7.7 should have been called --include-dir (hyphen, not underscore)
308        for compatibility with GNU grep. I have changed it to --include-dir, but
309        left --include_dir as an undocumented synonym, and the same for
310        --exclude-dir, though that is not available in GNU grep, at least as of
311        release 2.5.4.
312    
313    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
314        characters from a string of bytes) have been redefined so as not to use
315        loops, in order to improve performance in some environments. At the same
316        time, I abstracted some of the common code into auxiliary macros to save
317        repetition (this should not affect the compiled code).
318    
319    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
320        compile-time error is now given if \c is not followed by an ASCII
321        character, that is, a byte less than 128. (In EBCDIC mode, the code is
322        different, and any byte value is allowed.)
323    
324    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
325        START_OPTIMIZE option, which is now allowed at compile time - but just
326        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
327        to pcregrep and other applications that have no direct access to PCRE
328        options. The new /Y option in pcretest sets this option when calling
329        pcre_compile().
330    
331    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
332        back references. Groups containing recursive back references were forced to
333        be atomic by that change, but in the case of named groups, the amount of
334        memory required was incorrectly computed, leading to "Failed: internal
335        error: code overflow". This has been fixed.
336    
337    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
338        pcretest.c, to avoid build problems in some Borland environments.
339    
340    
341    Version 8.10 25-Jun-2010
342    ------------------------
343    
344    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
345        THEN.
346    
347    2.  (*ACCEPT) was not working when inside an atomic group.
348    
349    3.  Inside a character class, \B is treated as a literal by default, but
350        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
351        causes the error). The code is unchanged, but I tidied the documentation.
352    
353    4.  Inside a character class, PCRE always treated \R and \X as literals,
354        whereas Perl faults them if its -w option is set. I have changed PCRE so
355        that it faults them when PCRE_EXTRA is set.
356    
357    5.  Added support for \N, which always matches any character other than
358        newline. (It is the same as "." when PCRE_DOTALL is not set.)
359    
360    6.  When compiling pcregrep with newer versions of gcc which may have
361        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
362        declared with attribute warn_unused_result" were given. Just casting the
363        result to (void) does not stop the warnings; a more elaborate fudge is
364        needed. I've used a macro to implement this.
365    
366    7.  Minor change to pcretest.c to avoid a compiler warning.
367    
368    8.  Added four artifical Unicode properties to help with an option to make
369        \s etc use properties (see next item). The new properties are: Xan
370        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
371    
372    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
373        use Unicode properties. (*UCP) at the start of a pattern can be used to set
374        this option. Modified pcretest to add /W to test this facility. Added
375        REG_UCP to make it available via the POSIX interface.
376    
377    10. Added --line-buffered to pcregrep.
378    
379    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
380        studied, and the match started with a letter with a code point greater than
381        127 whose first byte was different to the first byte of the other case of
382        the letter, the other case of this starting letter was not recognized
383        (#976).
384    
385    12. If a pattern that was studied started with a repeated Unicode property
386        test, for example, \p{Nd}+, there was the theoretical possibility of
387        setting up an incorrect bitmap of starting bytes, but fortunately it could
388        not have actually happened in practice until change 8 above was made (it
389        added property types that matched character-matching opcodes).
390    
391    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
392        possible starting bytes for non-anchored patterns.
393    
394    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
395        \R, and also a number of cases that involve Unicode properties, both
396        explicit and implicit when PCRE_UCP is set.
397    
398    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
399        input, it could crash or give wrong results if characters with values
400        greater than 0xc0 were present in the subject string. (Detail: it assumed
401        UTF-8 input when processing these items.)
402    
403    16. Added a lot of (int) casts to avoid compiler warnings in systems where
404        size_t is 64-bit (#991).
405    
406    17. Added a check for running out of memory when PCRE is compiled with
407        --disable-stack-for-recursion (#990).
408    
409    18. If the last data line in a file for pcretest does not have a newline on
410        the end, a newline was missing in the output.
411    
412    19. The default pcre_chartables.c file recognizes only ASCII characters (values
413        less than 128) in its various bitmaps. However, there is a facility for
414        generating tables according to the current locale when PCRE is compiled. It
415        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
416        characters, are recognized by isspace() and therefore were getting set in
417        these tables, and indeed these tables seem to approximate to ISO 8859. This
418        caused a problem in UTF-8 mode when pcre_study() was used to create a list
419        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
420        which of course cannot start UTF-8 characters. I have changed the code so
421        that only real ASCII characters (less than 128) and the correct starting
422        bytes for UTF-8 encodings are set for characters greater than 127 when in
423        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
424        altogether.)
425    
426    20. Added the /T option to pcretest so as to be able to run tests with non-
427        standard character tables, thus making it possible to include the tests
428        used for 19 above in the standard set of tests.
429    
430    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
431        reference to a subpattern the other side of a comment that contains an
432        opening parenthesis caused either an internal compiling error, or a
433        reference to the wrong subpattern.
434    
435    
436    Version 8.02 19-Mar-2010
437    ------------------------
438    
439    1.  The Unicode data tables have been updated to Unicode 5.2.0.
440    
441    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
442        configured.
443    
444    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
445        original author of that file, following a query about its status.
446    
447    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
448        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
449    
450    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
451        quantifier applied to a forward-referencing subroutine call, could compile
452        incorrect code or give the error "internal error: previously-checked
453        referenced subpattern not found".
454    
455    6.  Both MS Visual Studio and Symbian OS have problems with initializing
456        variables to point to external functions. For these systems, therefore,
457        pcre_malloc etc. are now initialized to local functions that call the
458        relevant global functions.
459    
460    7.  There were two entries missing in the vectors called coptable and poptable
461        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
462        I've fixed the data, and added a kludgy way of testing at compile time that
463        the lengths are correct (equal to the number of opcodes).
464    
465    8.  Following on from 7, I added a similar kludge to check the length of the
466        eint vector in pcreposix.c.
467    
468    9.  Error texts for pcre_compile() are held as one long string to avoid too
469        much relocation at load time. To find a text, the string is searched,
470        counting zeros. There was no check for running off the end of the string,
471        which could happen if a new error number was added without updating the
472        string.
473    
474    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
475    
476    11. \K was not working if it appeared in an atomic group or in a group that
477        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
478        \K is "not well defined" if used in an assertion. PCRE now accepts it if
479        the assertion is positive, but not if it is negative.
480    
481    12. Change 11 fortuitously reduced the size of the stack frame used in the
482        "match()" function of pcre_exec.c by one pointer. Forthcoming
483        implementation of support for (*MARK) will need an extra pointer on the
484        stack; I have reserved it now, so that the stack frame size does not
485        decrease.
486    
487    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
488        item in branch that calls a recursion is a subroutine call - as in the
489        second branch in the above example - was incorrectly given the compile-
490        time error "recursive call could loop indefinitely" because pcre_compile()
491        was not correctly checking the subroutine for matching a non-empty string.
492    
493    14. The checks for overrunning compiling workspace could trigger after an
494        overrun had occurred. This is a "should never occur" error, but it can be
495        triggered by pathological patterns such as hundreds of nested parentheses.
496        The checks now trigger 100 bytes before the end of the workspace.
497    
498    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
499    
500    
501    Version 8.01 19-Jan-2010
502    ------------------------
503    
504    1.  If a pattern contained a conditional subpattern with only one branch (in
505        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
506        computed the wrong minimum data length (which is of course zero for such
507        subpatterns). This could cause incorrect "no match" results.
508    
509    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
510        the pattern is reset in the first branch, pcre_compile() failed with
511        "internal error: code overflow at offset...". This happened only when
512        the reset was to the original external option setting. (An optimization
513        abstracts leading options settings into an external setting, which was the
514        cause of this.)
515    
516    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
517        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
518        assertion pattern did not match (meaning that the assertion was true), it
519        was incorrectly treated as false if the SKIP had been reached during the
520        matching. This also applied to assertions used as conditions.
521    
522    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
523        assertion subpattern, including such a pattern used as a condition,
524        unpredictable results occurred, instead of the error return
525        PCRE_ERROR_DFA_UITEM.
526    
527    5.  The C++ GlobalReplace function was not working like Perl for the special
528        situation when an empty string is matched. It now does the fancy magic
529        stuff that is necessary.
530    
531    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
532        removed. (These were left over from very, very early versions of PCRE.)
533    
534    7.  Some cosmetic changes to the code to make life easier when compiling it
535        as part of something else:
536    
537        (a) Change DEBUG to PCRE_DEBUG.
538    
539        (b) In pcre_compile(), rename the member of the "branch_chain" structure
540            called "current" as "current_branch", to prevent a collision with the
541            Linux macro when compiled as a kernel module.
542    
543        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
544            prevent a collision with the Linux macro when compiled as a kernel
545            module.
546    
547    8.  In pcre_compile() there are some checks for integer overflows that used to
548        cast potentially large values to (double). This has been changed to that
549        when building, a check for int64_t is made, and if it is found, it is used
550        instead, thus avoiding the use of floating point arithmetic. (There is no
551        other use of FP in PCRE.) If int64_t is not found, the fallback is to
552        double.
553    
554    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
555        2005 (difference between two addresses compared to an unsigned value).
556    
557    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
558        custom one, because of the following reported problem in Windows:
559    
560          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
561              under Win32.
562          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
563              therefore missing the function definition.
564          - The compiler thus generates a "C" signature for the test function.
565          - The linker fails to find the "C" function.
566          - PCRE fails to configure if asked to do so against libbz2.
567    
568    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
569        messages were output:
570    
571          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
572          rerunning libtoolize, to keep the correct libtool macros in-tree.
573          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
574    
575        I have done both of these things.
576    
577    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
578        most of the time, it *can* run out if it is given a pattern that contains a
579        runaway infinite recursion. I updated the discussion in the pcrestack man
580        page.
581    
582    13. Now that we have gone to the x.xx style of version numbers, the minor
583        version may start with zero. Using 08 or 09 is a bad idea because users
584        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
585        interpreted as invalid octal numbers. I've updated the previous comment in
586        configure.ac, and also added a check that gives an error if 08 or 09 are
587        used.
588    
589    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
590        causing partial matching to fail when the end of the subject matched \W
591        in a UTF-8 pattern where \W was quantified with a minimum of 3.
592    
593    15. There were some discrepancies between the declarations in pcre_internal.h
594        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
595        their definitions. The declarations used "const uschar *" and the
596        definitions used USPTR. Even though USPTR is normally defined as "const
597        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
598        reported that: "This difference in casting confuses some C++ compilers, for
599        example, SunCC recognizes above declarations as different functions and
600        generates broken code for hbpcre." I have changed the declarations to use
601        USPTR.
602    
603    16. GNU libtool is named differently on some systems. The autogen.sh script now
604        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
605        (FreeBSD).
606    
607    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
608        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
609        comment: "Figure out how to create a longlong from a string: strtoll and
610        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
611        instance, but it only takes 2 args instead of 3!"
612    
613    18. A subtle bug concerned with back references has been fixed by a change of
614        specification, with a corresponding code fix. A pattern such as
615        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
616        refers, was giving matches when it shouldn't. For example, xa=xaaa would
617        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
618        same bug. Such groups have to be quantified to be useful, or contained
619        inside another quantified group. (If there's no repetition, the reference
620        can never match.) The problem arises because, having left the group and
621        moved on to the rest of the pattern, a later failure that backtracks into
622        the group uses the captured value from the final iteration of the group
623        rather than the correct earlier one. I have fixed this in PCRE by forcing
624        any group that contains a reference to itself to be an atomic group; that
625        is, there cannot be any backtracking into it once it has completed. This is
626        similar to recursive and subroutine calls.
627    
628    
629    Version 8.00 19-Oct-09
630    ----------------------
631    
632    1.  The table for translating pcre_compile() error codes into POSIX error codes
633        was out-of-date, and there was no check on the pcre_compile() error code
634        being within the table. This could lead to an OK return being given in
635        error.
636    
637    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
638        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
639        in a Windows environment.
640    
641    3.  The pcregrep --count option prints the count for each file even when it is
642        zero, as does GNU grep. However, pcregrep was also printing all files when
643        --files-with-matches was added. Now, when both options are given, it prints
644        counts only for those files that have at least one match. (GNU grep just
645        prints the file name in this circumstance, but including the count seems
646        more useful - otherwise, why use --count?) Also ensured that the
647        combination -clh just lists non-zero counts, with no names.
648    
649    4.  The long form of the pcregrep -F option was incorrectly implemented as
650        --fixed_strings instead of --fixed-strings. This is an incompatible change,
651        but it seems right to fix it, and I didn't think it was worth preserving
652        the old behaviour.
653    
654    5.  The command line items --regex=pattern and --regexp=pattern were not
655        recognized by pcregrep, which required --regex pattern or --regexp pattern
656        (with a space rather than an '='). The man page documented the '=' forms,
657        which are compatible with GNU grep; these now work.
658    
659    6.  No libpcreposix.pc file was created for pkg-config; there was just
660        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
661    
662    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
663        when UCP support is not needed, by modifying the Python script that
664        generates it from Unicode data files. This should not matter if the module
665        is correctly used as a library, but I received one complaint about 50K of
666        unwanted data. My guess is that the person linked everything into his
667        program rather than using a library. Anyway, it does no harm.
668    
669    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
670        was a minimum greater than 1 for a wide character in a possessive
671        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
672        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
673        character. Chaos in the form of incorrect output or a compiling loop could
674        result.
675    
676    9.  The restrictions on what a pattern can contain when partial matching is
677        requested for pcre_exec() have been removed. All patterns can now be
678        partially matched by this function. In addition, if there are at least two
679        slots in the offset vector, the offset of the earliest inspected character
680        for the match and the offset of the end of the subject are set in them when
681        PCRE_ERROR_PARTIAL is returned.
682    
683    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
684        synonymous with PCRE_PARTIAL, for backwards compatibility, and
685        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
686        and may be more useful for multi-segment matching.
687    
688    11. Partial matching with pcre_exec() is now more intuitive. A partial match
689        used to be given if ever the end of the subject was reached; now it is
690        given only if matching could not proceed because another character was
691        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
692        string "Z", which now yields "no match" instead of "partial match". In the
693        case of pcre_dfa_exec(), "no match" is given if every matching path for the
694        final character ended with (*FAIL).
695    
696    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
697        if the pattern had a "must contain" character that was already found in the
698        earlier partial match, unless partial matching was again requested. For
699        example, with the pattern /dog.(body)?/, the "must contain" character is
700        "g". If the first part-match was for the string "dog", restarting with
701        "sbody" failed. This bug has been fixed.
702    
703    13. The string returned by pcre_dfa_exec() after a partial match has been
704        changed so that it starts at the first inspected character rather than the
705        first character of the match. This makes a difference only if the pattern
706        starts with a lookbehind assertion or \b or \B (\K is not supported by
707        pcre_dfa_exec()). It's an incompatible change, but it makes the two
708        matching functions compatible, and I think it's the right thing to do.
709    
710    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
711        so that the demonstration program is easily available in environments where
712        PCRE has not been installed from source.
713    
714    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
715        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
716        library.
717    
718    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
719        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
720        is not the first non-POSIX option to be added. Clearly some people find
721        these options useful.
722    
723    17. If a caller to the POSIX matching function regexec() passes a non-zero
724        value for nmatch with a NULL value for pmatch, the value of
725        nmatch is forced to zero.
