/[pcre]/code/trunk/ChangeLog
ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 19 by nigel, Sat Feb 24 21:38:33 2007 UTC revision 622 by ph10, Mon Jul 18 10:46:51 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    
153    
154    Version 8.12 15-Jan-2011
155    ------------------------
156    
157    1.  Fixed some typos in the markup of the man pages, and wrote a script that
158        checks for such things as part of the documentation building process.
159    
160    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
161        --match-limit and --recursion-limit options (added for 8.11). In
162        particular, this made one of the standard tests fail. (The integer value
163        went into the wrong half of a long int.)
164    
165    3.  If the --colour option was given to pcregrep with -v (invert match), it
166        did strange things, either producing crazy output, or crashing. It should,
167        of course, ignore a request for colour when reporting lines that do not
168        match.
169    
170    4.  Another pcregrep bug caused similar problems if --colour was specified with
171        -M (multiline) and the pattern match finished with a line ending.
172    
173    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
174        matched in multiline mode, the following line was shown as part of the
175        match. This seems wrong, so I have changed it.
176    
177    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
178        the check for further matches in the same line (so they could be coloured)
179        to overrun the end of the current line. If another match was found, it was
180        incorrectly shown (and then shown again when found in the next line).
181    
182    7.  If pcregrep was compiled under Windows, there was a reference to the
183        function pcregrep_exit() before it was defined. I am assuming this was
184        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
185        reported by a user. I've moved the definition above the reference.
186    
187    
188    Version 8.11 10-Dec-2010
189    ------------------------
190    
191    1.  (*THEN) was not working properly if there were untried alternatives prior
192        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
193        backtracked to try for "b" instead of moving to the next alternative branch
194        at the same level (in this case, to look for "c"). The Perl documentation
195        is clear that when (*THEN) is backtracked onto, it goes to the "next
196        alternative in the innermost enclosing group".
197    
198    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
199        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
200        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
201        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
202        (*THEN).
203    
204    3.  If \s appeared in a character class, it removed the VT character from
205        the class, even if it had been included by some previous item, for example
206        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
207        of \s, but is part of the POSIX "space" class.)
208    
209    4.  A partial match never returns an empty string (because you can always
210        match an empty string at the end of the subject); however the checking for
211        an empty string was starting at the "start of match" point. This has been
212        changed to the "earliest inspected character" point, because the returned
213        data for a partial match starts at this character. This means that, for
214        example, /(?<=abc)def/ gives a partial match for the subject "abc"
215        (previously it gave "no match").
216    
217    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
218        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
219        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
220        has an implication that the given string is incomplete (because a partial
221        match is preferred over a full match). For this reason, these items now
222        give a partial match in this situation. [Aside: previously, the one case
223        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
224        match rather than a full match, which was wrong by the old rules, but is
225        now correct.]
226    
227    6.  There was a bug in the handling of #-introduced comments, recognized when
228        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
229        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
230        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
231        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
232        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
233        places in pcre_compile().
234    
235    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
236        comments when looking ahead for named forward references to subpatterns,
237        the only newline sequence it recognized was NL. It now handles newlines
238        according to the set newline convention.
239    
240    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
241        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
242        cater for a lack of strerror(). These oversights have been fixed.
243    
244    9.  Added --match-limit and --recursion-limit to pcregrep.
245    
246    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
247    
248    11. When the -o option was used, pcregrep was setting a return code of 1, even
249        when matches were found, and --line-buffered was not being honoured.
250    
251    12. Added an optional parentheses number to the -o and --only-matching options
252        of pcregrep.
253    
254    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
255        can match an empty string. The code to do it in pcretest and pcredemo
256        needed fixing:
257    
258        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
259            only one byte after an empty string match just before CRLF (this case
260            just got forgotten; "any" and "anycrlf" were OK).
261    
262        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
263            mode when an empty string match preceded an ASCII character followed by
264            a non-ASCII character. (The code for advancing by one character rather
265            than one byte was nonsense.)
266    
267        (c) The pcredemo.c sample program did not have any code at all to handle
268            the cases when CRLF is a valid newline sequence.
269    
270    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
271        as a starting offset was within the subject string. There is now a new
272        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
273        negative or greater than the length of the string. In order to test this,
274        pcretest is extended to allow the setting of negative starting offsets.
275    
276    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
277        starting offset points to the beginning of a UTF-8 character was
278        unnecessarily clumsy. I tidied it up.
279    
280    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
281        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
282    
283    17. Nobody had reported that the --include_dir option, which was added in
284        release 7.7 should have been called --include-dir (hyphen, not underscore)
285        for compatibility with GNU grep. I have changed it to --include-dir, but
286        left --include_dir as an undocumented synonym, and the same for
287        --exclude-dir, though that is not available in GNU grep, at least as of
288        release 2.5.4.
289    
290    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
291        characters from a string of bytes) have been redefined so as not to use
292        loops, in order to improve performance in some environments. At the same
293        time, I abstracted some of the common code into auxiliary macros to save
294        repetition (this should not affect the compiled code).
295    
296    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
297        compile-time error is now given if \c is not followed by an ASCII
298        character, that is, a byte less than 128. (In EBCDIC mode, the code is
299        different, and any byte value is allowed.)
300    
301    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
302        START_OPTIMIZE option, which is now allowed at compile time - but just
303        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
304        to pcregrep and other applications that have no direct access to PCRE
305        options. The new /Y option in pcretest sets this option when calling
306        pcre_compile().
307    
308    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
309        back references. Groups containing recursive back references were forced to
310        be atomic by that change, but in the case of named groups, the amount of
311        memory required was incorrectly computed, leading to "Failed: internal
312        error: code overflow". This has been fixed.
313    
314    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
315        pcretest.c, to avoid build problems in some Borland environments.
316    
317    
318    Version 8.10 25-Jun-2010
319    ------------------------
320    
321    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
322        THEN.
323    
324    2.  (*ACCEPT) was not working when inside an atomic group.
325    
326    3.  Inside a character class, \B is treated as a literal by default, but
327        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
328        causes the error). The code is unchanged, but I tidied the documentation.
329    
330    4.  Inside a character class, PCRE always treated \R and \X as literals,
331        whereas Perl faults them if its -w option is set. I have changed PCRE so
332        that it faults them when PCRE_EXTRA is set.
333    
334    5.  Added support for \N, which always matches any character other than
335        newline. (It is the same as "." when PCRE_DOTALL is not set.)
336    
337    6.  When compiling pcregrep with newer versions of gcc which may have
338        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
339        declared with attribute warn_unused_result" were given. Just casting the
340        result to (void) does not stop the warnings; a more elaborate fudge is
341        needed. I've used a macro to implement this.
342    
343    7.  Minor change to pcretest.c to avoid a compiler warning.
344    
345    8.  Added four artifical Unicode properties to help with an option to make
346        \s etc use properties (see next item). The new properties are: Xan
347        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
348    
349    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
350        use Unicode properties. (*UCP) at the start of a pattern can be used to set
351        this option. Modified pcretest to add /W to test this facility. Added
352        REG_UCP to make it available via the POSIX interface.
353    
354    10. Added --line-buffered to pcregrep.
355    
356    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
357        studied, and the match started with a letter with a code point greater than
358        127 whose first byte was different to the first byte of the other case of
359        the letter, the other case of this starting letter was not recognized
360        (#976).
361    
362    12. If a pattern that was studied started with a repeated Unicode property
363        test, for example, \p{Nd}+, there was the theoretical possibility of
364        setting up an incorrect bitmap of starting bytes, but fortunately it could
365        not have actually happened in practice until change 8 above was made (it
366        added property types that matched character-matching opcodes).
367    
368    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
369        possible starting bytes for non-anchored patterns.
370    
371    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
372        \R, and also a number of cases that involve Unicode properties, both
373        explicit and implicit when PCRE_UCP is set.
374    
375    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
376        input, it could crash or give wrong results if characters with values
377        greater than 0xc0 were present in the subject string. (Detail: it assumed
378        UTF-8 input when processing these items.)
379    
380    16. Added a lot of (int) casts to avoid compiler warnings in systems where
381        size_t is 64-bit (#991).
382    
383    17. Added a check for running out of memory when PCRE is compiled with
384        --disable-stack-for-recursion (#990).
385    
386    18. If the last data line in a file for pcretest does not have a newline on
387        the end, a newline was missing in the output.
388    
389    19. The default pcre_chartables.c file recognizes only ASCII characters (values
390        less than 128) in its various bitmaps. However, there is a facility for
391        generating tables according to the current locale when PCRE is compiled. It
392        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
393        characters, are recognized by isspace() and therefore were getting set in
394        these tables, and indeed these tables seem to approximate to ISO 8859. This
395        caused a problem in UTF-8 mode when pcre_study() was used to create a list
396        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
397        which of course cannot start UTF-8 characters. I have changed the code so
398        that only real ASCII characters (less than 128) and the correct starting
399        bytes for UTF-8 encodings are set for characters greater than 127 when in
400        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
401        altogether.)
402    
403    20. Added the /T option to pcretest so as to be able to run tests with non-
404        standard character tables, thus making it possible to include the tests
405        used for 19 above in the standard set of tests.
406    
407    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
408        reference to a subpattern the other side of a comment that contains an
409        opening parenthesis caused either an internal compiling error, or a
410        reference to the wrong subpattern.
411    
412    
413    Version 8.02 19-Mar-2010
414    ------------------------
415    
416    1.  The Unicode data tables have been updated to Unicode 5.2.0.
417    
418    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
419        configured.
420    
421    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
422        original author of that file, following a query about its status.
423    
424    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
425        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
426    
427    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
428        quantifier applied to a forward-referencing subroutine call, could compile
429        incorrect code or give the error "internal error: previously-checked
430        referenced subpattern not found".
431    
432    6.  Both MS Visual Studio and Symbian OS have problems with initializing
433        variables to point to external functions. For these systems, therefore,
434        pcre_malloc etc. are now initialized to local functions that call the
435        relevant global functions.
436    
437    7.  There were two entries missing in the vectors called coptable and poptable
438        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
439        I've fixed the data, and added a kludgy way of testing at compile time that
440        the lengths are correct (equal to the number of opcodes).
441    
442    8.  Following on from 7, I added a similar kludge to check the length of the
443        eint vector in pcreposix.c.
444    
445    9.  Error texts for pcre_compile() are held as one long string to avoid too
446        much relocation at load time. To find a text, the string is searched,
447        counting zeros. There was no check for running off the end of the string,
448        which could happen if a new error number was added without updating the
449        string.
450    
451    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
452    
453    11. \K was not working if it appeared in an atomic group or in a group that
454        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
455        \K is "not well defined" if used in an assertion. PCRE now accepts it if
456        the assertion is positive, but not if it is negative.
457    
458    12. Change 11 fortuitously reduced the size of the stack frame used in the
459        "match()" function of pcre_exec.c by one pointer. Forthcoming
460        implementation of support for (*MARK) will need an extra pointer on the
461        stack; I have reserved it now, so that the stack frame size does not
462        decrease.
463    
464    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
465        item in branch that calls a recursion is a subroutine call - as in the
466        second branch in the above example - was incorrectly given the compile-
467        time error "recursive call could loop indefinitely" because pcre_compile()
468        was not correctly checking the subroutine for matching a non-empty string.
469    
470    14. The checks for overrunning compiling workspace could trigger after an
471        overrun had occurred. This is a "should never occur" error, but it can be
472        triggered by pathological patterns such as hundreds of nested parentheses.
473        The checks now trigger 100 bytes before the end of the workspace.
474    
475    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
476    
477    
478    Version 8.01 19-Jan-2010
479    ------------------------
480    
481    1.  If a pattern contained a conditional subpattern with only one branch (in
482        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
483        computed the wrong minimum data length (which is of course zero for such
484        subpatterns). This could cause incorrect "no match" results.
485    
486    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
487        the pattern is reset in the first branch, pcre_compile() failed with
488        "internal error: code overflow at offset...". This happened only when
489        the reset was to the original external option setting. (An optimization
490        abstracts leading options settings into an external setting, which was the
491        cause of this.)
492    
493    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
494        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
495        assertion pattern did not match (meaning that the assertion was true), it
496        was incorrectly treated as false if the SKIP had been reached during the
497        matching. This also applied to assertions used as conditions.
498    
499    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
500        assertion subpattern, including such a pattern used as a condition,
501        unpredictable results occurred, instead of the error return
502        PCRE_ERROR_DFA_UITEM.
503    
504    5.  The C++ GlobalReplace function was not working like Perl for the special
505        situation when an empty string is matched. It now does the fancy magic
506        stuff that is necessary.
507    
508    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
509        removed. (These were left over from very, very early versions of PCRE.)
510    
511    7.  Some cosmetic changes to the code to make life easier when compiling it
512        as part of something else:
513    
514        (a) Change DEBUG to PCRE_DEBUG.
515    
516        (b) In pcre_compile(), rename the member of the "branch_chain" structure
517            called "current" as "current_branch", to prevent a collision with the
518            Linux macro when compiled as a kernel module.
519    
520        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
521            prevent a collision with the Linux macro when compiled as a kernel
522            module.
523    
524    8.  In pcre_compile() there are some checks for integer overflows that used to
525        cast potentially large values to (double). This has been changed to that
526        when building, a check for int64_t is made, and if it is found, it is used
527        instead, thus avoiding the use of floating point arithmetic. (There is no
528        other use of FP in PCRE.) If int64_t is not found, the fallback is to
529        double.
530    
531    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
532        2005 (difference between two addresses compared to an unsigned value).
533    
534    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
535        custom one, because of the following reported problem in Windows:
536    
537          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
538              under Win32.
539          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
540              therefore missing the function definition.
541          - The compiler thus generates a "C" signature for the test function.
542          - The linker fails to find the "C" function.
543          - PCRE fails to configure if asked to do so against libbz2.
544    
545    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
546        messages were output:
547    
548          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
549          rerunning libtoolize, to keep the correct libtool macros in-tree.
550          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
551    
552        I have done both of these things.
553    
554    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
555        most of the time, it *can* run out if it is given a pattern that contains a
556        runaway infinite recursion. I updated the discussion in the pcrestack man
557        page.
558    
559    13. Now that we have gone to the x.xx style of version numbers, the minor
560        version may start with zero. Using 08 or 09 is a bad idea because users
561        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
562        interpreted as invalid octal numbers. I've updated the previous comment in
563        configure.ac, and also added a check that gives an error if 08 or 09 are
564        used.
565    
566    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
567        causing partial matching to fail when the end of the subject matched \W
568        in a UTF-8 pattern where \W was quantified with a minimum of 3.
569    
570    15. There were some discrepancies between the declarations in pcre_internal.h
571        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
572        their definitions. The declarations used "const uschar *" and the
573        definitions used USPTR. Even though USPTR is normally defined as "const
574        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
575        reported that: "This difference in casting confuses some C++ compilers, for
576        example, SunCC recognizes above declarations as different functions and
577        generates broken code for hbpcre." I have changed the declarations to use
578        USPTR.
579    
580    16. GNU libtool is named differently on some systems. The autogen.sh script now
581        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
582        (FreeBSD).
583    
584    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
585        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
586        comment: "Figure out how to create a longlong from a string: strtoll and
587        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
588        instance, but it only takes 2 args instead of 3!"
589    
590    18. A subtle bug concerned with back references has been fixed by a change of
591        specification, with a corresponding code fix. A pattern such as
592        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
593        refers, was giving matches when it shouldn't. For example, xa=xaaa would
594        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
595        same bug. Such groups have to be quantified to be useful, or contained
596        inside another quantified group. (If there's no repetition, the reference
597        can never match.) The problem arises because, having left the group and
598        moved on to the rest of the pattern, a later failure that backtracks into
599        the group uses the captured value from the final iteration of the group
600        rather than the correct earlier one. I have fixed this in PCRE by forcing
601        any group that contains a reference to itself to be an atomic group; that
602        is, there cannot be any backtracking into it once it has completed. This is
603        similar to recursive and subroutine calls.
604    
605    
606    Version 8.00 19-Oct-09
607    ----------------------
608    
609    1.  The table for translating pcre_compile() error codes into POSIX error codes
610        was out-of-date, and there was no check on the pcre_compile() error code
611        being within the table. This could lead to an OK return being given in
612        error.
