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

Diff of /code/trunk/ChangeLog

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

revision 5 by nigel, Sat Feb 24 21:38:05 2007 UTC revision 614 by ph10, Sat Jul 9 10:48:16 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    
127    Version 8.12 15-Jan-2011
128    ------------------------
129    
130    1.  Fixed some typos in the markup of the man pages, and wrote a script that
131        checks for such things as part of the documentation building process.
132    
133    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
134        --match-limit and --recursion-limit options (added for 8.11). In
135        particular, this made one of the standard tests fail. (The integer value
136        went into the wrong half of a long int.)
137    
138    3.  If the --colour option was given to pcregrep with -v (invert match), it
139        did strange things, either producing crazy output, or crashing. It should,
140        of course, ignore a request for colour when reporting lines that do not
141        match.
142    
143    4.  Another pcregrep bug caused similar problems if --colour was specified with
144        -M (multiline) and the pattern match finished with a line ending.
145    
146    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
147        matched in multiline mode, the following line was shown as part of the
148        match. This seems wrong, so I have changed it.
149    
150    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
151        the check for further matches in the same line (so they could be coloured)
152        to overrun the end of the current line. If another match was found, it was
153        incorrectly shown (and then shown again when found in the next line).
154    
155    7.  If pcregrep was compiled under Windows, there was a reference to the
156        function pcregrep_exit() before it was defined. I am assuming this was
157        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
158        reported by a user. I've moved the definition above the reference.
159    
160    
161    Version 8.11 10-Dec-2010
162    ------------------------
163    
164    1.  (*THEN) was not working properly if there were untried alternatives prior
165        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
166        backtracked to try for "b" instead of moving to the next alternative branch
167        at the same level (in this case, to look for "c"). The Perl documentation
168        is clear that when (*THEN) is backtracked onto, it goes to the "next
169        alternative in the innermost enclosing group".
170    
171    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
172        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
173        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
174        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
175        (*THEN).
176    
177    3.  If \s appeared in a character class, it removed the VT character from
178        the class, even if it had been included by some previous item, for example
179        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
180        of \s, but is part of the POSIX "space" class.)
181    
182    4.  A partial match never returns an empty string (because you can always
183        match an empty string at the end of the subject); however the checking for
184        an empty string was starting at the "start of match" point. This has been
185        changed to the "earliest inspected character" point, because the returned
186        data for a partial match starts at this character. This means that, for
187        example, /(?<=abc)def/ gives a partial match for the subject "abc"
188        (previously it gave "no match").
189    
190    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
191        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
192        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
193        has an implication that the given string is incomplete (because a partial
194        match is preferred over a full match). For this reason, these items now
195        give a partial match in this situation. [Aside: previously, the one case
196        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
197        match rather than a full match, which was wrong by the old rules, but is
198        now correct.]
199    
200    6.  There was a bug in the handling of #-introduced comments, recognized when
201        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
202        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
203        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
204        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
205        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
206        places in pcre_compile().
207    
208    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
209        comments when looking ahead for named forward references to subpatterns,
210        the only newline sequence it recognized was NL. It now handles newlines
211        according to the set newline convention.
212    
213    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
214        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
215        cater for a lack of strerror(). These oversights have been fixed.
216    
217    9.  Added --match-limit and --recursion-limit to pcregrep.
218    
219    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
220    
221    11. When the -o option was used, pcregrep was setting a return code of 1, even
222        when matches were found, and --line-buffered was not being honoured.
223    
224    12. Added an optional parentheses number to the -o and --only-matching options
225        of pcregrep.
226    
227    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
228        can match an empty string. The code to do it in pcretest and pcredemo
229        needed fixing:
230    
231        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
232            only one byte after an empty string match just before CRLF (this case
233            just got forgotten; "any" and "anycrlf" were OK).
234    
235        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
236            mode when an empty string match preceded an ASCII character followed by
237            a non-ASCII character. (The code for advancing by one character rather
238            than one byte was nonsense.)
239    
240        (c) The pcredemo.c sample program did not have any code at all to handle
241            the cases when CRLF is a valid newline sequence.
242    
243    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
244        as a starting offset was within the subject string. There is now a new
245        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
246        negative or greater than the length of the string. In order to test this,
247        pcretest is extended to allow the setting of negative starting offsets.
248    
249    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
250        starting offset points to the beginning of a UTF-8 character was
251        unnecessarily clumsy. I tidied it up.
252    
253    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
254        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
255    
256    17. Nobody had reported that the --include_dir option, which was added in
257        release 7.7 should have been called --include-dir (hyphen, not underscore)
258        for compatibility with GNU grep. I have changed it to --include-dir, but
259        left --include_dir as an undocumented synonym, and the same for
260        --exclude-dir, though that is not available in GNU grep, at least as of
261        release 2.5.4.
262    
263    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
264        characters from a string of bytes) have been redefined so as not to use
265        loops, in order to improve performance in some environments. At the same
266        time, I abstracted some of the common code into auxiliary macros to save
267        repetition (this should not affect the compiled code).
268    
269    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
270        compile-time error is now given if \c is not followed by an ASCII
271        character, that is, a byte less than 128. (In EBCDIC mode, the code is
272        different, and any byte value is allowed.)
273    
274    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
275        START_OPTIMIZE option, which is now allowed at compile time - but just
276        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
277        to pcregrep and other applications that have no direct access to PCRE
278        options. The new /Y option in pcretest sets this option when calling
279        pcre_compile().
280    
281    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
282        back references. Groups containing recursive back references were forced to
283        be atomic by that change, but in the case of named groups, the amount of
284        memory required was incorrectly computed, leading to "Failed: internal
285        error: code overflow". This has been fixed.
286    
287    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
288        pcretest.c, to avoid build problems in some Borland environments.
289    
290    
291    Version 8.10 25-Jun-2010
292    ------------------------
293    
294    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
295        THEN.
296    
297    2.  (*ACCEPT) was not working when inside an atomic group.
298    
299    3.  Inside a character class, \B is treated as a literal by default, but
300        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
301        causes the error). The code is unchanged, but I tidied the documentation.
302    
303    4.  Inside a character class, PCRE always treated \R and \X as literals,
304        whereas Perl faults them if its -w option is set. I have changed PCRE so
305        that it faults them when PCRE_EXTRA is set.
306    
307    5.  Added support for \N, which always matches any character other than
308        newline. (It is the same as "." when PCRE_DOTALL is not set.)
309    
310    6.  When compiling pcregrep with newer versions of gcc which may have
311        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
312        declared with attribute warn_unused_result" were given. Just casting the
313        result to (void) does not stop the warnings; a more elaborate fudge is
314        needed. I've used a macro to implement this.
315    
316    7.  Minor change to pcretest.c to avoid a compiler warning.
317    
318    8.  Added four artifical Unicode properties to help with an option to make
319        \s etc use properties (see next item). The new properties are: Xan
320        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
321    
322    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
323        use Unicode properties. (*UCP) at the start of a pattern can be used to set
324        this option. Modified pcretest to add /W to test this facility. Added
325        REG_UCP to make it available via the POSIX interface.
326    
327    10. Added --line-buffered to pcregrep.
328    
329    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
330        studied, and the match started with a letter with a code point greater than
331        127 whose first byte was different to the first byte of the other case of
332        the letter, the other case of this starting letter was not recognized
333        (#976).
334    
335    12. If a pattern that was studied started with a repeated Unicode property
336        test, for example, \p{Nd}+, there was the theoretical possibility of
337        setting up an incorrect bitmap of starting bytes, but fortunately it could
338        not have actually happened in practice until change 8 above was made (it
339        added property types that matched character-matching opcodes).
340    
341    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
342        possible starting bytes for non-anchored patterns.
343    
344    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
345        \R, and also a number of cases that involve Unicode properties, both
346        explicit and implicit when PCRE_UCP is set.
347    
348    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
349        input, it could crash or give wrong results if characters with values
350        greater than 0xc0 were present in the subject string. (Detail: it assumed
351        UTF-8 input when processing these items.)
352    
353    16. Added a lot of (int) casts to avoid compiler warnings in systems where
354        size_t is 64-bit (#991).
355    
356    17. Added a check for running out of memory when PCRE is compiled with
357        --disable-stack-for-recursion (#990).
358    
359    18. If the last data line in a file for pcretest does not have a newline on
360        the end, a newline was missing in the output.
361    
362    19. The default pcre_chartables.c file recognizes only ASCII characters (values
363        less than 128) in its various bitmaps. However, there is a facility for
364        generating tables according to the current locale when PCRE is compiled. It
365        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
366        characters, are recognized by isspace() and therefore were getting set in
367        these tables, and indeed these tables seem to approximate to ISO 8859. This
368        caused a problem in UTF-8 mode when pcre_study() was used to create a list
369        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
370        which of course cannot start UTF-8 characters. I have changed the code so
371        that only real ASCII characters (less than 128) and the correct starting
372        bytes for UTF-8 encodings are set for characters greater than 127 when in
373        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
374        altogether.)
375    
376    20. Added the /T option to pcretest so as to be able to run tests with non-
377        standard character tables, thus making it possible to include the tests
378        used for 19 above in the standard set of tests.
379    
380    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
381        reference to a subpattern the other side of a comment that contains an
382        opening parenthesis caused either an internal compiling error, or a
383        reference to the wrong subpattern.
384    
385    
386    Version 8.02 19-Mar-2010
387    ------------------------
388    
389    1.  The Unicode data tables have been updated to Unicode 5.2.0.
390    
391    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
392        configured.
393    
394    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
395        original author of that file, following a query about its status.
396    
397    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
398        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
399    
400    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
401        quantifier applied to a forward-referencing subroutine call, could compile
402        incorrect code or give the error "internal error: previously-checked
403        referenced subpattern not found".
404    
405    6.  Both MS Visual Studio and Symbian OS have problems with initializing
406        variables to point to external functions. For these systems, therefore,
407        pcre_malloc etc. are now initialized to local functions that call the
408        relevant global functions.
409    
410    7.  There were two entries missing in the vectors called coptable and poptable
411        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
412        I've fixed the data, and added a kludgy way of testing at compile time that
413        the lengths are correct (equal to the number of opcodes).
414    
415    8.  Following on from 7, I added a similar kludge to check the length of the
416        eint vector in pcreposix.c.
417    
418    9.  Error texts for pcre_compile() are held as one long string to avoid too
419        much relocation at load time. To find a text, the string is searched,
420        counting zeros. There was no check for running off the end of the string,
421        which could happen if a new error number was added without updating the
422        string.
423    
424    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
425    
426    11. \K was not working if it appeared in an atomic group or in a group that
427        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
428        \K is "not well defined" if used in an assertion. PCRE now accepts it if
429        the assertion is positive, but not if it is negative.
430    
431    12. Change 11 fortuitously reduced the size of the stack frame used in the
432        "match()" function of pcre_exec.c by one pointer. Forthcoming
433        implementation of support for (*MARK) will need an extra pointer on the
434        stack; I have reserved it now, so that the stack frame size does not
435        decrease.
436    
437    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
438        item in branch that calls a recursion is a subroutine call - as in the
439        second branch in the above example - was incorrectly given the compile-
440        time error "recursive call could loop indefinitely" because pcre_compile()
441        was not correctly checking the subroutine for matching a non-empty string.
442    
443    14. The checks for overrunning compiling workspace could trigger after an
444        overrun had occurred. This is a "should never occur" error, but it can be
445        triggered by pathological patterns such as hundreds of nested parentheses.
446        The checks now trigger 100 bytes before the end of the workspace.
447    
448    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
449    
450    
451    Version 8.01 19-Jan-2010
452    ------------------------
453    
454    1.  If a pattern contained a conditional subpattern with only one branch (in
455        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
456        computed the wrong minimum data length (which is of course zero for such
457        subpatterns). This could cause incorrect "no match" results.
458    
459    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
460        the pattern is reset in the first branch, pcre_compile() failed with
461        "internal error: code overflow at offset...". This happened only when
462        the reset was to the original external option setting. (An optimization
463        abstracts leading options settings into an external setting, which was the
464        cause of this.)
465    
466    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
467        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
468        assertion pattern did not match (meaning that the assertion was true), it
469        was incorrectly treated as false if the SKIP had been reached during the
470        matching. This also applied to assertions used as conditions.
471    
472    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
473        assertion subpattern, including such a pattern used as a condition,
474        unpredictable results occurred, instead of the error return
475        PCRE_ERROR_DFA_UITEM.
476    
477    5.  The C++ GlobalReplace function was not working like Perl for the special
478        situation when an empty string is matched. It now does the fancy magic
479        stuff that is necessary.
480    
481    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
482        removed. (These were left over from very, very early versions of PCRE.)
483    
484    7.  Some cosmetic changes to the code to make life easier when compiling it
485        as part of something else:
486    
487        (a) Change DEBUG to PCRE_DEBUG.
488    
489        (b) In pcre_compile(), rename the member of the "branch_chain" structure
490            called "current" as "current_branch", to prevent a collision with the
491            Linux macro when compiled as a kernel module.
492    
493        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
494            prevent a collision with the Linux macro when compiled as a kernel
495            module.
496    
497    8.  In pcre_compile() there are some checks for integer overflows that used to
498        cast potentially large values to (double). This has been changed to that
499        when building, a check for int64_t is made, and if it is found, it is used
500        instead, thus avoiding the use of floating point arithmetic. (There is no
501        other use of FP in PCRE.) If int64_t is not found, the fallback is to
502        double.
503    
504    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
505        2005 (difference between two addresses compared to an unsigned value).
506    
507    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
508        custom one, because of the following reported problem in Windows:
509    
510          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
511              under Win32.
512          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
513              therefore missing the function definition.
514          - The compiler thus generates a "C" signature for the test function.
515          - The linker fails to find the "C" function.
516          - PCRE fails to configure if asked to do so against libbz2.
517    
518    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
519        messages were output:
520    
521          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
522          rerunning libtoolize, to keep the correct libtool macros in-tree.
523          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
524    
525        I have done both of these things.
526    
527    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
528        most of the time, it *can* run out if it is given a pattern that contains a
529        runaway infinite recursion. I updated the discussion in the pcrestack man
530        page.
531    
532    13. Now that we have gone to the x.xx style of version numbers, the minor
533        version may start with zero. Using 08 or 09 is a bad idea because users
534        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
535        interpreted as invalid octal numbers. I've updated the previous comment in
536        configure.ac, and also added a check that gives an error if 08 or 09 are
537        used.
538    
539    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
540        causing partial matching to fail when the end of the subject matched \W
541        in a UTF-8 pattern where \W was quantified with a minimum of 3.