726    
727    18. RunGrepTest did not have a test for the availability of the -u option of
728        the diff command, as RunTest does. It now checks in the same way as
729        RunTest, and also checks for the -b option.
730    
731    19. If an odd number of negated classes containing just a single character
732        interposed, within parentheses, between a forward reference to a named
733        subpattern and the definition of the subpattern, compilation crashed with
734        an internal error, complaining that it could not find the referenced
735        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
736        [The bug was that it was starting one character too far in when skipping
737        over the character class, thus treating the ] as data rather than
738        terminating the class. This meant it could skip too much.]
739    
740    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
741        /g option in pcretest when the pattern contains \K, which makes it possible
742        to have an empty string match not at the start, even when the pattern is
743        anchored. Updated pcretest and pcredemo to use this option.
744    
745    21. If the maximum number of capturing subpatterns in a recursion was greater
746        than the maximum at the outer level, the higher number was returned, but
747        with unset values at the outer level. The correct (outer level) value is
748        now given.
749    
750    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
751        PCRE did not set those parentheses (unlike Perl). I have now found a way to
752        make it do so. The string so far is captured, making this feature
753        compatible with Perl.
754    
755    23. The tests have been re-organized, adding tests 11 and 12, to make it
756        possible to check the Perl 5.10 features against Perl 5.10.
757    
758    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
759        pattern matches a fixed length string. PCRE did not allow this; now it
760        does. Neither allows recursion.
761    
762    25. I finally figured out how to implement a request to provide the minimum
763        length of subject string that was needed in order to match a given pattern.
764        (It was back references and recursion that I had previously got hung up
765        on.) This code has now been added to pcre_study(); it finds a lower bound
766        to the length of subject needed. It is not necessarily the greatest lower
767        bound, but using it to avoid searching strings that are too short does give
768        some useful speed-ups. The value is available to calling programs via
769        pcre_fullinfo().
770    
771    26. While implementing 25, I discovered to my embarrassment that pcretest had
772        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
773        study optimizations had never been tested with that matching function.
774        Oops. What is worse, even when it was passed study data, there was a bug in
775        pcre_dfa_exec() that meant it never actually used it. Double oops. There
776        were also very few tests of studied patterns with pcre_dfa_exec().
777    
778    27. If (?| is used to create subpatterns with duplicate numbers, they are now
779        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
780        on the other side of the coin, they are no longer allowed to have different
781        names, because these cannot be distinguished in PCRE, and this has caused
782        confusion. (This is a difference from Perl.)
783    
784    28. When duplicate subpattern names are present (necessarily with different
785        numbers, as required by 27 above), and a test is made by name in a
786        conditional pattern, either for a subpattern having been matched, or for
787        recursion in such a pattern, all the associated numbered subpatterns are
788        tested, and the overall condition is true if the condition is true for any
789        one of them. This is the way Perl works, and is also more like the way
790        testing by number works.
791    
792    
793    Version 7.9 11-Apr-09
794    ---------------------
795    
796    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
797        (pcretest), all targets were linked against these libraries. This included
798        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
799        libraries. This caused unwanted dependencies to be created. This problem
800        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
801        pcretest is linked with readline.
802    
803    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
804        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
805        moved outside it again, because FALSE and TRUE are already defined in AIX,
806        but BOOL is not.
807    
808    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
809        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
810    
811    4.  The pcregrep documentation said spaces were inserted as well as colons (or
812        hyphens) following file names and line numbers when outputting matching
813        lines. This is not true; no spaces are inserted. I have also clarified the
814        wording for the --colour (or --color) option.
815    
816    5.  In pcregrep, when --colour was used with -o, the list of matching strings
817        was not coloured; this is different to GNU grep, so I have changed it to be
818        the same.
819    
820    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
821        each matching line was coloured. Now it goes on to look for further matches
822        of any of the test patterns, which is the same behaviour as GNU grep.
823    
824    7.  A pattern that could match an empty string could cause pcregrep to loop; it
825        doesn't make sense to accept an empty string match in pcregrep, so I have
826        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
827        seems to be how GNU grep behaves.
828    
829    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
830        start or after a newline", because the conditional assertion was not being
831        correctly handled. The rule now is that both the assertion and what follows
832        in the first alternative must satisfy the test.
833    
834    9.  If auto-callout was enabled in a pattern with a conditional group whose
835        condition was an assertion, PCRE could crash during matching, both with
836        pcre_exec() and pcre_dfa_exec().
837    
838    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
839        used for matching.
840    
841    11. Unicode property support in character classes was not working for
842        characters (bytes) greater than 127 when not in UTF-8 mode.
843    
844    12. Added the -M command line option to pcretest.
845    
846    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
847    
848    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
849    
850    16. Added comments and documentation about mis-use of no_arg in the C++
851        wrapper.
852    
853    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
854        from Martin Jerabek that uses macro names for all relevant character and
855        string constants.
856    
857    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
858        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
859        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
860        these, but not everybody uses configure.
861    
862    19. A conditional group that had only one branch was not being correctly
863        recognized as an item that could match an empty string. This meant that an
864        enclosing group might also not be so recognized, causing infinite looping
865        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
866        with the subject "ab", where knowledge that the repeated group can match
867        nothing is needed in order to break the loop.
868    
869    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
870        exec(), but without supplying a callout function, matching went wrong.
871    
872    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
873        leak if the size of the offset vector was greater than 30. When the vector
874        is smaller, the saved offsets during recursion go onto a local stack
875        vector, but for larger vectors malloc() is used. It was failing to free
876        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
877        error, in fact).
878    
879    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
880        heapframe that is used only when UTF-8 support is enabled. This caused no
881        problem, but was untidy.
882    
883    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
884        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
885        included within another project.
886    
887    24. Steven Van Ingelgem's patches to add more options to the CMake support,
888        slightly modified by me:
889    
890          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
891              not building pcregrep.
892    
893          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
894              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
895    
896    25. Forward references, both numeric and by name, in patterns that made use of
897        duplicate group numbers, could behave incorrectly or give incorrect errors,
898        because when scanning forward to find the reference group, PCRE was not
899        taking into account the duplicate group numbers. A pattern such as
900        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
901    
902    26. Changed a few more instances of "const unsigned char *" to USPTR, making
903        the feature of a custom pointer more persuasive (as requested by a user).
904    
905    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
906        pcretest.c, inside #ifndefs, because it seems they are sometimes already
907        pre-defined.
908    
909    28. Added support for (*UTF8) at the start of a pattern.
910    
911    29. Arrange for flags added by the "release type" setting in CMake to be shown
912        in the configuration summary.
913    
914    
915    Version 7.8 05-Sep-08
916    ---------------------
917    
918    1.  Replaced UCP searching code with optimized version as implemented for Ad
919        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
920        stage table and inline lookup instead of a function, giving speed ups of 2
921        to 5 times on some simple patterns that I tested. Permission was given to
922        distribute the MultiStage2.py script that generates the tables (it's not in
923        the tarball, but is in the Subversion repository).
924    
925    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
926        scripts.
927    
928    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
929        a group with a zero qualifier. The result of the study could be incorrect,
930        or the function might crash, depending on the pattern.
931    
932    4.  Caseless matching was not working for non-ASCII characters in back
933        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
934        It now works when Unicode Property Support is available.
935    
936    5.  In pcretest, an escape such as \x{de} in the data was always generating
937        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
938        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
939        truncation.
940    
941    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
942    
943    7.  Added two (int) casts to pcregrep when printing the difference of two
944        pointers, in case they are 64-bit values.
945    
946    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
947        test 2 if it fails.
948    
949    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
950        and a #define of that name to empty if it is not externally set. This is to
951        allow users of MSVC to set it if necessary.
952    
953    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
954        the convenience functions in the pcre_get.c source file.
955    
956    11. An option change at the start of a pattern that had top-level alternatives
957        could cause overwriting and/or a crash. This command provoked a crash in
958        some environments:
959    
960          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
961    
962        This potential security problem was recorded as CVE-2008-2371.
963    
964    12. For a pattern where the match had to start at the beginning or immediately
965        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
966        pcre_dfa_exec() could read past the end of the passed subject if there was
967        no match. To help with detecting such bugs (e.g. with valgrind), I modified
968        pcretest so that it places the subject at the end of its malloc-ed buffer.
969    
970    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
971        exec() might read past the end of the data buffer in UTF-8 mode.
972    
973    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
974        the data contained the byte 0x85 as part of a UTF-8 character within its
975        first line. This applied both to normal and DFA matching.
976    
977    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
978        /^[^d]*?$/8 failed to match "abc".
979    
980    16. Added a missing copyright notice to pcrecpp_internal.h.
981    
982    17. Make it more clear in the documentation that values returned from
983        pcre_exec() in ovector are byte offsets, not character counts.
984    
985    18. Tidied a few places to stop certain compilers from issuing warnings.
986    
987    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
988        supplied by Stefan Weber. I made a further small update for 7.8 because
989        there is a change of source arrangements: the pcre_searchfuncs.c module is
990        replaced by pcre_ucd.c.
991    
992    
993    Version 7.7 07-May-08
994    ---------------------
995    
996    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
997        a string to a long long, pretend we don't even have a long long." This is
998        done by checking for the strtoq, strtoll, and _strtoi64 functions.
999    
1000    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
1001        pre-7.6 versions, which defined a global no_arg variable instead of putting
1002        it in the RE class. (See also #8 below.)
1003    
1004    3.  Remove a line of dead code, identified by coverity and reported by Nuno
1005        Lopes.
1006    
1007    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
1008    
1009        (1) The include/exclude patterns were being applied to the whole pathnames
1010            of files, instead of just to the final components.
1011    
1012        (2) If there was more than one level of directory, the subdirectories were
1013            skipped unless they satisfied the include/exclude conditions. This is
1014            inconsistent with GNU grep (and could even be seen as contrary to the
1015            pcregrep specification - which I improved to make it absolutely clear).
1016            The action now is always to scan all levels of directory, and just
1017            apply the include/exclude patterns to regular files.
1018    
1019    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
1020        --exclude_dir in the tests to avoid scanning .svn directories.
1021    
1022    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
1023        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
1024        doesn't support NULs in patterns.
1025    
1026    7.  Added some missing "const"s to declarations of static tables in
1027        pcre_compile.c and pcre_dfa_exec.c.
1028    
1029    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
1030        caused by fix #2  above. (Subsequently also a second patch to fix the
1031        first patch. And a third patch - this was a messy problem.)
1032    
1033    9.  Applied Craig's patch to remove the use of push_back().
1034    
1035    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
1036        matching function regexec().
1037    
1038    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
1039        which, however, unlike Perl's \g{...}, are subroutine calls, not back
1040        references. PCRE supports relative numbers with this syntax (I don't think
1041        Oniguruma does).
1042    
1043    12. Previously, a group with a zero repeat such as (...){0} was completely
1044        omitted from the compiled regex. However, this means that if the group
1045        was called as a subroutine from elsewhere in the pattern, things went wrong
1046        (an internal error was given). Such groups are now left in the compiled
1047        pattern, with a new opcode that causes them to be skipped at execution
1048        time.
1049    
1050    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
1051        to the way PCRE behaves:
1052    
1053        (a) A lone ] character is dis-allowed (Perl treats it as data).
1054    
1055        (b) A back reference to an unmatched subpattern matches an empty string
1056            (Perl fails the current match path).
1057    
1058        (c) A data ] in a character class must be notated as \] because if the
1059            first data character in a class is ], it defines an empty class. (In
1060            Perl it is not possible to have an empty class.) The empty class []
1061            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
1062            The negative empty class [^] matches any one character, independently
1063            of the DOTALL setting.
1064    
1065    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
1066        non-existent subpattern following a character class starting with ']' and
1067        containing () gave an internal compiling error instead of "reference to
1068        non-existent subpattern". Fortunately, when the pattern did exist, the
1069        compiled code was correct. (When scanning forwards to check for the
1070        existencd of the subpattern, it was treating the data ']' as terminating
1071        the class, so got the count wrong. When actually compiling, the reference
1072        was subsequently set up correctly.)
1073    
1074    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
1075        it was being rejected as not supported by pcre_dfa_exec(), even though
1076        other assertions are supported. I have made pcre_dfa_exec() support
1077        (*FAIL).
1078    
1079    16. The implementation of 13c above involved the invention of a new opcode,
1080        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
1081        cannot be changed at match time, I realized I could make a small
1082        improvement to matching performance by compiling OP_ALLANY instead of
1083        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
1084        on the OP_ANY path.
1085    
1086    17. Compiling pcretest on Windows with readline support failed without the
1087        following two fixes: (1) Make the unistd.h include conditional on
1088        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
1089    
1090    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
1091        ncurses library to be included for pcretest when ReadLine support is
1092        requested, but also to allow for it to be overridden. This patch came from
1093        Daniel Bergström.
1094    
1095    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
1096        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
1097        any errors with the current Unicode tables. Thanks to Peter Kankowski for
1098        spotting this.
1099    
1100    
1101    Version 7.6 28-Jan-08
1102    ---------------------
1103    
1104    1.  A character class containing a very large number of characters with
1105        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
1106        overflow.
1107    
1108    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
1109        HAVE_LONG_LONG is not defined.
1110    
1111    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
1112        bring them up to date and include new features. This patch includes:
1113    
1114        - Fixed PH's badly added libz and libbz2 support.
1115        - Fixed a problem with static linking.
1116        - Added pcredemo. [But later removed - see 7 below.]
1117        - Fixed dftables problem and added an option.
1118        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
1119            HAVE_LONG_LONG.
1120        - Added readline support for pcretest.
1121        - Added an listing of the option settings after cmake has run.
1122    
1123    4.  A user submitted a patch to Makefile that makes it easy to create
1124        "pcre.dll" under mingw when using Configure/Make. I added stuff to
1125        Makefile.am that cause it to include this special target, without
1126        affecting anything else. Note that the same mingw target plus all
1127        the other distribution libraries and programs are now supported
1128        when configuring with CMake (see 6 below) instead of with
1129        Configure/Make.
1130    
1131    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
1132        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
1133        exported in the Windows port". It has not yet been confirmed that the patch
1134        solves the problem, but it does no harm.
1135    
1136    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
1137        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
1138        with CMake, and also correct the comment about stack recursion.
1139    
1140    7.  Remove the automatic building of pcredemo from the ./configure system and
1141        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
1142        of a program that users should build themselves after PCRE is installed, so
1143        building it automatically is not really right. What is more, it gave
1144        trouble in some build environments.
1145    
1146    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
1147    
1148    
1149    Version 7.5 10-Jan-08
1150    ---------------------
1151    
1152    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
1153        values in parens when parsing an RE using the C++ wrapper."
1154    
1155    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
1156        Characters greater than 255 were excluded from the class instead of being
1157        included.
1158    
1159    3.  The same bug as (2) above applied to negated POSIX classes such as
1160        [:^space:].
1161    
1162    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
1163        defined or documented. It seems to have been a typo for PCRE_STATIC, so
1164        I have changed it.
1165    
1166    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
1167        first named subpattern) and a construct such as (?&a) would reference the
1168        first named subpattern whose name started with "a" (in other words, the
1169        length check was missing). Both these problems are fixed. "Subpattern name
1170        expected" is now given for (?&) (a zero-length name), and this patch also
1171        makes it give the same error for \k'' (previously it complained that that
1172        was a reference to a non-existent subpattern).