613    
614    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
615        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
616        in a Windows environment.
617    
618    3.  The pcregrep --count option prints the count for each file even when it is
619        zero, as does GNU grep. However, pcregrep was also printing all files when
620        --files-with-matches was added. Now, when both options are given, it prints
621        counts only for those files that have at least one match. (GNU grep just
622        prints the file name in this circumstance, but including the count seems
623        more useful - otherwise, why use --count?) Also ensured that the
624        combination -clh just lists non-zero counts, with no names.
625    
626    4.  The long form of the pcregrep -F option was incorrectly implemented as
627        --fixed_strings instead of --fixed-strings. This is an incompatible change,
628        but it seems right to fix it, and I didn't think it was worth preserving
629        the old behaviour.
630    
631    5.  The command line items --regex=pattern and --regexp=pattern were not
632        recognized by pcregrep, which required --regex pattern or --regexp pattern
633        (with a space rather than an '='). The man page documented the '=' forms,
634        which are compatible with GNU grep; these now work.
635    
636    6.  No libpcreposix.pc file was created for pkg-config; there was just
637        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
638    
639    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
640        when UCP support is not needed, by modifying the Python script that
641        generates it from Unicode data files. This should not matter if the module
642        is correctly used as a library, but I received one complaint about 50K of
643        unwanted data. My guess is that the person linked everything into his
644        program rather than using a library. Anyway, it does no harm.
645    
646    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
647        was a minimum greater than 1 for a wide character in a possessive
648        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
649        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
650        character. Chaos in the form of incorrect output or a compiling loop could
651        result.
652    
653    9.  The restrictions on what a pattern can contain when partial matching is
654        requested for pcre_exec() have been removed. All patterns can now be
655        partially matched by this function. In addition, if there are at least two
656        slots in the offset vector, the offset of the earliest inspected character
657        for the match and the offset of the end of the subject are set in them when
658        PCRE_ERROR_PARTIAL is returned.
659    
660    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
661        synonymous with PCRE_PARTIAL, for backwards compatibility, and
662        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
663        and may be more useful for multi-segment matching.
664    
665    11. Partial matching with pcre_exec() is now more intuitive. A partial match
666        used to be given if ever the end of the subject was reached; now it is
667        given only if matching could not proceed because another character was
668        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
669        string "Z", which now yields "no match" instead of "partial match". In the
670        case of pcre_dfa_exec(), "no match" is given if every matching path for the
671        final character ended with (*FAIL).
672    
673    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
674        if the pattern had a "must contain" character that was already found in the
675        earlier partial match, unless partial matching was again requested. For
676        example, with the pattern /dog.(body)?/, the "must contain" character is
677        "g". If the first part-match was for the string "dog", restarting with
678        "sbody" failed. This bug has been fixed.
679    
680    13. The string returned by pcre_dfa_exec() after a partial match has been
681        changed so that it starts at the first inspected character rather than the
682        first character of the match. This makes a difference only if the pattern
683        starts with a lookbehind assertion or \b or \B (\K is not supported by
684        pcre_dfa_exec()). It's an incompatible change, but it makes the two
685        matching functions compatible, and I think it's the right thing to do.
686    
687    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
688        so that the demonstration program is easily available in environments where
689        PCRE has not been installed from source.
690    
691    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
692        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
693        library.
694    
695    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
696        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
697        is not the first non-POSIX option to be added. Clearly some people find
698        these options useful.
699    
700    17. If a caller to the POSIX matching function regexec() passes a non-zero
701        value for nmatch with a NULL value for pmatch, the value of
702        nmatch is forced to zero.
703    
704    18. RunGrepTest did not have a test for the availability of the -u option of
705        the diff command, as RunTest does. It now checks in the same way as
706        RunTest, and also checks for the -b option.
707    
708    19. If an odd number of negated classes containing just a single character
709        interposed, within parentheses, between a forward reference to a named
710        subpattern and the definition of the subpattern, compilation crashed with
711        an internal error, complaining that it could not find the referenced
712        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
713        [The bug was that it was starting one character too far in when skipping
714        over the character class, thus treating the ] as data rather than
715        terminating the class. This meant it could skip too much.]
716    
717    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
718        /g option in pcretest when the pattern contains \K, which makes it possible
719        to have an empty string match not at the start, even when the pattern is
720        anchored. Updated pcretest and pcredemo to use this option.
721    
722    21. If the maximum number of capturing subpatterns in a recursion was greater
723        than the maximum at the outer level, the higher number was returned, but
724        with unset values at the outer level. The correct (outer level) value is
725        now given.
726    
727    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
728        PCRE did not set those parentheses (unlike Perl). I have now found a way to
729        make it do so. The string so far is captured, making this feature
730        compatible with Perl.
731    
732    23. The tests have been re-organized, adding tests 11 and 12, to make it
733        possible to check the Perl 5.10 features against Perl 5.10.
734    
735    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
736        pattern matches a fixed length string. PCRE did not allow this; now it
737        does. Neither allows recursion.
738    
739    25. I finally figured out how to implement a request to provide the minimum
740        length of subject string that was needed in order to match a given pattern.
741        (It was back references and recursion that I had previously got hung up
742        on.) This code has now been added to pcre_study(); it finds a lower bound
743        to the length of subject needed. It is not necessarily the greatest lower
744        bound, but using it to avoid searching strings that are too short does give
745        some useful speed-ups. The value is available to calling programs via
746        pcre_fullinfo().
747    
748    26. While implementing 25, I discovered to my embarrassment that pcretest had
749        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
750        study optimizations had never been tested with that matching function.
751        Oops. What is worse, even when it was passed study data, there was a bug in
752        pcre_dfa_exec() that meant it never actually used it. Double oops. There
753        were also very few tests of studied patterns with pcre_dfa_exec().
754    
755    27. If (?| is used to create subpatterns with duplicate numbers, they are now
756        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
757        on the other side of the coin, they are no longer allowed to have different
758        names, because these cannot be distinguished in PCRE, and this has caused
759        confusion. (This is a difference from Perl.)
760    
761    28. When duplicate subpattern names are present (necessarily with different
762        numbers, as required by 27 above), and a test is made by name in a
763        conditional pattern, either for a subpattern having been matched, or for
764        recursion in such a pattern, all the associated numbered subpatterns are
765        tested, and the overall condition is true if the condition is true for any
766        one of them. This is the way Perl works, and is also more like the way
767        testing by number works.
768    
769    
770    Version 7.9 11-Apr-09
771    ---------------------
772    
773    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
774        (pcretest), all targets were linked against these libraries. This included
775        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
776        libraries. This caused unwanted dependencies to be created. This problem
777        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
778        pcretest is linked with readline.
779    
780    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
781        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
782        moved outside it again, because FALSE and TRUE are already defined in AIX,
783        but BOOL is not.
784    
785    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
786        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
787    
788    4.  The pcregrep documentation said spaces were inserted as well as colons (or
789        hyphens) following file names and line numbers when outputting matching
790        lines. This is not true; no spaces are inserted. I have also clarified the
791        wording for the --colour (or --color) option.
792    
793    5.  In pcregrep, when --colour was used with -o, the list of matching strings
794        was not coloured; this is different to GNU grep, so I have changed it to be
795        the same.
796    
797    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
798        each matching line was coloured. Now it goes on to look for further matches
799        of any of the test patterns, which is the same behaviour as GNU grep.
800    
801    7.  A pattern that could match an empty string could cause pcregrep to loop; it
802        doesn't make sense to accept an empty string match in pcregrep, so I have
803        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
804        seems to be how GNU grep behaves.
805    
806    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
807        start or after a newline", because the conditional assertion was not being
808        correctly handled. The rule now is that both the assertion and what follows
809        in the first alternative must satisfy the test.
810    
811    9.  If auto-callout was enabled in a pattern with a conditional group whose
812        condition was an assertion, PCRE could crash during matching, both with
813        pcre_exec() and pcre_dfa_exec().
814    
815    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
816        used for matching.
817    
818    11. Unicode property support in character classes was not working for
819        characters (bytes) greater than 127 when not in UTF-8 mode.
820    
821    12. Added the -M command line option to pcretest.
822    
823    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
824    
825    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
826    
827    16. Added comments and documentation about mis-use of no_arg in the C++
828        wrapper.
829    
830    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
831        from Martin Jerabek that uses macro names for all relevant character and
832        string constants.
833    
834    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
835        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
836        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
837        these, but not everybody uses configure.
838    
839    19. A conditional group that had only one branch was not being correctly
840        recognized as an item that could match an empty string. This meant that an
841        enclosing group might also not be so recognized, causing infinite looping
842        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
843        with the subject "ab", where knowledge that the repeated group can match
844        nothing is needed in order to break the loop.
845    
846    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
847        exec(), but without supplying a callout function, matching went wrong.
848    
849    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
850        leak if the size of the offset vector was greater than 30. When the vector
851        is smaller, the saved offsets during recursion go onto a local stack
852        vector, but for larger vectors malloc() is used. It was failing to free
853        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
854        error, in fact).
855    
856    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
857        heapframe that is used only when UTF-8 support is enabled. This caused no
858        problem, but was untidy.
859    
860    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
861        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
862        included within another project.
863    
864    24. Steven Van Ingelgem's patches to add more options to the CMake support,
865        slightly modified by me:
866    
867          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
868              not building pcregrep.
869    
870          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
871              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
872    
873    25. Forward references, both numeric and by name, in patterns that made use of
874        duplicate group numbers, could behave incorrectly or give incorrect errors,
875        because when scanning forward to find the reference group, PCRE was not
876        taking into account the duplicate group numbers. A pattern such as
877        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
878    
879    26. Changed a few more instances of "const unsigned char *" to USPTR, making
880        the feature of a custom pointer more persuasive (as requested by a user).
881    
882    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
883        pcretest.c, inside #ifndefs, because it seems they are sometimes already
884        pre-defined.
885    
886    28. Added support for (*UTF8) at the start of a pattern.
887    
888    29. Arrange for flags added by the "release type" setting in CMake to be shown
889        in the configuration summary.
890    
891    
892    Version 7.8 05-Sep-08
893    ---------------------
894    
895    1.  Replaced UCP searching code with optimized version as implemented for Ad
896        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
897        stage table and inline lookup instead of a function, giving speed ups of 2
898        to 5 times on some simple patterns that I tested. Permission was given to
899        distribute the MultiStage2.py script that generates the tables (it's not in
900        the tarball, but is in the Subversion repository).
901    
902    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
903        scripts.
904    
905    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
906        a group with a zero qualifier. The result of the study could be incorrect,
907        or the function might crash, depending on the pattern.
908    
909    4.  Caseless matching was not working for non-ASCII characters in back
910        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
911        It now works when Unicode Property Support is available.
912    
913    5.  In pcretest, an escape such as \x{de} in the data was always generating
914        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
915        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
916        truncation.
917    
918    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
919    
920    7.  Added two (int) casts to pcregrep when printing the difference of two
921        pointers, in case they are 64-bit values.
922    
923    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
924        test 2 if it fails.
925    
926    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
927        and a #define of that name to empty if it is not externally set. This is to
928        allow users of MSVC to set it if necessary.
929    
930    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
931        the convenience functions in the pcre_get.c source file.
932    
933    11. An option change at the start of a pattern that had top-level alternatives
934        could cause overwriting and/or a crash. This command provoked a crash in
935        some environments:
936    
937          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
938    
939        This potential security problem was recorded as CVE-2008-2371.
940    
941    12. For a pattern where the match had to start at the beginning or immediately
942        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
943        pcre_dfa_exec() could read past the end of the passed subject if there was
944        no match. To help with detecting such bugs (e.g. with valgrind), I modified
945        pcretest so that it places the subject at the end of its malloc-ed buffer.
946    
947    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
948        exec() might read past the end of the data buffer in UTF-8 mode.
949    
950    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
951        the data contained the byte 0x85 as part of a UTF-8 character within its
952        first line. This applied both to normal and DFA matching.
953    
954    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
955        /^[^d]*?$/8 failed to match "abc".
956    
957    16. Added a missing copyright notice to pcrecpp_internal.h.
958    
959    17. Make it more clear in the documentation that values returned from
960        pcre_exec() in ovector are byte offsets, not character counts.
961    
962    18. Tidied a few places to stop certain compilers from issuing warnings.
963    
964    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
965        supplied by Stefan Weber. I made a further small update for 7.8 because
966        there is a change of source arrangements: the pcre_searchfuncs.c module is
967        replaced by pcre_ucd.c.
968    
969    
970    Version 7.7 07-May-08
971    ---------------------
972    
973    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
974        a string to a long long, pretend we don't even have a long long." This is
975        done by checking for the strtoq, strtoll, and _strtoi64 functions.
976    
977    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
978        pre-7.6 versions, which defined a global no_arg variable instead of putting
979        it in the RE class. (See also #8 below.)
980    
981    3.  Remove a line of dead code, identified by coverity and reported by Nuno
982        Lopes.
983    
984    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
985    
986        (1) The include/exclude patterns were being applied to the whole pathnames
987            of files, instead of just to the final components.
988    
989        (2) If there was more than one level of directory, the subdirectories were
990            skipped unless they satisfied the include/exclude conditions. This is
991            inconsistent with GNU grep (and could even be seen as contrary to the
992            pcregrep specification - which I improved to make it absolutely clear).
993            The action now is always to scan all levels of directory, and just
994            apply the include/exclude patterns to regular files.
995    
996    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
997        --exclude_dir in the tests to avoid scanning .svn directories.
998    
999    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
1000        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
1001        doesn't support NULs in patterns.
1002    
1003    7.  Added some missing "const"s to declarations of static tables in
1004        pcre_compile.c and pcre_dfa_exec.c.
1005    
1006    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
1007        caused by fix #2  above. (Subsequently also a second patch to fix the
1008        first patch. And a third patch - this was a messy problem.)
1009    
1010    9.  Applied Craig's patch to remove the use of push_back().
1011    
1012    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
1013        matching function regexec().
1014    
1015    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
1016        which, however, unlike Perl's \g{...}, are subroutine calls, not back
1017        references. PCRE supports relative numbers with this syntax (I don't think
1018        Oniguruma does).
1019    
1020    12. Previously, a group with a zero repeat such as (...){0} was completely
1021        omitted from the compiled regex. However, this means that if the group
1022        was called as a subroutine from elsewhere in the pattern, things went wrong
1023        (an internal error was given). Such groups are now left in the compiled
1024        pattern, with a new opcode that causes them to be skipped at execution
1025        time.
1026    
1027    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
1028        to the way PCRE behaves:
1029    
1030        (a) A lone ] character is dis-allowed (Perl treats it as data).
1031    
1032        (b) A back reference to an unmatched subpattern matches an empty string
1033            (Perl fails the current match path).
1034    
1035        (c) A data ] in a character class must be notated as \] because if the
1036            first data character in a class is ], it defines an empty class. (In
1037            Perl it is not possible to have an empty class.) The empty class []
1038            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
1039            The negative empty class [^] matches any one character, independently
1040            of the DOTALL setting.
1041    
1042    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
1043        non-existent subpattern following a character class starting with ']' and
1044        containing () gave an internal compiling error instead of "reference to
1045        non-existent subpattern". Fortunately, when the pattern did exist, the
1046        compiled code was correct. (When scanning forwards to check for the
1047        existencd of the subpattern, it was treating the data ']' as terminating
1048        the class, so got the count wrong. When actually compiling, the reference
1049        was subsequently set up correctly.)
1050    
1051    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
1052        it was being rejected as not supported by pcre_dfa_exec(), even though
1053        other assertions are supported. I have made pcre_dfa_exec() support
1054        (*FAIL).
1055    
1056    16. The implementation of 13c above involved the invention of a new opcode,
1057        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
1058        cannot be changed at match time, I realized I could make a small
1059        improvement to matching performance by compiling OP_ALLANY instead of
1060        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
1061        on the OP_ANY path.