542    
543    15. There were some discrepancies between the declarations in pcre_internal.h
544        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
545        their definitions. The declarations used "const uschar *" and the
546        definitions used USPTR. Even though USPTR is normally defined as "const
547        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
548        reported that: "This difference in casting confuses some C++ compilers, for
549        example, SunCC recognizes above declarations as different functions and
550        generates broken code for hbpcre." I have changed the declarations to use
551        USPTR.
552    
553    16. GNU libtool is named differently on some systems. The autogen.sh script now
554        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
555        (FreeBSD).
556    
557    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
558        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
559        comment: "Figure out how to create a longlong from a string: strtoll and
560        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
561        instance, but it only takes 2 args instead of 3!"
562    
563    18. A subtle bug concerned with back references has been fixed by a change of
564        specification, with a corresponding code fix. A pattern such as
565        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
566        refers, was giving matches when it shouldn't. For example, xa=xaaa would
567        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
568        same bug. Such groups have to be quantified to be useful, or contained
569        inside another quantified group. (If there's no repetition, the reference
570        can never match.) The problem arises because, having left the group and
571        moved on to the rest of the pattern, a later failure that backtracks into
572        the group uses the captured value from the final iteration of the group
573        rather than the correct earlier one. I have fixed this in PCRE by forcing
574        any group that contains a reference to itself to be an atomic group; that
575        is, there cannot be any backtracking into it once it has completed. This is
576        similar to recursive and subroutine calls.
577    
578    
579    Version 8.00 19-Oct-09
580    ----------------------
581    
582    1.  The table for translating pcre_compile() error codes into POSIX error codes
583        was out-of-date, and there was no check on the pcre_compile() error code
584        being within the table. This could lead to an OK return being given in
585        error.
586    
587    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
588        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
589        in a Windows environment.
590    
591    3.  The pcregrep --count option prints the count for each file even when it is
592        zero, as does GNU grep. However, pcregrep was also printing all files when
593        --files-with-matches was added. Now, when both options are given, it prints
594        counts only for those files that have at least one match. (GNU grep just
595        prints the file name in this circumstance, but including the count seems
596        more useful - otherwise, why use --count?) Also ensured that the
597        combination -clh just lists non-zero counts, with no names.
598    
599    4.  The long form of the pcregrep -F option was incorrectly implemented as
600        --fixed_strings instead of --fixed-strings. This is an incompatible change,
601        but it seems right to fix it, and I didn't think it was worth preserving
602        the old behaviour.
603    
604    5.  The command line items --regex=pattern and --regexp=pattern were not
605        recognized by pcregrep, which required --regex pattern or --regexp pattern
606        (with a space rather than an '='). The man page documented the '=' forms,
607        which are compatible with GNU grep; these now work.
608    
609    6.  No libpcreposix.pc file was created for pkg-config; there was just
610        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
611    
612    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
613        when UCP support is not needed, by modifying the Python script that
614        generates it from Unicode data files. This should not matter if the module
615        is correctly used as a library, but I received one complaint about 50K of
616        unwanted data. My guess is that the person linked everything into his
617        program rather than using a library. Anyway, it does no harm.
618    
619    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
620        was a minimum greater than 1 for a wide character in a possessive
621        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
622        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
623        character. Chaos in the form of incorrect output or a compiling loop could
624        result.
625    
626    9.  The restrictions on what a pattern can contain when partial matching is
627        requested for pcre_exec() have been removed. All patterns can now be
628        partially matched by this function. In addition, if there are at least two
629        slots in the offset vector, the offset of the earliest inspected character
630        for the match and the offset of the end of the subject are set in them when
631        PCRE_ERROR_PARTIAL is returned.
632    
633    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
634        synonymous with PCRE_PARTIAL, for backwards compatibility, and
635        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
636        and may be more useful for multi-segment matching.
637    
638    11. Partial matching with pcre_exec() is now more intuitive. A partial match
639        used to be given if ever the end of the subject was reached; now it is
640        given only if matching could not proceed because another character was
641        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
642        string "Z", which now yields "no match" instead of "partial match". In the
643        case of pcre_dfa_exec(), "no match" is given if every matching path for the
644        final character ended with (*FAIL).
645    
646    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
647        if the pattern had a "must contain" character that was already found in the
648        earlier partial match, unless partial matching was again requested. For
649        example, with the pattern /dog.(body)?/, the "must contain" character is
650        "g". If the first part-match was for the string "dog", restarting with
651        "sbody" failed. This bug has been fixed.
652    
653    13. The string returned by pcre_dfa_exec() after a partial match has been
654        changed so that it starts at the first inspected character rather than the
655        first character of the match. This makes a difference only if the pattern
656        starts with a lookbehind assertion or \b or \B (\K is not supported by
657        pcre_dfa_exec()). It's an incompatible change, but it makes the two
658        matching functions compatible, and I think it's the right thing to do.
659    
660    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
661        so that the demonstration program is easily available in environments where
662        PCRE has not been installed from source.
663    
664    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
665        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
666        library.
667    
668    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
669        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
670        is not the first non-POSIX option to be added. Clearly some people find
671        these options useful.
672    
673    17. If a caller to the POSIX matching function regexec() passes a non-zero
674        value for nmatch with a NULL value for pmatch, the value of
675        nmatch is forced to zero.
676    
677    18. RunGrepTest did not have a test for the availability of the -u option of
678        the diff command, as RunTest does. It now checks in the same way as
679        RunTest, and also checks for the -b option.
680    
681    19. If an odd number of negated classes containing just a single character
682        interposed, within parentheses, between a forward reference to a named
683        subpattern and the definition of the subpattern, compilation crashed with
684        an internal error, complaining that it could not find the referenced
685        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
686        [The bug was that it was starting one character too far in when skipping
687        over the character class, thus treating the ] as data rather than
688        terminating the class. This meant it could skip too much.]
689    
690    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
691        /g option in pcretest when the pattern contains \K, which makes it possible
692        to have an empty string match not at the start, even when the pattern is
693        anchored. Updated pcretest and pcredemo to use this option.
694    
695    21. If the maximum number of capturing subpatterns in a recursion was greater
696        than the maximum at the outer level, the higher number was returned, but
697        with unset values at the outer level. The correct (outer level) value is
698        now given.
699    
700    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
701        PCRE did not set those parentheses (unlike Perl). I have now found a way to
702        make it do so. The string so far is captured, making this feature
703        compatible with Perl.
704    
705    23. The tests have been re-organized, adding tests 11 and 12, to make it
706        possible to check the Perl 5.10 features against Perl 5.10.
707    
708    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
709        pattern matches a fixed length string. PCRE did not allow this; now it
710        does. Neither allows recursion.
711    
712    25. I finally figured out how to implement a request to provide the minimum
713        length of subject string that was needed in order to match a given pattern.
714        (It was back references and recursion that I had previously got hung up
715        on.) This code has now been added to pcre_study(); it finds a lower bound
716        to the length of subject needed. It is not necessarily the greatest lower
717        bound, but using it to avoid searching strings that are too short does give
718        some useful speed-ups. The value is available to calling programs via
719        pcre_fullinfo().
720    
721    26. While implementing 25, I discovered to my embarrassment that pcretest had
722        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
723        study optimizations had never been tested with that matching function.
724        Oops. What is worse, even when it was passed study data, there was a bug in
725        pcre_dfa_exec() that meant it never actually used it. Double oops. There
726        were also very few tests of studied patterns with pcre_dfa_exec().
727    
728    27. If (?| is used to create subpatterns with duplicate numbers, they are now
729        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
730        on the other side of the coin, they are no longer allowed to have different
731        names, because these cannot be distinguished in PCRE, and this has caused
732        confusion. (This is a difference from Perl.)
733    
734    28. When duplicate subpattern names are present (necessarily with different
735        numbers, as required by 27 above), and a test is made by name in a
736        conditional pattern, either for a subpattern having been matched, or for
737        recursion in such a pattern, all the associated numbered subpatterns are
738        tested, and the overall condition is true if the condition is true for any
739        one of them. This is the way Perl works, and is also more like the way
740        testing by number works.
741    
742    
743    Version 7.9 11-Apr-09
744    ---------------------
745    
746    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
747        (pcretest), all targets were linked against these libraries. This included
748        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
749        libraries. This caused unwanted dependencies to be created. This problem
750        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
751        pcretest is linked with readline.
752    
753    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
754        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
755        moved outside it again, because FALSE and TRUE are already defined in AIX,
756        but BOOL is not.
757    
758    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
759        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
760    
761    4.  The pcregrep documentation said spaces were inserted as well as colons (or
762        hyphens) following file names and line numbers when outputting matching
763        lines. This is not true; no spaces are inserted. I have also clarified the
764        wording for the --colour (or --color) option.
765    
766    5.  In pcregrep, when --colour was used with -o, the list of matching strings
767        was not coloured; this is different to GNU grep, so I have changed it to be
768        the same.
769    
770    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
771        each matching line was coloured. Now it goes on to look for further matches
772        of any of the test patterns, which is the same behaviour as GNU grep.
773    
774    7.  A pattern that could match an empty string could cause pcregrep to loop; it
775        doesn't make sense to accept an empty string match in pcregrep, so I have
776        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
777        seems to be how GNU grep behaves.
778    
779    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
780        start or after a newline", because the conditional assertion was not being
781        correctly handled. The rule now is that both the assertion and what follows
782        in the first alternative must satisfy the test.
783    
784    9.  If auto-callout was enabled in a pattern with a conditional group whose
785        condition was an assertion, PCRE could crash during matching, both with
786        pcre_exec() and pcre_dfa_exec().
787    
788    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
789        used for matching.
790    
791    11. Unicode property support in character classes was not working for
792        characters (bytes) greater than 127 when not in UTF-8 mode.
793    
794    12. Added the -M command line option to pcretest.
795    
796    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
797    
798    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
799    
800    16. Added comments and documentation about mis-use of no_arg in the C++
801        wrapper.
802    
803    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
804        from Martin Jerabek that uses macro names for all relevant character and
805        string constants.
806    
807    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
808        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
809        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
810        these, but not everybody uses configure.
811    
812    19. A conditional group that had only one branch was not being correctly
813        recognized as an item that could match an empty string. This meant that an
814        enclosing group might also not be so recognized, causing infinite looping
815        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
816        with the subject "ab", where knowledge that the repeated group can match
817        nothing is needed in order to break the loop.
818    
819    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
820        exec(), but without supplying a callout function, matching went wrong.
821    
822    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
823        leak if the size of the offset vector was greater than 30. When the vector
824        is smaller, the saved offsets during recursion go onto a local stack
825        vector, but for larger vectors malloc() is used. It was failing to free
826        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
827        error, in fact).
828    
829    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
830        heapframe that is used only when UTF-8 support is enabled. This caused no
831        problem, but was untidy.
832    
833    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
834        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
835        included within another project.
836    
837    24. Steven Van Ingelgem's patches to add more options to the CMake support,
838        slightly modified by me:
839    
840          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
841              not building pcregrep.
842    
843          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
844              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
845    
846    25. Forward references, both numeric and by name, in patterns that made use of
847        duplicate group numbers, could behave incorrectly or give incorrect errors,
848        because when scanning forward to find the reference group, PCRE was not
849        taking into account the duplicate group numbers. A pattern such as
850        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
851    
852    26. Changed a few more instances of "const unsigned char *" to USPTR, making
853        the feature of a custom pointer more persuasive (as requested by a user).
854    
855    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
856        pcretest.c, inside #ifndefs, because it seems they are sometimes already
857        pre-defined.
858    
859    28. Added support for (*UTF8) at the start of a pattern.
860    
861    29. Arrange for flags added by the "release type" setting in CMake to be shown
862        in the configuration summary.
863    
864    
865    Version 7.8 05-Sep-08
866    ---------------------
867    
868    1.  Replaced UCP searching code with optimized version as implemented for Ad
869        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
870        stage table and inline lookup instead of a function, giving speed ups of 2
871        to 5 times on some simple patterns that I tested. Permission was given to
872        distribute the MultiStage2.py script that generates the tables (it's not in
873        the tarball, but is in the Subversion repository).
874    
875    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
876        scripts.
877    
878    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
879        a group with a zero qualifier. The result of the study could be incorrect,
880        or the function might crash, depending on the pattern.
881    
882    4.  Caseless matching was not working for non-ASCII characters in back
883        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
884        It now works when Unicode Property Support is available.
885    
886    5.  In pcretest, an escape such as \x{de} in the data was always generating
887        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
888        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
889        truncation.
890    
891    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
892    
893    7.  Added two (int) casts to pcregrep when printing the difference of two
894        pointers, in case they are 64-bit values.
895    
896    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
897        test 2 if it fails.
898    
899    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
900        and a #define of that name to empty if it is not externally set. This is to
901        allow users of MSVC to set it if necessary.
902    
903    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
904        the convenience functions in the pcre_get.c source file.
905    
906    11. An option change at the start of a pattern that had top-level alternatives
907        could cause overwriting and/or a crash. This command provoked a crash in
908        some environments:
909    
910          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
911    
912        This potential security problem was recorded as CVE-2008-2371.
913    
914    12. For a pattern where the match had to start at the beginning or immediately
915        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
916        pcre_dfa_exec() could read past the end of the passed subject if there was
917        no match. To help with detecting such bugs (e.g. with valgrind), I modified
918        pcretest so that it places the subject at the end of its malloc-ed buffer.
919    
920    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
921        exec() might read past the end of the data buffer in UTF-8 mode.
922    
923    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
924        the data contained the byte 0x85 as part of a UTF-8 character within its
925        first line. This applied both to normal and DFA matching.
926    
927    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
928        /^[^d]*?$/8 failed to match "abc".
929    
930    16. Added a missing copyright notice to pcrecpp_internal.h.
931    
932    17. Make it more clear in the documentation that values returned from
933        pcre_exec() in ovector are byte offsets, not character counts.
934    
935    18. Tidied a few places to stop certain compilers from issuing warnings.
936    
937    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
938        supplied by Stefan Weber. I made a further small update for 7.8 because
939        there is a change of source arrangements: the pcre_searchfuncs.c module is
940        replaced by pcre_ucd.c.
941    
942    
943    Version 7.7 07-May-08
944    ---------------------
945    
946    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
947        a string to a long long, pretend we don't even have a long long." This is
948        done by checking for the strtoq, strtoll, and _strtoi64 functions.
949    
950    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
951        pre-7.6 versions, which defined a global no_arg variable instead of putting
952        it in the RE class. (See also #8 below.)