1173    
1174    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
1175        this is right because (?- can be followed by option settings as well as by
1176        digits. I have, however, made the messages clearer.
1177    
1178    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
1179        than the number used in the conditional) now cause a compile-time error.
1180        This is actually not compatible with Perl, which accepts such patterns, but
1181        treats the conditional as always being FALSE (as PCRE used to), but it
1182        seems to me that giving a diagnostic is better.
1183    
1184    8.  Change "alphameric" to the more common word "alphanumeric" in comments
1185        and messages.
1186    
1187    9.  Fix two occurrences of "backslash" in comments that should have been
1188        "backspace".
1189    
1190    10. Remove two redundant lines of code that can never be obeyed (their function
1191        was moved elsewhere).
1192    
1193    11. The program that makes PCRE's Unicode character property table had a bug
1194        which caused it to generate incorrect table entries for sequences of
1195        characters that have the same character type, but are in different scripts.
1196        It amalgamated them into a single range, with the script of the first of
1197        them. In other words, some characters were in the wrong script. There were
1198        thirteen such cases, affecting characters in the following ranges:
1199    
1200          U+002b0 - U+002c1
1201          U+0060c - U+0060d
1202          U+0061e - U+00612
1203          U+0064b - U+0065e
1204          U+0074d - U+0076d
1205          U+01800 - U+01805
1206          U+01d00 - U+01d77
1207          U+01d9b - U+01dbf
1208          U+0200b - U+0200f
1209          U+030fc - U+030fe
1210          U+03260 - U+0327f
1211          U+0fb46 - U+0fbb1
1212          U+10450 - U+1049d
1213    
1214    12. The -o option (show only the matching part of a line) for pcregrep was not
1215        compatible with GNU grep in that, if there was more than one match in a
1216        line, it showed only the first of them. It now behaves in the same way as
1217        GNU grep.
1218    
1219    13. If the -o and -v options were combined for pcregrep, it printed a blank
1220        line for every non-matching line. GNU grep prints nothing, and pcregrep now
1221        does the same. The return code can be used to tell if there were any
1222        non-matching lines.
1223    
1224    14. Added --file-offsets and --line-offsets to pcregrep.
1225    
1226    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
1227        infinitely looping recursion. The bug was that positive lookaheads were not
1228        being skipped when checking for a possible empty match (negative lookaheads
1229        and both kinds of lookbehind were skipped).
1230    
1231    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
1232        inclusion of <windows.h> to before rather than after the definition of
1233        INVALID_FILE_ATTRIBUTES (patch from David Byron).
1234    
1235    17. Specifying a possessive quantifier with a specific limit for a Unicode
1236        character property caused pcre_compile() to compile bad code, which led at
1237        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
1238        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
1239        caused the error; without that there was no problem.
1240    
1241    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
1242    
1243    19. Added --enable-pcretest-libreadline.
1244    
1245    20. In pcrecpp.cc, the variable 'count' was incremented twice in
1246        RE::GlobalReplace(). As a result, the number of replacements returned was
1247        double what it should be. I removed one of the increments, but Craig sent a
1248        later patch that removed the other one (the right fix) and added unit tests
1249        that check the return values (which was not done before).
1250    
1251    21. Several CMake things:
1252    
1253        (1) Arranged that, when cmake is used on Unix, the libraries end up with
1254            the names libpcre and libpcreposix, not just pcre and pcreposix.
1255    
1256        (2) The above change means that pcretest and pcregrep are now correctly
1257            linked with the newly-built libraries, not previously installed ones.
1258    
1259        (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
1260    
1261    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
1262        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
1263        UTF-8 newline character). The key issue is that the pattern starts .*;
1264        this means that the match must be either at the beginning, or after a
1265        newline. The bug was in the code for advancing after a failed match and
1266        checking that the new position followed a newline. It was not taking
1267        account of UTF-8 characters correctly.
1268    
1269    23. PCRE was behaving differently from Perl in the way it recognized POSIX
1270        character classes. PCRE was not treating the sequence [:...:] as a
1271        character class unless the ... were all letters. Perl, however, seems to
1272        allow any characters between [: and :], though of course it rejects as
1273        unknown any "names" that contain non-letters, because all the known class
1274        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
1275        for example, whereas PCRE did not - it did not recognize a POSIX character
1276        class. This seemed a bit dangerous, so the code has been changed to be
1277        closer to Perl. The behaviour is not identical to Perl, because PCRE will
1278        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
1279        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
1280        Perl does, and where it didn't before.
1281    
1282    24. Rewrite so as to remove the single use of %n from pcregrep because in some
1283        Windows environments %n is disabled by default.
1284    
1285    
1286    Version 7.4 21-Sep-07
1287    ---------------------
1288    
1289    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
1290        means that a class such as [\s] counted as "explicit reference to CR or
1291        LF". That isn't really right - the whole point of the change was to try to
1292        help when there was an actual mention of one of the two characters. So now
1293        the change happens only if \r or \n (or a literal CR or LF) character is
1294        encountered.
1295    
1296    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
1297        of both had grown to the point where there were only 3 bits left.
1298        Fortunately, there was spare space in the data structure, and so I have
1299        moved the internal flags into a new 16-bit field to free up more option
1300        bits.
1301    
1302    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
1303        but did not set the internal JCHANGED flag - either of these is enough to
1304        control the way the "get" function works - but the PCRE_INFO_JCHANGED
1305        facility is supposed to tell if (?J) was ever used, so now (?J) at the
1306        start sets both bits.
1307    
1308    4.  Added options (at build time, compile time, exec time) to change \R from
1309        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
1310    
1311    5.  doc/pcresyntax.html was missing from the distribution.
1312    
1313    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
1314        compatibility, even though it is no longer used.
1315    
1316    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
1317        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
1318        windows.h file is present (where different names are used). [This was
1319        reversed later after testing - see 16 below.]
1320    
1321    8.  Changed all #include <config.h> to #include "config.h". There were also
1322        some further <pcre.h> cases that I changed to "pcre.h".
1323    
1324    9.  When pcregrep was used with the --colour option, it missed the line ending
1325        sequence off the lines that it output.
1326    
1327    10. It was pointed out to me that arrays of string pointers cause lots of
1328        relocations when a shared library is dynamically loaded. A technique of
1329        using a single long string with a table of offsets can drastically reduce
1330        these. I have refactored PCRE in four places to do this. The result is
1331        dramatic:
1332    
1333          Originally:                          290
1334          After changing UCP table:            187
1335          After changing error message table:   43
1336          After changing table of "verbs"       36
1337          After changing table of Posix names   22
1338    
1339        Thanks to the folks working on Gregex for glib for this insight.
1340    
1341    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
1342        unicode-properties was also set.
1343    
1344    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
1345    
1346    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
1347        checked only for CRLF.
1348    
1349    14. Added casts to pcretest.c to avoid compiler warnings.
1350    
1351    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
1352    
1353    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
1354        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
1355        entirely. This removes changes made in 7 above.
1356    
1357    17. The CMake files have been updated, and there is now more information about
1358        building with CMake in the NON-UNIX-USE document.
1359    
1360    
1361    Version 7.3 28-Aug-07
1362    ---------------------
1363    
1364     1. In the rejigging of the build system that eventually resulted in 7.1, the
1365        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
1366        brackets there is not right, since it causes compilers to look for an
1367        installed pcre.h, not the version that is in the source that is being
1368        compiled (which of course may be different). I have changed it back to:
1369    
1370          #include "pcre.h"
1371    
1372        I have a vague recollection that the change was concerned with compiling in
1373        different directories, but in the new build system, that is taken care of
1374        by the VPATH setting the Makefile.
1375    
1376     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
1377        when the subject happened to end in the byte 0x85 (e.g. if the last
1378        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
1379        characters but of course it shouldn't be taken as a newline when it is part
1380        of another character. The bug was that, for an unlimited repeat of . in
1381        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
1382        characters when looking for a newline.
1383    
1384     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
1385    
1386     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
1387        in debug output.
1388    
1389     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
1390        long printing in the pcrecpp unittest when running under MinGW.
1391    
1392     6. ESC_K was left out of the EBCDIC table.
1393    
1394     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
1395        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
1396        limit also applies to "virtual nesting" when a pattern is recursive, and in
1397        this case 1000 isn't so big. I have been able to remove this limit at the
1398        expense of backing off one optimization in certain circumstances. Normally,
1399        when pcre_exec() would call its internal match() function recursively and
1400        immediately return the result unconditionally, it uses a "tail recursion"
1401        feature to save stack. However, when a subpattern that can match an empty
1402        string has an unlimited repetition quantifier, it no longer makes this
1403        optimization. That gives it a stack frame in which to save the data for
1404        checking that an empty string has been matched. Previously this was taken
1405        from the 1000-entry workspace that had been reserved. So now there is no
1406        explicit limit, but more stack is used.
1407    
1408     8. Applied Daniel's patches to solve problems with the import/export magic
1409        syntax that is required for Windows, and which was going wrong for the
1410        pcreposix and pcrecpp parts of the library. These were overlooked when this
1411        problem was solved for the main library.
1412    
1413     9. There were some crude static tests to avoid integer overflow when computing
1414        the size of patterns that contain repeated groups with explicit upper
1415        limits. As the maximum quantifier is 65535, the maximum group length was
1416        set at 30,000 so that the product of these two numbers did not overflow a
1417        32-bit integer. However, it turns out that people want to use groups that
1418        are longer than 30,000 bytes (though not repeat them that many times).
1419        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1420        made it possible to implement the integer overflow checks in a much more
1421        dynamic way, which I have now done. The artificial limitation on group
1422        length has been removed - we now have only the limit on the total length of
1423        the compiled pattern, which depends on the LINK_SIZE setting.
1424    
1425    10. Fixed a bug in the documentation for get/copy named substring when
1426        duplicate names are permitted. If none of the named substrings are set, the
1427        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1428        empty string.
1429    
1430    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1431        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1432        because the ] is interpreted as the first data character and the
1433        terminating ] is not found. PCRE has been made compatible with Perl in this
1434        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1435        cause memory overwriting.
1436    
1437    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1438        string has been matched (to stop an infinite loop). It was not recognizing
1439        a conditional subpattern that could match an empty string if that
1440        subpattern was within another subpattern. For example, it looped when
1441        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
1442        condition was not nested. This bug has been fixed.
1443    
1444    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1445        past the start of the subject in the presence of bytes with the top bit
1446        set, for example "\x8aBCD".
1447    
1448    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1449        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1450    
1451    14. Optimized (?!) to (*FAIL).
1452    
1453    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1454        This restricts code points to be within the range 0 to 0x10FFFF, excluding
1455        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1456        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1457        does: it's just the validity check that is more restrictive.
1458    
1459    16. Inserted checks for integer overflows during escape sequence (backslash)
1460        processing, and also fixed erroneous offset values for syntax errors during
1461        backslash processing.
1462    
1463    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1464        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1465    
1466    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1467        caused an overrun.
1468    
1469    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1470        something other than just ASCII characters) inside a group that had an
1471        unlimited repeat caused a loop at compile time (while checking to see
1472        whether the group could match an empty string).
1473    
1474    20. Debugging a pattern containing \p or \P could cause a crash. For example,
1475        [\P{Any}] did so. (Error in the code for printing property names.)
1476    
1477    21. An orphan \E inside a character class could cause a crash.
1478    
1479    22. A repeated capturing bracket such as (A)? could cause a wild memory
1480        reference during compilation.
1481    
1482    23. There are several functions in pcre_compile() that scan along a compiled
1483        expression for various reasons (e.g. to see if it's fixed length for look
1484        behind). There were bugs in these functions when a repeated \p or \P was
1485        present in the pattern. These operators have additional parameters compared
1486        with \d, etc, and these were not being taken into account when moving along
1487        the compiled data. Specifically:
1488    
1489        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1490            length.
1491    
1492        (b) An item such as \pL+ within a repeated group could cause crashes or
1493            loops.
1494    
1495        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1496            "reference to non-existent subpattern" error.
1497    
1498        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1499    
1500    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1501        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1502    
1503    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1504    
1505    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1506        character were causing crashes (broken optimization).
1507    
1508    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1509        \p or \P) caused a compile-time loop.
1510    
1511    28. More problems have arisen in unanchored patterns when CRLF is a valid line
1512        break. For example, the unstudied pattern [\r\n]A does not match the string
1513        "\r\nA" because change 7.0/46 below moves the current point on by two
1514        characters after failing to match at the start. However, the pattern \nA
1515        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
1516        the same is true. There doesn't seem any very clean way out of this, but
1517        what I have chosen to do makes the common cases work: PCRE now takes note
1518        of whether there can be an explicit match for \r or \n anywhere in the
1519        pattern, and if so, 7.0/46 no longer applies. As part of this change,
1520        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1521        pattern has explicit CR or LF references.
1522    
1523    29. Added (*CR) etc for changing newline setting at start of pattern.
1524    
1525    
1526    Version 7.2 19-Jun-07
1527    ---------------------
1528    
1529     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
1530        which is apparently normally available under Windows.
1531    
1532     2. Re-jig the pcregrep tests with different newline settings in an attempt
1533        to make them independent of the local environment's newline setting.
1534    
1535     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
1536    
1537     4. Some of the "internals" tests were previously cut out when the link size
1538        was not 2, because the output contained actual offsets. The recent new
1539        "Z" feature of pcretest means that these can be cut out, making the tests
1540        usable with all link sizes.
1541    
1542     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
1543        stack recursion. This gives a massive performance boost under BSD, but just
1544        a small improvement under Linux. However, it saves one field in the frame
1545        in all cases.
1546    
1547     6. Added more features from the forthcoming Perl 5.10:
1548    
1549        (a) (?-n) (where n is a string of digits) is a relative subroutine or
1550            recursion call. It refers to the nth most recently opened parentheses.
1551    
1552        (b) (?+n) is also a relative subroutine call; it refers to the nth next
1553            to be opened parentheses.
1554    
1555        (c) Conditions that refer to capturing parentheses can be specified
1556            relatively, for example, (?(-2)... or (?(+3)...
1557    
1558        (d) \K resets the start of the current match so that everything before
1559            is not part of it.
1560    
1561        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1562    
1563        (f) \g{name} is another synonym - part of Perl 5.10's unification of
1564            reference syntax.
1565    
1566        (g) (?| introduces a group in which the numbering of parentheses in each
1567            alternative starts with the same number.
1568    
1569        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
1570    
1571     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
1572        PCRE_INFO_JCHANGED.
1573    
1574     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
1575        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
1576        for detecting groups that can match an empty string.
1577    
1578     9. A pattern with a very large number of alternatives (more than several
1579        hundred) was running out of internal workspace during the pre-compile
1580        phase, where pcre_compile() figures out how much memory will be needed. A
1581        bit of new cunning has reduced the workspace needed for groups with
1582        alternatives. The 1000-alternative test pattern now uses 12 bytes of
1583        workspace instead of running out of the 4096 that are available.
1584    
1585    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
1586    
1587    11. Applied patch from Google to remove an optimization that didn't quite work.
1588        The report of the bug said:
1589    
1590          pcrecpp::RE("a*").FullMatch("aaa") matches, while
1591          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
1592          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
1593    
1594    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
1595        it matched the wrong number of bytes.