1062    
1063    17. Compiling pcretest on Windows with readline support failed without the
1064        following two fixes: (1) Make the unistd.h include conditional on
1065        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
1066    
1067    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
1068        ncurses library to be included for pcretest when ReadLine support is
1069        requested, but also to allow for it to be overridden. This patch came from
1070        Daniel Bergström.
1071    
1072    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
1073        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
1074        any errors with the current Unicode tables. Thanks to Peter Kankowski for
1075        spotting this.
1076    
1077    
1078    Version 7.6 28-Jan-08
1079    ---------------------
1080    
1081    1.  A character class containing a very large number of characters with
1082        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
1083        overflow.
1084    
1085    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
1086        HAVE_LONG_LONG is not defined.
1087    
1088    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
1089        bring them up to date and include new features. This patch includes:
1090    
1091        - Fixed PH's badly added libz and libbz2 support.
1092        - Fixed a problem with static linking.
1093        - Added pcredemo. [But later removed - see 7 below.]
1094        - Fixed dftables problem and added an option.
1095        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
1096            HAVE_LONG_LONG.
1097        - Added readline support for pcretest.
1098        - Added an listing of the option settings after cmake has run.
1099    
1100    4.  A user submitted a patch to Makefile that makes it easy to create
1101        "pcre.dll" under mingw when using Configure/Make. I added stuff to
1102        Makefile.am that cause it to include this special target, without
1103        affecting anything else. Note that the same mingw target plus all
1104        the other distribution libraries and programs are now supported
1105        when configuring with CMake (see 6 below) instead of with
1106        Configure/Make.
1107    
1108    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
1109        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
1110        exported in the Windows port". It has not yet been confirmed that the patch
1111        solves the problem, but it does no harm.
1112    
1113    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
1114        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
1115        with CMake, and also correct the comment about stack recursion.
1116    
1117    7.  Remove the automatic building of pcredemo from the ./configure system and
1118        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
1119        of a program that users should build themselves after PCRE is installed, so
1120        building it automatically is not really right. What is more, it gave
1121        trouble in some build environments.
1122    
1123    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
1124    
1125    
1126    Version 7.5 10-Jan-08
1127    ---------------------
1128    
1129    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
1130        values in parens when parsing an RE using the C++ wrapper."
1131    
1132    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
1133        Characters greater than 255 were excluded from the class instead of being
1134        included.
1135    
1136    3.  The same bug as (2) above applied to negated POSIX classes such as
1137        [:^space:].
1138    
1139    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
1140        defined or documented. It seems to have been a typo for PCRE_STATIC, so
1141        I have changed it.
1142    
1143    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
1144        first named subpattern) and a construct such as (?&a) would reference the
1145        first named subpattern whose name started with "a" (in other words, the
1146        length check was missing). Both these problems are fixed. "Subpattern name
1147        expected" is now given for (?&) (a zero-length name), and this patch also
1148        makes it give the same error for \k'' (previously it complained that that
1149        was a reference to a non-existent subpattern).
1150    
1151    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
1152        this is right because (?- can be followed by option settings as well as by
1153        digits. I have, however, made the messages clearer.
1154    
1155    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
1156        than the number used in the conditional) now cause a compile-time error.
1157        This is actually not compatible with Perl, which accepts such patterns, but
1158        treats the conditional as always being FALSE (as PCRE used to), but it
1159        seems to me that giving a diagnostic is better.
1160    
1161    8.  Change "alphameric" to the more common word "alphanumeric" in comments
1162        and messages.
1163    
1164    9.  Fix two occurrences of "backslash" in comments that should have been
1165        "backspace".
1166    
1167    10. Remove two redundant lines of code that can never be obeyed (their function
1168        was moved elsewhere).
1169    
1170    11. The program that makes PCRE's Unicode character property table had a bug
1171        which caused it to generate incorrect table entries for sequences of
1172        characters that have the same character type, but are in different scripts.
1173        It amalgamated them into a single range, with the script of the first of
1174        them. In other words, some characters were in the wrong script. There were
1175        thirteen such cases, affecting characters in the following ranges:
1176    
1177          U+002b0 - U+002c1
1178          U+0060c - U+0060d
1179          U+0061e - U+00612
1180          U+0064b - U+0065e
1181          U+0074d - U+0076d
1182          U+01800 - U+01805
1183          U+01d00 - U+01d77
1184          U+01d9b - U+01dbf
1185          U+0200b - U+0200f
1186          U+030fc - U+030fe
1187          U+03260 - U+0327f
1188          U+0fb46 - U+0fbb1
1189          U+10450 - U+1049d
1190    
1191    12. The -o option (show only the matching part of a line) for pcregrep was not
1192        compatible with GNU grep in that, if there was more than one match in a
1193        line, it showed only the first of them. It now behaves in the same way as
1194        GNU grep.
1195    
1196    13. If the -o and -v options were combined for pcregrep, it printed a blank
1197        line for every non-matching line. GNU grep prints nothing, and pcregrep now
1198        does the same. The return code can be used to tell if there were any
1199        non-matching lines.
1200    
1201    14. Added --file-offsets and --line-offsets to pcregrep.
1202    
1203    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
1204        infinitely looping recursion. The bug was that positive lookaheads were not
1205        being skipped when checking for a possible empty match (negative lookaheads
1206        and both kinds of lookbehind were skipped).
1207    
1208    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
1209        inclusion of <windows.h> to before rather than after the definition of
1210        INVALID_FILE_ATTRIBUTES (patch from David Byron).
1211    
1212    17. Specifying a possessive quantifier with a specific limit for a Unicode
1213        character property caused pcre_compile() to compile bad code, which led at
1214        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
1215        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
1216        caused the error; without that there was no problem.
1217    
1218    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
1219    
1220    19. Added --enable-pcretest-libreadline.
1221    
1222    20. In pcrecpp.cc, the variable 'count' was incremented twice in
1223        RE::GlobalReplace(). As a result, the number of replacements returned was
1224        double what it should be. I removed one of the increments, but Craig sent a
1225        later patch that removed the other one (the right fix) and added unit tests
1226        that check the return values (which was not done before).
1227    
1228    21. Several CMake things:
1229    
1230        (1) Arranged that, when cmake is used on Unix, the libraries end up with
1231            the names libpcre and libpcreposix, not just pcre and pcreposix.
1232    
1233        (2) The above change means that pcretest and pcregrep are now correctly
1234            linked with the newly-built libraries, not previously installed ones.
1235    
1236        (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
1237    
1238    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
1239        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
1240        UTF-8 newline character). The key issue is that the pattern starts .*;
1241        this means that the match must be either at the beginning, or after a
1242        newline. The bug was in the code for advancing after a failed match and
1243        checking that the new position followed a newline. It was not taking
1244        account of UTF-8 characters correctly.
1245    
1246    23. PCRE was behaving differently from Perl in the way it recognized POSIX
1247        character classes. PCRE was not treating the sequence [:...:] as a
1248        character class unless the ... were all letters. Perl, however, seems to
1249        allow any characters between [: and :], though of course it rejects as
1250        unknown any "names" that contain non-letters, because all the known class
1251        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
1252        for example, whereas PCRE did not - it did not recognize a POSIX character
1253        class. This seemed a bit dangerous, so the code has been changed to be
1254        closer to Perl. The behaviour is not identical to Perl, because PCRE will
1255        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
1256        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
1257        Perl does, and where it didn't before.
1258    
1259    24. Rewrite so as to remove the single use of %n from pcregrep because in some
1260        Windows environments %n is disabled by default.
1261    
1262    
1263    Version 7.4 21-Sep-07
1264    ---------------------
1265    
1266    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
1267        means that a class such as [\s] counted as "explicit reference to CR or
1268        LF". That isn't really right - the whole point of the change was to try to
1269        help when there was an actual mention of one of the two characters. So now
1270        the change happens only if \r or \n (or a literal CR or LF) character is
1271        encountered.
1272    
1273    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
1274        of both had grown to the point where there were only 3 bits left.
1275        Fortunately, there was spare space in the data structure, and so I have
1276        moved the internal flags into a new 16-bit field to free up more option
1277        bits.
1278    
1279    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
1280        but did not set the internal JCHANGED flag - either of these is enough to
1281        control the way the "get" function works - but the PCRE_INFO_JCHANGED
1282        facility is supposed to tell if (?J) was ever used, so now (?J) at the
1283        start sets both bits.
1284    
1285    4.  Added options (at build time, compile time, exec time) to change \R from
1286        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
1287    
1288    5.  doc/pcresyntax.html was missing from the distribution.
1289    
1290    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
1291        compatibility, even though it is no longer used.
1292    
1293    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
1294        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
1295        windows.h file is present (where different names are used). [This was
1296        reversed later after testing - see 16 below.]
1297    
1298    8.  Changed all #include <config.h> to #include "config.h". There were also
1299        some further <pcre.h> cases that I changed to "pcre.h".
1300    
1301    9.  When pcregrep was used with the --colour option, it missed the line ending
1302        sequence off the lines that it output.
1303    
1304    10. It was pointed out to me that arrays of string pointers cause lots of
1305        relocations when a shared library is dynamically loaded. A technique of
1306        using a single long string with a table of offsets can drastically reduce
1307        these. I have refactored PCRE in four places to do this. The result is
1308        dramatic:
1309    
1310          Originally:                          290
1311          After changing UCP table:            187
1312          After changing error message table:   43
1313          After changing table of "verbs"       36
1314          After changing table of Posix names   22
1315    
1316        Thanks to the folks working on Gregex for glib for this insight.
1317    
1318    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
1319        unicode-properties was also set.
1320    
1321    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
1322    
1323    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
1324        checked only for CRLF.
1325    
1326    14. Added casts to pcretest.c to avoid compiler warnings.
1327    
1328    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
1329    
1330    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
1331        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
1332        entirely. This removes changes made in 7 above.
1333    
1334    17. The CMake files have been updated, and there is now more information about
1335        building with CMake in the NON-UNIX-USE document.
1336    
1337    
1338    Version 7.3 28-Aug-07
1339    ---------------------
1340    
1341     1. In the rejigging of the build system that eventually resulted in 7.1, the
1342        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
1343        brackets there is not right, since it causes compilers to look for an
1344        installed pcre.h, not the version that is in the source that is being
1345        compiled (which of course may be different). I have changed it back to:
1346    
1347          #include "pcre.h"
1348    
1349        I have a vague recollection that the change was concerned with compiling in
1350        different directories, but in the new build system, that is taken care of
1351        by the VPATH setting the Makefile.
1352    
1353     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
1354        when the subject happened to end in the byte 0x85 (e.g. if the last
1355        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
1356        characters but of course it shouldn't be taken as a newline when it is part
1357        of another character. The bug was that, for an unlimited repeat of . in
1358        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
1359        characters when looking for a newline.
1360    
1361     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
1362    
1363     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
1364        in debug output.
1365    
1366     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
1367        long printing in the pcrecpp unittest when running under MinGW.
1368    
1369     6. ESC_K was left out of the EBCDIC table.
1370    
1371     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
1372        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
1373        limit also applies to "virtual nesting" when a pattern is recursive, and in
1374        this case 1000 isn't so big. I have been able to remove this limit at the
1375        expense of backing off one optimization in certain circumstances. Normally,
1376        when pcre_exec() would call its internal match() function recursively and
1377        immediately return the result unconditionally, it uses a "tail recursion"
1378        feature to save stack. However, when a subpattern that can match an empty
1379        string has an unlimited repetition quantifier, it no longer makes this
1380        optimization. That gives it a stack frame in which to save the data for
1381        checking that an empty string has been matched. Previously this was taken
1382        from the 1000-entry workspace that had been reserved. So now there is no
1383        explicit limit, but more stack is used.
1384    
1385     8. Applied Daniel's patches to solve problems with the import/export magic
1386        syntax that is required for Windows, and which was going wrong for the
1387        pcreposix and pcrecpp parts of the library. These were overlooked when this
1388        problem was solved for the main library.
1389    
1390     9. There were some crude static tests to avoid integer overflow when computing
1391        the size of patterns that contain repeated groups with explicit upper
1392        limits. As the maximum quantifier is 65535, the maximum group length was
1393        set at 30,000 so that the product of these two numbers did not overflow a
1394        32-bit integer. However, it turns out that people want to use groups that
1395        are longer than 30,000 bytes (though not repeat them that many times).
1396        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1397        made it possible to implement the integer overflow checks in a much more
1398        dynamic way, which I have now done. The artificial limitation on group
1399        length has been removed - we now have only the limit on the total length of
1400        the compiled pattern, which depends on the LINK_SIZE setting.
1401    
1402    10. Fixed a bug in the documentation for get/copy named substring when
1403        duplicate names are permitted. If none of the named substrings are set, the
1404        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1405        empty string.
1406    
1407    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1408        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1409        because the ] is interpreted as the first data character and the
1410        terminating ] is not found. PCRE has been made compatible with Perl in this
1411        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1412        cause memory overwriting.
1413    
1414    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1415        string has been matched (to stop an infinite loop). It was not recognizing
1416        a conditional subpattern that could match an empty string if that
1417        subpattern was within another subpattern. For example, it looped when
1418        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
1419        condition was not nested. This bug has been fixed.
1420    
1421    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1422        past the start of the subject in the presence of bytes with the top bit
1423        set, for example "\x8aBCD".
1424    
1425    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1426        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1427    
1428    14. Optimized (?!) to (*FAIL).
1429    
1430    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1431        This restricts code points to be within the range 0 to 0x10FFFF, excluding
1432        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1433        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1434        does: it's just the validity check that is more restrictive.
1435    
1436    16. Inserted checks for integer overflows during escape sequence (backslash)
1437        processing, and also fixed erroneous offset values for syntax errors during
1438        backslash processing.
1439    
1440    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1441        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1442    
1443    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1444        caused an overrun.
1445    
1446    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1447        something other than just ASCII characters) inside a group that had an
1448        unlimited repeat caused a loop at compile time (while checking to see
1449        whether the group could match an empty string).
1450    
1451    20. Debugging a pattern containing \p or \P could cause a crash. For example,
1452        [\P{Any}] did so. (Error in the code for printing property names.)
1453    
1454    21. An orphan \E inside a character class could cause a crash.
1455    
1456    22. A repeated capturing bracket such as (A)? could cause a wild memory
1457        reference during compilation.
1458    
1459    23. There are several functions in pcre_compile() that scan along a compiled
1460        expression for various reasons (e.g. to see if it's fixed length for look
1461        behind). There were bugs in these functions when a repeated \p or \P was
1462        present in the pattern. These operators have additional parameters compared
1463        with \d, etc, and these were not being taken into account when moving along
1464        the compiled data. Specifically:
1465    
1466        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1467            length.
1468    
1469        (b) An item such as \pL+ within a repeated group could cause crashes or
1470            loops.
1471    
1472        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1473            "reference to non-existent subpattern" error.
1474    
1475        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1476    
1477    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1478        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1479    
1480    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1481    
1482    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1483        character were causing crashes (broken optimization).
1484    
1485    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1486        \p or \P) caused a compile-time loop.
1487    
1488    28. More problems have arisen in unanchored patterns when CRLF is a valid line
1489        break. For example, the unstudied pattern [\r\n]A does not match the string
1490        "\r\nA" because change 7.0/46 below moves the current point on by two
1491        characters after failing to match at the start. However, the pattern \nA
1492        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
1493        the same is true. There doesn't seem any very clean way out of this, but
1494        what I have chosen to do makes the common cases work: PCRE now takes note
1495        of whether there can be an explicit match for \r or \n anywhere in the
1496        pattern, and if so, 7.0/46 no longer applies. As part of this change,
1497        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1498        pattern has explicit CR or LF references.
1499    
1500    29. Added (*CR) etc for changing newline setting at start of pattern.
1501    
1502    
1503    Version 7.2 19-Jun-07
1504    ---------------------
1505    
1506     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
1507        which is apparently normally available under Windows.
1508    
1509     2. Re-jig the pcregrep tests with different newline settings in an attempt
1510        to make them independent of the local environment's newline setting.
1511    
1512     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
1513    
1514     4. Some of the "internals" tests were previously cut out when the link size
1515        was not 2, because the output contained actual offsets. The recent new
1516        "Z" feature of pcretest means that these can be cut out, making the tests
1517        usable with all link sizes.