953    
954    3.  Remove a line of dead code, identified by coverity and reported by Nuno
955        Lopes.
956    
957    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
958    
959        (1) The include/exclude patterns were being applied to the whole pathnames
960            of files, instead of just to the final components.
961    
962        (2) If there was more than one level of directory, the subdirectories were
963            skipped unless they satisfied the include/exclude conditions. This is
964            inconsistent with GNU grep (and could even be seen as contrary to the
965            pcregrep specification - which I improved to make it absolutely clear).
966            The action now is always to scan all levels of directory, and just
967            apply the include/exclude patterns to regular files.
968    
969    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
970        --exclude_dir in the tests to avoid scanning .svn directories.
971    
972    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
973        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
974        doesn't support NULs in patterns.
975    
976    7.  Added some missing "const"s to declarations of static tables in
977        pcre_compile.c and pcre_dfa_exec.c.
978    
979    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
980        caused by fix #2  above. (Subsequently also a second patch to fix the
981        first patch. And a third patch - this was a messy problem.)
982    
983    9.  Applied Craig's patch to remove the use of push_back().
984    
985    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
986        matching function regexec().
987    
988    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
989        which, however, unlike Perl's \g{...}, are subroutine calls, not back
990        references. PCRE supports relative numbers with this syntax (I don't think
991        Oniguruma does).
992    
993    12. Previously, a group with a zero repeat such as (...){0} was completely
994        omitted from the compiled regex. However, this means that if the group
995        was called as a subroutine from elsewhere in the pattern, things went wrong
996        (an internal error was given). Such groups are now left in the compiled
997        pattern, with a new opcode that causes them to be skipped at execution
998        time.
999    
1000    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
1001        to the way PCRE behaves:
1002    
1003        (a) A lone ] character is dis-allowed (Perl treats it as data).
1004    
1005        (b) A back reference to an unmatched subpattern matches an empty string
1006            (Perl fails the current match path).
1007    
1008        (c) A data ] in a character class must be notated as \] because if the
1009            first data character in a class is ], it defines an empty class. (In
1010            Perl it is not possible to have an empty class.) The empty class []
1011            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
1012            The negative empty class [^] matches any one character, independently
1013            of the DOTALL setting.
1014    
1015    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
1016        non-existent subpattern following a character class starting with ']' and
1017        containing () gave an internal compiling error instead of "reference to
1018        non-existent subpattern". Fortunately, when the pattern did exist, the
1019        compiled code was correct. (When scanning forwards to check for the
1020        existencd of the subpattern, it was treating the data ']' as terminating
1021        the class, so got the count wrong. When actually compiling, the reference
1022        was subsequently set up correctly.)
1023    
1024    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
1025        it was being rejected as not supported by pcre_dfa_exec(), even though
1026        other assertions are supported. I have made pcre_dfa_exec() support
1027        (*FAIL).
1028    
1029    16. The implementation of 13c above involved the invention of a new opcode,
1030        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
1031        cannot be changed at match time, I realized I could make a small
1032        improvement to matching performance by compiling OP_ALLANY instead of
1033        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
1034        on the OP_ANY path.
1035    
1036    17. Compiling pcretest on Windows with readline support failed without the
1037        following two fixes: (1) Make the unistd.h include conditional on
1038        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
1039    
1040    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
1041        ncurses library to be included for pcretest when ReadLine support is
1042        requested, but also to allow for it to be overridden. This patch came from
1043        Daniel Bergström.
1044    
1045    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
1046        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
1047        any errors with the current Unicode tables. Thanks to Peter Kankowski for
1048        spotting this.
1049    
1050    
1051    Version 7.6 28-Jan-08
1052    ---------------------
1053    
1054    1.  A character class containing a very large number of characters with
1055        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
1056        overflow.
1057    
1058    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
1059        HAVE_LONG_LONG is not defined.
1060    
1061    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
1062        bring them up to date and include new features. This patch includes:
1063    
1064        - Fixed PH's badly added libz and libbz2 support.
1065        - Fixed a problem with static linking.
1066        - Added pcredemo. [But later removed - see 7 below.]
1067        - Fixed dftables problem and added an option.
1068        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
1069            HAVE_LONG_LONG.
1070        - Added readline support for pcretest.
1071        - Added an listing of the option settings after cmake has run.
1072    
1073    4.  A user submitted a patch to Makefile that makes it easy to create
1074        "pcre.dll" under mingw when using Configure/Make. I added stuff to
1075        Makefile.am that cause it to include this special target, without
1076        affecting anything else. Note that the same mingw target plus all
1077        the other distribution libraries and programs are now supported
1078        when configuring with CMake (see 6 below) instead of with
1079        Configure/Make.
1080    
1081    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
1082        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
1083        exported in the Windows port". It has not yet been confirmed that the patch
1084        solves the problem, but it does no harm.
1085    
1086    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
1087        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
1088        with CMake, and also correct the comment about stack recursion.
1089    
1090    7.  Remove the automatic building of pcredemo from the ./configure system and
1091        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
1092        of a program that users should build themselves after PCRE is installed, so
1093        building it automatically is not really right. What is more, it gave
1094        trouble in some build environments.
1095    
1096    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
1097    
1098    
1099    Version 7.5 10-Jan-08
1100    ---------------------
1101    
1102    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
1103        values in parens when parsing an RE using the C++ wrapper."
1104    
1105    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
1106        Characters greater than 255 were excluded from the class instead of being
1107        included.
1108    
1109    3.  The same bug as (2) above applied to negated POSIX classes such as
1110        [:^space:].
1111    
1112    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
1113        defined or documented. It seems to have been a typo for PCRE_STATIC, so
1114        I have changed it.
1115    
1116    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
1117        first named subpattern) and a construct such as (?&a) would reference the
1118        first named subpattern whose name started with "a" (in other words, the
1119        length check was missing). Both these problems are fixed. "Subpattern name
1120        expected" is now given for (?&) (a zero-length name), and this patch also
1121        makes it give the same error for \k'' (previously it complained that that
1122        was a reference to a non-existent subpattern).
1123    
1124    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
1125        this is right because (?- can be followed by option settings as well as by
1126        digits. I have, however, made the messages clearer.
1127    
1128    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
1129        than the number used in the conditional) now cause a compile-time error.
1130        This is actually not compatible with Perl, which accepts such patterns, but
1131        treats the conditional as always being FALSE (as PCRE used to), but it
1132        seems to me that giving a diagnostic is better.
1133    
1134    8.  Change "alphameric" to the more common word "alphanumeric" in comments
1135        and messages.
1136    
1137    9.  Fix two occurrences of "backslash" in comments that should have been
1138        "backspace".
1139    
1140    10. Remove two redundant lines of code that can never be obeyed (their function
1141        was moved elsewhere).
1142    
1143    11. The program that makes PCRE's Unicode character property table had a bug
1144        which caused it to generate incorrect table entries for sequences of
1145        characters that have the same character type, but are in different scripts.
1146        It amalgamated them into a single range, with the script of the first of
1147        them. In other words, some characters were in the wrong script. There were
1148        thirteen such cases, affecting characters in the following ranges:
1149    
1150          U+002b0 - U+002c1
1151          U+0060c - U+0060d
1152          U+0061e - U+00612
1153          U+0064b - U+0065e
1154          U+0074d - U+0076d
1155          U+01800 - U+01805
1156          U+01d00 - U+01d77
1157          U+01d9b - U+01dbf
1158          U+0200b - U+0200f
1159          U+030fc - U+030fe
1160          U+03260 - U+0327f
1161          U+0fb46 - U+0fbb1
1162          U+10450 - U+1049d
1163    
1164    12. The -o option (show only the matching part of a line) for pcregrep was not
1165        compatible with GNU grep in that, if there was more than one match in a
1166        line, it showed only the first of them. It now behaves in the same way as
1167        GNU grep.
1168    
1169    13. If the -o and -v options were combined for pcregrep, it printed a blank
1170        line for every non-matching line. GNU grep prints nothing, and pcregrep now
1171        does the same. The return code can be used to tell if there were any
1172        non-matching lines.
1173    
1174    14. Added --file-offsets and --line-offsets to pcregrep.
1175    
1176    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
1177        infinitely looping recursion. The bug was that positive lookaheads were not
1178        being skipped when checking for a possible empty match (negative lookaheads
1179        and both kinds of lookbehind were skipped).
1180    
1181    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
1182        inclusion of <windows.h> to before rather than after the definition of
1183        INVALID_FILE_ATTRIBUTES (patch from David Byron).
1184    
1185    17. Specifying a possessive quantifier with a specific limit for a Unicode
1186        character property caused pcre_compile() to compile bad code, which led at
1187        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
1188        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
1189        caused the error; without that there was no problem.
1190    
1191    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
1192    
1193    19. Added --enable-pcretest-libreadline.
1194    
1195    20. In pcrecpp.cc, the variable 'count' was incremented twice in
1196        RE::GlobalReplace(). As a result, the number of replacements returned was
1197        double what it should be. I removed one of the increments, but Craig sent a
1198        later patch that removed the other one (the right fix) and added unit tests
1199        that check the return values (which was not done before).
1200    
1201    21. Several CMake things:
1202    
1203        (1) Arranged that, when cmake is used on Unix, the libraries end up with
1204            the names libpcre and libpcreposix, not just pcre and pcreposix.
1205    
1206        (2) The above change means that pcretest and pcregrep are now correctly
1207            linked with the newly-built libraries, not previously installed ones.
1208    
1209        (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
1210    
1211    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
1212        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
1213        UTF-8 newline character). The key issue is that the pattern starts .*;
1214        this means that the match must be either at the beginning, or after a
1215        newline. The bug was in the code for advancing after a failed match and
1216        checking that the new position followed a newline. It was not taking
1217        account of UTF-8 characters correctly.
1218    
1219    23. PCRE was behaving differently from Perl in the way it recognized POSIX
1220        character classes. PCRE was not treating the sequence [:...:] as a
1221        character class unless the ... were all letters. Perl, however, seems to
1222        allow any characters between [: and :], though of course it rejects as
1223        unknown any "names" that contain non-letters, because all the known class
1224        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
1225        for example, whereas PCRE did not - it did not recognize a POSIX character
1226        class. This seemed a bit dangerous, so the code has been changed to be
1227        closer to Perl. The behaviour is not identical to Perl, because PCRE will
1228        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
1229        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
1230        Perl does, and where it didn't before.
1231    
1232    24. Rewrite so as to remove the single use of %n from pcregrep because in some
1233        Windows environments %n is disabled by default.
1234    
1235    
1236    Version 7.4 21-Sep-07
1237    ---------------------
1238    
1239    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
1240        means that a class such as [\s] counted as "explicit reference to CR or
1241        LF". That isn't really right - the whole point of the change was to try to
1242        help when there was an actual mention of one of the two characters. So now
1243        the change happens only if \r or \n (or a literal CR or LF) character is
1244        encountered.
1245    
1246    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
1247        of both had grown to the point where there were only 3 bits left.
1248        Fortunately, there was spare space in the data structure, and so I have
1249        moved the internal flags into a new 16-bit field to free up more option
1250        bits.
1251    
1252    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
1253        but did not set the internal JCHANGED flag - either of these is enough to
1254        control the way the "get" function works - but the PCRE_INFO_JCHANGED
1255        facility is supposed to tell if (?J) was ever used, so now (?J) at the
1256        start sets both bits.
1257    
1258    4.  Added options (at build time, compile time, exec time) to change \R from
1259        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
1260    
1261    5.  doc/pcresyntax.html was missing from the distribution.
1262    
1263    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
1264        compatibility, even though it is no longer used.
1265    
1266    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
1267        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
1268        windows.h file is present (where different names are used). [This was
1269        reversed later after testing - see 16 below.]
1270    
1271    8.  Changed all #include <config.h> to #include "config.h". There were also
1272        some further <pcre.h> cases that I changed to "pcre.h".
1273    
1274    9.  When pcregrep was used with the --colour option, it missed the line ending
1275        sequence off the lines that it output.
1276    
1277    10. It was pointed out to me that arrays of string pointers cause lots of
1278        relocations when a shared library is dynamically loaded. A technique of
1279        using a single long string with a table of offsets can drastically reduce
1280        these. I have refactored PCRE in four places to do this. The result is
1281        dramatic:
1282    
1283          Originally:                          290
1284          After changing UCP table:            187
1285          After changing error message table:   43
1286          After changing table of "verbs"       36
1287          After changing table of Posix names   22
1288    
1289        Thanks to the folks working on Gregex for glib for this insight.
1290    
1291    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
1292        unicode-properties was also set.
1293    
1294    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
1295    
1296    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
1297        checked only for CRLF.
1298    
1299    14. Added casts to pcretest.c to avoid compiler warnings.
1300    
1301    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
1302    
1303    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
1304        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
1305        entirely. This removes changes made in 7 above.
1306    
1307    17. The CMake files have been updated, and there is now more information about
1308        building with CMake in the NON-UNIX-USE document.
1309    
1310    
1311    Version 7.3 28-Aug-07
1312    ---------------------
1313    
1314     1. In the rejigging of the build system that eventually resulted in 7.1, the
1315        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
1316        brackets there is not right, since it causes compilers to look for an
1317        installed pcre.h, not the version that is in the source that is being
1318        compiled (which of course may be different). I have changed it back to:
1319    
1320          #include "pcre.h"
1321    
1322        I have a vague recollection that the change was concerned with compiling in
1323        different directories, but in the new build system, that is taken care of
1324        by the VPATH setting the Makefile.
1325    
1326     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
1327        when the subject happened to end in the byte 0x85 (e.g. if the last
1328        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
1329        characters but of course it shouldn't be taken as a newline when it is part
1330        of another character. The bug was that, for an unlimited repeat of . in
1331        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
1332        characters when looking for a newline.
1333    
1334     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
1335    
1336     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
1337        in debug output.
1338    
1339     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
1340        long printing in the pcrecpp unittest when running under MinGW.
1341    
1342     6. ESC_K was left out of the EBCDIC table.
1343    
1344     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
1345        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
1346        limit also applies to "virtual nesting" when a pattern is recursive, and in
1347        this case 1000 isn't so big. I have been able to remove this limit at the
1348        expense of backing off one optimization in certain circumstances. Normally,
1349        when pcre_exec() would call its internal match() function recursively and
1350        immediately return the result unconditionally, it uses a "tail recursion"
1351        feature to save stack. However, when a subpattern that can match an empty
1352        string has an unlimited repetition quantifier, it no longer makes this
1353        optimization. That gives it a stack frame in which to save the data for
1354        checking that an empty string has been matched. Previously this was taken
1355        from the 1000-entry workspace that had been reserved. So now there is no
1356        explicit limit, but more stack is used.