1596    
1597    
1598    Version 7.1 24-Apr-07
1599    ---------------------
1600    
1601     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
1602        that is more "standard", making use of automake and other Autotools. There
1603        is some re-arrangement of the files and adjustment of comments consequent
1604        on this.
1605    
1606     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
1607        for recursive directory scanning broke on some systems because the files
1608        are not scanned in any specific order and on different systems the order
1609        was different. A call to "sort" has been inserted into RunGrepTest for the
1610        approprate test as a short-term fix. In the longer term there may be an
1611        alternative.
1612    
1613     3. I had an email from Eric Raymond about problems translating some of PCRE's
1614        man pages to HTML (despite the fact that I distribute HTML pages, some
1615        people do their own conversions for various reasons). The problems
1616        concerned the use of low-level troff macros .br and .in. I have therefore
1617        removed all such uses from the man pages (some were redundant, some could
1618        be replaced by .nf/.fi pairs). The 132html script that I use to generate
1619        HTML has been updated to handle .nf/.fi and to complain if it encounters
1620        .br or .in.
1621    
1622     4. Updated comments in configure.ac that get placed in config.h.in and also
1623        arranged for config.h to be included in the distribution, with the name
1624        config.h.generic, for the benefit of those who have to compile without
1625        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
1626    
1627     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
1628        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
1629        makevp.bat for the current PCRE, using the additional files
1630        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
1631    
1632     6. A Windows user reported a minor discrepancy with test 2, which turned out
1633        to be caused by a trailing space on an input line that had got lost in his
1634        copy. The trailing space was an accident, so I've just removed it.
1635    
1636     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
1637        that is needed.
1638    
1639     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
1640        as "const" (a) because they are and (b) because it helps the PHP
1641        maintainers who have recently made a script to detect big data structures
1642        in the php code that should be moved to the .rodata section. I remembered
1643        to update Builducptable as well, so it won't revert if ucptable.h is ever
1644        re-created.
1645    
1646     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
1647        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
1648        order to be able to cut out the UTF-8 tables in the latter when UTF-8
1649        support is not required. This saves 1.5-2K of code, which is important in
1650        some applications.
1651    
1652        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
1653        so as not to refer to the tables, even though these functions will never be
1654        called when UTF-8 support is disabled. Otherwise there are problems with a
1655        shared library.
1656    
1657    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
1658    
1659        (a) It was defining its arguments as char * instead of void *.
1660    
1661        (b) It was assuming that all moves were upwards in memory; this was true
1662            a long time ago when I wrote it, but is no longer the case.
1663    
1664        The emulated memove() is provided for those environments that have neither
1665        memmove() nor bcopy(). I didn't think anyone used it these days, but that
1666        is clearly not the case, as these two bugs were recently reported.
1667    
1668    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
1669        and Detrail to create the HTML documentation, the .txt form of the man
1670        pages, and it removes trailing spaces from listed files. It also creates
1671        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
1672        case, it wraps all the #defines with #ifndefs. This script should be run
1673        before "make dist".
1674    
1675    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1676        with Unicode property support.
1677    
1678        (a) For a maximizing quantifier, if the two different cases of the
1679            character were of different lengths in their UTF-8 codings (there are
1680            some cases like this - I found 11), and the matching function had to
1681            back up over a mixture of the two cases, it incorrectly assumed they
1682            were both the same length.
1683    
1684        (b) When PCRE was configured to use the heap rather than the stack for
1685            recursion during matching, it was not correctly preserving the data for
1686            the other case of a UTF-8 character when checking ahead for a match
1687            while processing a minimizing repeat. If the check also involved
1688            matching a wide character, but failed, corruption could cause an
1689            erroneous result when trying to check for a repeat of the original
1690            character.
1691    
1692    13. Some tidying changes to the testing mechanism:
1693    
1694        (a) The RunTest script now detects the internal link size and whether there
1695            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1696            values substituted by "configure". (The RunGrepTest script already did
1697            this for UTF-8.) The configure.ac script no longer substitutes the
1698            relevant variables.
1699    
1700        (b) The debugging options /B and /D in pcretest show the compiled bytecode
1701            with length and offset values. This means that the output is different
1702            for different internal link sizes. Test 2 is skipped for link sizes
1703            other than 2 because of this, bypassing the problem. Unfortunately,
1704            there was also a test in test 3 (the locale tests) that used /B and
1705            failed for link sizes other than 2. Rather than cut the whole test out,
1706            I have added a new /Z option to pcretest that replaces the length and
1707            offset values with spaces. This is now used to make test 3 independent
1708            of link size. (Test 2 will be tidied up later.)
1709    
1710    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1711        segmentation fault instead of returning the appropriate error message.
1712    
1713    15. In multiline mode when the newline sequence was set to "any", the pattern
1714        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1715        This doesn't seem right; it now treats the CRLF combination as the line
1716        ending, and so does not match in that case. It's only a pattern such as ^$
1717        that would hit this one: something like ^ABC$ would have failed after \r
1718        and then tried again after \r\n.
1719    
1720    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1721        in an attempt to make files that differ only in their line terminators
1722        compare equal. This works on Linux.
1723    
1724    17. Under certain error circumstances pcregrep might try to free random memory
1725        as it exited. This is now fixed, thanks to valgrind.
1726    
1727    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1728        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1729        was because its rules for how to advance for /g after matching an empty
1730        string at the end of a line did not allow for this case. They now check for
1731        it specially.
1732    
1733    20. pcretest is supposed to handle patterns and data of any length, by
1734        extending its buffers when necessary. It was getting this wrong when the
1735        buffer for a data line had to be extended.
1736    
1737    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1738        CRLF as a newline sequence.
1739    
1740    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1741        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1742        I have nevertheless tidied it up.
1743    
1744    23. Added some casts to kill warnings from HP-UX ia64 compiler.
1745    
1746    24. Added a man page for pcre-config.
1747    
1748    
1749    Version 7.0 19-Dec-06
1750    ---------------------
1751    
1752     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1753        moving to gcc 4.1.1.
1754    
1755     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1756        sys/time.h, which is documented as needed for this function. It doesn't
1757        seem to matter on Linux, but it showed up on some releases of OS X.
1758    
1759     3. It seems that there are systems where bytes whose values are greater than
1760        127 match isprint() in the "C" locale. The "C" locale should be the
1761        default when a C program starts up. In most systems, only ASCII printing
1762        characters match isprint(). This difference caused the output from pcretest
1763        to vary, making some of the tests fail. I have changed pcretest so that:
1764    
1765        (a) When it is outputting text in the compiled version of a pattern, bytes
1766            other than 32-126 are always shown as hex escapes.
1767    
1768        (b) When it is outputting text that is a matched part of a subject string,
1769            it does the same, unless a different locale has been set for the match
1770            (using the /L modifier). In this case, it uses isprint() to decide.
1771    
1772     4. Fixed a major bug that caused incorrect computation of the amount of memory
1773        required for a compiled pattern when options that changed within the
1774        pattern affected the logic of the preliminary scan that determines the
1775        length. The relevant options are -x, and -i in UTF-8 mode. The result was
1776        that the computed length was too small. The symptoms of this bug were
1777        either the PCRE error "internal error: code overflow" from pcre_compile(),
1778        or a glibc crash with a message such as "pcretest: free(): invalid next
1779        size (fast)". Examples of patterns that provoked this bug (shown in
1780        pcretest format) are:
1781    
1782          /(?-x: )/x
1783          /(?x)(?-x: \s*#\s*)/
1784          /((?i)[\x{c0}])/8
1785          /(?i:[\x{c0}])/8
1786    
1787        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1788        is now done differently.
1789    
1790     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1791        wrapper classes; (b) implement a new function in the C++ scanner that is
1792        more efficient than the old way of doing things because it avoids levels of
1793        recursion in the regex matching; (c) add a paragraph to the documentation
1794        for the FullMatch() function.
1795    
1796     6. The escape sequence \n was being treated as whatever was defined as
1797        "newline". Not only was this contrary to the documentation, which states
1798        that \n is character 10 (hex 0A), but it also went horribly wrong when
1799        "newline" was defined as CRLF. This has been fixed.
1800    
1801     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1802        was being set to -1 for the "end of line" case (supposedly a value that no
1803        character can have). Though this value is never used (the check for end of
1804        line is "zero bytes in current character"), it caused compiler complaints.
1805        I've changed it to 0xffffffff.
1806    
1807     8. In pcre_version.c, the version string was being built by a sequence of
1808        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1809        string (as it is for production releases) called a macro with an empty
1810        argument. The C standard says the result of this is undefined. The gcc
1811        compiler treats it as an empty string (which was what was wanted) but it is
1812        reported that Visual C gives an error. The source has been hacked around to
1813        avoid this problem.
1814    
1815     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1816        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1817        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1818        of them did).
1819    
1820    10. Originally, pcretest opened its input and output without "b"; then I was
1821        told that "b" was needed in some environments, so it was added for release
1822        5.0 to both the input and output. (It makes no difference on Unix-like
1823        systems.) Later I was told that it is wrong for the input on Windows. I've
1824        now abstracted the modes into two macros, to make it easier to fiddle with
1825        them, and removed "b" from the input mode under Windows.
1826    
1827    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1828    
1829    12. Added -help and --help to pcretest as an official way of being reminded
1830        of the options.
1831    
1832    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1833        and pcrecpp.cc because they annoy compilers at high warning levels.
1834    
1835    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1836    
1837    15. Fixed an occurrence of == in configure.ac that should have been = (shell
1838        scripts are not C programs :-) and which was not noticed because it works
1839        on Linux.
1840    
1841    16. pcretest is supposed to handle any length of pattern and data line (as one
1842        line or as a continued sequence of lines) by extending its input buffer if
1843        necessary. This feature was broken for very long pattern lines, leading to
1844        a string of junk being passed to pcre_compile() if the pattern was longer
1845        than about 50K.
1846    
1847    17. I have done a major re-factoring of the way pcre_compile() computes the
1848        amount of memory needed for a compiled pattern. Previously, there was code
1849        that made a preliminary scan of the pattern in order to do this. That was
1850        OK when PCRE was new, but as the facilities have expanded, it has become
1851        harder and harder to keep it in step with the real compile phase, and there
1852        have been a number of bugs (see for example, 4 above). I have now found a
1853        cunning way of running the real compile function in a "fake" mode that
1854        enables it to compute how much memory it would need, while actually only
1855        ever using a few hundred bytes of working memory and without too many
1856        tests of the mode. This should make future maintenance and development
1857        easier. A side effect of this work is that the limit of 200 on the nesting
1858        depth of parentheses has been removed (though this was never a serious
1859        limitation, I suspect). However, there is a downside: pcre_compile() now
1860        runs more slowly than before (30% or more, depending on the pattern). I
1861        hope this isn't a big issue. There is no effect on runtime performance.
1862    
1863    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1864        newline (only possible for the last line of a file) and it was a
1865        pattern that set a locale (followed by /Lsomething), pcretest crashed.
1866    
1867    19. Added additional timing features to pcretest. (1) The -tm option now times
1868        matching only, not compiling. (2) Both -t and -tm can be followed, as a
1869        separate command line item, by a number that specifies the number of
1870        repeats to use when timing. The default is 50000; this gives better
1871        precision, but takes uncomfortably long for very large patterns.
1872    
1873    20. Extended pcre_study() to be more clever in cases where a branch of a
1874        subpattern has no definite first character. For example, (a*|b*)[cd] would
1875        previously give no result from pcre_study(). Now it recognizes that the
1876        first character must be a, b, c, or d.
1877    
1878    21. There was an incorrect error "recursive call could loop indefinitely" if
1879        a subpattern (or the entire pattern) that was being tested for matching an
1880        empty string contained only one non-empty item after a nested subpattern.
1881        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1882        incorrectly, because the \d was being skipped in the check.
1883    
1884    22. The pcretest program now has a new pattern option /B and a command line
1885        option -b, which is equivalent to adding /B to every pattern. This causes
1886        it to show the compiled bytecode, without the additional information that
1887        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1888        is the same as /B/I).
1889    
1890    23. A new optimization is now able automatically to treat some sequences such
1891        as a*b as a*+b. More specifically, if something simple (such as a character
1892        or a simple class like \d) has an unlimited quantifier, and is followed by
1893        something that cannot possibly match the quantified thing, the quantifier
1894        is automatically "possessified".
1895    
1896    24. A recursive reference to a subpattern whose number was greater than 39
1897        went wrong under certain circumstances in UTF-8 mode. This bug could also
1898        have affected the operation of pcre_study().
1899    
1900    25. Realized that a little bit of performance could be had by replacing
1901        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1902    
1903    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1904    
1905    27. Possessive quantifiers such as a++ were previously implemented by turning
1906        them into atomic groups such as ($>a+). Now they have their own opcodes,
1907        which improves performance. This includes the automatically created ones
1908        from 23 above.
1909    
1910    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1911        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1912        the first matched character to be a colon. This applied both to named and
1913        numbered groups.
1914    
1915    29. The ucpinternal.h header file was missing its idempotency #ifdef.
1916    
1917    30. I was sent a "project" file called libpcre.a.dev which I understand makes
1918        building PCRE on Windows easier, so I have included it in the distribution.
1919    
1920    31. There is now a check in pcretest against a ridiculously large number being
1921        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1922        loop, the loop is abandoned.
1923    
1924    32. Forward references to subpatterns in conditions such as (?(2)...) where
1925        subpattern 2 is defined later cause pcre_compile() to search forwards in
1926        the pattern for the relevant set of parentheses. This search went wrong
1927        when there were unescaped parentheses in a character class, parentheses
1928        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1929    
1930    33. "Subroutine" calls and backreferences were previously restricted to
1931        referencing subpatterns earlier in the regex. This restriction has now
1932        been removed.
1933    
1934    34. Added a number of extra features that are going to be in Perl 5.10. On the
1935        whole, these are just syntactic alternatives for features that PCRE had
1936        previously implemented using the Python syntax or my own invention. The
1937        other formats are all retained for compatibility.
1938    
1939        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1940            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1941            also .NET compatible.
1942    
1943        (b) A recursion or subroutine call to a named group can now be defined as
1944            (?&name) as well as (?P>name).
1945    
1946        (c) A backreference to a named group can now be defined as \k<name> or
1947            \k'name' as well as (?P=name). The new forms, as well as being in Perl
1948            5.10, are also .NET compatible.
1949    
1950        (d) A conditional reference to a named group can now use the syntax
1951            (?(<name>) or (?('name') as well as (?(name).
1952    
1953        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1954            groups (named and numbered) that are never evaluated inline, but can be
1955            called as "subroutines" from elsewhere. In effect, the DEFINE condition
1956            is always false. There may be only one alternative in such a group.
1957    
1958        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1959            as the simple (?(R). The condition is true only if the most recent
1960            recursion is that of the given number or name. It does not search out
1961            through the entire recursion stack.
1962    
1963        (g) The escape \gN or \g{N} has been added, where N is a positive or
1964            negative number, specifying an absolute or relative reference.
1965    
1966    35. Tidied to get rid of some further signed/unsigned compiler warnings and
1967        some "unreachable code" warnings.
1968    
1969    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1970        things, this adds five new scripts.
1971    
1972    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1973        There were also incompatibilities regarding the handling of \Q..\E inside
1974        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1975        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1976    
1977    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1978        matches an empty string, and forcibly breaks the loop. There were bugs in
1979        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
1980        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
1981        separate and independent bugs (that affected different cases) have been
1982        fixed.