1518    
1519     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
1520        stack recursion. This gives a massive performance boost under BSD, but just
1521        a small improvement under Linux. However, it saves one field in the frame
1522        in all cases.
1523    
1524     6. Added more features from the forthcoming Perl 5.10:
1525    
1526        (a) (?-n) (where n is a string of digits) is a relative subroutine or
1527            recursion call. It refers to the nth most recently opened parentheses.
1528    
1529        (b) (?+n) is also a relative subroutine call; it refers to the nth next
1530            to be opened parentheses.
1531    
1532        (c) Conditions that refer to capturing parentheses can be specified
1533            relatively, for example, (?(-2)... or (?(+3)...
1534    
1535        (d) \K resets the start of the current match so that everything before
1536            is not part of it.
1537    
1538        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1539    
1540        (f) \g{name} is another synonym - part of Perl 5.10's unification of
1541            reference syntax.
1542    
1543        (g) (?| introduces a group in which the numbering of parentheses in each
1544            alternative starts with the same number.
1545    
1546        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
1547    
1548     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
1549        PCRE_INFO_JCHANGED.
1550    
1551     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
1552        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
1553        for detecting groups that can match an empty string.
1554    
1555     9. A pattern with a very large number of alternatives (more than several
1556        hundred) was running out of internal workspace during the pre-compile
1557        phase, where pcre_compile() figures out how much memory will be needed. A
1558        bit of new cunning has reduced the workspace needed for groups with
1559        alternatives. The 1000-alternative test pattern now uses 12 bytes of
1560        workspace instead of running out of the 4096 that are available.
1561    
1562    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
1563    
1564    11. Applied patch from Google to remove an optimization that didn't quite work.
1565        The report of the bug said:
1566    
1567          pcrecpp::RE("a*").FullMatch("aaa") matches, while
1568          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
1569          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
1570    
1571    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
1572        it matched the wrong number of bytes.
1573    
1574    
1575    Version 7.1 24-Apr-07
1576    ---------------------
1577    
1578     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
1579        that is more "standard", making use of automake and other Autotools. There
1580        is some re-arrangement of the files and adjustment of comments consequent
1581        on this.
1582    
1583     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
1584        for recursive directory scanning broke on some systems because the files
1585        are not scanned in any specific order and on different systems the order
1586        was different. A call to "sort" has been inserted into RunGrepTest for the
1587        approprate test as a short-term fix. In the longer term there may be an
1588        alternative.
1589    
1590     3. I had an email from Eric Raymond about problems translating some of PCRE's
1591        man pages to HTML (despite the fact that I distribute HTML pages, some
1592        people do their own conversions for various reasons). The problems
1593        concerned the use of low-level troff macros .br and .in. I have therefore
1594        removed all such uses from the man pages (some were redundant, some could
1595        be replaced by .nf/.fi pairs). The 132html script that I use to generate
1596        HTML has been updated to handle .nf/.fi and to complain if it encounters
1597        .br or .in.
1598    
1599     4. Updated comments in configure.ac that get placed in config.h.in and also
1600        arranged for config.h to be included in the distribution, with the name
1601        config.h.generic, for the benefit of those who have to compile without
1602        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
1603    
1604     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
1605        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
1606        makevp.bat for the current PCRE, using the additional files
1607        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
1608    
1609     6. A Windows user reported a minor discrepancy with test 2, which turned out
1610        to be caused by a trailing space on an input line that had got lost in his
1611        copy. The trailing space was an accident, so I've just removed it.
1612    
1613     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
1614        that is needed.
1615    
1616     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
1617        as "const" (a) because they are and (b) because it helps the PHP
1618        maintainers who have recently made a script to detect big data structures
1619        in the php code that should be moved to the .rodata section. I remembered
1620        to update Builducptable as well, so it won't revert if ucptable.h is ever
1621        re-created.
1622    
1623     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
1624        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
1625        order to be able to cut out the UTF-8 tables in the latter when UTF-8
1626        support is not required. This saves 1.5-2K of code, which is important in
1627        some applications.
1628    
1629        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
1630        so as not to refer to the tables, even though these functions will never be
1631        called when UTF-8 support is disabled. Otherwise there are problems with a
1632        shared library.
1633    
1634    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
1635    
1636        (a) It was defining its arguments as char * instead of void *.
1637    
1638        (b) It was assuming that all moves were upwards in memory; this was true
1639            a long time ago when I wrote it, but is no longer the case.
1640    
1641        The emulated memove() is provided for those environments that have neither
1642        memmove() nor bcopy(). I didn't think anyone used it these days, but that
1643        is clearly not the case, as these two bugs were recently reported.
1644    
1645    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
1646        and Detrail to create the HTML documentation, the .txt form of the man
1647        pages, and it removes trailing spaces from listed files. It also creates
1648        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
1649        case, it wraps all the #defines with #ifndefs. This script should be run
1650        before "make dist".
1651    
1652    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1653        with Unicode property support.
1654    
1655        (a) For a maximizing quantifier, if the two different cases of the
1656            character were of different lengths in their UTF-8 codings (there are
1657            some cases like this - I found 11), and the matching function had to
1658            back up over a mixture of the two cases, it incorrectly assumed they
1659            were both the same length.
1660    
1661        (b) When PCRE was configured to use the heap rather than the stack for
1662            recursion during matching, it was not correctly preserving the data for
1663            the other case of a UTF-8 character when checking ahead for a match
1664            while processing a minimizing repeat. If the check also involved
1665            matching a wide character, but failed, corruption could cause an
1666            erroneous result when trying to check for a repeat of the original
1667            character.
1668    
1669    13. Some tidying changes to the testing mechanism:
1670    
1671        (a) The RunTest script now detects the internal link size and whether there
1672            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1673            values substituted by "configure". (The RunGrepTest script already did
1674            this for UTF-8.) The configure.ac script no longer substitutes the
1675            relevant variables.
1676    
1677        (b) The debugging options /B and /D in pcretest show the compiled bytecode
1678            with length and offset values. This means that the output is different
1679            for different internal link sizes. Test 2 is skipped for link sizes
1680            other than 2 because of this, bypassing the problem. Unfortunately,
1681            there was also a test in test 3 (the locale tests) that used /B and
1682            failed for link sizes other than 2. Rather than cut the whole test out,
1683            I have added a new /Z option to pcretest that replaces the length and
1684            offset values with spaces. This is now used to make test 3 independent
1685            of link size. (Test 2 will be tidied up later.)
1686    
1687    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1688        segmentation fault instead of returning the appropriate error message.
1689    
1690    15. In multiline mode when the newline sequence was set to "any", the pattern
1691        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1692        This doesn't seem right; it now treats the CRLF combination as the line
1693        ending, and so does not match in that case. It's only a pattern such as ^$
1694        that would hit this one: something like ^ABC$ would have failed after \r
1695        and then tried again after \r\n.
1696    
1697    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1698        in an attempt to make files that differ only in their line terminators
1699        compare equal. This works on Linux.
1700    
1701    17. Under certain error circumstances pcregrep might try to free random memory
1702        as it exited. This is now fixed, thanks to valgrind.
1703    
1704    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1705        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1706        was because its rules for how to advance for /g after matching an empty
1707        string at the end of a line did not allow for this case. They now check for
1708        it specially.
1709    
1710    20. pcretest is supposed to handle patterns and data of any length, by
1711        extending its buffers when necessary. It was getting this wrong when the
1712        buffer for a data line had to be extended.
1713    
1714    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1715        CRLF as a newline sequence.
1716    
1717    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1718        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1719        I have nevertheless tidied it up.
1720    
1721    23. Added some casts to kill warnings from HP-UX ia64 compiler.
1722    
1723    24. Added a man page for pcre-config.
1724    
1725    
1726    Version 7.0 19-Dec-06
1727    ---------------------
1728    
1729     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1730        moving to gcc 4.1.1.
1731    
1732     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1733        sys/time.h, which is documented as needed for this function. It doesn't
1734        seem to matter on Linux, but it showed up on some releases of OS X.
1735    
1736     3. It seems that there are systems where bytes whose values are greater than
1737        127 match isprint() in the "C" locale. The "C" locale should be the
1738        default when a C program starts up. In most systems, only ASCII printing
1739        characters match isprint(). This difference caused the output from pcretest
1740        to vary, making some of the tests fail. I have changed pcretest so that:
1741    
1742        (a) When it is outputting text in the compiled version of a pattern, bytes
1743            other than 32-126 are always shown as hex escapes.
1744    
1745        (b) When it is outputting text that is a matched part of a subject string,
1746            it does the same, unless a different locale has been set for the match
1747            (using the /L modifier). In this case, it uses isprint() to decide.
1748    
1749     4. Fixed a major bug that caused incorrect computation of the amount of memory
1750        required for a compiled pattern when options that changed within the
1751        pattern affected the logic of the preliminary scan that determines the
1752        length. The relevant options are -x, and -i in UTF-8 mode. The result was
1753        that the computed length was too small. The symptoms of this bug were
1754        either the PCRE error "internal error: code overflow" from pcre_compile(),
1755        or a glibc crash with a message such as "pcretest: free(): invalid next
1756        size (fast)". Examples of patterns that provoked this bug (shown in
1757        pcretest format) are:
1758    
1759          /(?-x: )/x
1760          /(?x)(?-x: \s*#\s*)/
1761          /((?i)[\x{c0}])/8
1762          /(?i:[\x{c0}])/8
1763    
1764        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1765        is now done differently.
1766    
1767     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1768        wrapper classes; (b) implement a new function in the C++ scanner that is
1769        more efficient than the old way of doing things because it avoids levels of
1770        recursion in the regex matching; (c) add a paragraph to the documentation
1771        for the FullMatch() function.
1772    
1773     6. The escape sequence \n was being treated as whatever was defined as
1774        "newline". Not only was this contrary to the documentation, which states
1775        that \n is character 10 (hex 0A), but it also went horribly wrong when
1776        "newline" was defined as CRLF. This has been fixed.
1777    
1778     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1779        was being set to -1 for the "end of line" case (supposedly a value that no
1780        character can have). Though this value is never used (the check for end of
1781        line is "zero bytes in current character"), it caused compiler complaints.
1782        I've changed it to 0xffffffff.
1783    
1784     8. In pcre_version.c, the version string was being built by a sequence of
1785        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1786        string (as it is for production releases) called a macro with an empty
1787        argument. The C standard says the result of this is undefined. The gcc
1788        compiler treats it as an empty string (which was what was wanted) but it is
1789        reported that Visual C gives an error. The source has been hacked around to
1790        avoid this problem.
1791    
1792     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1793        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1794        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1795        of them did).
1796    
1797    10. Originally, pcretest opened its input and output without "b"; then I was
1798        told that "b" was needed in some environments, so it was added for release
1799        5.0 to both the input and output. (It makes no difference on Unix-like
1800        systems.) Later I was told that it is wrong for the input on Windows. I've
1801        now abstracted the modes into two macros, to make it easier to fiddle with
1802        them, and removed "b" from the input mode under Windows.
1803    
1804    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1805    
1806    12. Added -help and --help to pcretest as an official way of being reminded
1807        of the options.
1808    
1809    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1810        and pcrecpp.cc because they annoy compilers at high warning levels.
1811    
1812    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1813    
1814    15. Fixed an occurrence of == in configure.ac that should have been = (shell
1815        scripts are not C programs :-) and which was not noticed because it works
1816        on Linux.
1817    
1818    16. pcretest is supposed to handle any length of pattern and data line (as one
1819        line or as a continued sequence of lines) by extending its input buffer if
1820        necessary. This feature was broken for very long pattern lines, leading to
1821        a string of junk being passed to pcre_compile() if the pattern was longer
1822        than about 50K.
1823    
1824    17. I have done a major re-factoring of the way pcre_compile() computes the
1825        amount of memory needed for a compiled pattern. Previously, there was code
1826        that made a preliminary scan of the pattern in order to do this. That was
1827        OK when PCRE was new, but as the facilities have expanded, it has become
1828        harder and harder to keep it in step with the real compile phase, and there
1829        have been a number of bugs (see for example, 4 above). I have now found a
1830        cunning way of running the real compile function in a "fake" mode that
1831        enables it to compute how much memory it would need, while actually only
1832        ever using a few hundred bytes of working memory and without too many
1833        tests of the mode. This should make future maintenance and development
1834        easier. A side effect of this work is that the limit of 200 on the nesting
1835        depth of parentheses has been removed (though this was never a serious
1836        limitation, I suspect). However, there is a downside: pcre_compile() now
1837        runs more slowly than before (30% or more, depending on the pattern). I
1838        hope this isn't a big issue. There is no effect on runtime performance.
1839    
1840    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1841        newline (only possible for the last line of a file) and it was a
1842        pattern that set a locale (followed by /Lsomething), pcretest crashed.
1843    
1844    19. Added additional timing features to pcretest. (1) The -tm option now times
1845        matching only, not compiling. (2) Both -t and -tm can be followed, as a
1846        separate command line item, by a number that specifies the number of
1847        repeats to use when timing. The default is 50000; this gives better
1848        precision, but takes uncomfortably long for very large patterns.
1849    
1850    20. Extended pcre_study() to be more clever in cases where a branch of a
1851        subpattern has no definite first character. For example, (a*|b*)[cd] would
1852        previously give no result from pcre_study(). Now it recognizes that the
1853        first character must be a, b, c, or d.
1854    
1855    21. There was an incorrect error "recursive call could loop indefinitely" if
1856        a subpattern (or the entire pattern) that was being tested for matching an
1857        empty string contained only one non-empty item after a nested subpattern.
1858        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1859        incorrectly, because the \d was being skipped in the check.
1860    
1861    22. The pcretest program now has a new pattern option /B and a command line
1862        option -b, which is equivalent to adding /B to every pattern. This causes
1863        it to show the compiled bytecode, without the additional information that
1864        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1865        is the same as /B/I).
1866    
1867    23. A new optimization is now able automatically to treat some sequences such
1868        as a*b as a*+b. More specifically, if something simple (such as a character
1869        or a simple class like \d) has an unlimited quantifier, and is followed by
1870        something that cannot possibly match the quantified thing, the quantifier
1871        is automatically "possessified".
1872    
1873    24. A recursive reference to a subpattern whose number was greater than 39
1874        went wrong under certain circumstances in UTF-8 mode. This bug could also
1875        have affected the operation of pcre_study().
1876    
1877    25. Realized that a little bit of performance could be had by replacing
1878        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1879    
1880    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1881    
1882    27. Possessive quantifiers such as a++ were previously implemented by turning
1883        them into atomic groups such as ($>a+). Now they have their own opcodes,
1884        which improves performance. This includes the automatically created ones
1885        from 23 above.
1886    
1887    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1888        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1889        the first matched character to be a colon. This applied both to named and
1890        numbered groups.
1891    
1892    29. The ucpinternal.h header file was missing its idempotency #ifdef.
1893    
1894    30. I was sent a "project" file called libpcre.a.dev which I understand makes
1895        building PCRE on Windows easier, so I have included it in the distribution.
1896    
1897    31. There is now a check in pcretest against a ridiculously large number being
1898        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1899        loop, the loop is abandoned.
1900    
1901    32. Forward references to subpatterns in conditions such as (?(2)...) where
1902        subpattern 2 is defined later cause pcre_compile() to search forwards in
1903        the pattern for the relevant set of parentheses. This search went wrong
1904        when there were unescaped parentheses in a character class, parentheses
1905        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1906    
1907    33. "Subroutine" calls and backreferences were previously restricted to
1908        referencing subpatterns earlier in the regex. This restriction has now
1909        been removed.
1910    
1911    34. Added a number of extra features that are going to be in Perl 5.10. On the
1912        whole, these are just syntactic alternatives for features that PCRE had
1913        previously implemented using the Python syntax or my own invention. The
1914        other formats are all retained for compatibility.
1915    
1916        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1917            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1918            also .NET compatible.
1919    
1920        (b) A recursion or subroutine call to a named group can now be defined as
1921            (?&name) as well as (?P>name).
1922    
1923        (c) A backreference to a named group can now be defined as \k<name> or
1924            \k'name' as well as (?P=name). The new forms, as well as being in Perl
1925            5.10, are also .NET compatible.