1357    
1358     8. Applied Daniel's patches to solve problems with the import/export magic
1359        syntax that is required for Windows, and which was going wrong for the
1360        pcreposix and pcrecpp parts of the library. These were overlooked when this
1361        problem was solved for the main library.
1362    
1363     9. There were some crude static tests to avoid integer overflow when computing
1364        the size of patterns that contain repeated groups with explicit upper
1365        limits. As the maximum quantifier is 65535, the maximum group length was
1366        set at 30,000 so that the product of these two numbers did not overflow a
1367        32-bit integer. However, it turns out that people want to use groups that
1368        are longer than 30,000 bytes (though not repeat them that many times).
1369        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1370        made it possible to implement the integer overflow checks in a much more
1371        dynamic way, which I have now done. The artificial limitation on group
1372        length has been removed - we now have only the limit on the total length of
1373        the compiled pattern, which depends on the LINK_SIZE setting.
1374    
1375    10. Fixed a bug in the documentation for get/copy named substring when
1376        duplicate names are permitted. If none of the named substrings are set, the
1377        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1378        empty string.
1379    
1380    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1381        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1382        because the ] is interpreted as the first data character and the
1383        terminating ] is not found. PCRE has been made compatible with Perl in this
1384        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1385        cause memory overwriting.
1386    
1387    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1388        string has been matched (to stop an infinite loop). It was not recognizing
1389        a conditional subpattern that could match an empty string if that
1390        subpattern was within another subpattern. For example, it looped when
1391        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
1392        condition was not nested. This bug has been fixed.
1393    
1394    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1395        past the start of the subject in the presence of bytes with the top bit
1396        set, for example "\x8aBCD".
1397    
1398    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1399        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1400    
1401    14. Optimized (?!) to (*FAIL).
1402    
1403    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1404        This restricts code points to be within the range 0 to 0x10FFFF, excluding
1405        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1406        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1407        does: it's just the validity check that is more restrictive.
1408    
1409    16. Inserted checks for integer overflows during escape sequence (backslash)
1410        processing, and also fixed erroneous offset values for syntax errors during
1411        backslash processing.
1412    
1413    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1414        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1415    
1416    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1417        caused an overrun.
1418    
1419    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1420        something other than just ASCII characters) inside a group that had an
1421        unlimited repeat caused a loop at compile time (while checking to see
1422        whether the group could match an empty string).
1423    
1424    20. Debugging a pattern containing \p or \P could cause a crash. For example,
1425        [\P{Any}] did so. (Error in the code for printing property names.)
1426    
1427    21. An orphan \E inside a character class could cause a crash.
1428    
1429    22. A repeated capturing bracket such as (A)? could cause a wild memory
1430        reference during compilation.
1431    
1432    23. There are several functions in pcre_compile() that scan along a compiled
1433        expression for various reasons (e.g. to see if it's fixed length for look
1434        behind). There were bugs in these functions when a repeated \p or \P was
1435        present in the pattern. These operators have additional parameters compared
1436        with \d, etc, and these were not being taken into account when moving along
1437        the compiled data. Specifically:
1438    
1439        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1440            length.
1441    
1442        (b) An item such as \pL+ within a repeated group could cause crashes or
1443            loops.
1444    
1445        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1446            "reference to non-existent subpattern" error.
1447    
1448        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1449    
1450    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1451        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1452    
1453    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1454    
1455    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1456        character were causing crashes (broken optimization).
1457    
1458    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1459        \p or \P) caused a compile-time loop.
1460    
1461    28. More problems have arisen in unanchored patterns when CRLF is a valid line
1462        break. For example, the unstudied pattern [\r\n]A does not match the string
1463        "\r\nA" because change 7.0/46 below moves the current point on by two
1464        characters after failing to match at the start. However, the pattern \nA
1465        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
1466        the same is true. There doesn't seem any very clean way out of this, but
1467        what I have chosen to do makes the common cases work: PCRE now takes note
1468        of whether there can be an explicit match for \r or \n anywhere in the
1469        pattern, and if so, 7.0/46 no longer applies. As part of this change,
1470        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1471        pattern has explicit CR or LF references.
1472    
1473    29. Added (*CR) etc for changing newline setting at start of pattern.
1474    
1475    
1476    Version 7.2 19-Jun-07
1477    ---------------------
1478    
1479     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
1480        which is apparently normally available under Windows.
1481    
1482     2. Re-jig the pcregrep tests with different newline settings in an attempt
1483        to make them independent of the local environment's newline setting.
1484    
1485     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
1486    
1487     4. Some of the "internals" tests were previously cut out when the link size
1488        was not 2, because the output contained actual offsets. The recent new
1489        "Z" feature of pcretest means that these can be cut out, making the tests
1490        usable with all link sizes.
1491    
1492     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
1493        stack recursion. This gives a massive performance boost under BSD, but just
1494        a small improvement under Linux. However, it saves one field in the frame
1495        in all cases.
1496    
1497     6. Added more features from the forthcoming Perl 5.10:
1498    
1499        (a) (?-n) (where n is a string of digits) is a relative subroutine or
1500            recursion call. It refers to the nth most recently opened parentheses.
1501    
1502        (b) (?+n) is also a relative subroutine call; it refers to the nth next
1503            to be opened parentheses.
1504    
1505        (c) Conditions that refer to capturing parentheses can be specified
1506            relatively, for example, (?(-2)... or (?(+3)...
1507    
1508        (d) \K resets the start of the current match so that everything before
1509            is not part of it.
1510    
1511        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1512    
1513        (f) \g{name} is another synonym - part of Perl 5.10's unification of
1514            reference syntax.
1515    
1516        (g) (?| introduces a group in which the numbering of parentheses in each
1517            alternative starts with the same number.
1518    
1519        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
1520    
1521     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
1522        PCRE_INFO_JCHANGED.
1523    
1524     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
1525        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
1526        for detecting groups that can match an empty string.
1527    
1528     9. A pattern with a very large number of alternatives (more than several
1529        hundred) was running out of internal workspace during the pre-compile
1530        phase, where pcre_compile() figures out how much memory will be needed. A
1531        bit of new cunning has reduced the workspace needed for groups with
1532        alternatives. The 1000-alternative test pattern now uses 12 bytes of
1533        workspace instead of running out of the 4096 that are available.
1534    
1535    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
1536    
1537    11. Applied patch from Google to remove an optimization that didn't quite work.
1538        The report of the bug said:
1539    
1540          pcrecpp::RE("a*").FullMatch("aaa") matches, while
1541          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
1542          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
1543    
1544    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
1545        it matched the wrong number of bytes.
1546    
1547    
1548    Version 7.1 24-Apr-07
1549    ---------------------
1550    
1551     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
1552        that is more "standard", making use of automake and other Autotools. There
1553        is some re-arrangement of the files and adjustment of comments consequent
1554        on this.
1555    
1556     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
1557        for recursive directory scanning broke on some systems because the files
1558        are not scanned in any specific order and on different systems the order
1559        was different. A call to "sort" has been inserted into RunGrepTest for the
1560        approprate test as a short-term fix. In the longer term there may be an
1561        alternative.
1562    
1563     3. I had an email from Eric Raymond about problems translating some of PCRE's
1564        man pages to HTML (despite the fact that I distribute HTML pages, some
1565        people do their own conversions for various reasons). The problems
1566        concerned the use of low-level troff macros .br and .in. I have therefore
1567        removed all such uses from the man pages (some were redundant, some could
1568        be replaced by .nf/.fi pairs). The 132html script that I use to generate
1569        HTML has been updated to handle .nf/.fi and to complain if it encounters
1570        .br or .in.
1571    
1572     4. Updated comments in configure.ac that get placed in config.h.in and also
1573        arranged for config.h to be included in the distribution, with the name
1574        config.h.generic, for the benefit of those who have to compile without
1575        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
1576    
1577     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
1578        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
1579        makevp.bat for the current PCRE, using the additional files
1580        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
1581    
1582     6. A Windows user reported a minor discrepancy with test 2, which turned out
1583        to be caused by a trailing space on an input line that had got lost in his
1584        copy. The trailing space was an accident, so I've just removed it.
1585    
1586     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
1587        that is needed.
1588    
1589     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
1590        as "const" (a) because they are and (b) because it helps the PHP
1591        maintainers who have recently made a script to detect big data structures
1592        in the php code that should be moved to the .rodata section. I remembered
1593        to update Builducptable as well, so it won't revert if ucptable.h is ever
1594        re-created.
1595    
1596     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
1597        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
1598        order to be able to cut out the UTF-8 tables in the latter when UTF-8
1599        support is not required. This saves 1.5-2K of code, which is important in
1600        some applications.
1601    
1602        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
1603        so as not to refer to the tables, even though these functions will never be
1604        called when UTF-8 support is disabled. Otherwise there are problems with a
1605        shared library.
1606    
1607    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
1608    
1609        (a) It was defining its arguments as char * instead of void *.
1610    
1611        (b) It was assuming that all moves were upwards in memory; this was true
1612            a long time ago when I wrote it, but is no longer the case.
1613    
1614        The emulated memove() is provided for those environments that have neither
1615        memmove() nor bcopy(). I didn't think anyone used it these days, but that
1616        is clearly not the case, as these two bugs were recently reported.
1617    
1618    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
1619        and Detrail to create the HTML documentation, the .txt form of the man
1620        pages, and it removes trailing spaces from listed files. It also creates
1621        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
1622        case, it wraps all the #defines with #ifndefs. This script should be run
1623        before "make dist".
1624    
1625    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1626        with Unicode property support.
1627    
1628        (a) For a maximizing quantifier, if the two different cases of the
1629            character were of different lengths in their UTF-8 codings (there are
1630            some cases like this - I found 11), and the matching function had to
1631            back up over a mixture of the two cases, it incorrectly assumed they
1632            were both the same length.
1633    
1634        (b) When PCRE was configured to use the heap rather than the stack for
1635            recursion during matching, it was not correctly preserving the data for
1636            the other case of a UTF-8 character when checking ahead for a match
1637            while processing a minimizing repeat. If the check also involved
1638            matching a wide character, but failed, corruption could cause an
1639            erroneous result when trying to check for a repeat of the original
1640            character.
1641    
1642    13. Some tidying changes to the testing mechanism:
1643    
1644        (a) The RunTest script now detects the internal link size and whether there
1645            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1646            values substituted by "configure". (The RunGrepTest script already did
1647            this for UTF-8.) The configure.ac script no longer substitutes the
1648            relevant variables.
1649    
1650        (b) The debugging options /B and /D in pcretest show the compiled bytecode
1651            with length and offset values. This means that the output is different
1652            for different internal link sizes. Test 2 is skipped for link sizes
1653            other than 2 because of this, bypassing the problem. Unfortunately,
1654            there was also a test in test 3 (the locale tests) that used /B and
1655            failed for link sizes other than 2. Rather than cut the whole test out,
1656            I have added a new /Z option to pcretest that replaces the length and
1657            offset values with spaces. This is now used to make test 3 independent
1658            of link size. (Test 2 will be tidied up later.)
1659    
1660    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1661        segmentation fault instead of returning the appropriate error message.
1662    
1663    15. In multiline mode when the newline sequence was set to "any", the pattern
1664        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1665        This doesn't seem right; it now treats the CRLF combination as the line
1666        ending, and so does not match in that case. It's only a pattern such as ^$
1667        that would hit this one: something like ^ABC$ would have failed after \r
1668        and then tried again after \r\n.
1669    
1670    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1671        in an attempt to make files that differ only in their line terminators
1672        compare equal. This works on Linux.
1673    
1674    17. Under certain error circumstances pcregrep might try to free random memory
1675        as it exited. This is now fixed, thanks to valgrind.
1676    
1677    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1678        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1679        was because its rules for how to advance for /g after matching an empty
1680        string at the end of a line did not allow for this case. They now check for
1681        it specially.
1682    
1683    20. pcretest is supposed to handle patterns and data of any length, by
1684        extending its buffers when necessary. It was getting this wrong when the
1685        buffer for a data line had to be extended.
1686    
1687    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1688        CRLF as a newline sequence.
1689    
1690    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1691        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1692        I have nevertheless tidied it up.
1693    
1694    23. Added some casts to kill warnings from HP-UX ia64 compiler.
1695    
1696    24. Added a man page for pcre-config.
1697    
1698    
1699    Version 7.0 19-Dec-06
1700    ---------------------
1701    
1702     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1703        moving to gcc 4.1.1.
1704    
1705     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1706        sys/time.h, which is documented as needed for this function. It doesn't
1707        seem to matter on Linux, but it showed up on some releases of OS X.
1708    
1709     3. It seems that there are systems where bytes whose values are greater than
1710        127 match isprint() in the "C" locale. The "C" locale should be the
1711        default when a C program starts up. In most systems, only ASCII printing
1712        characters match isprint(). This difference caused the output from pcretest
1713        to vary, making some of the tests fail. I have changed pcretest so that:
1714    
1715        (a) When it is outputting text in the compiled version of a pattern, bytes
1716            other than 32-126 are always shown as hex escapes.
1717    
1718        (b) When it is outputting text that is a matched part of a subject string,
1719            it does the same, unless a different locale has been set for the match
1720            (using the /L modifier). In this case, it uses isprint() to decide.
1721    
1722     4. Fixed a major bug that caused incorrect computation of the amount of memory
1723        required for a compiled pattern when options that changed within the
1724        pattern affected the logic of the preliminary scan that determines the
1725        length. The relevant options are -x, and -i in UTF-8 mode. The result was
1726        that the computed length was too small. The symptoms of this bug were
1727        either the PCRE error "internal error: code overflow" from pcre_compile(),
1728        or a glibc crash with a message such as "pcretest: free(): invalid next
1729        size (fast)". Examples of patterns that provoked this bug (shown in
1730        pcretest format) are:
1731    
1732          /(?-x: )/x
1733          /(?x)(?-x: \s*#\s*)/
1734          /((?i)[\x{c0}])/8
1735          /(?i:[\x{c0}])/8
1736    
1737        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1738        is now done differently.
1739    
1740     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1741        wrapper classes; (b) implement a new function in the C++ scanner that is
1742        more efficient than the old way of doing things because it avoids levels of
1743        recursion in the regex matching; (c) add a paragraph to the documentation
1744        for the FullMatch() function.