1983    
1984    39. Refactored the code to abolish the use of different opcodes for small
1985        capturing bracket numbers. This is a tidy that I avoided doing when I
1986        removed the limit on the number of capturing brackets for 3.5 back in 2001.
1987        The new approach is not only tidier, it makes it possible to reduce the
1988        memory needed to fix the previous bug (38).
1989    
1990    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1991        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1992        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1993        mode.
1994    
1995    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1996        report.
1997    
1998    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1999        copy construction and assignment in the C++ wrapper.
2000    
2001    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
2002        couple of bugs that could have given wrong results in the "--newline=crlf"
2003        case.
2004    
2005    44. Added a number of casts and did some reorganization of signed/unsigned int
2006        variables following suggestions from Dair Grant. Also renamed the variable
2007        "this" as "item" because it is a C++ keyword.
2008    
2009    45. Arranged for dftables to add
2010    
2011          #include "pcre_internal.h"
2012    
2013        to pcre_chartables.c because without it, gcc 4.x may remove the array
2014        definition from the final binary if PCRE is built into a static library and
2015        dead code stripping is activated.
2016    
2017    46. For an unanchored pattern, if a match attempt fails at the start of a
2018        newline sequence, and the newline setting is CRLF or ANY, and the next two
2019        characters are CRLF, advance by two characters instead of one.
2020    
2021    
2022    Version 6.7 04-Jul-06
2023    ---------------------
2024    
2025     1. In order to handle tests when input lines are enormously long, pcretest has
2026        been re-factored so that it automatically extends its buffers when
2027        necessary. The code is crude, but this _is_ just a test program. The
2028        default size has been increased from 32K to 50K.
2029    
2030     2. The code in pcre_study() was using the value of the re argument before
2031        testing it for NULL. (Of course, in any sensible call of the function, it
2032        won't be NULL.)
2033    
2034     3. The memmove() emulation function in pcre_internal.h, which is used on
2035        systems that lack both memmove() and bcopy() - that is, hardly ever -
2036        was missing a "static" storage class specifier.
2037    
2038     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
2039        containing an extended class (one that cannot be represented by a bitmap
2040        because it contains high-valued characters or Unicode property items, e.g.
2041        [\pZ]). Almost always one would set UTF-8 mode when processing such a
2042        pattern, but PCRE should not loop if you do not (it no longer does).
2043        [Detail: two cases were found: (a) a repeated subpattern containing an
2044        extended class; (b) a recursive reference to a subpattern that followed a
2045        previous extended class. It wasn't skipping over the extended class
2046        correctly when UTF-8 mode was not set.]
2047    
2048     5. A negated single-character class was not being recognized as fixed-length
2049        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
2050        compile error "lookbehind assertion is not fixed length".
2051    
2052     6. The RunPerlTest auxiliary script was showing an unexpected difference
2053        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
2054        write a Perl script that can interpret lines of an input file either as
2055        byte characters or as UTF-8, which is what "perltest" was being required to
2056        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
2057        can't do is switch easily at run time between having the "use utf8;" pragma
2058        or not. In the end, I fudged it by using the RunPerlTest script to insert
2059        "use utf8;" explicitly for the UTF-8 tests.
2060    
2061     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
2062        the end of the subject string, contrary to the documentation and to what
2063        Perl does. This was true of both matching functions. Now it matches only at
2064        the start of the subject and immediately after *internal* newlines.
2065    
2066     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
2067        a pointer to an int instead of a pointer to an unsigned long int. This
2068        caused problems on 64-bit systems.
2069    
2070     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
2071        instance of the 'standard' template library not being so standard".
2072    
2073    10. There was no check on the number of named subpatterns nor the maximum
2074        length of a subpattern name. The product of these values is used to compute
2075        the size of the memory block for a compiled pattern. By supplying a very
2076        long subpattern name and a large number of named subpatterns, the size
2077        computation could be caused to overflow. This is now prevented by limiting
2078        the length of names to 32 characters, and the number of named subpatterns
2079        to 10,000.
2080    
2081    11. Subpatterns that are repeated with specific counts have to be replicated in
2082        the compiled pattern. The size of memory for this was computed from the
2083        length of the subpattern and the repeat count. The latter is limited to
2084        65535, but there was no limit on the former, meaning that integer overflow
2085        could in principle occur. The compiled length of a repeated subpattern is
2086        now limited to 30,000 bytes in order to prevent this.
2087    
2088    12. Added the optional facility to have named substrings with the same name.
2089    
2090    13. Added the ability to use a named substring as a condition, using the
2091        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
2092        are numbers (not recommended). Forward references are permitted.
2093    
2094    14. Added forward references in named backreferences (if you see what I mean).
2095    
2096    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
2097        pattern could run off the end of the subject. For example, the pattern
2098        "(?s)(.{1,5})"8 did this with the subject "ab".
2099    
2100    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
2101        PCRE_CASELESS was set when matching characters that were quantified with ?
2102        or *.
2103    
2104    17. A character class other than a single negated character that had a minimum
2105        but no maximum quantifier - for example [ab]{6,} - was not handled
2106        correctly by pce_dfa_exec(). It would match only one character.
2107    
2108    18. A valid (though odd) pattern that looked like a POSIX character
2109        class but used an invalid character after [ (for example [[,abc,]]) caused
2110        pcre_compile() to give the error "Failed: internal error: code overflow" or
2111        in some cases to crash with a glibc free() error. This could even happen if
2112        the pattern terminated after [[ but there just happened to be a sequence of
2113        letters, a binary zero, and a closing ] in the memory that followed.
2114    
2115    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
2116        over the years. Originally (before any Unicode support), just the bottom 8
2117        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
2118        output from "man perlunicode" includes this:
2119    
2120          The regular expression compiler produces polymorphic opcodes.  That
2121          is, the pattern adapts to the data and automatically switches to
2122          the Unicode character scheme when presented with Unicode data--or
2123          instead uses a traditional byte scheme when presented with byte
2124          data.
2125    
2126        Sadly, a wide octal escape does not cause a switch, and in a string with
2127        no other multibyte characters, these octal escapes are treated as before.
2128        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
2129        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
2130        Unicode string.
2131    
2132        I have not perpetrated such confusion in PCRE. Up till now, it took just
2133        the bottom 8 bits, as in old Perl. I have now made octal escapes with
2134        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
2135        translate to the appropriate multibyte character.
2136    
2137    29. Applied some refactoring to reduce the number of warnings from Microsoft
2138        and Borland compilers. This has included removing the fudge introduced
2139        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
2140        a warning about an unused variable.
2141    
2142    21. PCRE has not included VT (character 0x0b) in the set of whitespace
2143        characters since release 4.0, because Perl (from release 5.004) does not.
2144        [Or at least, is documented not to: some releases seem to be in conflict
2145        with the documentation.] However, when a pattern was studied with
2146        pcre_study() and all its branches started with \s, PCRE still included VT
2147        as a possible starting character. Of course, this did no harm; it just
2148        caused an unnecessary match attempt.
2149    
2150    22. Removed a now-redundant internal flag bit that recorded the fact that case
2151        dependency changed within the pattern. This was once needed for "required
2152        byte" processing, but is no longer used. This recovers a now-scarce options
2153        bit. Also moved the least significant internal flag bit to the most-
2154        significant bit of the word, which was not previously used (hangover from
2155        the days when it was an int rather than a uint) to free up another bit for
2156        the future.
2157    
2158    23. Added support for CRLF line endings as well as CR and LF. As well as the
2159        default being selectable at build time, it can now be changed at runtime
2160        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
2161        specify that it is scanning data with non-default line endings.
2162    
2163    24. Changed the definition of CXXLINK to make it agree with the definition of
2164        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
2165    
2166    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
2167        recursions. This makes a big different to stack usage for some patterns.
2168    
2169    26. If a subpattern containing a named recursion or subroutine reference such
2170        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
2171        the space required for the compiled pattern went wrong and gave too small a
2172        value. Depending on the environment, this could lead to "Failed: internal
2173        error: code overflow at offset 49" or "glibc detected double free or
2174        corruption" errors.
2175    
2176    27. Applied patches from Google (a) to support the new newline modes and (b) to
2177        advance over multibyte UTF-8 characters in GlobalReplace.
2178    
2179    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
2180        difference for some implementation of PCRE in some Windows version.
2181    
2182    29. Added some extra testing facilities to pcretest:
2183    
2184        \q<number>   in a data line sets the "match limit" value
2185        \Q<number>   in a data line sets the "match recursion limt" value
2186        -S <number>  sets the stack size, where <number> is in megabytes
2187    
2188        The -S option isn't available for Windows.
2189    
2190    
2191    Version 6.6 06-Feb-06
2192    ---------------------
2193    
2194     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
2195        in pcreposix.h. I have copied the definition from pcre.h.
2196    
2197     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
2198        because pcre.h is no longer a built file.
2199    
2200     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
2201        not normally included in the compiled code.
2202    
2203    
2204    Version 6.5 01-Feb-06
2205    ---------------------
2206    
2207     1. When using the partial match feature with pcre_dfa_exec(), it was not
2208        anchoring the second and subsequent partial matches at the new starting
2209        point. This could lead to incorrect results. For example, with the pattern
2210        /1234/, partially matching against "123" and then "a4" gave a match.
2211    
2212     2. Changes to pcregrep:
2213    
2214        (a) All non-match returns from pcre_exec() were being treated as failures
2215            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
2216            error message is output. Some extra information is given for the
2217            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
2218            probably the only errors that are likely to be caused by users (by
2219            specifying a regex that has nested indefinite repeats, for instance).
2220            If there are more than 20 of these errors, pcregrep is abandoned.
2221    
2222        (b) A binary zero was treated as data while matching, but terminated the
2223            output line if it was written out. This has been fixed: binary zeroes
2224            are now no different to any other data bytes.
2225    
2226        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
2227            used to set a locale for matching. The --locale=xxxx long option has
2228            been added (no short equivalent) to specify a locale explicitly on the
2229            pcregrep command, overriding the environment variables.
2230    
2231        (d) When -B was used with -n, some line numbers in the output were one less
2232            than they should have been.
2233    
2234        (e) Added the -o (--only-matching) option.
2235    
2236        (f) If -A or -C was used with -c (count only), some lines of context were
2237            accidentally printed for the final match.
2238    
2239        (g) Added the -H (--with-filename) option.
2240    
2241        (h) The combination of options -rh failed to suppress file names for files
2242            that were found from directory arguments.
2243    
2244        (i) Added the -D (--devices) and -d (--directories) options.
2245    
2246        (j) Added the -F (--fixed-strings) option.
2247    
2248        (k) Allow "-" to be used as a file name for -f as well as for a data file.
2249    
2250        (l) Added the --colo(u)r option.
2251    
2252        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
2253            is not present by default.
2254    
2255     3. A nasty bug was discovered in the handling of recursive patterns, that is,
2256        items such as (?R) or (?1), when the recursion could match a number of
2257        alternatives. If it matched one of the alternatives, but subsequently,
2258        outside the recursion, there was a failure, the code tried to back up into
2259        the recursion. However, because of the way PCRE is implemented, this is not
2260        possible, and the result was an incorrect result from the match.
2261    
2262        In order to prevent this happening, the specification of recursion has
2263        been changed so that all such subpatterns are automatically treated as
2264        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
2265    
2266     4. I had overlooked the fact that, in some locales, there are characters for
2267        which isalpha() is true but neither isupper() nor islower() are true. In
2268        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
2269        and ordfeminine) are like this. This affected the treatment of \w and \W
2270        when they appeared in character classes, but not when they appeared outside
2271        a character class. The bit map for "word" characters is now created
2272        separately from the results of isalnum() instead of just taking it from the
2273        upper, lower, and digit maps. (Plus the underscore character, of course.)
2274    
2275     5. The above bug also affected the handling of POSIX character classes such as
2276        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
2277        permanent tables. Instead, the bit maps for such a class were previously
2278        created as the appropriate unions of the upper, lower, and digit bitmaps.
2279        Now they are created by subtraction from the [[:word:]] class, which has
2280        its own bitmap.
2281    
2282     6. The [[:blank:]] character class matches horizontal, but not vertical space.
2283        It is created by subtracting the vertical space characters (\x09, \x0a,
2284        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
2285        subtraction was done in the overall bitmap for a character class, meaning
2286        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
2287        be recognized. This bug has been fixed.
2288    
2289     7. Patches from the folks at Google:
2290    
2291          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
2292          real life, but is still worth protecting against".
2293    
2294          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
2295          regular expressions".
2296    
2297          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
2298          have it.
2299    
2300          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
2301          "configure" and the latter not, in order to fix a problem somebody had
2302          with compiling the Arg class on HP-UX.
2303    
2304          (e) Improve the error-handling of the C++ wrapper a little bit.
2305    
2306          (f) New tests for checking recursion limiting.
2307    
2308     8. The pcre_memmove() function, which is used only if the environment does not
2309        have a standard memmove() function (and is therefore rarely compiled),
2310        contained two bugs: (a) use of int instead of size_t, and (b) it was not
2311        returning a result (though PCRE never actually uses the result).
2312    
2313     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
2314        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
2315        returned instead of calling malloc() with an overflowing number that would
2316        most likely cause subsequent chaos.
2317    
2318    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
2319    
2320    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
2321        with this option is matched, the nmatch and pmatch options of regexec() are
2322        ignored.
2323    
2324    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
2325        provided in case anyone wants to the the POSIX interface with UTF-8
2326        strings.
2327    
2328    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
2329        C++ linking (needed for some HP-UX environments).
2330    
2331    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
2332        (unused parameter) and in the pcre_printint() function (omitted "default"
2333        switch label when the default is to do nothing).
2334    
2335    15. Added some code to make it possible, when PCRE is compiled as a C++
2336        library, to replace subject pointers for pcre_exec() with a smart pointer
2337        class, thus making it possible to process discontinuous strings.
2338    
2339    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
2340        much the same function. They were added by different people who were trying
2341        to make PCRE easy to compile on non-Unix systems. It has been suggested
2342        that PCRE_EXPORT be abolished now that there is more automatic apparatus
2343        for compiling on Windows systems. I have therefore replaced it with
2344        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
2345        defaults to "extern" for C or "extern C" for C++, which works fine on
2346        Unix-like systems. It is now possible to override the value of PCRE_DATA_
2347        SCOPE with something explicit in config.h. In addition:
2348    
2349        (a) pcreposix.h still had just "extern" instead of either of these macros;
2350            I have replaced it with PCRE_DATA_SCOPE.
2351    
2352        (b) Functions such as _pcre_xclass(), which are internal to the library,
2353            but external in the C sense, all had PCRE_EXPORT in their definitions.
2354            This is apparently wrong for the Windows case, so I have removed it.
2355            (It makes no difference on Unix-like systems.)
2356    
2357    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
2358        of recursive calls to match(). This is different to MATCH_LIMIT because
2359        that limits the total number of calls to match(), not all of which increase
2360        the depth of recursion. Limiting the recursion depth limits the amount of
2361        stack (or heap if NO_RECURSE is set) that is used. The default can be set
2362        when PCRE is compiled, and changed at run time. A patch from Google adds
2363        this functionality to the C++ interface.
2364    
2365    18. Changes to the handling of Unicode character properties:
2366    
2367        (a) Updated the table to Unicode 4.1.0.