1926    
1927        (d) A conditional reference to a named group can now use the syntax
1928            (?(<name>) or (?('name') as well as (?(name).
1929    
1930        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1931            groups (named and numbered) that are never evaluated inline, but can be
1932            called as "subroutines" from elsewhere. In effect, the DEFINE condition
1933            is always false. There may be only one alternative in such a group.
1934    
1935        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1936            as the simple (?(R). The condition is true only if the most recent
1937            recursion is that of the given number or name. It does not search out
1938            through the entire recursion stack.
1939    
1940        (g) The escape \gN or \g{N} has been added, where N is a positive or
1941            negative number, specifying an absolute or relative reference.
1942    
1943    35. Tidied to get rid of some further signed/unsigned compiler warnings and
1944        some "unreachable code" warnings.
1945    
1946    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1947        things, this adds five new scripts.
1948    
1949    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1950        There were also incompatibilities regarding the handling of \Q..\E inside
1951        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1952        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1953    
1954    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1955        matches an empty string, and forcibly breaks the loop. There were bugs in
1956        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
1957        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
1958        separate and independent bugs (that affected different cases) have been
1959        fixed.
1960    
1961    39. Refactored the code to abolish the use of different opcodes for small
1962        capturing bracket numbers. This is a tidy that I avoided doing when I
1963        removed the limit on the number of capturing brackets for 3.5 back in 2001.
1964        The new approach is not only tidier, it makes it possible to reduce the
1965        memory needed to fix the previous bug (38).
1966    
1967    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1968        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1969        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1970        mode.
1971    
1972    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1973        report.
1974    
1975    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1976        copy construction and assignment in the C++ wrapper.
1977    
1978    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1979        couple of bugs that could have given wrong results in the "--newline=crlf"
1980        case.
1981    
1982    44. Added a number of casts and did some reorganization of signed/unsigned int
1983        variables following suggestions from Dair Grant. Also renamed the variable
1984        "this" as "item" because it is a C++ keyword.
1985    
1986    45. Arranged for dftables to add
1987    
1988          #include "pcre_internal.h"
1989    
1990        to pcre_chartables.c because without it, gcc 4.x may remove the array
1991        definition from the final binary if PCRE is built into a static library and
1992        dead code stripping is activated.
1993    
1994    46. For an unanchored pattern, if a match attempt fails at the start of a
1995        newline sequence, and the newline setting is CRLF or ANY, and the next two
1996        characters are CRLF, advance by two characters instead of one.
1997    
1998    
1999    Version 6.7 04-Jul-06
2000    ---------------------
2001    
2002     1. In order to handle tests when input lines are enormously long, pcretest has
2003        been re-factored so that it automatically extends its buffers when
2004        necessary. The code is crude, but this _is_ just a test program. The
2005        default size has been increased from 32K to 50K.
2006    
2007     2. The code in pcre_study() was using the value of the re argument before
2008        testing it for NULL. (Of course, in any sensible call of the function, it
2009        won't be NULL.)
2010    
2011     3. The memmove() emulation function in pcre_internal.h, which is used on
2012        systems that lack both memmove() and bcopy() - that is, hardly ever -
2013        was missing a "static" storage class specifier.
2014    
2015     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
2016        containing an extended class (one that cannot be represented by a bitmap
2017        because it contains high-valued characters or Unicode property items, e.g.
2018        [\pZ]). Almost always one would set UTF-8 mode when processing such a
2019        pattern, but PCRE should not loop if you do not (it no longer does).
2020        [Detail: two cases were found: (a) a repeated subpattern containing an
2021        extended class; (b) a recursive reference to a subpattern that followed a
2022        previous extended class. It wasn't skipping over the extended class
2023        correctly when UTF-8 mode was not set.]
2024    
2025     5. A negated single-character class was not being recognized as fixed-length
2026        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
2027        compile error "lookbehind assertion is not fixed length".
2028    
2029     6. The RunPerlTest auxiliary script was showing an unexpected difference
2030        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
2031        write a Perl script that can interpret lines of an input file either as
2032        byte characters or as UTF-8, which is what "perltest" was being required to
2033        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
2034        can't do is switch easily at run time between having the "use utf8;" pragma
2035        or not. In the end, I fudged it by using the RunPerlTest script to insert
2036        "use utf8;" explicitly for the UTF-8 tests.
2037    
2038     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
2039        the end of the subject string, contrary to the documentation and to what
2040        Perl does. This was true of both matching functions. Now it matches only at
2041        the start of the subject and immediately after *internal* newlines.
2042    
2043     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
2044        a pointer to an int instead of a pointer to an unsigned long int. This
2045        caused problems on 64-bit systems.
2046    
2047     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
2048        instance of the 'standard' template library not being so standard".
2049    
2050    10. There was no check on the number of named subpatterns nor the maximum
2051        length of a subpattern name. The product of these values is used to compute
2052        the size of the memory block for a compiled pattern. By supplying a very
2053        long subpattern name and a large number of named subpatterns, the size
2054        computation could be caused to overflow. This is now prevented by limiting
2055        the length of names to 32 characters, and the number of named subpatterns
2056        to 10,000.
2057    
2058    11. Subpatterns that are repeated with specific counts have to be replicated in
2059        the compiled pattern. The size of memory for this was computed from the
2060        length of the subpattern and the repeat count. The latter is limited to
2061        65535, but there was no limit on the former, meaning that integer overflow
2062        could in principle occur. The compiled length of a repeated subpattern is
2063        now limited to 30,000 bytes in order to prevent this.
2064    
2065    12. Added the optional facility to have named substrings with the same name.
2066    
2067    13. Added the ability to use a named substring as a condition, using the
2068        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
2069        are numbers (not recommended). Forward references are permitted.
2070    
2071    14. Added forward references in named backreferences (if you see what I mean).
2072    
2073    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
2074        pattern could run off the end of the subject. For example, the pattern
2075        "(?s)(.{1,5})"8 did this with the subject "ab".
2076    
2077    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
2078        PCRE_CASELESS was set when matching characters that were quantified with ?
2079        or *.
2080    
2081    17. A character class other than a single negated character that had a minimum
2082        but no maximum quantifier - for example [ab]{6,} - was not handled
2083        correctly by pce_dfa_exec(). It would match only one character.
2084    
2085    18. A valid (though odd) pattern that looked like a POSIX character
2086        class but used an invalid character after [ (for example [[,abc,]]) caused
2087        pcre_compile() to give the error "Failed: internal error: code overflow" or
2088        in some cases to crash with a glibc free() error. This could even happen if
2089        the pattern terminated after [[ but there just happened to be a sequence of
2090        letters, a binary zero, and a closing ] in the memory that followed.
2091    
2092    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
2093        over the years. Originally (before any Unicode support), just the bottom 8
2094        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
2095        output from "man perlunicode" includes this:
2096    
2097          The regular expression compiler produces polymorphic opcodes.  That
2098          is, the pattern adapts to the data and automatically switches to
2099          the Unicode character scheme when presented with Unicode data--or
2100          instead uses a traditional byte scheme when presented with byte
2101          data.
2102    
2103        Sadly, a wide octal escape does not cause a switch, and in a string with
2104        no other multibyte characters, these octal escapes are treated as before.
2105        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
2106        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
2107        Unicode string.
2108    
2109        I have not perpetrated such confusion in PCRE. Up till now, it took just
2110        the bottom 8 bits, as in old Perl. I have now made octal escapes with
2111        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
2112        translate to the appropriate multibyte character.
2113    
2114    29. Applied some refactoring to reduce the number of warnings from Microsoft
2115        and Borland compilers. This has included removing the fudge introduced
2116        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
2117        a warning about an unused variable.
2118    
2119    21. PCRE has not included VT (character 0x0b) in the set of whitespace
2120        characters since release 4.0, because Perl (from release 5.004) does not.
2121        [Or at least, is documented not to: some releases seem to be in conflict
2122        with the documentation.] However, when a pattern was studied with
2123        pcre_study() and all its branches started with \s, PCRE still included VT
2124        as a possible starting character. Of course, this did no harm; it just
2125        caused an unnecessary match attempt.
2126    
2127    22. Removed a now-redundant internal flag bit that recorded the fact that case
2128        dependency changed within the pattern. This was once needed for "required
2129        byte" processing, but is no longer used. This recovers a now-scarce options
2130        bit. Also moved the least significant internal flag bit to the most-
2131        significant bit of the word, which was not previously used (hangover from
2132        the days when it was an int rather than a uint) to free up another bit for
2133        the future.
2134    
2135    23. Added support for CRLF line endings as well as CR and LF. As well as the
2136        default being selectable at build time, it can now be changed at runtime
2137        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
2138        specify that it is scanning data with non-default line endings.
2139    
2140    24. Changed the definition of CXXLINK to make it agree with the definition of
2141        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
2142    
2143    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
2144        recursions. This makes a big different to stack usage for some patterns.
2145    
2146    26. If a subpattern containing a named recursion or subroutine reference such
2147        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
2148        the space required for the compiled pattern went wrong and gave too small a
2149        value. Depending on the environment, this could lead to "Failed: internal
2150        error: code overflow at offset 49" or "glibc detected double free or
2151        corruption" errors.
2152    
2153    27. Applied patches from Google (a) to support the new newline modes and (b) to
2154        advance over multibyte UTF-8 characters in GlobalReplace.
2155    
2156    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
2157        difference for some implementation of PCRE in some Windows version.
2158    
2159    29. Added some extra testing facilities to pcretest:
2160    
2161        \q<number>   in a data line sets the "match limit" value
2162        \Q<number>   in a data line sets the "match recursion limt" value
2163        -S <number>  sets the stack size, where <number> is in megabytes
2164    
2165        The -S option isn't available for Windows.
2166    
2167    
2168    Version 6.6 06-Feb-06
2169    ---------------------
2170    
2171     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
2172        in pcreposix.h. I have copied the definition from pcre.h.
2173    
2174     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
2175        because pcre.h is no longer a built file.
2176    
2177     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
2178        not normally included in the compiled code.
2179    
2180    
2181    Version 6.5 01-Feb-06
2182    ---------------------
2183    
2184     1. When using the partial match feature with pcre_dfa_exec(), it was not
2185        anchoring the second and subsequent partial matches at the new starting
2186        point. This could lead to incorrect results. For example, with the pattern
2187        /1234/, partially matching against "123" and then "a4" gave a match.
2188    
2189     2. Changes to pcregrep:
2190    
2191        (a) All non-match returns from pcre_exec() were being treated as failures
2192            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
2193            error message is output. Some extra information is given for the
2194            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
2195            probably the only errors that are likely to be caused by users (by
2196            specifying a regex that has nested indefinite repeats, for instance).
2197            If there are more than 20 of these errors, pcregrep is abandoned.
2198    
2199        (b) A binary zero was treated as data while matching, but terminated the
2200            output line if it was written out. This has been fixed: binary zeroes
2201            are now no different to any other data bytes.
2202    
2203        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
2204            used to set a locale for matching. The --locale=xxxx long option has
2205            been added (no short equivalent) to specify a locale explicitly on the
2206            pcregrep command, overriding the environment variables.
2207    
2208        (d) When -B was used with -n, some line numbers in the output were one less
2209            than they should have been.
2210    
2211        (e) Added the -o (--only-matching) option.
2212    
2213        (f) If -A or -C was used with -c (count only), some lines of context were
2214            accidentally printed for the final match.
2215    
2216        (g) Added the -H (--with-filename) option.
2217    
2218        (h) The combination of options -rh failed to suppress file names for files
2219            that were found from directory arguments.
2220    
2221        (i) Added the -D (--devices) and -d (--directories) options.
2222    
2223        (j) Added the -F (--fixed-strings) option.
2224    
2225        (k) Allow "-" to be used as a file name for -f as well as for a data file.
2226    
2227        (l) Added the --colo(u)r option.
2228    
2229        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
2230            is not present by default.
2231    
2232     3. A nasty bug was discovered in the handling of recursive patterns, that is,
2233        items such as (?R) or (?1), when the recursion could match a number of
2234        alternatives. If it matched one of the alternatives, but subsequently,
2235        outside the recursion, there was a failure, the code tried to back up into
2236        the recursion. However, because of the way PCRE is implemented, this is not
2237        possible, and the result was an incorrect result from the match.
2238    
2239        In order to prevent this happening, the specification of recursion has
2240        been changed so that all such subpatterns are automatically treated as
2241        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
2242    
2243     4. I had overlooked the fact that, in some locales, there are characters for
2244        which isalpha() is true but neither isupper() nor islower() are true. In
2245        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
2246        and ordfeminine) are like this. This affected the treatment of \w and \W
2247        when they appeared in character classes, but not when they appeared outside
2248        a character class. The bit map for "word" characters is now created
2249        separately from the results of isalnum() instead of just taking it from the
2250        upper, lower, and digit maps. (Plus the underscore character, of course.)
2251    
2252     5. The above bug also affected the handling of POSIX character classes such as
2253        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
2254        permanent tables. Instead, the bit maps for such a class were previously
2255        created as the appropriate unions of the upper, lower, and digit bitmaps.
2256        Now they are created by subtraction from the [[:word:]] class, which has
2257        its own bitmap.
2258    
2259     6. The [[:blank:]] character class matches horizontal, but not vertical space.
2260        It is created by subtracting the vertical space characters (\x09, \x0a,
2261        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
2262        subtraction was done in the overall bitmap for a character class, meaning
2263        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
2264        be recognized. This bug has been fixed.
2265    
2266     7. Patches from the folks at Google:
2267    
2268          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
2269          real life, but is still worth protecting against".
2270    
2271          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
2272          regular expressions".
2273    
2274          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
2275          have it.
2276    
2277          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
2278          "configure" and the latter not, in order to fix a problem somebody had
2279          with compiling the Arg class on HP-UX.
2280    
2281          (e) Improve the error-handling of the C++ wrapper a little bit.
2282    
2283          (f) New tests for checking recursion limiting.
2284    
2285     8. The pcre_memmove() function, which is used only if the environment does not
2286        have a standard memmove() function (and is therefore rarely compiled),
2287        contained two bugs: (a) use of int instead of size_t, and (b) it was not
2288        returning a result (though PCRE never actually uses the result).
2289    
2290     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
2291        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
2292        returned instead of calling malloc() with an overflowing number that would
2293        most likely cause subsequent chaos.
2294    
2295    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
2296    
2297    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
2298        with this option is matched, the nmatch and pmatch options of regexec() are
2299        ignored.
2300    
2301    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
2302        provided in case anyone wants to the the POSIX interface with UTF-8
2303        strings.
2304    
2305    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
2306        C++ linking (needed for some HP-UX environments).
2307    
2308    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
2309        (unused parameter) and in the pcre_printint() function (omitted "default"
2310        switch label when the default is to do nothing).
2311    
2312    15. Added some code to make it possible, when PCRE is compiled as a C++
2313        library, to replace subject pointers for pcre_exec() with a smart pointer
2314        class, thus making it possible to process discontinuous strings.
2315    
2316    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
2317        much the same function. They were added by different people who were trying
2318        to make PCRE easy to compile on non-Unix systems. It has been suggested
2319        that PCRE_EXPORT be abolished now that there is more automatic apparatus
2320        for compiling on Windows systems. I have therefore replaced it with
2321        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
2322        defaults to "extern" for C or "extern C" for C++, which works fine on
2323        Unix-like systems. It is now possible to override the value of PCRE_DATA_
2324        SCOPE with something explicit in config.h. In addition:
2325    
2326        (a) pcreposix.h still had just "extern" instead of either of these macros;
2327            I have replaced it with PCRE_DATA_SCOPE.
2328    
2329        (b) Functions such as _pcre_xclass(), which are internal to the library,
2330            but external in the C sense, all had PCRE_EXPORT in their definitions.
2331            This is apparently wrong for the Windows case, so I have removed it.
2332            (It makes no difference on Unix-like systems.)
2333    
2334    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
2335        of recursive calls to match(). This is different to MATCH_LIMIT because
2336        that limits the total number of calls to match(), not all of which increase
2337        the depth of recursion. Limiting the recursion depth limits the amount of
2338        stack (or heap if NO_RECURSE is set) that is used. The default can be set
2339        when PCRE is compiled, and changed at run time. A patch from Google adds
2340        this functionality to the C++ interface.