1745    
1746     6. The escape sequence \n was being treated as whatever was defined as
1747        "newline". Not only was this contrary to the documentation, which states
1748        that \n is character 10 (hex 0A), but it also went horribly wrong when
1749        "newline" was defined as CRLF. This has been fixed.
1750    
1751     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1752        was being set to -1 for the "end of line" case (supposedly a value that no
1753        character can have). Though this value is never used (the check for end of
1754        line is "zero bytes in current character"), it caused compiler complaints.
1755        I've changed it to 0xffffffff.
1756    
1757     8. In pcre_version.c, the version string was being built by a sequence of
1758        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1759        string (as it is for production releases) called a macro with an empty
1760        argument. The C standard says the result of this is undefined. The gcc
1761        compiler treats it as an empty string (which was what was wanted) but it is
1762        reported that Visual C gives an error. The source has been hacked around to
1763        avoid this problem.
1764    
1765     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1766        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1767        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1768        of them did).
1769    
1770    10. Originally, pcretest opened its input and output without "b"; then I was
1771        told that "b" was needed in some environments, so it was added for release
1772        5.0 to both the input and output. (It makes no difference on Unix-like
1773        systems.) Later I was told that it is wrong for the input on Windows. I've
1774        now abstracted the modes into two macros, to make it easier to fiddle with
1775        them, and removed "b" from the input mode under Windows.
1776    
1777    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1778    
1779    12. Added -help and --help to pcretest as an official way of being reminded
1780        of the options.
1781    
1782    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1783        and pcrecpp.cc because they annoy compilers at high warning levels.
1784    
1785    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1786    
1787    15. Fixed an occurrence of == in configure.ac that should have been = (shell
1788        scripts are not C programs :-) and which was not noticed because it works
1789        on Linux.
1790    
1791    16. pcretest is supposed to handle any length of pattern and data line (as one
1792        line or as a continued sequence of lines) by extending its input buffer if
1793        necessary. This feature was broken for very long pattern lines, leading to
1794        a string of junk being passed to pcre_compile() if the pattern was longer
1795        than about 50K.
1796    
1797    17. I have done a major re-factoring of the way pcre_compile() computes the
1798        amount of memory needed for a compiled pattern. Previously, there was code
1799        that made a preliminary scan of the pattern in order to do this. That was
1800        OK when PCRE was new, but as the facilities have expanded, it has become
1801        harder and harder to keep it in step with the real compile phase, and there
1802        have been a number of bugs (see for example, 4 above). I have now found a
1803        cunning way of running the real compile function in a "fake" mode that
1804        enables it to compute how much memory it would need, while actually only
1805        ever using a few hundred bytes of working memory and without too many
1806        tests of the mode. This should make future maintenance and development
1807        easier. A side effect of this work is that the limit of 200 on the nesting
1808        depth of parentheses has been removed (though this was never a serious
1809        limitation, I suspect). However, there is a downside: pcre_compile() now
1810        runs more slowly than before (30% or more, depending on the pattern). I
1811        hope this isn't a big issue. There is no effect on runtime performance.
1812    
1813    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1814        newline (only possible for the last line of a file) and it was a
1815        pattern that set a locale (followed by /Lsomething), pcretest crashed.
1816    
1817    19. Added additional timing features to pcretest. (1) The -tm option now times
1818        matching only, not compiling. (2) Both -t and -tm can be followed, as a
1819        separate command line item, by a number that specifies the number of
1820        repeats to use when timing. The default is 50000; this gives better
1821        precision, but takes uncomfortably long for very large patterns.
1822    
1823    20. Extended pcre_study() to be more clever in cases where a branch of a
1824        subpattern has no definite first character. For example, (a*|b*)[cd] would
1825        previously give no result from pcre_study(). Now it recognizes that the
1826        first character must be a, b, c, or d.
1827    
1828    21. There was an incorrect error "recursive call could loop indefinitely" if
1829        a subpattern (or the entire pattern) that was being tested for matching an
1830        empty string contained only one non-empty item after a nested subpattern.
1831        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1832        incorrectly, because the \d was being skipped in the check.
1833    
1834    22. The pcretest program now has a new pattern option /B and a command line
1835        option -b, which is equivalent to adding /B to every pattern. This causes
1836        it to show the compiled bytecode, without the additional information that
1837        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1838        is the same as /B/I).
1839    
1840    23. A new optimization is now able automatically to treat some sequences such
1841        as a*b as a*+b. More specifically, if something simple (such as a character
1842        or a simple class like \d) has an unlimited quantifier, and is followed by
1843        something that cannot possibly match the quantified thing, the quantifier
1844        is automatically "possessified".
1845    
1846    24. A recursive reference to a subpattern whose number was greater than 39
1847        went wrong under certain circumstances in UTF-8 mode. This bug could also
1848        have affected the operation of pcre_study().
1849    
1850    25. Realized that a little bit of performance could be had by replacing
1851        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1852    
1853    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1854    
1855    27. Possessive quantifiers such as a++ were previously implemented by turning
1856        them into atomic groups such as ($>a+). Now they have their own opcodes,
1857        which improves performance. This includes the automatically created ones
1858        from 23 above.
1859    
1860    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1861        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1862        the first matched character to be a colon. This applied both to named and
1863        numbered groups.
1864    
1865    29. The ucpinternal.h header file was missing its idempotency #ifdef.
1866    
1867    30. I was sent a "project" file called libpcre.a.dev which I understand makes
1868        building PCRE on Windows easier, so I have included it in the distribution.
1869    
1870    31. There is now a check in pcretest against a ridiculously large number being
1871        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1872        loop, the loop is abandoned.
1873    
1874    32. Forward references to subpatterns in conditions such as (?(2)...) where
1875        subpattern 2 is defined later cause pcre_compile() to search forwards in
1876        the pattern for the relevant set of parentheses. This search went wrong
1877        when there were unescaped parentheses in a character class, parentheses
1878        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1879    
1880    33. "Subroutine" calls and backreferences were previously restricted to
1881        referencing subpatterns earlier in the regex. This restriction has now
1882        been removed.
1883    
1884    34. Added a number of extra features that are going to be in Perl 5.10. On the
1885        whole, these are just syntactic alternatives for features that PCRE had
1886        previously implemented using the Python syntax or my own invention. The
1887        other formats are all retained for compatibility.
1888    
1889        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1890            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1891            also .NET compatible.
1892    
1893        (b) A recursion or subroutine call to a named group can now be defined as
1894            (?&name) as well as (?P>name).
1895    
1896        (c) A backreference to a named group can now be defined as \k<name> or
1897            \k'name' as well as (?P=name). The new forms, as well as being in Perl
1898            5.10, are also .NET compatible.
1899    
1900        (d) A conditional reference to a named group can now use the syntax
1901            (?(<name>) or (?('name') as well as (?(name).
1902    
1903        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1904            groups (named and numbered) that are never evaluated inline, but can be
1905            called as "subroutines" from elsewhere. In effect, the DEFINE condition
1906            is always false. There may be only one alternative in such a group.
1907    
1908        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1909            as the simple (?(R). The condition is true only if the most recent
1910            recursion is that of the given number or name. It does not search out
1911            through the entire recursion stack.
1912    
1913        (g) The escape \gN or \g{N} has been added, where N is a positive or
1914            negative number, specifying an absolute or relative reference.
1915    
1916    35. Tidied to get rid of some further signed/unsigned compiler warnings and
1917        some "unreachable code" warnings.
1918    
1919    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1920        things, this adds five new scripts.
1921    
1922    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1923        There were also incompatibilities regarding the handling of \Q..\E inside
1924        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1925        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1926    
1927    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1928        matches an empty string, and forcibly breaks the loop. There were bugs in
1929        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
1930        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
1931        separate and independent bugs (that affected different cases) have been
1932        fixed.
1933    
1934    39. Refactored the code to abolish the use of different opcodes for small
1935        capturing bracket numbers. This is a tidy that I avoided doing when I
1936        removed the limit on the number of capturing brackets for 3.5 back in 2001.
1937        The new approach is not only tidier, it makes it possible to reduce the
1938        memory needed to fix the previous bug (38).
1939    
1940    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1941        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1942        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1943        mode.
1944    
1945    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1946        report.
1947    
1948    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1949        copy construction and assignment in the C++ wrapper.
1950    
1951    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1952        couple of bugs that could have given wrong results in the "--newline=crlf"
1953        case.
1954    
1955    44. Added a number of casts and did some reorganization of signed/unsigned int
1956        variables following suggestions from Dair Grant. Also renamed the variable
1957        "this" as "item" because it is a C++ keyword.
1958    
1959    45. Arranged for dftables to add
1960    
1961          #include "pcre_internal.h"
1962    
1963        to pcre_chartables.c because without it, gcc 4.x may remove the array
1964        definition from the final binary if PCRE is built into a static library and
1965        dead code stripping is activated.
1966    
1967    46. For an unanchored pattern, if a match attempt fails at the start of a
1968        newline sequence, and the newline setting is CRLF or ANY, and the next two
1969        characters are CRLF, advance by two characters instead of one.
1970    
1971    
1972    Version 6.7 04-Jul-06
1973    ---------------------
1974    
1975     1. In order to handle tests when input lines are enormously long, pcretest has
1976        been re-factored so that it automatically extends its buffers when
1977        necessary. The code is crude, but this _is_ just a test program. The
1978        default size has been increased from 32K to 50K.
1979    
1980     2. The code in pcre_study() was using the value of the re argument before
1981        testing it for NULL. (Of course, in any sensible call of the function, it
1982        won't be NULL.)
1983    
1984     3. The memmove() emulation function in pcre_internal.h, which is used on
1985        systems that lack both memmove() and bcopy() - that is, hardly ever -
1986        was missing a "static" storage class specifier.
1987    
1988     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1989        containing an extended class (one that cannot be represented by a bitmap
1990        because it contains high-valued characters or Unicode property items, e.g.
1991        [\pZ]). Almost always one would set UTF-8 mode when processing such a
1992        pattern, but PCRE should not loop if you do not (it no longer does).
1993        [Detail: two cases were found: (a) a repeated subpattern containing an
1994        extended class; (b) a recursive reference to a subpattern that followed a
1995        previous extended class. It wasn't skipping over the extended class
1996        correctly when UTF-8 mode was not set.]
1997    
1998     5. A negated single-character class was not being recognized as fixed-length
1999        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
2000        compile error "lookbehind assertion is not fixed length".
2001    
2002     6. The RunPerlTest auxiliary script was showing an unexpected difference
2003        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
2004        write a Perl script that can interpret lines of an input file either as
2005        byte characters or as UTF-8, which is what "perltest" was being required to
2006        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
2007        can't do is switch easily at run time between having the "use utf8;" pragma
2008        or not. In the end, I fudged it by using the RunPerlTest script to insert
2009        "use utf8;" explicitly for the UTF-8 tests.
2010    
2011     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
2012        the end of the subject string, contrary to the documentation and to what
2013        Perl does. This was true of both matching functions. Now it matches only at
2014        the start of the subject and immediately after *internal* newlines.
2015    
2016     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
2017        a pointer to an int instead of a pointer to an unsigned long int. This
2018        caused problems on 64-bit systems.
2019    
2020     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
2021        instance of the 'standard' template library not being so standard".
2022    
2023    10. There was no check on the number of named subpatterns nor the maximum
2024        length of a subpattern name. The product of these values is used to compute
2025        the size of the memory block for a compiled pattern. By supplying a very
2026        long subpattern name and a large number of named subpatterns, the size
2027        computation could be caused to overflow. This is now prevented by limiting
2028        the length of names to 32 characters, and the number of named subpatterns
2029        to 10,000.
2030    
2031    11. Subpatterns that are repeated with specific counts have to be replicated in
2032        the compiled pattern. The size of memory for this was computed from the
2033        length of the subpattern and the repeat count. The latter is limited to
2034        65535, but there was no limit on the former, meaning that integer overflow
2035        could in principle occur. The compiled length of a repeated subpattern is
2036        now limited to 30,000 bytes in order to prevent this.
2037    
2038    12. Added the optional facility to have named substrings with the same name.
2039    
2040    13. Added the ability to use a named substring as a condition, using the
2041        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
2042        are numbers (not recommended). Forward references are permitted.
2043    
2044    14. Added forward references in named backreferences (if you see what I mean).
2045    
2046    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
2047        pattern could run off the end of the subject. For example, the pattern
2048        "(?s)(.{1,5})"8 did this with the subject "ab".
2049    
2050    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
2051        PCRE_CASELESS was set when matching characters that were quantified with ?
2052        or *.
2053    
2054    17. A character class other than a single negated character that had a minimum
2055        but no maximum quantifier - for example [ab]{6,} - was not handled
2056        correctly by pce_dfa_exec(). It would match only one character.
2057    
2058    18. A valid (though odd) pattern that looked like a POSIX character
2059        class but used an invalid character after [ (for example [[,abc,]]) caused
2060        pcre_compile() to give the error "Failed: internal error: code overflow" or
2061        in some cases to crash with a glibc free() error. This could even happen if
2062        the pattern terminated after [[ but there just happened to be a sequence of
2063        letters, a binary zero, and a closing ] in the memory that followed.
2064    
2065    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
2066        over the years. Originally (before any Unicode support), just the bottom 8
2067        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
2068        output from "man perlunicode" includes this:
2069    
2070          The regular expression compiler produces polymorphic opcodes.  That
2071          is, the pattern adapts to the data and automatically switches to
2072          the Unicode character scheme when presented with Unicode data--or
2073          instead uses a traditional byte scheme when presented with byte
2074          data.
2075    
2076        Sadly, a wide octal escape does not cause a switch, and in a string with
2077        no other multibyte characters, these octal escapes are treated as before.
2078        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
2079        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
2080        Unicode string.
2081    
2082        I have not perpetrated such confusion in PCRE. Up till now, it took just
2083        the bottom 8 bits, as in old Perl. I have now made octal escapes with
2084        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
2085        translate to the appropriate multibyte character.
2086    
2087    29. Applied some refactoring to reduce the number of warnings from Microsoft
2088        and Borland compilers. This has included removing the fudge introduced
2089        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
2090        a warning about an unused variable.
2091    
2092    21. PCRE has not included VT (character 0x0b) in the set of whitespace
2093        characters since release 4.0, because Perl (from release 5.004) does not.
2094        [Or at least, is documented not to: some releases seem to be in conflict
2095        with the documentation.] However, when a pattern was studied with
2096        pcre_study() and all its branches started with \s, PCRE still included VT
2097        as a possible starting character. Of course, this did no harm; it just
2098        caused an unnecessary match attempt.