2368    
2369        (b) Recognize characters that are not in the table as "Cn" (undefined).
2370    
2371        (c) I revised the way the table is implemented to a much improved format
2372            which includes recognition of ranges. It now supports the ranges that
2373            are defined in UnicodeData.txt, and it also amalgamates other
2374            characters into ranges. This has reduced the number of entries in the
2375            table from around 16,000 to around 3,000, thus reducing its size
2376            considerably. I realized I did not need to use a tree structure after
2377            all - a binary chop search is just as efficient. Having reduced the
2378            number of entries, I extended their size from 6 bytes to 8 bytes to
2379            allow for more data.
2380    
2381        (d) Added support for Unicode script names via properties such as \p{Han}.
2382    
2383    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
2384        matching that character.
2385    
2386    20. When matching a repeated Unicode property with a minimum greater than zero,
2387        (for example \pL{2,}), PCRE could look past the end of the subject if it
2388        reached it while seeking the minimum number of characters. This could
2389        happen only if some of the characters were more than one byte long, because
2390        there is a check for at least the minimum number of bytes.
2391    
2392    21. Refactored the implementation of \p and \P so as to be more general, to
2393        allow for more different types of property in future. This has changed the
2394        compiled form incompatibly. Anybody with saved compiled patterns that use
2395        \p or \P will have to recompile them.
2396    
2397    22. Added "Any" and "L&" to the supported property types.
2398    
2399    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
2400        but give a compile time error if the value is greater than 0xff.
2401    
2402    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
2403        accidentally not being installed or uninstalled.
2404    
2405    25. The pcre.h file was built from pcre.h.in, but the only changes that were
2406        made were to insert the current release number. This seemed silly, because
2407        it made things harder for people building PCRE on systems that don't run
2408        "configure". I have turned pcre.h into a distributed file, no longer built
2409        by "configure", with the version identification directly included. There is
2410        no longer a pcre.h.in file.
2411    
2412        However, this change necessitated a change to the pcre-config script as
2413        well. It is built from pcre-config.in, and one of the substitutions was the
2414        release number. I have updated configure.ac so that ./configure now finds
2415        the release number by grepping pcre.h.
2416    
2417    26. Added the ability to run the tests under valgrind.
2418    
2419    
2420    Version 6.4 05-Sep-05
2421    ---------------------
2422    
2423     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
2424        "--" to be printed when multiple files were scanned, even when none of the
2425        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
2426        consider it to be a bug, and have restored the previous behaviour.
2427    
2428     2. A couple of code tidies to get rid of compiler warnings.
2429    
2430     3. The pcretest program used to cheat by referring to symbols in the library
2431        whose names begin with _pcre_. These are internal symbols that are not
2432        really supposed to be visible externally, and in some environments it is
2433        possible to suppress them. The cheating is now confined to including
2434        certain files from the library's source, which is a bit cleaner.
2435    
2436     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
2437        file's purpose clearer.
2438    
2439     5. Reorganized pcre_ucp_findchar().
2440    
2441    
2442    Version 6.3 15-Aug-05
2443    ---------------------
2444    
2445     1. The file libpcre.pc.in did not have general read permission in the tarball.
2446    
2447     2. There were some problems when building without C++ support:
2448    
2449        (a) If C++ support was not built, "make install" and "make test" still
2450            tried to test it.
2451    
2452        (b) There were problems when the value of CXX was explicitly set. Some
2453            changes have been made to try to fix these, and ...
2454    
2455        (c) --disable-cpp can now be used to explicitly disable C++ support.
2456    
2457        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
2458            backslash in a target when C++ was disabled. This confuses some
2459            versions of "make", apparently. Using an intermediate variable solves
2460            this. (Same for CPP_LOBJ.)
2461    
2462     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
2463        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
2464        necessary on certain architectures.
2465    
2466     4. Added a setting of -export-symbols-regex to the link command to remove
2467        those symbols that are exported in the C sense, but actually are local
2468        within the library, and not documented. Their names all begin with
2469        "_pcre_". This is not a perfect job, because (a) we have to except some
2470        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
2471        available (and never for static libraries). I have made a note to try to
2472        find a way round (a) in the future.
2473    
2474    
2475    Version 6.2 01-Aug-05
2476    ---------------------
2477    
2478     1. There was no test for integer overflow of quantifier values. A construction
2479        such as {1111111111111111} would give undefined results. What is worse, if
2480        a minimum quantifier for a parenthesized subpattern overflowed and became
2481        negative, the calculation of the memory size went wrong. This could have
2482        led to memory overwriting.
2483    
2484     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
2485    
2486     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
2487        operating environments where this matters.
2488    
2489     4. Applied Giuseppe Maxia's patch to add additional features for controlling
2490        PCRE options from within the C++ wrapper.
2491    
2492     5. Named capturing subpatterns were not being correctly counted when a pattern
2493        was compiled. This caused two problems: (a) If there were more than 100
2494        such subpatterns, the calculation of the memory needed for the whole
2495        compiled pattern went wrong, leading to an overflow error. (b) Numerical
2496        back references of the form \12, where the number was greater than 9, were
2497        not recognized as back references, even though there were sufficient
2498        previous subpatterns.
2499    
2500     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
2501        versions of gcc, e.g. 2.95.4.
2502    
2503    
2504    Version 6.1 21-Jun-05
2505    ---------------------
2506    
2507     1. There was one reference to the variable "posix" in pcretest.c that was not
2508        surrounded by "#if !defined NOPOSIX".
2509    
2510     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
2511        the cross-check on the old pcre_info() function, for the benefit of the
2512        cut-down version of PCRE that is currently imported into Exim.
2513    
2514     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
2515        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
2516        patterns that start (?i) but I don't think that matters. The use of (?i) is
2517        just an example; this all applies to the other options as well.
2518    
2519     4. Since libtool seems to echo the compile commands it is issuing, the output
2520        from "make" can be reduced a bit by putting "@" in front of each libtool
2521        compile command.
2522    
2523     5. Patch from the folks at Google for configure.in to be a bit more thorough
2524        in checking for a suitable C++ installation before trying to compile the
2525        C++ stuff. This should fix a reported problem when a compiler was present,
2526        but no suitable headers.
2527    
2528     6. The man pages all had just "PCRE" as their title. I have changed them to
2529        be the relevant file name. I have also arranged that these names are
2530        retained in the file doc/pcre.txt, which is a concatenation in text format
2531        of all the man pages except the little individual ones for each function.
2532    
2533     7. The NON-UNIX-USE file had not been updated for the different set of source
2534        files that come with release 6. I also added a few comments about the C++
2535        wrapper.
2536    
2537    
2538    Version 6.0 07-Jun-05
2539    ---------------------
2540    
2541     1. Some minor internal re-organization to help with my DFA experiments.
2542    
2543     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
2544        didn't matter for the library itself when fully configured, but did matter
2545        when compiling without UCP support, or within Exim, where the ucp files are
2546        not imported.
2547    
2548     3. Refactoring of the library code to split up the various functions into
2549        different source modules. The addition of the new DFA matching code (see
2550        below) to a single monolithic source would have made it really too
2551        unwieldy, quite apart from causing all the code to be include in a
2552        statically linked application, when only some functions are used. This is
2553        relevant even without the DFA addition now that patterns can be compiled in
2554        one application and matched in another.
2555    
2556        The downside of splitting up is that there have to be some external
2557        functions and data tables that are used internally in different modules of
2558        the library but which are not part of the API. These have all had their
2559        names changed to start with "_pcre_" so that they are unlikely to clash
2560        with other external names.
2561    
2562     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
2563        a different (DFA) algorithm. Although it is slower than the original
2564        function, it does have some advantages for certain types of matching
2565        problem.
2566    
2567     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
2568        including restarting after a partial match.
2569    
2570     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
2571        defined when compiling for Windows was sent to me. I have put it into the
2572        code, though I have no means of testing or verifying it.
2573    
2574     7. Added the pcre_refcount() auxiliary function.
2575    
2576     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
2577        match before or at the first newline in the subject string. In pcretest,
2578        the /f option on a pattern can be used to set this.
2579    
2580     9. A repeated \w when used in UTF-8 mode with characters greater than 256
2581        would behave wrongly. This has been present in PCRE since release 4.0.
2582    
2583    10. A number of changes to the pcregrep command:
2584    
2585        (a) Refactored how -x works; insert ^(...)$ instead of setting
2586            PCRE_ANCHORED and checking the length, in preparation for adding
2587            something similar for -w.
2588    
2589        (b) Added the -w (match as a word) option.
2590    
2591        (c) Refactored the way lines are read and buffered so as to have more
2592            than one at a time available.
2593    
2594        (d) Implemented a pcregrep test script.
2595    
2596        (e) Added the -M (multiline match) option. This allows patterns to match
2597            over several lines of the subject. The buffering ensures that at least
2598            8K, or the rest of the document (whichever is the shorter) is available
2599            for matching (and similarly the previous 8K for lookbehind assertions).
2600    
2601        (f) Changed the --help output so that it now says
2602    
2603              -w, --word-regex(p)
2604    
2605            instead of two lines, one with "regex" and the other with "regexp"
2606            because that confused at least one person since the short forms are the
2607            same. (This required a bit of code, as the output is generated
2608            automatically from a table. It wasn't just a text change.)
2609    
2610        (g) -- can be used to terminate pcregrep options if the next thing isn't an
2611            option but starts with a hyphen. Could be a pattern or a path name
2612            starting with a hyphen, for instance.
2613    
2614        (h) "-" can be given as a file name to represent stdin.
2615    
2616        (i) When file names are being printed, "(standard input)" is used for
2617            the standard input, for compatibility with GNU grep. Previously
2618            "<stdin>" was used.
2619    
2620        (j) The option --label=xxx can be used to supply a name to be used for
2621            stdin when file names are being printed. There is no short form.
2622    
2623        (k) Re-factored the options decoding logic because we are going to add
2624            two more options that take data. Such options can now be given in four
2625            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
2626    
2627        (l) Added the -A, -B, and -C options for requesting that lines of context
2628            around matches be printed.
2629    
2630        (m) Added the -L option to print the names of files that do not contain
2631            any matching lines, that is, the complement of -l.
2632    
2633        (n) The return code is 2 if any file cannot be opened, but pcregrep does
2634            continue to scan other files.
2635    
2636        (o) The -s option was incorrectly implemented. For compatibility with other
2637            greps, it now suppresses the error message for a non-existent or non-
2638            accessible file (but not the return code). There is a new option called
2639            -q that suppresses the output of matching lines, which was what -s was
2640            previously doing.
2641    
2642        (p) Added --include and --exclude options to specify files for inclusion
2643            and exclusion when recursing.
2644    
2645    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
2646        Hopefully, it now does.
2647    
2648    12. Missing cast in pcre_study().
2649    
2650    13. Added an "uninstall" target to the makefile.
2651    
2652    14. Replaced "extern" in the function prototypes in Makefile.in with
2653        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
2654        world, but is set differently for Windows.
2655    
2656    15. Added a second compiling function called pcre_compile2(). The only
2657        difference is that it has an extra argument, which is a pointer to an
2658        integer error code. When there is a compile-time failure, this is set
2659        non-zero, in addition to the error test pointer being set to point to an
2660        error message. The new argument may be NULL if no error number is required
2661        (but then you may as well call pcre_compile(), which is now just a
2662        wrapper). This facility is provided because some applications need a
2663        numeric error indication, but it has also enabled me to tidy up the way
2664        compile-time errors are handled in the POSIX wrapper.
2665    
2666    16. Added VPATH=.libs to the makefile; this should help when building with one
2667        prefix path and installing with another. (Or so I'm told by someone who
2668        knows more about this stuff than I do.)
2669    
2670    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
2671        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
2672        match everything, including newlines. This is not POSIX-compatible, but
2673        somebody wanted the feature. From pcretest it can be activated by using
2674        both the P and the s flags.
2675    
2676    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
2677    
2678    19. libpcre.pc was being incorrectly installed as executable.
2679    
2680    20. A couple of places in pcretest check for end-of-line by looking for '\n';
2681        it now also looks for '\r' so that it will work unmodified on Windows.
2682    
2683    21. Added Google's contributed C++ wrapper to the distribution.
2684    
2685    22. Added some untidy missing memory free() calls in pcretest, to keep
2686        Electric Fence happy when testing.
2687    
2688    
2689    
2690    Version 5.0 13-Sep-04
2691    ---------------------
2692    
2693     1. Internal change: literal characters are no longer packed up into items
2694        containing multiple characters in a single byte-string. Each character
2695        is now matched using a separate opcode. However, there may be more than one
2696        byte in the character in UTF-8 mode.
2697    
2698     2. The pcre_callout_block structure has two new fields: pattern_position and
2699        next_item_length. These contain the offset in the pattern to the next match
2700        item, and its length, respectively.
2701    
2702     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
2703        insertion of callouts before each pattern item. Added the /C option to
2704        pcretest to make use of this.
2705    
2706     4. On the advice of a Windows user, the lines
2707    
2708          #if defined(_WIN32) || defined(WIN32)
2709          _setmode( _fileno( stdout ), 0x8000 );
2710          #endif  /* defined(_WIN32) || defined(WIN32) */
2711    
2712        have been added to the source of pcretest. This apparently does useful
2713        magic in relation to line terminators.
2714    
2715     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
2716        for the benefit of those environments where the "b" makes a difference.
2717    
2718     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
2719        to know about it. I have put a hack into configure.in that adds in code
2720        to set GCC=yes if CC=icc. This seems to end up at a point in the
2721        generated configure script that is early enough to affect the setting of
2722        compiler options, which is what is needed, but I have no means of testing
2723        whether it really works. (The user who reported this had patched the
2724        generated configure script, which of course I cannot do.)
2725    
2726        LATER: After change 22 below (new libtool files), the configure script
2727        seems to know about icc (and also ecc). Therefore, I have commented out
2728        this hack in configure.in.
2729    
2730     7. Added support for pkg-config (2 patches were sent in).
2731    
2732     8. Negated POSIX character classes that used a combination of internal tables
2733        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2734        [[:^ascii]]. Typically, they would match almost any characters. The other
2735        POSIX classes were not broken in this way.
2736    
2737     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2738        to find the match, as PCRE was deluded into thinking that the match had to
2739        start at the start point or following a newline. The same bug applied to
2740        patterns with negative forward assertions or any backward assertions
2741        preceding ".*" at the start, unless the pattern required a fixed first
2742        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2743    
2744    10. In UTF-8 mode, when moving forwards in the subject after a failed match
2745        starting at the last subject character, bytes beyond the end of the subject
2746        string were read.
2747    
2748    11. Renamed the variable "class" as "classbits" to make life easier for C++
2749        users. (Previously there was a macro definition, but it apparently wasn't
2750        enough.)
2751    
2752    12. Added the new field "tables" to the extra data so that tables can be passed
2753        in at exec time, or the internal tables can be re-selected. This allows
2754        a compiled regex to be saved and re-used at a later time by a different
2755        program that might have everything at different addresses.
2756    
2757    13. Modified the pcre-config script so that, when run on Solaris, it shows a
2758        -R library as well as a -L library.
2759    
2760    14. The debugging options of pcretest (-d on the command line or D on a
2761        pattern) showed incorrect output for anything following an extended class
2762        that contained multibyte characters and which was followed by a quantifier.