2341    
2342    18. Changes to the handling of Unicode character properties:
2343    
2344        (a) Updated the table to Unicode 4.1.0.
2345    
2346        (b) Recognize characters that are not in the table as "Cn" (undefined).
2347    
2348        (c) I revised the way the table is implemented to a much improved format
2349            which includes recognition of ranges. It now supports the ranges that
2350            are defined in UnicodeData.txt, and it also amalgamates other
2351            characters into ranges. This has reduced the number of entries in the
2352            table from around 16,000 to around 3,000, thus reducing its size
2353            considerably. I realized I did not need to use a tree structure after
2354            all - a binary chop search is just as efficient. Having reduced the
2355            number of entries, I extended their size from 6 bytes to 8 bytes to
2356            allow for more data.
2357    
2358        (d) Added support for Unicode script names via properties such as \p{Han}.
2359    
2360    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
2361        matching that character.
2362    
2363    20. When matching a repeated Unicode property with a minimum greater than zero,
2364        (for example \pL{2,}), PCRE could look past the end of the subject if it
2365        reached it while seeking the minimum number of characters. This could
2366        happen only if some of the characters were more than one byte long, because
2367        there is a check for at least the minimum number of bytes.
2368    
2369    21. Refactored the implementation of \p and \P so as to be more general, to
2370        allow for more different types of property in future. This has changed the
2371        compiled form incompatibly. Anybody with saved compiled patterns that use
2372        \p or \P will have to recompile them.
2373    
2374    22. Added "Any" and "L&" to the supported property types.
2375    
2376    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
2377        but give a compile time error if the value is greater than 0xff.
2378    
2379    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
2380        accidentally not being installed or uninstalled.
2381    
2382    25. The pcre.h file was built from pcre.h.in, but the only changes that were
2383        made were to insert the current release number. This seemed silly, because
2384        it made things harder for people building PCRE on systems that don't run
2385        "configure". I have turned pcre.h into a distributed file, no longer built
2386        by "configure", with the version identification directly included. There is
2387        no longer a pcre.h.in file.
2388    
2389        However, this change necessitated a change to the pcre-config script as
2390        well. It is built from pcre-config.in, and one of the substitutions was the
2391        release number. I have updated configure.ac so that ./configure now finds
2392        the release number by grepping pcre.h.
2393    
2394    26. Added the ability to run the tests under valgrind.
2395    
2396    
2397    Version 6.4 05-Sep-05
2398    ---------------------
2399    
2400     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
2401        "--" to be printed when multiple files were scanned, even when none of the
2402        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
2403        consider it to be a bug, and have restored the previous behaviour.
2404    
2405     2. A couple of code tidies to get rid of compiler warnings.
2406    
2407     3. The pcretest program used to cheat by referring to symbols in the library
2408        whose names begin with _pcre_. These are internal symbols that are not
2409        really supposed to be visible externally, and in some environments it is
2410        possible to suppress them. The cheating is now confined to including
2411        certain files from the library's source, which is a bit cleaner.
2412    
2413     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
2414        file's purpose clearer.
2415    
2416     5. Reorganized pcre_ucp_findchar().
2417    
2418    
2419    Version 6.3 15-Aug-05
2420    ---------------------
2421    
2422     1. The file libpcre.pc.in did not have general read permission in the tarball.
2423    
2424     2. There were some problems when building without C++ support:
2425    
2426        (a) If C++ support was not built, "make install" and "make test" still
2427            tried to test it.
2428    
2429        (b) There were problems when the value of CXX was explicitly set. Some
2430            changes have been made to try to fix these, and ...
2431    
2432        (c) --disable-cpp can now be used to explicitly disable C++ support.
2433    
2434        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
2435            backslash in a target when C++ was disabled. This confuses some
2436            versions of "make", apparently. Using an intermediate variable solves
2437            this. (Same for CPP_LOBJ.)
2438    
2439     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
2440        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
2441        necessary on certain architectures.
2442    
2443     4. Added a setting of -export-symbols-regex to the link command to remove
2444        those symbols that are exported in the C sense, but actually are local
2445        within the library, and not documented. Their names all begin with
2446        "_pcre_". This is not a perfect job, because (a) we have to except some
2447        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
2448        available (and never for static libraries). I have made a note to try to
2449        find a way round (a) in the future.
2450    
2451    
2452    Version 6.2 01-Aug-05
2453    ---------------------
2454    
2455     1. There was no test for integer overflow of quantifier values. A construction
2456        such as {1111111111111111} would give undefined results. What is worse, if
2457        a minimum quantifier for a parenthesized subpattern overflowed and became
2458        negative, the calculation of the memory size went wrong. This could have
2459        led to memory overwriting.
2460    
2461     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
2462    
2463     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
2464        operating environments where this matters.
2465    
2466     4. Applied Giuseppe Maxia's patch to add additional features for controlling
2467        PCRE options from within the C++ wrapper.
2468    
2469     5. Named capturing subpatterns were not being correctly counted when a pattern
2470        was compiled. This caused two problems: (a) If there were more than 100
2471        such subpatterns, the calculation of the memory needed for the whole
2472        compiled pattern went wrong, leading to an overflow error. (b) Numerical
2473        back references of the form \12, where the number was greater than 9, were
2474        not recognized as back references, even though there were sufficient
2475        previous subpatterns.
2476    
2477     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
2478        versions of gcc, e.g. 2.95.4.
2479    
2480    
2481    Version 6.1 21-Jun-05
2482    ---------------------
2483    
2484     1. There was one reference to the variable "posix" in pcretest.c that was not
2485        surrounded by "#if !defined NOPOSIX".
2486    
2487     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
2488        the cross-check on the old pcre_info() function, for the benefit of the
2489        cut-down version of PCRE that is currently imported into Exim.
2490    
2491     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
2492        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
2493        patterns that start (?i) but I don't think that matters. The use of (?i) is
2494        just an example; this all applies to the other options as well.
2495    
2496     4. Since libtool seems to echo the compile commands it is issuing, the output
2497        from "make" can be reduced a bit by putting "@" in front of each libtool
2498        compile command.
2499    
2500     5. Patch from the folks at Google for configure.in to be a bit more thorough
2501        in checking for a suitable C++ installation before trying to compile the
2502        C++ stuff. This should fix a reported problem when a compiler was present,
2503        but no suitable headers.
2504    
2505     6. The man pages all had just "PCRE" as their title. I have changed them to
2506        be the relevant file name. I have also arranged that these names are
2507        retained in the file doc/pcre.txt, which is a concatenation in text format
2508        of all the man pages except the little individual ones for each function.
2509    
2510     7. The NON-UNIX-USE file had not been updated for the different set of source
2511        files that come with release 6. I also added a few comments about the C++
2512        wrapper.
2513    
2514    
2515    Version 6.0 07-Jun-05
2516    ---------------------
2517    
2518     1. Some minor internal re-organization to help with my DFA experiments.
2519    
2520     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
2521        didn't matter for the library itself when fully configured, but did matter
2522        when compiling without UCP support, or within Exim, where the ucp files are
2523        not imported.
2524    
2525     3. Refactoring of the library code to split up the various functions into
2526        different source modules. The addition of the new DFA matching code (see
2527        below) to a single monolithic source would have made it really too
2528        unwieldy, quite apart from causing all the code to be include in a
2529        statically linked application, when only some functions are used. This is
2530        relevant even without the DFA addition now that patterns can be compiled in
2531        one application and matched in another.
2532    
2533        The downside of splitting up is that there have to be some external
2534        functions and data tables that are used internally in different modules of
2535        the library but which are not part of the API. These have all had their
2536        names changed to start with "_pcre_" so that they are unlikely to clash
2537        with other external names.
2538    
2539     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
2540        a different (DFA) algorithm. Although it is slower than the original
2541        function, it does have some advantages for certain types of matching
2542        problem.
2543    
2544     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
2545        including restarting after a partial match.
2546    
2547     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
2548        defined when compiling for Windows was sent to me. I have put it into the
2549        code, though I have no means of testing or verifying it.
2550    
2551     7. Added the pcre_refcount() auxiliary function.
2552    
2553     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
2554        match before or at the first newline in the subject string. In pcretest,
2555        the /f option on a pattern can be used to set this.
2556    
2557     9. A repeated \w when used in UTF-8 mode with characters greater than 256
2558        would behave wrongly. This has been present in PCRE since release 4.0.
2559    
2560    10. A number of changes to the pcregrep command:
2561    
2562        (a) Refactored how -x works; insert ^(...)$ instead of setting
2563            PCRE_ANCHORED and checking the length, in preparation for adding
2564            something similar for -w.
2565    
2566        (b) Added the -w (match as a word) option.
2567    
2568        (c) Refactored the way lines are read and buffered so as to have more
2569            than one at a time available.
2570    
2571        (d) Implemented a pcregrep test script.
2572    
2573        (e) Added the -M (multiline match) option. This allows patterns to match
2574            over several lines of the subject. The buffering ensures that at least
2575            8K, or the rest of the document (whichever is the shorter) is available
2576            for matching (and similarly the previous 8K for lookbehind assertions).
2577    
2578        (f) Changed the --help output so that it now says
2579    
2580              -w, --word-regex(p)
2581    
2582            instead of two lines, one with "regex" and the other with "regexp"
2583            because that confused at least one person since the short forms are the
2584            same. (This required a bit of code, as the output is generated
2585            automatically from a table. It wasn't just a text change.)
2586    
2587        (g) -- can be used to terminate pcregrep options if the next thing isn't an
2588            option but starts with a hyphen. Could be a pattern or a path name
2589            starting with a hyphen, for instance.
2590    
2591        (h) "-" can be given as a file name to represent stdin.
2592    
2593        (i) When file names are being printed, "(standard input)" is used for
2594            the standard input, for compatibility with GNU grep. Previously
2595            "<stdin>" was used.
2596    
2597        (j) The option --label=xxx can be used to supply a name to be used for
2598            stdin when file names are being printed. There is no short form.
2599    
2600        (k) Re-factored the options decoding logic because we are going to add
2601            two more options that take data. Such options can now be given in four
2602            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
2603    
2604        (l) Added the -A, -B, and -C options for requesting that lines of context
2605            around matches be printed.
2606    
2607        (m) Added the -L option to print the names of files that do not contain
2608            any matching lines, that is, the complement of -l.
2609    
2610        (n) The return code is 2 if any file cannot be opened, but pcregrep does
2611            continue to scan other files.
2612    
2613        (o) The -s option was incorrectly implemented. For compatibility with other
2614            greps, it now suppresses the error message for a non-existent or non-
2615            accessible file (but not the return code). There is a new option called
2616            -q that suppresses the output of matching lines, which was what -s was
2617            previously doing.
2618    
2619        (p) Added --include and --exclude options to specify files for inclusion
2620            and exclusion when recursing.
2621    
2622    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
2623        Hopefully, it now does.
2624    
2625    12. Missing cast in pcre_study().
2626    
2627    13. Added an "uninstall" target to the makefile.
2628    
2629    14. Replaced "extern" in the function prototypes in Makefile.in with
2630        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
2631        world, but is set differently for Windows.
2632    
2633    15. Added a second compiling function called pcre_compile2(). The only
2634        difference is that it has an extra argument, which is a pointer to an
2635        integer error code. When there is a compile-time failure, this is set
2636        non-zero, in addition to the error test pointer being set to point to an
2637        error message. The new argument may be NULL if no error number is required
2638        (but then you may as well call pcre_compile(), which is now just a
2639        wrapper). This facility is provided because some applications need a
2640        numeric error indication, but it has also enabled me to tidy up the way
2641        compile-time errors are handled in the POSIX wrapper.
2642    
2643    16. Added VPATH=.libs to the makefile; this should help when building with one
2644        prefix path and installing with another. (Or so I'm told by someone who
2645        knows more about this stuff than I do.)
2646    
2647    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
2648        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
2649        match everything, including newlines. This is not POSIX-compatible, but
2650        somebody wanted the feature. From pcretest it can be activated by using
2651        both the P and the s flags.
2652    
2653    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
2654    
2655    19. libpcre.pc was being incorrectly installed as executable.
2656    
2657    20. A couple of places in pcretest check for end-of-line by looking for '\n';
2658        it now also looks for '\r' so that it will work unmodified on Windows.
2659    
2660    21. Added Google's contributed C++ wrapper to the distribution.
2661    
2662    22. Added some untidy missing memory free() calls in pcretest, to keep
2663        Electric Fence happy when testing.
2664    
2665    
2666    
2667    Version 5.0 13-Sep-04
2668    ---------------------
2669    
2670     1. Internal change: literal characters are no longer packed up into items
2671        containing multiple characters in a single byte-string. Each character
2672        is now matched using a separate opcode. However, there may be more than one
2673        byte in the character in UTF-8 mode.
2674    
2675     2. The pcre_callout_block structure has two new fields: pattern_position and
2676        next_item_length. These contain the offset in the pattern to the next match
2677        item, and its length, respectively.
2678    
2679     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
2680        insertion of callouts before each pattern item. Added the /C option to
2681        pcretest to make use of this.
2682    
2683     4. On the advice of a Windows user, the lines
2684    
2685          #if defined(_WIN32) || defined(WIN32)
2686          _setmode( _fileno( stdout ), 0x8000 );
2687          #endif  /* defined(_WIN32) || defined(WIN32) */
2688    
2689        have been added to the source of pcretest. This apparently does useful
2690        magic in relation to line terminators.
2691    
2692     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
2693        for the benefit of those environments where the "b" makes a difference.
2694    
2695     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
2696        to know about it. I have put a hack into configure.in that adds in code
2697        to set GCC=yes if CC=icc. This seems to end up at a point in the
2698        generated configure script that is early enough to affect the setting of
2699        compiler options, which is what is needed, but I have no means of testing
2700        whether it really works. (The user who reported this had patched the
2701        generated configure script, which of course I cannot do.)
2702    
2703        LATER: After change 22 below (new libtool files), the configure script
2704        seems to know about icc (and also ecc). Therefore, I have commented out
2705        this hack in configure.in.
2706    
2707     7. Added support for pkg-config (2 patches were sent in).
2708    
2709     8. Negated POSIX character classes that used a combination of internal tables
2710        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2711        [[:^ascii]]. Typically, they would match almost any characters. The other
2712        POSIX classes were not broken in this way.
2713    
2714     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2715        to find the match, as PCRE was deluded into thinking that the match had to
2716        start at the start point or following a newline. The same bug applied to
2717        patterns with negative forward assertions or any backward assertions
2718        preceding ".*" at the start, unless the pattern required a fixed first
2719        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2720    
2721    10. In UTF-8 mode, when moving forwards in the subject after a failed match
2722        starting at the last subject character, bytes beyond the end of the subject
2723        string were read.
2724    
2725    11. Renamed the variable "class" as "classbits" to make life easier for C++
2726        users. (Previously there was a macro definition, but it apparently wasn't
2727        enough.)
2728    
2729    12. Added the new field "tables" to the extra data so that tables can be passed
2730        in at exec time, or the internal tables can be re-selected. This allows
2731        a compiled regex to be saved and re-used at a later time by a different
2732        program that might have everything at different addresses.
2733    
2734    13. Modified the pcre-config script so that, when run on Solaris, it shows a
2735        -R library as well as a -L library.
2736    
2737    14. The debugging options of pcretest (-d on the command line or D on a
2738        pattern) showed incorrect output for anything following an extended class
2739        that contained multibyte characters and which was followed by a quantifier.
2740    
2741    15. Added optional support for general category Unicode character properties
2742        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2743        support. It adds about 90K to the size of the library. The meanings of the
2744        inbuilt class escapes such as \d and \s have NOT been changed.
2745    
2746    16. Updated pcredemo.c to include calls to free() to release the memory for the
2747        compiled pattern.
2748    
2749    17. The generated file chartables.c was being created in the source directory
2750        instead of in the building directory. This caused the build to fail if the
2751        source directory was different from the building directory, and was
2752        read-only.