2099    
2100    22. Removed a now-redundant internal flag bit that recorded the fact that case
2101        dependency changed within the pattern. This was once needed for "required
2102        byte" processing, but is no longer used. This recovers a now-scarce options
2103        bit. Also moved the least significant internal flag bit to the most-
2104        significant bit of the word, which was not previously used (hangover from
2105        the days when it was an int rather than a uint) to free up another bit for
2106        the future.
2107    
2108    23. Added support for CRLF line endings as well as CR and LF. As well as the
2109        default being selectable at build time, it can now be changed at runtime
2110        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
2111        specify that it is scanning data with non-default line endings.
2112    
2113    24. Changed the definition of CXXLINK to make it agree with the definition of
2114        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
2115    
2116    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
2117        recursions. This makes a big different to stack usage for some patterns.
2118    
2119    26. If a subpattern containing a named recursion or subroutine reference such
2120        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
2121        the space required for the compiled pattern went wrong and gave too small a
2122        value. Depending on the environment, this could lead to "Failed: internal
2123        error: code overflow at offset 49" or "glibc detected double free or
2124        corruption" errors.
2125    
2126    27. Applied patches from Google (a) to support the new newline modes and (b) to
2127        advance over multibyte UTF-8 characters in GlobalReplace.
2128    
2129    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
2130        difference for some implementation of PCRE in some Windows version.
2131    
2132    29. Added some extra testing facilities to pcretest:
2133    
2134        \q<number>   in a data line sets the "match limit" value
2135        \Q<number>   in a data line sets the "match recursion limt" value
2136        -S <number>  sets the stack size, where <number> is in megabytes
2137    
2138        The -S option isn't available for Windows.
2139    
2140    
2141    Version 6.6 06-Feb-06
2142    ---------------------
2143    
2144     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
2145        in pcreposix.h. I have copied the definition from pcre.h.
2146    
2147     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
2148        because pcre.h is no longer a built file.
2149    
2150     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
2151        not normally included in the compiled code.
2152    
2153    
2154    Version 6.5 01-Feb-06
2155    ---------------------
2156    
2157     1. When using the partial match feature with pcre_dfa_exec(), it was not
2158        anchoring the second and subsequent partial matches at the new starting
2159        point. This could lead to incorrect results. For example, with the pattern
2160        /1234/, partially matching against "123" and then "a4" gave a match.
2161    
2162     2. Changes to pcregrep:
2163    
2164        (a) All non-match returns from pcre_exec() were being treated as failures
2165            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
2166            error message is output. Some extra information is given for the
2167            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
2168            probably the only errors that are likely to be caused by users (by
2169            specifying a regex that has nested indefinite repeats, for instance).
2170            If there are more than 20 of these errors, pcregrep is abandoned.
2171    
2172        (b) A binary zero was treated as data while matching, but terminated the
2173            output line if it was written out. This has been fixed: binary zeroes
2174            are now no different to any other data bytes.
2175    
2176        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
2177            used to set a locale for matching. The --locale=xxxx long option has
2178            been added (no short equivalent) to specify a locale explicitly on the
2179            pcregrep command, overriding the environment variables.
2180    
2181        (d) When -B was used with -n, some line numbers in the output were one less
2182            than they should have been.
2183    
2184        (e) Added the -o (--only-matching) option.
2185    
2186        (f) If -A or -C was used with -c (count only), some lines of context were
2187            accidentally printed for the final match.
2188    
2189        (g) Added the -H (--with-filename) option.
2190    
2191        (h) The combination of options -rh failed to suppress file names for files
2192            that were found from directory arguments.
2193    
2194        (i) Added the -D (--devices) and -d (--directories) options.
2195    
2196        (j) Added the -F (--fixed-strings) option.
2197    
2198        (k) Allow "-" to be used as a file name for -f as well as for a data file.
2199    
2200        (l) Added the --colo(u)r option.
2201    
2202        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
2203            is not present by default.
2204    
2205     3. A nasty bug was discovered in the handling of recursive patterns, that is,
2206        items such as (?R) or (?1), when the recursion could match a number of
2207        alternatives. If it matched one of the alternatives, but subsequently,
2208        outside the recursion, there was a failure, the code tried to back up into
2209        the recursion. However, because of the way PCRE is implemented, this is not
2210        possible, and the result was an incorrect result from the match.
2211    
2212        In order to prevent this happening, the specification of recursion has
2213        been changed so that all such subpatterns are automatically treated as
2214        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
2215    
2216     4. I had overlooked the fact that, in some locales, there are characters for
2217        which isalpha() is true but neither isupper() nor islower() are true. In
2218        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
2219        and ordfeminine) are like this. This affected the treatment of \w and \W
2220        when they appeared in character classes, but not when they appeared outside
2221        a character class. The bit map for "word" characters is now created
2222        separately from the results of isalnum() instead of just taking it from the
2223        upper, lower, and digit maps. (Plus the underscore character, of course.)
2224    
2225     5. The above bug also affected the handling of POSIX character classes such as
2226        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
2227        permanent tables. Instead, the bit maps for such a class were previously
2228        created as the appropriate unions of the upper, lower, and digit bitmaps.
2229        Now they are created by subtraction from the [[:word:]] class, which has
2230        its own bitmap.
2231    
2232     6. The [[:blank:]] character class matches horizontal, but not vertical space.
2233        It is created by subtracting the vertical space characters (\x09, \x0a,
2234        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
2235        subtraction was done in the overall bitmap for a character class, meaning
2236        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
2237        be recognized. This bug has been fixed.
2238    
2239     7. Patches from the folks at Google:
2240    
2241          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
2242          real life, but is still worth protecting against".
2243    
2244          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
2245          regular expressions".
2246    
2247          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
2248          have it.
2249    
2250          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
2251          "configure" and the latter not, in order to fix a problem somebody had
2252          with compiling the Arg class on HP-UX.
2253    
2254          (e) Improve the error-handling of the C++ wrapper a little bit.
2255    
2256          (f) New tests for checking recursion limiting.
2257    
2258     8. The pcre_memmove() function, which is used only if the environment does not
2259        have a standard memmove() function (and is therefore rarely compiled),
2260        contained two bugs: (a) use of int instead of size_t, and (b) it was not
2261        returning a result (though PCRE never actually uses the result).
2262    
2263     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
2264        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
2265        returned instead of calling malloc() with an overflowing number that would
2266        most likely cause subsequent chaos.
2267    
2268    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
2269    
2270    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
2271        with this option is matched, the nmatch and pmatch options of regexec() are
2272        ignored.
2273    
2274    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
2275        provided in case anyone wants to the the POSIX interface with UTF-8
2276        strings.
2277    
2278    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
2279        C++ linking (needed for some HP-UX environments).
2280    
2281    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
2282        (unused parameter) and in the pcre_printint() function (omitted "default"
2283        switch label when the default is to do nothing).
2284    
2285    15. Added some code to make it possible, when PCRE is compiled as a C++
2286        library, to replace subject pointers for pcre_exec() with a smart pointer
2287        class, thus making it possible to process discontinuous strings.
2288    
2289    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
2290        much the same function. They were added by different people who were trying
2291        to make PCRE easy to compile on non-Unix systems. It has been suggested
2292        that PCRE_EXPORT be abolished now that there is more automatic apparatus
2293        for compiling on Windows systems. I have therefore replaced it with
2294        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
2295        defaults to "extern" for C or "extern C" for C++, which works fine on
2296        Unix-like systems. It is now possible to override the value of PCRE_DATA_
2297        SCOPE with something explicit in config.h. In addition:
2298    
2299        (a) pcreposix.h still had just "extern" instead of either of these macros;
2300            I have replaced it with PCRE_DATA_SCOPE.
2301    
2302        (b) Functions such as _pcre_xclass(), which are internal to the library,
2303            but external in the C sense, all had PCRE_EXPORT in their definitions.
2304            This is apparently wrong for the Windows case, so I have removed it.
2305            (It makes no difference on Unix-like systems.)
2306    
2307    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
2308        of recursive calls to match(). This is different to MATCH_LIMIT because
2309        that limits the total number of calls to match(), not all of which increase
2310        the depth of recursion. Limiting the recursion depth limits the amount of
2311        stack (or heap if NO_RECURSE is set) that is used. The default can be set
2312        when PCRE is compiled, and changed at run time. A patch from Google adds
2313        this functionality to the C++ interface.
2314    
2315    18. Changes to the handling of Unicode character properties:
2316    
2317        (a) Updated the table to Unicode 4.1.0.
2318    
2319        (b) Recognize characters that are not in the table as "Cn" (undefined).
2320    
2321        (c) I revised the way the table is implemented to a much improved format
2322            which includes recognition of ranges. It now supports the ranges that
2323            are defined in UnicodeData.txt, and it also amalgamates other
2324            characters into ranges. This has reduced the number of entries in the
2325            table from around 16,000 to around 3,000, thus reducing its size
2326            considerably. I realized I did not need to use a tree structure after
2327            all - a binary chop search is just as efficient. Having reduced the
2328            number of entries, I extended their size from 6 bytes to 8 bytes to
2329            allow for more data.
2330    
2331        (d) Added support for Unicode script names via properties such as \p{Han}.
2332    
2333    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
2334        matching that character.
2335    
2336    20. When matching a repeated Unicode property with a minimum greater than zero,
2337        (for example \pL{2,}), PCRE could look past the end of the subject if it
2338        reached it while seeking the minimum number of characters. This could
2339        happen only if some of the characters were more than one byte long, because
2340        there is a check for at least the minimum number of bytes.
2341    
2342    21. Refactored the implementation of \p and \P so as to be more general, to
2343        allow for more different types of property in future. This has changed the
2344        compiled form incompatibly. Anybody with saved compiled patterns that use
2345        \p or \P will have to recompile them.
2346    
2347    22. Added "Any" and "L&" to the supported property types.
2348    
2349    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
2350        but give a compile time error if the value is greater than 0xff.
2351    
2352    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
2353        accidentally not being installed or uninstalled.
2354    
2355    25. The pcre.h file was built from pcre.h.in, but the only changes that were
2356        made were to insert the current release number. This seemed silly, because
2357        it made things harder for people building PCRE on systems that don't run
2358        "configure". I have turned pcre.h into a distributed file, no longer built
2359        by "configure", with the version identification directly included. There is
2360        no longer a pcre.h.in file.
2361    
2362        However, this change necessitated a change to the pcre-config script as
2363        well. It is built from pcre-config.in, and one of the substitutions was the
2364        release number. I have updated configure.ac so that ./configure now finds
2365        the release number by grepping pcre.h.
2366    
2367    26. Added the ability to run the tests under valgrind.
2368    
2369    
2370    Version 6.4 05-Sep-05
2371    ---------------------
2372    
2373     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
2374        "--" to be printed when multiple files were scanned, even when none of the
2375        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
2376        consider it to be a bug, and have restored the previous behaviour.
2377    
2378     2. A couple of code tidies to get rid of compiler warnings.
2379    
2380     3. The pcretest program used to cheat by referring to symbols in the library
2381        whose names begin with _pcre_. These are internal symbols that are not
2382        really supposed to be visible externally, and in some environments it is
2383        possible to suppress them. The cheating is now confined to including
2384        certain files from the library's source, which is a bit cleaner.
2385    
2386     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
2387        file's purpose clearer.
2388    
2389     5. Reorganized pcre_ucp_findchar().
2390    
2391    
2392    Version 6.3 15-Aug-05
2393    ---------------------
2394    
2395     1. The file libpcre.pc.in did not have general read permission in the tarball.
2396    
2397     2. There were some problems when building without C++ support:
2398    
2399        (a) If C++ support was not built, "make install" and "make test" still
2400            tried to test it.
2401    
2402        (b) There were problems when the value of CXX was explicitly set. Some
2403            changes have been made to try to fix these, and ...
2404    
2405        (c) --disable-cpp can now be used to explicitly disable C++ support.
2406    
2407        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
2408            backslash in a target when C++ was disabled. This confuses some
2409            versions of "make", apparently. Using an intermediate variable solves
2410            this. (Same for CPP_LOBJ.)
2411    
2412     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
2413        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
2414        necessary on certain architectures.
2415    
2416     4. Added a setting of -export-symbols-regex to the link command to remove
2417        those symbols that are exported in the C sense, but actually are local
2418        within the library, and not documented. Their names all begin with
2419        "_pcre_". This is not a perfect job, because (a) we have to except some
2420        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
2421        available (and never for static libraries). I have made a note to try to
2422        find a way round (a) in the future.
2423    
2424    
2425    Version 6.2 01-Aug-05
2426    ---------------------
2427    
2428     1. There was no test for integer overflow of quantifier values. A construction
2429        such as {1111111111111111} would give undefined results. What is worse, if
2430        a minimum quantifier for a parenthesized subpattern overflowed and became
2431        negative, the calculation of the memory size went wrong. This could have
2432        led to memory overwriting.
2433    
2434     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
2435    
2436     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
2437        operating environments where this matters.
2438    
2439     4. Applied Giuseppe Maxia's patch to add additional features for controlling
2440        PCRE options from within the C++ wrapper.
2441    
2442     5. Named capturing subpatterns were not being correctly counted when a pattern
2443        was compiled. This caused two problems: (a) If there were more than 100
2444        such subpatterns, the calculation of the memory needed for the whole
2445        compiled pattern went wrong, leading to an overflow error. (b) Numerical
2446        back references of the form \12, where the number was greater than 9, were
2447        not recognized as back references, even though there were sufficient
2448        previous subpatterns.
2449    
2450     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
2451        versions of gcc, e.g. 2.95.4.
2452    
2453    
2454    Version 6.1 21-Jun-05
2455    ---------------------
2456    
2457     1. There was one reference to the variable "posix" in pcretest.c that was not
2458        surrounded by "#if !defined NOPOSIX".
2459    
2460     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
2461        the cross-check on the old pcre_info() function, for the benefit of the
2462        cut-down version of PCRE that is currently imported into Exim.
2463    
2464     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
2465        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
2466        patterns that start (?i) but I don't think that matters. The use of (?i) is
2467        just an example; this all applies to the other options as well.
2468    
2469     4. Since libtool seems to echo the compile commands it is issuing, the output
2470        from "make" can be reduced a bit by putting "@" in front of each libtool
2471        compile command.
2472    
2473     5. Patch from the folks at Google for configure.in to be a bit more thorough
2474        in checking for a suitable C++ installation before trying to compile the
2475        C++ stuff. This should fix a reported problem when a compiler was present,
2476        but no suitable headers.