2763    
2764    15. Added optional support for general category Unicode character properties
2765        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2766        support. It adds about 90K to the size of the library. The meanings of the
2767        inbuilt class escapes such as \d and \s have NOT been changed.
2768    
2769    16. Updated pcredemo.c to include calls to free() to release the memory for the
2770        compiled pattern.
2771    
2772    17. The generated file chartables.c was being created in the source directory
2773        instead of in the building directory. This caused the build to fail if the
2774        source directory was different from the building directory, and was
2775        read-only.
2776    
2777    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2778        file. No doubt somebody will tell me if they don't make sense... Also added
2779        Dan Mooney's comments about building on OpenVMS.
2780    
2781    19. Added support for partial matching via the PCRE_PARTIAL option for
2782        pcre_exec() and the \P data escape in pcretest.
2783    
2784    20. Extended pcretest with 3 new pattern features:
2785    
2786        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
2787              write the compiled pattern to the file whose name is "rest-of-line".
2788              This is a straight binary dump of the data, with the saved pointer to
2789              the character tables forced to be NULL. The study data, if any, is
2790              written too. After writing, pcretest reads a new pattern.
2791    
2792        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2793              compiled pattern from the given file. There must not be any
2794              occurrences of "<" in the file name (pretty unlikely); if there are,
2795              pcretest will instead treat the initial "<" as a pattern delimiter.
2796              After reading in the pattern, pcretest goes on to read data lines as
2797              usual.
2798    
2799        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2800              and 16-bit fields in a compiled pattern, to simulate a pattern that
2801              was compiled on a host of opposite endianness.
2802    
2803    21. The pcre-exec() function can now cope with patterns that were compiled on
2804        hosts of opposite endianness, with this restriction:
2805    
2806          As for any compiled expression that is saved and used later, the tables
2807          pointer field cannot be preserved; the extra_data field in the arguments
2808          to pcre_exec() should be used to pass in a tables address if a value
2809          other than the default internal tables were used at compile time.
2810    
2811    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2812        now diagnosed as an error. Previously, most of the time, a negative number
2813        would have been treated as zero, but if in addition "ovector" was passed as
2814        NULL, a crash could occur.
2815    
2816    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2817        new versions from the libtool 1.5 distribution (the last one is a copy of
2818        a file called libtool.m4). This seems to have fixed the need to patch
2819        "configure" to support Darwin 1.3 (which I used to do). However, I still
2820        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2821        workstation).
2822    
2823    24. Changed the PCRE licence to be the more standard "BSD" licence.
2824    
2825    
2826    Version 4.5 01-Dec-03
2827    ---------------------
2828    
2829     1. There has been some re-arrangement of the code for the match() function so
2830        that it can be compiled in a version that does not call itself recursively.
2831        Instead, it keeps those local variables that need separate instances for
2832        each "recursion" in a frame on the heap, and gets/frees frames whenever it
2833        needs to "recurse". Keeping track of where control must go is done by means
2834        of setjmp/longjmp. The whole thing is implemented by a set of macros that
2835        hide most of the details from the main code, and operates only if
2836        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2837        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2838        operating.
2839    
2840        To make it easier for callers to provide specially tailored get/free
2841        functions for this usage, two new functions, pcre_stack_malloc, and
2842        pcre_stack_free, are used. They are always called in strict stacking order,
2843        and the size of block requested is always the same.
2844    
2845        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2846        PCRE has been compiled to use the stack or the heap for recursion. The
2847        -C option of pcretest uses this to show which version is compiled.
2848    
2849        A new data escape \S, is added to pcretest; it causes the amounts of store
2850        obtained and freed by both kinds of malloc/free at match time to be added
2851        to the output.
2852    
2853     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2854        what's available on my current Linux desktop machine.
2855    
2856     3. When matching a UTF-8 string, the test for a valid string at the start has
2857        been extended. If start_offset is not zero, PCRE now checks that it points
2858        to a byte that is the start of a UTF-8 character. If not, it returns
2859        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2860        this is necessary because there may be backward assertions in the pattern.
2861        When matching the same subject several times, it may save resources to use
2862        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2863    
2864     4. The code for checking the validity of UTF-8 strings has been tightened so
2865        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2866        containing "overlong sequences".
2867    
2868     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2869        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2870        should have been "&", but it just so happened that all the cases this let
2871        through by mistake were picked up later in the function.
2872    
2873     6. I had used a variable called "isblank" - this is a C99 function, causing
2874        some compilers to warn. To avoid this, I renamed it (as "blankclass").
2875    
2876     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2877        prompting; (b) run "./pcretest /dev/null" at the start of the test script
2878        so the version is shown; (c) stop "make test" echoing "./RunTest".
2879    
2880     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2881    
2882     9. The prototype for memmove() for systems that don't have it was using
2883        size_t, but the inclusion of the header that defines size_t was later. I've
2884        moved the #includes for the C headers earlier to avoid this.
2885    
2886    10. Added some adjustments to the code to make it easier to compiler on certain
2887        special systems:
2888    
2889          (a) Some "const" qualifiers were missing.
2890          (b) Added the macro EXPORT before all exported functions; by default this
2891              is defined to be empty.
2892          (c) Changed the dftables auxiliary program (that builds chartables.c) so
2893              that it reads its output file name as an argument instead of writing
2894              to the standard output and assuming this can be redirected.
2895    
2896    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2897        class containing characters with values greater than 255, PCRE compilation
2898        went into a loop.
2899    
2900    12. A recursive reference to a subpattern that was within another subpattern
2901        that had a minimum quantifier of zero caused PCRE to crash. For example,
2902        (x(y(?2))z)? provoked this bug with a subject that got as far as the
2903        recursion. If the recursively-called subpattern itself had a zero repeat,
2904        that was OK.
2905    
2906    13. In pcretest, the buffer for reading a data line was set at 30K, but the
2907        buffer into which it was copied (for escape processing) was still set at
2908        1024, so long lines caused crashes.
2909    
2910    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2911        "internal error: code overflow...". This applied to any character class
2912        that was followed by a possessive quantifier.
2913    
2914    15. Modified the Makefile to add libpcre.la as a prerequisite for
2915        libpcreposix.la because I was told this is needed for a parallel build to
2916        work.
2917    
2918    16. If a pattern that contained .* following optional items at the start was
2919        studied, the wrong optimizing data was generated, leading to matching
2920        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2921        matching string must start with a or b or c. The correct conclusion for
2922        this pattern is that a match can start with any character.
2923    
2924    
2925    Version 4.4 13-Aug-03
2926    ---------------------
2927    
2928     1. In UTF-8 mode, a character class containing characters with values between
2929        127 and 255 was not handled correctly if the compiled pattern was studied.
2930        In fixing this, I have also improved the studying algorithm for such
2931        classes (slightly).
2932    
2933     2. Three internal functions had redundant arguments passed to them. Removal
2934        might give a very teeny performance improvement.
2935    
2936     3. Documentation bug: the value of the capture_top field in a callout is *one
2937        more than* the number of the hightest numbered captured substring.
2938    
2939     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2940        in incorrectly linking with a previously installed version. They now link
2941        explicitly with libpcre.la.
2942    
2943     5. configure.in no longer needs to recognize Cygwin specially.
2944    
2945     6. A problem in pcre.in for Windows platforms is fixed.
2946    
2947     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2948        pcretest, it used to include the size of the study block as part of its
2949        output. Unfortunately, the structure contains a field that has a different
2950        size on different hardware architectures. This meant that the tests that
2951        showed this size failed. As the block is currently always of a fixed size,
2952        this information isn't actually particularly useful in pcretest output, so
2953        I have just removed it.
2954    
2955     8. Three pre-processor statements accidentally did not start in column 1.
2956        Sadly, there are *still* compilers around that complain, even though
2957        standard C has not required this for well over a decade. Sigh.
2958    
2959     9. In pcretest, the code for checking callouts passed small integers in the
2960        callout_data field, which is a void * field. However, some picky compilers
2961        complained about the casts involved for this on 64-bit systems. Now
2962        pcretest passes the address of the small integer instead, which should get
2963        rid of the warnings.
2964    
2965    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2966        both compile and run time, and gives an error if an invalid UTF-8 sequence
2967        is found. There is a option for disabling this check in cases where the
2968        string is known to be correct and/or the maximum performance is wanted.
2969    
2970    11. In response to a bug report, I changed one line in Makefile.in from
2971    
2972            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2973        to
2974            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2975    
2976        to look similar to other lines, but I have no way of telling whether this
2977        is the right thing to do, as I do not use Windows. No doubt I'll get told
2978        if it's wrong...
2979    
2980    
2981    Version 4.3 21-May-03
2982    ---------------------
2983    
2984    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2985       Makefile.
2986    
2987    2. Some refactoring to improve the quality of the code:
2988    
2989       (i)   The utf8_table... variables are now declared "const".
2990    
2991       (ii)  The code for \cx, which used the "case flipping" table to upper case
2992             lower case letters, now just substracts 32. This is ASCII-specific,
2993             but the whole concept of \cx is ASCII-specific, so it seems
2994             reasonable.
2995    
2996       (iii) PCRE was using its character types table to recognize decimal and
2997             hexadecimal digits in the pattern. This is silly, because it handles
2998             only 0-9, a-f, and A-F, but the character types table is locale-
2999             specific, which means strange things might happen. A private
3000             table is now used for this - though it costs 256 bytes, a table is
3001             much faster than multiple explicit tests. Of course, the standard
3002             character types table is still used for matching digits in subject
3003             strings against \d.
3004    
3005       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
3006             ending in _t are). So I've renamed it as ESC_tee.
3007    
3008    3. The first argument for regexec() in the POSIX wrapper should have been
3009       defined as "const".
3010    
3011    4. Changed pcretest to use malloc() for its buffers so that they can be
3012       Electric Fenced for debugging.
3013    
3014    5. There were several places in the code where, in UTF-8 mode, PCRE would try
3015       to read one or more bytes before the start of the subject string. Often this
3016       had no effect on PCRE's behaviour, but in some circumstances it could
3017       provoke a segmentation fault.
3018    
3019    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
3020       to try to read one or more bytes before the start of the subject string.
3021    
3022    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
3023       UTF-8 support could misbehave in various ways if the subject string
3024       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
3025       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
3026       back over UTF-8 characters.)
3027    
3028    
3029    Version 4.2 14-Apr-03
3030    ---------------------
3031    
3032    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
3033    
3034    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
3035         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
3036         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
3037         [WIN_PREFIX]: new variable, "cyg" for Cygwin
3038         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
3039           and BUILD_EXEEXT
3040         Note: automatic setting of the BUILD variables is not yet working
3041         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
3042           compile-time but not at link-time
3043         [LINK]: use for linking executables only
3044         make different versions for Windows and non-Windows
3045         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
3046           libraries
3047         [LINK_FOR_BUILD]: new variable
3048         [OBJEXT]: use throughout
3049         [EXEEXT]: use throughout
3050         <winshared>: new target
3051         <wininstall>: new target
3052         <dftables.o>: use native compiler
3053         <dftables>: use native linker
3054         <install>: handle Windows platform correctly
3055         <clean>: ditto
3056         <check>: ditto
3057         copy DLL to top builddir before testing
3058    
3059       As part of these changes, -no-undefined was removed again. This was reported
3060       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
3061       in any case.
3062    
3063    3. Some tidies to get rid of compiler warnings:
3064    
3065       . In the match_data structure, match_limit was an unsigned long int, whereas
3066         match_call_count was an int. I've made them both unsigned long ints.
3067    
3068       . In pcretest the fact that a const uschar * doesn't automatically cast to
3069         a void * provoked a warning.
3070    
3071       . Turning on some more compiler warnings threw up some "shadow" variables
3072         and a few more missing casts.
3073    
3074    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
3075       option, a class that contained a single character with a value between 128
3076       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
3077    
3078    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
3079       option, a class that contained several characters, but with at least one
3080       whose value was between 128 and 255 caused PCRE to crash.
3081    
3082    
3083    Version 4.1 12-Mar-03
3084    ---------------------
3085    
3086    1. Compiling with gcc -pedantic found a couple of places where casts were
3087    needed, and a string in dftables.c that was longer than standard compilers are
3088    required to support.
3089    
3090    2. Compiling with Sun's compiler found a few more places where the code could
3091    be tidied up in order to avoid warnings.
3092    
3093    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
3094    first of these names is deprecated in the latest Autoconf in favour of the name
3095    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
3096    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
3097    analogy I have changed it to CFLAGS_FOR_BUILD.
3098    
3099    4. Added -no-undefined to the linking command in the Makefile, because this is
3100    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
3101    linking step for the pcreposix library.
3102    
3103    5. PCRE was failing to diagnose the case of two named groups with the same
3104    name.
3105    
3106    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
3107    literal character that is needed in the subject for a match, and scans along to
3108    ensure that it is present before embarking on the full matching process. This
3109    saves time in cases of nested unlimited repeats that are never going to match.
3110    Problem: the scan can take a lot of time if the subject is very long (e.g.
3111    megabytes), thus penalizing straightforward matches. It is now done only if the
3112    amount of subject to be scanned is less than 1000 bytes.
3113    
3114    7. A lesser problem with the same optimization is that it was recording the
3115    first character of an anchored pattern as "needed", thus provoking a search
3116    right along the subject, even when the first match of the pattern was going to
3117    fail. The "needed" character is now not set for anchored patterns, unless it
3118    follows something in the pattern that is of non-fixed length. Thus, it still
3119    fulfils its original purpose of finding quick non-matches in cases of nested
3120    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
3121    
3122    
3123    Version 4.0 17-Feb-03
3124    ---------------------
3125    
3126    1. If a comment in an extended regex that started immediately after a meta-item
3127    extended to the end of string, PCRE compiled incorrect data. This could lead to
3128    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
3129    
3130    2. Moved to autoconf 2.53 and libtool 1.4.2.
3131    
3132    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
3133    the special perltest8 script is no longer needed - all the tests can be run
3134    from a single perltest script.
3135    
3136    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
3137    by \s. It has now been removed in PCRE. This means it isn't recognized as
3138    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
3139    class [:space:] *does* include VT, thereby creating a mess.
3140    
3141    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
3142    space and tab.
3143    
3144    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
3145    its new features into the main test script, reducing the number of scripts.
3146    
3147    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
3148    were backward compatible, and made the (?i) apply to the whole pattern, as if
3149    /i were given. Now it behaves more logically, and applies the option setting
3150    only to what follows. PCRE has been changed to follow suit. However, if it
3151    finds options settings right at the start of the pattern, it extracts them into
3152    the global options, as before. Thus, they show up in the info data.
3153    
3154    8. Added support for the \Q...\E escape sequence. Characters in between are
3155    treated as literals. This is slightly different from Perl in that $ and @ are
3156    also handled as literals inside the quotes. In Perl, they will cause variable
3157    interpolation. Note the following examples:
3158    
3159        Pattern            PCRE matches      Perl matches
3160    
3161        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
3162        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
3163        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
3164    
3165    For compatibility with Perl, \Q...\E sequences are recognized inside character
3166    classes as well as outside them.
3167    
3168    9. Re-organized 3 code statements in pcretest to avoid "overflow in
3169    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
3170    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
3171    signed/unsigned warnings.
3172    
3173    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
3174    option for pcretest, so I've replaced it by a simple function that does just
3175    that job.