2753    
2754    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2755        file. No doubt somebody will tell me if they don't make sense... Also added
2756        Dan Mooney's comments about building on OpenVMS.
2757    
2758    19. Added support for partial matching via the PCRE_PARTIAL option for
2759        pcre_exec() and the \P data escape in pcretest.
2760    
2761    20. Extended pcretest with 3 new pattern features:
2762    
2763        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
2764              write the compiled pattern to the file whose name is "rest-of-line".
2765              This is a straight binary dump of the data, with the saved pointer to
2766              the character tables forced to be NULL. The study data, if any, is
2767              written too. After writing, pcretest reads a new pattern.
2768    
2769        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2770              compiled pattern from the given file. There must not be any
2771              occurrences of "<" in the file name (pretty unlikely); if there are,
2772              pcretest will instead treat the initial "<" as a pattern delimiter.
2773              After reading in the pattern, pcretest goes on to read data lines as
2774              usual.
2775    
2776        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2777              and 16-bit fields in a compiled pattern, to simulate a pattern that
2778              was compiled on a host of opposite endianness.
2779    
2780    21. The pcre-exec() function can now cope with patterns that were compiled on
2781        hosts of opposite endianness, with this restriction:
2782    
2783          As for any compiled expression that is saved and used later, the tables
2784          pointer field cannot be preserved; the extra_data field in the arguments
2785          to pcre_exec() should be used to pass in a tables address if a value
2786          other than the default internal tables were used at compile time.
2787    
2788    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2789        now diagnosed as an error. Previously, most of the time, a negative number
2790        would have been treated as zero, but if in addition "ovector" was passed as
2791        NULL, a crash could occur.
2792    
2793    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2794        new versions from the libtool 1.5 distribution (the last one is a copy of
2795        a file called libtool.m4). This seems to have fixed the need to patch
2796        "configure" to support Darwin 1.3 (which I used to do). However, I still
2797        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2798        workstation).
2799    
2800    24. Changed the PCRE licence to be the more standard "BSD" licence.
2801    
2802    
2803    Version 4.5 01-Dec-03
2804    ---------------------
2805    
2806     1. There has been some re-arrangement of the code for the match() function so
2807        that it can be compiled in a version that does not call itself recursively.
2808        Instead, it keeps those local variables that need separate instances for
2809        each "recursion" in a frame on the heap, and gets/frees frames whenever it
2810        needs to "recurse". Keeping track of where control must go is done by means
2811        of setjmp/longjmp. The whole thing is implemented by a set of macros that
2812        hide most of the details from the main code, and operates only if
2813        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2814        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2815        operating.
2816    
2817        To make it easier for callers to provide specially tailored get/free
2818        functions for this usage, two new functions, pcre_stack_malloc, and
2819        pcre_stack_free, are used. They are always called in strict stacking order,
2820        and the size of block requested is always the same.
2821    
2822        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2823        PCRE has been compiled to use the stack or the heap for recursion. The
2824        -C option of pcretest uses this to show which version is compiled.
2825    
2826        A new data escape \S, is added to pcretest; it causes the amounts of store
2827        obtained and freed by both kinds of malloc/free at match time to be added
2828        to the output.
2829    
2830     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2831        what's available on my current Linux desktop machine.
2832    
2833     3. When matching a UTF-8 string, the test for a valid string at the start has
2834        been extended. If start_offset is not zero, PCRE now checks that it points
2835        to a byte that is the start of a UTF-8 character. If not, it returns
2836        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2837        this is necessary because there may be backward assertions in the pattern.
2838        When matching the same subject several times, it may save resources to use
2839        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2840    
2841     4. The code for checking the validity of UTF-8 strings has been tightened so
2842        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2843        containing "overlong sequences".
2844    
2845     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2846        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2847        should have been "&", but it just so happened that all the cases this let
2848        through by mistake were picked up later in the function.
2849    
2850     6. I had used a variable called "isblank" - this is a C99 function, causing
2851        some compilers to warn. To avoid this, I renamed it (as "blankclass").
2852    
2853     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2854        prompting; (b) run "./pcretest /dev/null" at the start of the test script
2855        so the version is shown; (c) stop "make test" echoing "./RunTest".
2856    
2857     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2858    
2859     9. The prototype for memmove() for systems that don't have it was using
2860        size_t, but the inclusion of the header that defines size_t was later. I've
2861        moved the #includes for the C headers earlier to avoid this.
2862    
2863    10. Added some adjustments to the code to make it easier to compiler on certain
2864        special systems:
2865    
2866          (a) Some "const" qualifiers were missing.
2867          (b) Added the macro EXPORT before all exported functions; by default this
2868              is defined to be empty.
2869          (c) Changed the dftables auxiliary program (that builds chartables.c) so
2870              that it reads its output file name as an argument instead of writing
2871              to the standard output and assuming this can be redirected.
2872    
2873    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2874        class containing characters with values greater than 255, PCRE compilation
2875        went into a loop.
2876    
2877    12. A recursive reference to a subpattern that was within another subpattern
2878        that had a minimum quantifier of zero caused PCRE to crash. For example,
2879        (x(y(?2))z)? provoked this bug with a subject that got as far as the
2880        recursion. If the recursively-called subpattern itself had a zero repeat,
2881        that was OK.
2882    
2883    13. In pcretest, the buffer for reading a data line was set at 30K, but the
2884        buffer into which it was copied (for escape processing) was still set at
2885        1024, so long lines caused crashes.
2886    
2887    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2888        "internal error: code overflow...". This applied to any character class
2889        that was followed by a possessive quantifier.
2890    
2891    15. Modified the Makefile to add libpcre.la as a prerequisite for
2892        libpcreposix.la because I was told this is needed for a parallel build to
2893        work.
2894    
2895    16. If a pattern that contained .* following optional items at the start was
2896        studied, the wrong optimizing data was generated, leading to matching
2897        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2898        matching string must start with a or b or c. The correct conclusion for
2899        this pattern is that a match can start with any character.
2900    
2901    
2902    Version 4.4 13-Aug-03
2903    ---------------------
2904    
2905     1. In UTF-8 mode, a character class containing characters with values between
2906        127 and 255 was not handled correctly if the compiled pattern was studied.
2907        In fixing this, I have also improved the studying algorithm for such
2908        classes (slightly).
2909    
2910     2. Three internal functions had redundant arguments passed to them. Removal
2911        might give a very teeny performance improvement.
2912    
2913     3. Documentation bug: the value of the capture_top field in a callout is *one
2914        more than* the number of the hightest numbered captured substring.
2915    
2916     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2917        in incorrectly linking with a previously installed version. They now link
2918        explicitly with libpcre.la.
2919    
2920     5. configure.in no longer needs to recognize Cygwin specially.
2921    
2922     6. A problem in pcre.in for Windows platforms is fixed.
2923    
2924     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2925        pcretest, it used to include the size of the study block as part of its
2926        output. Unfortunately, the structure contains a field that has a different
2927        size on different hardware architectures. This meant that the tests that
2928        showed this size failed. As the block is currently always of a fixed size,
2929        this information isn't actually particularly useful in pcretest output, so
2930        I have just removed it.
2931    
2932     8. Three pre-processor statements accidentally did not start in column 1.
2933        Sadly, there are *still* compilers around that complain, even though
2934        standard C has not required this for well over a decade. Sigh.
2935    
2936     9. In pcretest, the code for checking callouts passed small integers in the
2937        callout_data field, which is a void * field. However, some picky compilers
2938        complained about the casts involved for this on 64-bit systems. Now
2939        pcretest passes the address of the small integer instead, which should get
2940        rid of the warnings.
2941    
2942    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2943        both compile and run time, and gives an error if an invalid UTF-8 sequence
2944        is found. There is a option for disabling this check in cases where the
2945        string is known to be correct and/or the maximum performance is wanted.
2946    
2947    11. In response to a bug report, I changed one line in Makefile.in from
2948    
2949            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2950        to
2951            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2952    
2953        to look similar to other lines, but I have no way of telling whether this
2954        is the right thing to do, as I do not use Windows. No doubt I'll get told
2955        if it's wrong...
2956    
2957    
2958    Version 4.3 21-May-03
2959    ---------------------
2960    
2961    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2962       Makefile.
2963    
2964    2. Some refactoring to improve the quality of the code:
2965    
2966       (i)   The utf8_table... variables are now declared "const".
2967    
2968       (ii)  The code for \cx, which used the "case flipping" table to upper case
2969             lower case letters, now just substracts 32. This is ASCII-specific,
2970             but the whole concept of \cx is ASCII-specific, so it seems
2971             reasonable.
2972    
2973       (iii) PCRE was using its character types table to recognize decimal and
2974             hexadecimal digits in the pattern. This is silly, because it handles
2975             only 0-9, a-f, and A-F, but the character types table is locale-
2976             specific, which means strange things might happen. A private
2977             table is now used for this - though it costs 256 bytes, a table is
2978             much faster than multiple explicit tests. Of course, the standard
2979             character types table is still used for matching digits in subject
2980             strings against \d.
2981    
2982       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2983             ending in _t are). So I've renamed it as ESC_tee.
2984    
2985    3. The first argument for regexec() in the POSIX wrapper should have been
2986       defined as "const".
2987    
2988    4. Changed pcretest to use malloc() for its buffers so that they can be
2989       Electric Fenced for debugging.
2990    
2991    5. There were several places in the code where, in UTF-8 mode, PCRE would try
2992       to read one or more bytes before the start of the subject string. Often this
2993       had no effect on PCRE's behaviour, but in some circumstances it could
2994       provoke a segmentation fault.
2995    
2996    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2997       to try to read one or more bytes before the start of the subject string.
2998    
2999    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
3000       UTF-8 support could misbehave in various ways if the subject string
3001       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
3002       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
3003       back over UTF-8 characters.)
3004    
3005    
3006    Version 4.2 14-Apr-03
3007    ---------------------
3008    
3009    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
3010    
3011    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
3012         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
3013         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
3014         [WIN_PREFIX]: new variable, "cyg" for Cygwin
3015         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
3016           and BUILD_EXEEXT
3017         Note: automatic setting of the BUILD variables is not yet working
3018         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
3019           compile-time but not at link-time
3020         [LINK]: use for linking executables only
3021         make different versions for Windows and non-Windows
3022         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
3023           libraries
3024         [LINK_FOR_BUILD]: new variable
3025         [OBJEXT]: use throughout
3026         [EXEEXT]: use throughout
3027         <winshared>: new target
3028         <wininstall>: new target
3029         <dftables.o>: use native compiler
3030         <dftables>: use native linker
3031         <install>: handle Windows platform correctly
3032         <clean>: ditto
3033         <check>: ditto
3034         copy DLL to top builddir before testing
3035    
3036       As part of these changes, -no-undefined was removed again. This was reported
3037       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
3038       in any case.
3039    
3040    3. Some tidies to get rid of compiler warnings:
3041    
3042       . In the match_data structure, match_limit was an unsigned long int, whereas
3043         match_call_count was an int. I've made them both unsigned long ints.
3044    
3045       . In pcretest the fact that a const uschar * doesn't automatically cast to
3046         a void * provoked a warning.
3047    
3048       . Turning on some more compiler warnings threw up some "shadow" variables
3049         and a few more missing casts.
3050    
3051    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
3052       option, a class that contained a single character with a value between 128
3053       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
3054    
3055    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
3056       option, a class that contained several characters, but with at least one
3057       whose value was between 128 and 255 caused PCRE to crash.
3058    
3059    
3060    Version 4.1 12-Mar-03
3061    ---------------------
3062    
3063    1. Compiling with gcc -pedantic found a couple of places where casts were
3064    needed, and a string in dftables.c that was longer than standard compilers are
3065    required to support.
3066    
3067    2. Compiling with Sun's compiler found a few more places where the code could
3068    be tidied up in order to avoid warnings.
3069    
3070    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
3071    first of these names is deprecated in the latest Autoconf in favour of the name
3072    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
3073    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
3074    analogy I have changed it to CFLAGS_FOR_BUILD.
3075    
3076    4. Added -no-undefined to the linking command in the Makefile, because this is
3077    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
3078    linking step for the pcreposix library.
3079    
3080    5. PCRE was failing to diagnose the case of two named groups with the same
3081    name.
3082    
3083    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
3084    literal character that is needed in the subject for a match, and scans along to
3085    ensure that it is present before embarking on the full matching process. This
3086    saves time in cases of nested unlimited repeats that are never going to match.
3087    Problem: the scan can take a lot of time if the subject is very long (e.g.
3088    megabytes), thus penalizing straightforward matches. It is now done only if the
3089    amount of subject to be scanned is less than 1000 bytes.
3090    
3091    7. A lesser problem with the same optimization is that it was recording the
3092    first character of an anchored pattern as "needed", thus provoking a search
3093    right along the subject, even when the first match of the pattern was going to
3094    fail. The "needed" character is now not set for anchored patterns, unless it
3095    follows something in the pattern that is of non-fixed length. Thus, it still
3096    fulfils its original purpose of finding quick non-matches in cases of nested
3097    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
3098    
3099    
3100    Version 4.0 17-Feb-03
3101    ---------------------
3102    
3103    1. If a comment in an extended regex that started immediately after a meta-item
3104    extended to the end of string, PCRE compiled incorrect data. This could lead to
3105    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
3106    
3107    2. Moved to autoconf 2.53 and libtool 1.4.2.
3108    
3109    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
3110    the special perltest8 script is no longer needed - all the tests can be run
3111    from a single perltest script.
3112    
3113    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
3114    by \s. It has now been removed in PCRE. This means it isn't recognized as
3115    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
3116    class [:space:] *does* include VT, thereby creating a mess.
3117    
3118    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
3119    space and tab.
3120    
3121    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
3122    its new features into the main test script, reducing the number of scripts.
3123    
3124    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
3125    were backward compatible, and made the (?i) apply to the whole pattern, as if
3126    /i were given. Now it behaves more logically, and applies the option setting
3127    only to what follows. PCRE has been changed to follow suit. However, if it
3128    finds options settings right at the start of the pattern, it extracts them into
3129    the global options, as before. Thus, they show up in the info data.
3130    
3131    8. Added support for the \Q...\E escape sequence. Characters in between are
3132    treated as literals. This is slightly different from Perl in that $ and @ are
3133    also handled as literals inside the quotes. In Perl, they will cause variable
3134    interpolation. Note the following examples:
3135    
3136        Pattern            PCRE matches      Perl matches
3137    
3138        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
3139        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
3140        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
3141    
3142    For compatibility with Perl, \Q...\E sequences are recognized inside character
3143    classes as well as outside them.
3144    
3145    9. Re-organized 3 code statements in pcretest to avoid "overflow in
3146    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
3147    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
3148    signed/unsigned warnings.
3149    
3150    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
3151    option for pcretest, so I've replaced it by a simple function that does just
3152    that job.
3153    
3154    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
3155    "pcregrep -".
3156    
3157    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
3158    Java package. This provides some syntactic sugar for simple cases of what my
3159    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
3160    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
3161    item, you can use this simplified notation. Note that only makes sense with
3162    greedy quantifiers. Consequently, the use of the possessive quantifier forces
3163    greediness, whatever the setting of the PCRE_UNGREEDY option.
3164    
3165    13. A change of greediness default within a pattern was not taking effect at
3166    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
3167    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
3168    was abstracted outside.
3169    
3170    14. PCRE now supports the \G assertion. It is true when the current matching
3171    position is at the start point of the match. This differs from \A when the
3172    starting offset is non-zero. Used with the /g option of pcretest (or similar
3173    code), it works in the same way as it does for Perl's /g option. If all
3174    alternatives of a regex begin with \G, the expression is anchored to the start
3175    match position, and the "anchored" flag is set in the compiled expression.
3176    
3177    15. Some bugs concerning the handling of certain option changes within patterns
3178    have been fixed. These applied to options other than (?ims). For example,
3179    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
3180    been the other way round. Some of this was related to change 7 above.
3181    
3182    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
3183    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
3184    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
3185    POSIX classes only within a class (e.g. /[[:alpha:]]/).
3186    
3187    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
3188    mode. Unlike ".", it always matches newline, whatever the setting of
3189    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
3190    assertions. Perl allows it, but it doesn't (in general) work because it can't
3191    calculate the length of the lookbehind. At least, that's the case for Perl
3192    5.8.0 - I've been told they are going to document that it doesn't work in
3193    future.