2477    
2478     6. The man pages all had just "PCRE" as their title. I have changed them to
2479        be the relevant file name. I have also arranged that these names are
2480        retained in the file doc/pcre.txt, which is a concatenation in text format
2481        of all the man pages except the little individual ones for each function.
2482    
2483     7. The NON-UNIX-USE file had not been updated for the different set of source
2484        files that come with release 6. I also added a few comments about the C++
2485        wrapper.
2486    
2487    
2488    Version 6.0 07-Jun-05
2489    ---------------------
2490    
2491     1. Some minor internal re-organization to help with my DFA experiments.
2492    
2493     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
2494        didn't matter for the library itself when fully configured, but did matter
2495        when compiling without UCP support, or within Exim, where the ucp files are
2496        not imported.
2497    
2498     3. Refactoring of the library code to split up the various functions into
2499        different source modules. The addition of the new DFA matching code (see
2500        below) to a single monolithic source would have made it really too
2501        unwieldy, quite apart from causing all the code to be include in a
2502        statically linked application, when only some functions are used. This is
2503        relevant even without the DFA addition now that patterns can be compiled in
2504        one application and matched in another.
2505    
2506        The downside of splitting up is that there have to be some external
2507        functions and data tables that are used internally in different modules of
2508        the library but which are not part of the API. These have all had their
2509        names changed to start with "_pcre_" so that they are unlikely to clash
2510        with other external names.
2511    
2512     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
2513        a different (DFA) algorithm. Although it is slower than the original
2514        function, it does have some advantages for certain types of matching
2515        problem.
2516    
2517     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
2518        including restarting after a partial match.
2519    
2520     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
2521        defined when compiling for Windows was sent to me. I have put it into the
2522        code, though I have no means of testing or verifying it.
2523    
2524     7. Added the pcre_refcount() auxiliary function.
2525    
2526     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
2527        match before or at the first newline in the subject string. In pcretest,
2528        the /f option on a pattern can be used to set this.
2529    
2530     9. A repeated \w when used in UTF-8 mode with characters greater than 256
2531        would behave wrongly. This has been present in PCRE since release 4.0.
2532    
2533    10. A number of changes to the pcregrep command:
2534    
2535        (a) Refactored how -x works; insert ^(...)$ instead of setting
2536            PCRE_ANCHORED and checking the length, in preparation for adding
2537            something similar for -w.
2538    
2539        (b) Added the -w (match as a word) option.
2540    
2541        (c) Refactored the way lines are read and buffered so as to have more
2542            than one at a time available.
2543    
2544        (d) Implemented a pcregrep test script.
2545    
2546        (e) Added the -M (multiline match) option. This allows patterns to match
2547            over several lines of the subject. The buffering ensures that at least
2548            8K, or the rest of the document (whichever is the shorter) is available
2549            for matching (and similarly the previous 8K for lookbehind assertions).
2550    
2551        (f) Changed the --help output so that it now says
2552    
2553              -w, --word-regex(p)
2554    
2555            instead of two lines, one with "regex" and the other with "regexp"
2556            because that confused at least one person since the short forms are the
2557            same. (This required a bit of code, as the output is generated
2558            automatically from a table. It wasn't just a text change.)
2559    
2560        (g) -- can be used to terminate pcregrep options if the next thing isn't an
2561            option but starts with a hyphen. Could be a pattern or a path name
2562            starting with a hyphen, for instance.
2563    
2564        (h) "-" can be given as a file name to represent stdin.
2565    
2566        (i) When file names are being printed, "(standard input)" is used for
2567            the standard input, for compatibility with GNU grep. Previously
2568            "<stdin>" was used.
2569    
2570        (j) The option --label=xxx can be used to supply a name to be used for
2571            stdin when file names are being printed. There is no short form.
2572    
2573        (k) Re-factored the options decoding logic because we are going to add
2574            two more options that take data. Such options can now be given in four
2575            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
2576    
2577        (l) Added the -A, -B, and -C options for requesting that lines of context
2578            around matches be printed.
2579    
2580        (m) Added the -L option to print the names of files that do not contain
2581            any matching lines, that is, the complement of -l.
2582    
2583        (n) The return code is 2 if any file cannot be opened, but pcregrep does
2584            continue to scan other files.
2585    
2586        (o) The -s option was incorrectly implemented. For compatibility with other
2587            greps, it now suppresses the error message for a non-existent or non-
2588            accessible file (but not the return code). There is a new option called
2589            -q that suppresses the output of matching lines, which was what -s was
2590            previously doing.
2591    
2592        (p) Added --include and --exclude options to specify files for inclusion
2593            and exclusion when recursing.
2594    
2595    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
2596        Hopefully, it now does.
2597    
2598    12. Missing cast in pcre_study().
2599    
2600    13. Added an "uninstall" target to the makefile.
2601    
2602    14. Replaced "extern" in the function prototypes in Makefile.in with
2603        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
2604        world, but is set differently for Windows.
2605    
2606    15. Added a second compiling function called pcre_compile2(). The only
2607        difference is that it has an extra argument, which is a pointer to an
2608        integer error code. When there is a compile-time failure, this is set
2609        non-zero, in addition to the error test pointer being set to point to an
2610        error message. The new argument may be NULL if no error number is required
2611        (but then you may as well call pcre_compile(), which is now just a
2612        wrapper). This facility is provided because some applications need a
2613        numeric error indication, but it has also enabled me to tidy up the way
2614        compile-time errors are handled in the POSIX wrapper.
2615    
2616    16. Added VPATH=.libs to the makefile; this should help when building with one
2617        prefix path and installing with another. (Or so I'm told by someone who
2618        knows more about this stuff than I do.)
2619    
2620    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
2621        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
2622        match everything, including newlines. This is not POSIX-compatible, but
2623        somebody wanted the feature. From pcretest it can be activated by using
2624        both the P and the s flags.
2625    
2626    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
2627    
2628    19. libpcre.pc was being incorrectly installed as executable.
2629    
2630    20. A couple of places in pcretest check for end-of-line by looking for '\n';
2631        it now also looks for '\r' so that it will work unmodified on Windows.
2632    
2633    21. Added Google's contributed C++ wrapper to the distribution.
2634    
2635    22. Added some untidy missing memory free() calls in pcretest, to keep
2636        Electric Fence happy when testing.
2637    
2638    
2639    
2640    Version 5.0 13-Sep-04
2641    ---------------------
2642    
2643     1. Internal change: literal characters are no longer packed up into items
2644        containing multiple characters in a single byte-string. Each character
2645        is now matched using a separate opcode. However, there may be more than one
2646        byte in the character in UTF-8 mode.
2647    
2648     2. The pcre_callout_block structure has two new fields: pattern_position and
2649        next_item_length. These contain the offset in the pattern to the next match
2650        item, and its length, respectively.
2651    
2652     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
2653        insertion of callouts before each pattern item. Added the /C option to
2654        pcretest to make use of this.
2655    
2656     4. On the advice of a Windows user, the lines
2657    
2658          #if defined(_WIN32) || defined(WIN32)
2659          _setmode( _fileno( stdout ), 0x8000 );
2660          #endif  /* defined(_WIN32) || defined(WIN32) */
2661    
2662        have been added to the source of pcretest. This apparently does useful
2663        magic in relation to line terminators.
2664    
2665     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
2666        for the benefit of those environments where the "b" makes a difference.
2667    
2668     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
2669        to know about it. I have put a hack into configure.in that adds in code
2670        to set GCC=yes if CC=icc. This seems to end up at a point in the
2671        generated configure script that is early enough to affect the setting of
2672        compiler options, which is what is needed, but I have no means of testing
2673        whether it really works. (The user who reported this had patched the
2674        generated configure script, which of course I cannot do.)
2675    
2676        LATER: After change 22 below (new libtool files), the configure script
2677        seems to know about icc (and also ecc). Therefore, I have commented out
2678        this hack in configure.in.
2679    
2680     7. Added support for pkg-config (2 patches were sent in).
2681    
2682     8. Negated POSIX character classes that used a combination of internal tables
2683        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2684        [[:^ascii]]. Typically, they would match almost any characters. The other
2685        POSIX classes were not broken in this way.
2686    
2687     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2688        to find the match, as PCRE was deluded into thinking that the match had to
2689        start at the start point or following a newline. The same bug applied to
2690        patterns with negative forward assertions or any backward assertions
2691        preceding ".*" at the start, unless the pattern required a fixed first
2692        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2693    
2694    10. In UTF-8 mode, when moving forwards in the subject after a failed match
2695        starting at the last subject character, bytes beyond the end of the subject
2696        string were read.
2697    
2698    11. Renamed the variable "class" as "classbits" to make life easier for C++
2699        users. (Previously there was a macro definition, but it apparently wasn't
2700        enough.)
2701    
2702    12. Added the new field "tables" to the extra data so that tables can be passed
2703        in at exec time, or the internal tables can be re-selected. This allows
2704        a compiled regex to be saved and re-used at a later time by a different
2705        program that might have everything at different addresses.
2706    
2707    13. Modified the pcre-config script so that, when run on Solaris, it shows a
2708        -R library as well as a -L library.
2709    
2710    14. The debugging options of pcretest (-d on the command line or D on a
2711        pattern) showed incorrect output for anything following an extended class
2712        that contained multibyte characters and which was followed by a quantifier.
2713    
2714    15. Added optional support for general category Unicode character properties
2715        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2716        support. It adds about 90K to the size of the library. The meanings of the
2717        inbuilt class escapes such as \d and \s have NOT been changed.
2718    
2719    16. Updated pcredemo.c to include calls to free() to release the memory for the
2720        compiled pattern.
2721    
2722    17. The generated file chartables.c was being created in the source directory
2723        instead of in the building directory. This caused the build to fail if the
2724        source directory was different from the building directory, and was
2725        read-only.
2726    
2727    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2728        file. No doubt somebody will tell me if they don't make sense... Also added
2729        Dan Mooney's comments about building on OpenVMS.
2730    
2731    19. Added support for partial matching via the PCRE_PARTIAL option for
2732        pcre_exec() and the \P data escape in pcretest.
2733    
2734    20. Extended pcretest with 3 new pattern features:
2735    
2736        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
2737              write the compiled pattern to the file whose name is "rest-of-line".
2738              This is a straight binary dump of the data, with the saved pointer to
2739              the character tables forced to be NULL. The study data, if any, is
2740              written too. After writing, pcretest reads a new pattern.
2741    
2742        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2743              compiled pattern from the given file. There must not be any
2744              occurrences of "<" in the file name (pretty unlikely); if there are,
2745              pcretest will instead treat the initial "<" as a pattern delimiter.
2746              After reading in the pattern, pcretest goes on to read data lines as
2747              usual.
2748    
2749        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2750              and 16-bit fields in a compiled pattern, to simulate a pattern that
2751              was compiled on a host of opposite endianness.
2752    
2753    21. The pcre-exec() function can now cope with patterns that were compiled on
2754        hosts of opposite endianness, with this restriction:
2755    
2756          As for any compiled expression that is saved and used later, the tables
2757          pointer field cannot be preserved; the extra_data field in the arguments
2758          to pcre_exec() should be used to pass in a tables address if a value
2759          other than the default internal tables were used at compile time.
2760    
2761    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2762        now diagnosed as an error. Previously, most of the time, a negative number
2763        would have been treated as zero, but if in addition "ovector" was passed as
2764        NULL, a crash could occur.
2765    
2766    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2767        new versions from the libtool 1.5 distribution (the last one is a copy of
2768        a file called libtool.m4). This seems to have fixed the need to patch
2769        "configure" to support Darwin 1.3 (which I used to do). However, I still
2770        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2771        workstation).
2772    
2773    24. Changed the PCRE licence to be the more standard "BSD" licence.
2774    
2775    
2776    Version 4.5 01-Dec-03
2777    ---------------------
2778    
2779     1. There has been some re-arrangement of the code for the match() function so
2780        that it can be compiled in a version that does not call itself recursively.
2781        Instead, it keeps those local variables that need separate instances for
2782        each "recursion" in a frame on the heap, and gets/frees frames whenever it
2783        needs to "recurse". Keeping track of where control must go is done by means
2784        of setjmp/longjmp. The whole thing is implemented by a set of macros that
2785        hide most of the details from the main code, and operates only if
2786        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2787        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2788        operating.
2789    
2790        To make it easier for callers to provide specially tailored get/free
2791        functions for this usage, two new functions, pcre_stack_malloc, and
2792        pcre_stack_free, are used. They are always called in strict stacking order,
2793        and the size of block requested is always the same.
2794    
2795        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2796        PCRE has been compiled to use the stack or the heap for recursion. The
2797        -C option of pcretest uses this to show which version is compiled.
2798    
2799        A new data escape \S, is added to pcretest; it causes the amounts of store
2800        obtained and freed by both kinds of malloc/free at match time to be added
2801        to the output.
2802    
2803     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2804        what's available on my current Linux desktop machine.
2805    
2806     3. When matching a UTF-8 string, the test for a valid string at the start has
2807        been extended. If start_offset is not zero, PCRE now checks that it points
2808        to a byte that is the start of a UTF-8 character. If not, it returns
2809        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2810        this is necessary because there may be backward assertions in the pattern.
2811        When matching the same subject several times, it may save resources to use
2812        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2813    
2814     4. The code for checking the validity of UTF-8 strings has been tightened so
2815        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2816        containing "overlong sequences".
2817    
2818     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2819        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2820        should have been "&", but it just so happened that all the cases this let
2821        through by mistake were picked up later in the function.
2822    
2823     6. I had used a variable called "isblank" - this is a C99 function, causing
2824        some compilers to warn. To avoid this, I renamed it (as "blankclass").
2825    
2826     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2827        prompting; (b) run "./pcretest /dev/null" at the start of the test script
2828        so the version is shown; (c) stop "make test" echoing "./RunTest".
2829    
2830     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2831    
2832     9. The prototype for memmove() for systems that don't have it was using
2833        size_t, but the inclusion of the header that defines size_t was later. I've
2834        moved the #includes for the C headers earlier to avoid this.
2835    
2836    10. Added some adjustments to the code to make it easier to compiler on certain
2837        special systems:
2838    
2839          (a) Some "const" qualifiers were missing.
2840          (b) Added the macro EXPORT before all exported functions; by default this
2841              is defined to be empty.
2842          (c) Changed the dftables auxiliary program (that builds chartables.c) so
2843              that it reads its output file name as an argument instead of writing
2844              to the standard output and assuming this can be redirected.
2845    
2846    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2847        class containing characters with values greater than 255, PCRE compilation
2848        went into a loop.