3176    
3177    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
3178    "pcregrep -".
3179    
3180    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
3181    Java package. This provides some syntactic sugar for simple cases of what my
3182    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
3183    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
3184    item, you can use this simplified notation. Note that only makes sense with
3185    greedy quantifiers. Consequently, the use of the possessive quantifier forces
3186    greediness, whatever the setting of the PCRE_UNGREEDY option.
3187    
3188    13. A change of greediness default within a pattern was not taking effect at
3189    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
3190    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
3191    was abstracted outside.
3192    
3193    14. PCRE now supports the \G assertion. It is true when the current matching
3194    position is at the start point of the match. This differs from \A when the
3195    starting offset is non-zero. Used with the /g option of pcretest (or similar
3196    code), it works in the same way as it does for Perl's /g option. If all
3197    alternatives of a regex begin with \G, the expression is anchored to the start
3198    match position, and the "anchored" flag is set in the compiled expression.
3199    
3200    15. Some bugs concerning the handling of certain option changes within patterns
3201    have been fixed. These applied to options other than (?ims). For example,
3202    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
3203    been the other way round. Some of this was related to change 7 above.
3204    
3205    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
3206    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
3207    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
3208    POSIX classes only within a class (e.g. /[[:alpha:]]/).
3209    
3210    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
3211    mode. Unlike ".", it always matches newline, whatever the setting of
3212    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
3213    assertions. Perl allows it, but it doesn't (in general) work because it can't
3214    calculate the length of the lookbehind. At least, that's the case for Perl
3215    5.8.0 - I've been told they are going to document that it doesn't work in
3216    future.
3217    
3218    18. Added an error diagnosis for escapes that PCRE does not support: these are
3219    \L, \l, \N, \P, \p, \U, \u, and \X.
3220    
3221    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
3222    reading past the end of the pattern in cases such as /[abcd/.
3223    
3224    20. PCRE was getting more memory than necessary for patterns with classes that
3225    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
3226    
3227    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
3228    compiling PCRE for use with Virtual Pascal.
3229    
3230    22. Small fix to the Makefile to make it work properly if the build is done
3231    outside the source tree.
3232    
3233    23. Added a new extension: a condition to go with recursion. If a conditional
3234    subpattern starts with (?(R) the "true" branch is used if recursion has
3235    happened, whereas the "false" branch is used only at the top level.
3236    
3237    24. When there was a very long string of literal characters (over 255 bytes
3238    without UTF support, over 250 bytes with UTF support), the computation of how
3239    much memory was required could be incorrect, leading to segfaults or other
3240    strange effects.
3241    
3242    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
3243    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
3244    there was a subsequent back reference to those brackets. This meant that, for
3245    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
3246    possible to check for precisely this case. All we can do is abandon the
3247    optimization if .* occurs inside capturing brackets when there are any back
3248    references whatsoever. (See below for a better fix that came later.)
3249    
3250    26. The handling of the optimization for finding the first character of a
3251    non-anchored pattern, and for finding a character that is required later in the
3252    match were failing in some cases. This didn't break the matching; it just
3253    failed to optimize when it could. The way this is done has been re-implemented.
3254    
3255    27. Fixed typo in error message for invalid (?R item (it said "(?p").
3256    
3257    28. Added a new feature that provides some of the functionality that Perl
3258    provides with (?{...}). The facility is termed a "callout". The way it is done
3259    in PCRE is for the caller to provide an optional function, by setting
3260    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
3261    global variable. By default it is unset, which disables all calling out. To get
3262    the function called, the regex must include (?C) at appropriate points. This
3263    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
3264    This provides a means of identifying different callout points. When PCRE
3265    reaches such a point in the regex, if pcre_callout has been set, the external
3266    function is called. It is provided with data in a structure called
3267    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
3268    matching continues; if it returns a non-zero value, the match at the current
3269    point fails. However, backtracking will occur if possible. [This was changed
3270    later and other features added - see item 49 below.]
3271    
3272    29. pcretest is upgraded to test the callout functionality. It provides a
3273    callout function that displays information. By default, it shows the start of
3274    the match and the current position in the text. There are some new data escapes
3275    to vary what happens:
3276    
3277        \C+         in addition, show current contents of captured substrings
3278        \C-         do not supply a callout function
3279        \C!n        return 1 when callout number n is reached
3280        \C!n!m      return 1 when callout number n is reached for the mth time
3281    
3282    30. If pcregrep was called with the -l option and just a single file name, it
3283    output "<stdin>" if a match was found, instead of the file name.
3284    
3285    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
3286    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
3287    pcre_exec(). This saves a malloc/free per call. The default value of
3288    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
3289    when configuring.
3290    
3291    32. The default maximum size of a compiled pattern is 64K. There have been a
3292    few cases of people hitting this limit. The code now uses macros to handle the
3293    storing of links as offsets within the compiled pattern. It defaults to 2-byte
3294    links, but this can be changed to 3 or 4 bytes by --with-link-size when
3295    configuring. Tests 2 and 5 work only with 2-byte links because they output
3296    debugging information about compiled patterns.
3297    
3298    33. Internal code re-arrangements:
3299    
3300    (a) Moved the debugging function for printing out a compiled regex into
3301        its own source file (printint.c) and used #include to pull it into
3302        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
3303        separate copies.
3304    
3305    (b) Defined the list of op-code names for debugging as a macro in
3306        internal.h so that it is next to the definition of the opcodes.
3307    
3308    (c) Defined a table of op-code lengths for simpler skipping along compiled
3309        code. This is again a macro in internal.h so that it is next to the
3310        definition of the opcodes.
3311    
3312    34. Added support for recursive calls to individual subpatterns, along the
3313    lines of Robin Houston's patch (but implemented somewhat differently).
3314    
3315    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
3316    allow it to read and process whole directories in Win32. This code was
3317    contributed by Lionel Fourquaux; it has not been tested by me.
3318    
3319    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
3320    used to name a group. Names consist of alphanumerics and underscores, and must
3321    be unique. Back references use the syntax (?P=name) and recursive calls use
3322    (?P>name) which is a PCRE extension to the Python extension. Groups still have
3323    numbers. The function pcre_fullinfo() can be used after compilation to extract
3324    a name/number map. There are three relevant calls:
3325    
3326      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
3327      PCRE_INFO_NAMECOUNT            yields the number of entries
3328      PCRE_INFO_NAMETABLE            yields a pointer to the map.
3329    
3330    The map is a vector of fixed-size entries. The size of each entry depends on
3331    the length of the longest name used. The first two bytes of each entry are the
3332    group number, most significant byte first. There follows the corresponding
3333    name, zero terminated. The names are in alphabetical order.
3334    
3335    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
3336    case instead of 255. Making it the same both with and without UTF-8 support
3337    means that the same test output works with both.
3338    
3339    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
3340    calling malloc() with a zero argument.
3341    
3342    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
3343    optimization. I've improved things by keeping a bitmap of backreferences with
3344    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
3345    fact referenced, the optimization can be applied. It is unlikely that a
3346    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
3347    the match to follow \n) will appear inside brackets with a number greater than
3348    31, but if it does, any back reference > 31 suppresses the optimization.
3349    
3350    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
3351    of disabling numbered capturing parentheses. Any opening parenthesis that is
3352    not followed by ? behaves as if it were followed by ?: but named parentheses
3353    can still be used for capturing (and they will acquire numbers in the usual
3354    way).
3355    
3356    41. Redesigned the return codes from the match() function into yes/no/error so
3357    that errors can be passed back from deep inside the nested calls. A malloc
3358    failure while inside a recursive subpattern call now causes the
3359    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
3360    
3361    42. It is now possible to set a limit on the number of times the match()
3362    function is called in a call to pcre_exec(). This facility makes it possible to
3363    limit the amount of recursion and backtracking, though not in a directly
3364    obvious way, because the match() function is used in a number of different
3365    circumstances. The count starts from zero for each position in the subject
3366    string (for non-anchored patterns). The default limit is, for compatibility, a
3367    large number, namely 10 000 000. You can change this in two ways:
3368    
3369    (a) When configuring PCRE before making, you can use --with-match-limit=n
3370        to set a default value for the compiled library.
3371    
3372    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
3373        a different value is set. See 45 below.
3374    
3375    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
3376    
3377    43. Added a new function pcre_config(int, void *) to enable run-time extraction
3378    of things that can be changed at compile time. The first argument specifies
3379    what is wanted and the second points to where the information is to be placed.
3380    The current list of available information is:
3381    
3382      PCRE_CONFIG_UTF8
3383    
3384    The output is an integer that is set to one if UTF-8 support is available;
3385    otherwise it is set to zero.
3386    
3387      PCRE_CONFIG_NEWLINE
3388    
3389    The output is an integer that it set to the value of the code that is used for
3390    newline. It is either LF (10) or CR (13).
3391    
3392      PCRE_CONFIG_LINK_SIZE
3393    
3394    The output is an integer that contains the number of bytes used for internal
3395    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
3396    
3397      PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
3398    
3399    The output is an integer that contains the threshold above which the POSIX
3400    interface uses malloc() for output vectors. See item 31 above.
3401    
3402      PCRE_CONFIG_MATCH_LIMIT
3403    
3404    The output is an unsigned integer that contains the default limit of the number
3405    of match() calls in a pcre_exec() execution. See 42 above.
3406    
3407    44. pcretest has been upgraded by the addition of the -C option. This causes it
3408    to extract all the available output from the new pcre_config() function, and to
3409    output it. The program then exits immediately.
3410    
3411    45. A need has arisen to pass over additional data with calls to pcre_exec() in
3412    order to support additional features. One way would have been to define
3413    pcre_exec2() (for example) with extra arguments, but this would not have been
3414    extensible, and would also have required all calls to the original function to
3415    be mapped to the new one. Instead, I have chosen to extend the mechanism that
3416    is used for passing in "extra" data from pcre_study().
3417    
3418    The pcre_extra structure is now exposed and defined in pcre.h. It currently
3419    contains the following fields:
3420    
3421      flags         a bitmap indicating which of the following fields are set
3422      study_data    opaque data from pcre_study()
3423      match_limit   a way of specifying a limit on match() calls for a specific
3424                      call to pcre_exec()
3425      callout_data  data for callouts (see 49 below)
3426    
3427    The flag bits are also defined in pcre.h, and are
3428    
3429      PCRE_EXTRA_STUDY_DATA
3430      PCRE_EXTRA_MATCH_LIMIT
3431      PCRE_EXTRA_CALLOUT_DATA
3432    
3433    The pcre_study() function now returns one of these new pcre_extra blocks, with
3434    the actual study data pointed to by the study_data field, and the
3435    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
3436    before. That is, this change is entirely upwards-compatible and requires no
3437    change to existing code.
3438    
3439    If you want to pass in additional data to pcre_exec(), you can either place it
3440    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
3441    block.
3442    
3443    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
3444    data string contains the escape sequence \M, pcretest calls pcre_exec() several
3445    times with different match limits, until it finds the minimum value needed for
3446    pcre_exec() to complete. The value is then output. This can be instructive; for
3447    most simple matches the number is quite small, but for pathological cases it
3448    gets very large very quickly.
3449    
3450    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
3451    returns the size of the data block pointed to by the study_data field in a
3452    pcre_extra block, that is, the value that was passed as the argument to
3453    pcre_malloc() when PCRE was getting memory in which to place the information
3454    created by pcre_study(). The fourth argument should point to a size_t variable.
3455    pcretest has been extended so that this information is shown after a successful
3456    pcre_study() call when information about the compiled regex is being displayed.
3457    
3458    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
3459    because what follows is always an absolute path. (Later: it turns out that this
3460    is more than cosmetic for MinGW, because it doesn't like empty path
3461    components.)
3462    
3463    49. Some changes have been made to the callout feature (see 28 above):
3464    
3465    (i)  A callout function now has three choices for what it returns:
3466    
3467           0  =>  success, carry on matching
3468         > 0  =>  failure at this point, but backtrack if possible
3469         < 0  =>  serious error, return this value from pcre_exec()
3470    
3471         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
3472         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
3473         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
3474         use by callout functions. It will never be used by PCRE itself.
3475    
3476    (ii) The pcre_extra structure (see 45 above) has a void * field called
3477         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
3478         pcre_callout_block structure has a field of the same name. The contents of
3479         the field passed in the pcre_extra structure are passed to the callout
3480         function in the corresponding field in the callout block. This makes it
3481         easier to use the same callout-containing regex from multiple threads. For
3482         testing, the pcretest program has a new data escape
3483    
3484           \C*n        pass the number n (may be negative) as callout_data
3485    
3486         If the callout function in pcretest receives a non-zero value as
3487         callout_data, it returns that value.
3488    
3489    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
3490    there were some redundant $(CFLAGS) in commands that are now specified as
3491    $(LINK), which already includes $(CFLAGS).
3492    
3493    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
3494    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
3495    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
3496    one-byte characters throughout. Note that case-insensitive matching applies
3497    only to characters whose values are less than 256. PCRE doesn't support the
3498    notion of cases for higher-valued characters.
3499    
3500    (i)   A character class whose characters are all within 0-255 is handled as
3501          a bit map, and the map is inverted for negative classes. Previously, a
3502          character > 255 always failed to match such a class; however it should
3503          match if the class was a negative one (e.g. [^ab]). This has been fixed.
3504    
3505    (ii)  A negated character class with a single character < 255 is coded as
3506          "not this character" (OP_NOT). This wasn't working properly when the test
3507          character was multibyte, either singly or repeated.
3508    
3509    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
3510          mode, for example: \x{100}{2,3}.
3511    
3512    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
3513          singly or repeated) now correctly test multibyte characters. However,
3514          PCRE doesn't recognize any characters with values greater than 255 as
3515          digits, spaces, or word characters. Such characters always match \D, \S,
3516          and \W, and never match \d, \s, or \w.
3517    
3518    (v)   Classes may now contain characters and character ranges with values
3519          greater than 255. For example: [ab\x{100}-\x{400}].
3520    
3521    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
3522          PCRE in UTF-8 mode.
3523    
3524    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
3525    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
3526    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
3527    value.)
3528    
3529    53. The single man page has become too large. I have therefore split it up into
3530    a number of separate man pages. These also give rise to individual HTML pages;
3531    these are now put in a separate directory, and there is an index.html page that
3532    lists them all. Some hyperlinking between the pages has been installed.
3533    
3534    54. Added convenience functions for handling named capturing parentheses.
3535    
3536    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
3537    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
3538    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
3539    are faulted.
3540    
3541    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
3542    calling configure. These values are used when compiling the dftables.c program
3543    which is run to generate the source of the default character tables. They
3544    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
3545    you will need to set these values.
3546    
3547    57. Updated the building process for Windows DLL, as provided by Fred Cox.
3548    
3549    
3550    Version 3.9 02-Jan-02
3551    ---------------------
3552    
3553    1. A bit of extraneous text had somehow crept into the pcregrep documentation.
3554    
3555    2. If --disable-static was given, the building process failed when trying to
3556    build pcretest and pcregrep. (For some reason it was using libtool to compile
3557    them, which is not right, as they aren't part of the library.)
3558    
3559    
3560  Version 3.8 18-Dec-01  Version 3.8 18-Dec-01
3561  ---------------------  ---------------------
3562    

Legend:
Removed from v.59  
changed lines
  Added in v.632

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12