3194    
3195    18. Added an error diagnosis for escapes that PCRE does not support: these are
3196    \L, \l, \N, \P, \p, \U, \u, and \X.
3197    
3198    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
3199    reading past the end of the pattern in cases such as /[abcd/.
3200    
3201    20. PCRE was getting more memory than necessary for patterns with classes that
3202    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
3203    
3204    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
3205    compiling PCRE for use with Virtual Pascal.
3206    
3207    22. Small fix to the Makefile to make it work properly if the build is done
3208    outside the source tree.
3209    
3210    23. Added a new extension: a condition to go with recursion. If a conditional
3211    subpattern starts with (?(R) the "true" branch is used if recursion has
3212    happened, whereas the "false" branch is used only at the top level.
3213    
3214    24. When there was a very long string of literal characters (over 255 bytes
3215    without UTF support, over 250 bytes with UTF support), the computation of how
3216    much memory was required could be incorrect, leading to segfaults or other
3217    strange effects.
3218    
3219    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
3220    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
3221    there was a subsequent back reference to those brackets. This meant that, for
3222    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
3223    possible to check for precisely this case. All we can do is abandon the
3224    optimization if .* occurs inside capturing brackets when there are any back
3225    references whatsoever. (See below for a better fix that came later.)
3226    
3227    26. The handling of the optimization for finding the first character of a
3228    non-anchored pattern, and for finding a character that is required later in the
3229    match were failing in some cases. This didn't break the matching; it just
3230    failed to optimize when it could. The way this is done has been re-implemented.
3231    
3232    27. Fixed typo in error message for invalid (?R item (it said "(?p").
3233    
3234    28. Added a new feature that provides some of the functionality that Perl
3235    provides with (?{...}). The facility is termed a "callout". The way it is done
3236    in PCRE is for the caller to provide an optional function, by setting
3237    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
3238    global variable. By default it is unset, which disables all calling out. To get
3239    the function called, the regex must include (?C) at appropriate points. This
3240    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
3241    This provides a means of identifying different callout points. When PCRE
3242    reaches such a point in the regex, if pcre_callout has been set, the external
3243    function is called. It is provided with data in a structure called
3244    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
3245    matching continues; if it returns a non-zero value, the match at the current
3246    point fails. However, backtracking will occur if possible. [This was changed
3247    later and other features added - see item 49 below.]
3248    
3249    29. pcretest is upgraded to test the callout functionality. It provides a
3250    callout function that displays information. By default, it shows the start of
3251    the match and the current position in the text. There are some new data escapes
3252    to vary what happens:
3253    
3254        \C+         in addition, show current contents of captured substrings
3255        \C-         do not supply a callout function
3256        \C!n        return 1 when callout number n is reached
3257        \C!n!m      return 1 when callout number n is reached for the mth time
3258    
3259    30. If pcregrep was called with the -l option and just a single file name, it
3260    output "<stdin>" if a match was found, instead of the file name.
3261    
3262    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
3263    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
3264    pcre_exec(). This saves a malloc/free per call. The default value of
3265    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
3266    when configuring.
3267    
3268    32. The default maximum size of a compiled pattern is 64K. There have been a
3269    few cases of people hitting this limit. The code now uses macros to handle the
3270    storing of links as offsets within the compiled pattern. It defaults to 2-byte
3271    links, but this can be changed to 3 or 4 bytes by --with-link-size when
3272    configuring. Tests 2 and 5 work only with 2-byte links because they output
3273    debugging information about compiled patterns.
3274    
3275    33. Internal code re-arrangements:
3276    
3277    (a) Moved the debugging function for printing out a compiled regex into
3278        its own source file (printint.c) and used #include to pull it into
3279        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
3280        separate copies.
3281    
3282    (b) Defined the list of op-code names for debugging as a macro in
3283        internal.h so that it is next to the definition of the opcodes.
3284    
3285    (c) Defined a table of op-code lengths for simpler skipping along compiled
3286        code. This is again a macro in internal.h so that it is next to the
3287        definition of the opcodes.
3288    
3289    34. Added support for recursive calls to individual subpatterns, along the
3290    lines of Robin Houston's patch (but implemented somewhat differently).
3291    
3292    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
3293    allow it to read and process whole directories in Win32. This code was
3294    contributed by Lionel Fourquaux; it has not been tested by me.
3295    
3296    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
3297    used to name a group. Names consist of alphanumerics and underscores, and must
3298    be unique. Back references use the syntax (?P=name) and recursive calls use
3299    (?P>name) which is a PCRE extension to the Python extension. Groups still have
3300    numbers. The function pcre_fullinfo() can be used after compilation to extract
3301    a name/number map. There are three relevant calls:
3302    
3303      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
3304      PCRE_INFO_NAMECOUNT            yields the number of entries
3305      PCRE_INFO_NAMETABLE            yields a pointer to the map.
3306    
3307    The map is a vector of fixed-size entries. The size of each entry depends on
3308    the length of the longest name used. The first two bytes of each entry are the
3309    group number, most significant byte first. There follows the corresponding
3310    name, zero terminated. The names are in alphabetical order.
3311    
3312    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
3313    case instead of 255. Making it the same both with and without UTF-8 support
3314    means that the same test output works with both.
3315    
3316    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
3317    calling malloc() with a zero argument.
3318    
3319    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
3320    optimization. I've improved things by keeping a bitmap of backreferences with
3321    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
3322    fact referenced, the optimization can be applied. It is unlikely that a
3323    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
3324    the match to follow \n) will appear inside brackets with a number greater than
3325    31, but if it does, any back reference > 31 suppresses the optimization.
3326    
3327    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
3328    of disabling numbered capturing parentheses. Any opening parenthesis that is
3329    not followed by ? behaves as if it were followed by ?: but named parentheses
3330    can still be used for capturing (and they will acquire numbers in the usual
3331    way).
3332    
3333    41. Redesigned the return codes from the match() function into yes/no/error so
3334    that errors can be passed back from deep inside the nested calls. A malloc
3335    failure while inside a recursive subpattern call now causes the
3336    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
3337    
3338    42. It is now possible to set a limit on the number of times the match()
3339    function is called in a call to pcre_exec(). This facility makes it possible to
3340    limit the amount of recursion and backtracking, though not in a directly
3341    obvious way, because the match() function is used in a number of different
3342    circumstances. The count starts from zero for each position in the subject
3343    string (for non-anchored patterns). The default limit is, for compatibility, a
3344    large number, namely 10 000 000. You can change this in two ways:
3345    
3346    (a) When configuring PCRE before making, you can use --with-match-limit=n
3347        to set a default value for the compiled library.
3348    
3349    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
3350        a different value is set. See 45 below.
3351    
3352    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
3353    
3354    43. Added a new function pcre_config(int, void *) to enable run-time extraction
3355    of things that can be changed at compile time. The first argument specifies
3356    what is wanted and the second points to where the information is to be placed.
3357    The current list of available information is:
3358    
3359      PCRE_CONFIG_UTF8
3360    
3361    The output is an integer that is set to one if UTF-8 support is available;
3362    otherwise it is set to zero.
3363    
3364      PCRE_CONFIG_NEWLINE
3365    
3366    The output is an integer that it set to the value of the code that is used for
3367    newline. It is either LF (10) or CR (13).
3368    
3369      PCRE_CONFIG_LINK_SIZE
3370    
3371    The output is an integer that contains the number of bytes used for internal
3372    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
3373    
3374      PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
3375    
3376    The output is an integer that contains the threshold above which the POSIX
3377    interface uses malloc() for output vectors. See item 31 above.
3378    
3379      PCRE_CONFIG_MATCH_LIMIT
3380    
3381    The output is an unsigned integer that contains the default limit of the number
3382    of match() calls in a pcre_exec() execution. See 42 above.
3383    
3384    44. pcretest has been upgraded by the addition of the -C option. This causes it
3385    to extract all the available output from the new pcre_config() function, and to
3386    output it. The program then exits immediately.
3387    
3388    45. A need has arisen to pass over additional data with calls to pcre_exec() in
3389    order to support additional features. One way would have been to define
3390    pcre_exec2() (for example) with extra arguments, but this would not have been
3391    extensible, and would also have required all calls to the original function to
3392    be mapped to the new one. Instead, I have chosen to extend the mechanism that
3393    is used for passing in "extra" data from pcre_study().
3394    
3395    The pcre_extra structure is now exposed and defined in pcre.h. It currently
3396    contains the following fields:
3397    
3398      flags         a bitmap indicating which of the following fields are set
3399      study_data    opaque data from pcre_study()
3400      match_limit   a way of specifying a limit on match() calls for a specific
3401                      call to pcre_exec()
3402      callout_data  data for callouts (see 49 below)
3403    
3404    The flag bits are also defined in pcre.h, and are
3405    
3406      PCRE_EXTRA_STUDY_DATA
3407      PCRE_EXTRA_MATCH_LIMIT
3408      PCRE_EXTRA_CALLOUT_DATA
3409    
3410    The pcre_study() function now returns one of these new pcre_extra blocks, with
3411    the actual study data pointed to by the study_data field, and the
3412    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
3413    before. That is, this change is entirely upwards-compatible and requires no
3414    change to existing code.
3415    
3416    If you want to pass in additional data to pcre_exec(), you can either place it
3417    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
3418    block.
3419    
3420    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
3421    data string contains the escape sequence \M, pcretest calls pcre_exec() several
3422    times with different match limits, until it finds the minimum value needed for
3423    pcre_exec() to complete. The value is then output. This can be instructive; for
3424    most simple matches the number is quite small, but for pathological cases it
3425    gets very large very quickly.
3426    
3427    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
3428    returns the size of the data block pointed to by the study_data field in a
3429    pcre_extra block, that is, the value that was passed as the argument to
3430    pcre_malloc() when PCRE was getting memory in which to place the information
3431    created by pcre_study(). The fourth argument should point to a size_t variable.
3432    pcretest has been extended so that this information is shown after a successful
3433    pcre_study() call when information about the compiled regex is being displayed.
3434    
3435    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
3436    because what follows is always an absolute path. (Later: it turns out that this
3437    is more than cosmetic for MinGW, because it doesn't like empty path
3438    components.)
3439    
3440    49. Some changes have been made to the callout feature (see 28 above):
3441    
3442    (i)  A callout function now has three choices for what it returns:
3443    
3444           0  =>  success, carry on matching
3445         > 0  =>  failure at this point, but backtrack if possible
3446         < 0  =>  serious error, return this value from pcre_exec()
3447    
3448         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
3449         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
3450         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
3451         use by callout functions. It will never be used by PCRE itself.
3452    
3453    (ii) The pcre_extra structure (see 45 above) has a void * field called
3454         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
3455         pcre_callout_block structure has a field of the same name. The contents of
3456         the field passed in the pcre_extra structure are passed to the callout
3457         function in the corresponding field in the callout block. This makes it
3458         easier to use the same callout-containing regex from multiple threads. For
3459         testing, the pcretest program has a new data escape
3460    
3461           \C*n        pass the number n (may be negative) as callout_data
3462    
3463         If the callout function in pcretest receives a non-zero value as
3464         callout_data, it returns that value.
3465    
3466    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
3467    there were some redundant $(CFLAGS) in commands that are now specified as
3468    $(LINK), which already includes $(CFLAGS).
3469    
3470    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
3471    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
3472    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
3473    one-byte characters throughout. Note that case-insensitive matching applies
3474    only to characters whose values are less than 256. PCRE doesn't support the
3475    notion of cases for higher-valued characters.
3476    
3477    (i)   A character class whose characters are all within 0-255 is handled as
3478          a bit map, and the map is inverted for negative classes. Previously, a
3479          character > 255 always failed to match such a class; however it should
3480          match if the class was a negative one (e.g. [^ab]). This has been fixed.
3481    
3482    (ii)  A negated character class with a single character < 255 is coded as
3483          "not this character" (OP_NOT). This wasn't working properly when the test
3484          character was multibyte, either singly or repeated.
3485    
3486    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
3487          mode, for example: \x{100}{2,3}.
3488    
3489    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
3490          singly or repeated) now correctly test multibyte characters. However,
3491          PCRE doesn't recognize any characters with values greater than 255 as
3492          digits, spaces, or word characters. Such characters always match \D, \S,
3493          and \W, and never match \d, \s, or \w.
3494    
3495    (v)   Classes may now contain characters and character ranges with values
3496          greater than 255. For example: [ab\x{100}-\x{400}].
3497    
3498    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
3499          PCRE in UTF-8 mode.
3500    
3501    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
3502    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
3503    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
3504    value.)
3505    
3506    53. The single man page has become too large. I have therefore split it up into
3507    a number of separate man pages. These also give rise to individual HTML pages;
3508    these are now put in a separate directory, and there is an index.html page that
3509    lists them all. Some hyperlinking between the pages has been installed.
3510    
3511    54. Added convenience functions for handling named capturing parentheses.
3512    
3513    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
3514    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
3515    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
3516    are faulted.
3517    
3518    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
3519    calling configure. These values are used when compiling the dftables.c program
3520    which is run to generate the source of the default character tables. They
3521    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
3522    you will need to set these values.
3523    
3524    57. Updated the building process for Windows DLL, as provided by Fred Cox.
3525    
3526    
3527    Version 3.9 02-Jan-02
3528    ---------------------
3529    
3530    1. A bit of extraneous text had somehow crept into the pcregrep documentation.
3531    
3532    2. If --disable-static was given, the building process failed when trying to
3533    build pcretest and pcregrep. (For some reason it was using libtool to compile
3534    them, which is not right, as they aren't part of the library.)
3535    
3536    
3537    Version 3.8 18-Dec-01
3538    ---------------------
3539    
3540    1. The experimental UTF-8 code was completely screwed up. It was packing the
3541    bytes in the wrong order. How dumb can you get?
3542    
3543    
3544    Version 3.7 29-Oct-01
3545    ---------------------
3546    
3547    1. In updating pcretest to check change 1 of version 3.6, I screwed up.
3548    This caused pcretest, when used on the test data, to segfault. Unfortunately,
3549    this didn't happen under Solaris 8, where I normally test things.
3550    
3551    2. The Makefile had to be changed to make it work on BSD systems, where 'make'
3552    doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
3553    isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
3554    this fix an hour or so after the initial 3.7 release.)
3555    
3556    
3557    Version 3.6 23-Oct-01
3558    ---------------------
3559    
3560    1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
3561    offsets passed as NULL with zero offset count.
3562    
3563    2. The config.guess and config.sub files had not been updated when I moved to
3564    the latest autoconf.
3565    
3566    
3567    Version 3.5 15-Aug-01
3568    ---------------------
3569    
3570    1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
3571    had been forgotten.
3572    
3573    2. By using declared but undefined structures, we can avoid using "void"
3574    definitions in pcre.h while keeping the internal definitions of the structures
3575    private.
3576    
3577    3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
3578    user point of view, this means that both static and shared libraries are built
3579    by default, but this can be individually controlled. More of the work of
3580    handling this static/shared cases is now inside libtool instead of PCRE's make
3581    file.
3582    
3583    4. The pcretest utility is now installed along with pcregrep because it is
3584    useful for users (to test regexs) and by doing this, it automatically gets
3585    relinked by libtool. The documentation has been turned into a man page, so
3586    there are now .1, .txt, and .html versions in /doc.
3587    
3588    5. Upgrades to pcregrep:
3589       (i)   Added long-form option names like gnu grep.
3590       (ii)  Added --help to list all options with an explanatory phrase.
3591       (iii) Added -r, --recursive to recurse into sub-directories.
3592       (iv)  Added -f, --file to read patterns from a file.
3593    
3594    6. pcre_exec() was referring to its "code" argument before testing that
3595    argument for NULL (and giving an error if it was NULL).
3596    
3597    7. Upgraded Makefile.in to allow for compiling in a different directory from
3598    the source directory.
3599    
3600    8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
3601    options bits, the pointer it was passed was to an int instead of to an unsigned
3602    long int. This mattered only on 64-bit systems.
3603    
3604    9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
3605    generated) instead of pcre.in, which it its source. Also made the same change