2849    
2850    12. A recursive reference to a subpattern that was within another subpattern
2851        that had a minimum quantifier of zero caused PCRE to crash. For example,
2852        (x(y(?2))z)? provoked this bug with a subject that got as far as the
2853        recursion. If the recursively-called subpattern itself had a zero repeat,
2854        that was OK.
2855    
2856    13. In pcretest, the buffer for reading a data line was set at 30K, but the
2857        buffer into which it was copied (for escape processing) was still set at
2858        1024, so long lines caused crashes.
2859    
2860    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2861        "internal error: code overflow...". This applied to any character class
2862        that was followed by a possessive quantifier.
2863    
2864    15. Modified the Makefile to add libpcre.la as a prerequisite for
2865        libpcreposix.la because I was told this is needed for a parallel build to
2866        work.
2867    
2868    16. If a pattern that contained .* following optional items at the start was
2869        studied, the wrong optimizing data was generated, leading to matching
2870        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2871        matching string must start with a or b or c. The correct conclusion for
2872        this pattern is that a match can start with any character.
2873    
2874    
2875    Version 4.4 13-Aug-03
2876    ---------------------
2877    
2878     1. In UTF-8 mode, a character class containing characters with values between
2879        127 and 255 was not handled correctly if the compiled pattern was studied.
2880        In fixing this, I have also improved the studying algorithm for such
2881        classes (slightly).
2882    
2883     2. Three internal functions had redundant arguments passed to them. Removal
2884        might give a very teeny performance improvement.
2885    
2886     3. Documentation bug: the value of the capture_top field in a callout is *one
2887        more than* the number of the hightest numbered captured substring.
2888    
2889     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2890        in incorrectly linking with a previously installed version. They now link
2891        explicitly with libpcre.la.
2892    
2893     5. configure.in no longer needs to recognize Cygwin specially.
2894    
2895     6. A problem in pcre.in for Windows platforms is fixed.
2896    
2897     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2898        pcretest, it used to include the size of the study block as part of its
2899        output. Unfortunately, the structure contains a field that has a different
2900        size on different hardware architectures. This meant that the tests that
2901        showed this size failed. As the block is currently always of a fixed size,
2902        this information isn't actually particularly useful in pcretest output, so
2903        I have just removed it.
2904    
2905     8. Three pre-processor statements accidentally did not start in column 1.
2906        Sadly, there are *still* compilers around that complain, even though
2907        standard C has not required this for well over a decade. Sigh.
2908    
2909     9. In pcretest, the code for checking callouts passed small integers in the
2910        callout_data field, which is a void * field. However, some picky compilers
2911        complained about the casts involved for this on 64-bit systems. Now
2912        pcretest passes the address of the small integer instead, which should get
2913        rid of the warnings.
2914    
2915    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2916        both compile and run time, and gives an error if an invalid UTF-8 sequence
2917        is found. There is a option for disabling this check in cases where the
2918        string is known to be correct and/or the maximum performance is wanted.
2919    
2920    11. In response to a bug report, I changed one line in Makefile.in from
2921    
2922            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2923        to
2924            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2925    
2926        to look similar to other lines, but I have no way of telling whether this
2927        is the right thing to do, as I do not use Windows. No doubt I'll get told
2928        if it's wrong...
2929    
2930    
2931    Version 4.3 21-May-03
2932    ---------------------
2933    
2934    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2935       Makefile.
2936    
2937    2. Some refactoring to improve the quality of the code:
2938    
2939       (i)   The utf8_table... variables are now declared "const".
2940    
2941       (ii)  The code for \cx, which used the "case flipping" table to upper case
2942             lower case letters, now just substracts 32. This is ASCII-specific,
2943             but the whole concept of \cx is ASCII-specific, so it seems
2944             reasonable.
2945    
2946       (iii) PCRE was using its character types table to recognize decimal and
2947             hexadecimal digits in the pattern. This is silly, because it handles
2948             only 0-9, a-f, and A-F, but the character types table is locale-
2949             specific, which means strange things might happen. A private
2950             table is now used for this - though it costs 256 bytes, a table is
2951             much faster than multiple explicit tests. Of course, the standard
2952             character types table is still used for matching digits in subject
2953             strings against \d.
2954    
2955       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2956             ending in _t are). So I've renamed it as ESC_tee.
2957    
2958    3. The first argument for regexec() in the POSIX wrapper should have been
2959       defined as "const".
2960    
2961    4. Changed pcretest to use malloc() for its buffers so that they can be
2962       Electric Fenced for debugging.
2963    
2964    5. There were several places in the code where, in UTF-8 mode, PCRE would try
2965       to read one or more bytes before the start of the subject string. Often this
2966       had no effect on PCRE's behaviour, but in some circumstances it could
2967       provoke a segmentation fault.
2968    
2969    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2970       to try to read one or more bytes before the start of the subject string.
2971    
2972    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
2973       UTF-8 support could misbehave in various ways if the subject string
2974       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
2975       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
2976       back over UTF-8 characters.)
2977    
2978    
2979    Version 4.2 14-Apr-03
2980    ---------------------
2981    
2982    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
2983    
2984    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
2985         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
2986         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
2987         [WIN_PREFIX]: new variable, "cyg" for Cygwin
2988         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
2989           and BUILD_EXEEXT
2990         Note: automatic setting of the BUILD variables is not yet working
2991         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
2992           compile-time but not at link-time
2993         [LINK]: use for linking executables only
2994         make different versions for Windows and non-Windows
2995         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
2996           libraries
2997         [LINK_FOR_BUILD]: new variable
2998         [OBJEXT]: use throughout
2999         [EXEEXT]: use throughout
3000         <winshared>: new target
3001         <wininstall>: new target
3002         <dftables.o>: use native compiler
3003         <dftables>: use native linker
3004         <install>: handle Windows platform correctly
3005         <clean>: ditto
3006         <check>: ditto
3007         copy DLL to top builddir before testing
3008    
3009       As part of these changes, -no-undefined was removed again. This was reported
3010       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
3011       in any case.
3012    
3013    3. Some tidies to get rid of compiler warnings:
3014    
3015       . In the match_data structure, match_limit was an unsigned long int, whereas
3016         match_call_count was an int. I've made them both unsigned long ints.
3017    
3018       . In pcretest the fact that a const uschar * doesn't automatically cast to
3019         a void * provoked a warning.
3020    
3021       . Turning on some more compiler warnings threw up some "shadow" variables
3022         and a few more missing casts.
3023    
3024    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
3025       option, a class that contained a single character with a value between 128
3026       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
3027    
3028    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
3029       option, a class that contained several characters, but with at least one
3030       whose value was between 128 and 255 caused PCRE to crash.
3031    
3032    
3033    Version 4.1 12-Mar-03
3034    ---------------------
3035    
3036    1. Compiling with gcc -pedantic found a couple of places where casts were
3037    needed, and a string in dftables.c that was longer than standard compilers are
3038    required to support.
3039    
3040    2. Compiling with Sun's compiler found a few more places where the code could
3041    be tidied up in order to avoid warnings.
3042    
3043    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
3044    first of these names is deprecated in the latest Autoconf in favour of the name
3045    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
3046    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
3047    analogy I have changed it to CFLAGS_FOR_BUILD.
3048    
3049    4. Added -no-undefined to the linking command in the Makefile, because this is
3050    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
3051    linking step for the pcreposix library.
3052    
3053    5. PCRE was failing to diagnose the case of two named groups with the same
3054    name.
3055    
3056    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
3057    literal character that is needed in the subject for a match, and scans along to
3058    ensure that it is present before embarking on the full matching process. This
3059    saves time in cases of nested unlimited repeats that are never going to match.
3060    Problem: the scan can take a lot of time if the subject is very long (e.g.
3061    megabytes), thus penalizing straightforward matches. It is now done only if the
3062    amount of subject to be scanned is less than 1000 bytes.
3063    
3064    7. A lesser problem with the same optimization is that it was recording the
3065    first character of an anchored pattern as "needed", thus provoking a search
3066    right along the subject, even when the first match of the pattern was going to
3067    fail. The "needed" character is now not set for anchored patterns, unless it
3068    follows something in the pattern that is of non-fixed length. Thus, it still
3069    fulfils its original purpose of finding quick non-matches in cases of nested
3070    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
3071    
3072    
3073    Version 4.0 17-Feb-03
3074    ---------------------
3075    
3076    1. If a comment in an extended regex that started immediately after a meta-item
3077    extended to the end of string, PCRE compiled incorrect data. This could lead to
3078    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
3079    
3080    2. Moved to autoconf 2.53 and libtool 1.4.2.
3081    
3082    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
3083    the special perltest8 script is no longer needed - all the tests can be run
3084    from a single perltest script.
3085    
3086    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
3087    by \s. It has now been removed in PCRE. This means it isn't recognized as
3088    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
3089    class [:space:] *does* include VT, thereby creating a mess.
3090    
3091    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
3092    space and tab.
3093    
3094    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
3095    its new features into the main test script, reducing the number of scripts.
3096    
3097    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
3098    were backward compatible, and made the (?i) apply to the whole pattern, as if
3099    /i were given. Now it behaves more logically, and applies the option setting
3100    only to what follows. PCRE has been changed to follow suit. However, if it
3101    finds options settings right at the start of the pattern, it extracts them into
3102    the global options, as before. Thus, they show up in the info data.
3103    
3104    8. Added support for the \Q...\E escape sequence. Characters in between are
3105    treated as literals. This is slightly different from Perl in that $ and @ are
3106    also handled as literals inside the quotes. In Perl, they will cause variable
3107    interpolation. Note the following examples:
3108    
3109        Pattern            PCRE matches      Perl matches
3110    
3111        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
3112        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
3113        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
3114    
3115    For compatibility with Perl, \Q...\E sequences are recognized inside character
3116    classes as well as outside them.
3117    
3118    9. Re-organized 3 code statements in pcretest to avoid "overflow in
3119    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
3120    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
3121    signed/unsigned warnings.
3122    
3123    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
3124    option for pcretest, so I've replaced it by a simple function that does just
3125    that job.
3126    
3127    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
3128    "pcregrep -".
3129    
3130    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
3131    Java package. This provides some syntactic sugar for simple cases of what my
3132    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
3133    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
3134    item, you can use this simplified notation. Note that only makes sense with
3135    greedy quantifiers. Consequently, the use of the possessive quantifier forces
3136    greediness, whatever the setting of the PCRE_UNGREEDY option.
3137    
3138    13. A change of greediness default within a pattern was not taking effect at
3139    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
3140    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
3141    was abstracted outside.
3142    
3143    14. PCRE now supports the \G assertion. It is true when the current matching
3144    position is at the start point of the match. This differs from \A when the
3145    starting offset is non-zero. Used with the /g option of pcretest (or similar
3146    code), it works in the same way as it does for Perl's /g option. If all
3147    alternatives of a regex begin with \G, the expression is anchored to the start
3148    match position, and the "anchored" flag is set in the compiled expression.
3149    
3150    15. Some bugs concerning the handling of certain option changes within patterns
3151    have been fixed. These applied to options other than (?ims). For example,
3152    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
3153    been the other way round. Some of this was related to change 7 above.
3154    
3155    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
3156    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
3157    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
3158    POSIX classes only within a class (e.g. /[[:alpha:]]/).
3159    
3160    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
3161    mode. Unlike ".", it always matches newline, whatever the setting of
3162    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
3163    assertions. Perl allows it, but it doesn't (in general) work because it can't
3164    calculate the length of the lookbehind. At least, that's the case for Perl
3165    5.8.0 - I've been told they are going to document that it doesn't work in
3166    future.
3167    
3168    18. Added an error diagnosis for escapes that PCRE does not support: these are
3169    \L, \l, \N, \P, \p, \U, \u, and \X.
3170    
3171    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
3172    reading past the end of the pattern in cases such as /[abcd/.
3173    
3174    20. PCRE was getting more memory than necessary for patterns with classes that
3175    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
3176    
3177    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
3178    compiling PCRE for use with Virtual Pascal.
3179    
3180    22. Small fix to the Makefile to make it work properly if the build is done
3181    outside the source tree.
3182    
3183    23. Added a new extension: a condition to go with recursion. If a conditional
3184    subpattern starts with (?(R) the "true" branch is used if recursion has
3185    happened, whereas the "false" branch is used only at the top level.
3186    
3187    24. When there was a very long string of literal characters (over 255 bytes
3188    without UTF support, over 250 bytes with UTF support), the computation of how
3189    much memory was required could be incorrect, leading to segfaults or other
3190    strange effects.
3191    
3192    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
3193    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
3194    there was a subsequent back reference to those brackets. This meant that, for
3195    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
3196    possible to check for precisely this case. All we can do is abandon the
3197    optimization if .* occurs inside capturing brackets when there are any back
3198    references whatsoever. (See below for a better fix that came later.)
3199    
3200    26. The handling of the optimization for finding the first character of a
3201    non-anchored pattern, and for finding a character that is required later in the
3202    match were failing in some cases. This didn't break the matching; it just
3203    failed to optimize when it could. The way this is done has been re-implemented.
3204    
3205    27. Fixed typo in error message for invalid (?R item (it said "(?p").
3206    
3207    28. Added a new feature that provides some of the functionality that Perl
3208    provides with (?{...}). The facility is termed a "callout". The way it is done
3209    in PCRE is for the caller to provide an optional function, by setting
3210    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
3211    global variable. By default it is unset, which disables all calling out. To get
3212    the function called, the regex must include (?C) at appropriate points. This
3213    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
3214    This provides a means of identifying different callout points. When PCRE
3215    reaches such a point in the regex, if pcre_callout has been set, the external
3216    function is called. It is provided with data in a structure called
3217    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
3218    matching continues; if it returns a non-zero value, the match at the current
3219    point fails. However, backtracking will occur if possible. [This was changed
3220    later and other features added - see item 49 below.]
3221    
3222    29. pcretest is upgraded to test the callout functionality. It provides a
3223    callout function that displays information. By default, it shows the start of
3224    the match and the current position in the text. There are some new data escapes
3225    to vary what happens:
3226    
3227        \C+         in addition, show current contents of captured substrings
3228        \C-         do not supply a callout function
3229        \C!n        return 1 when callout number n is reached
3230        \C!n!m      return 1 when callout number n is reached for the mth time
3231    
3232    30. If pcregrep was called with the -l option and just a single file name, it
3233    output "<stdin>" if a match was found, instead of the file name.
3234    
3235    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
3236    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
3237    pcre_exec(). This saves a malloc/free per call. The default value of
3238    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
3239    when configuring.
3240    
3241    32. The default maximum size of a compiled pattern is 64K. There have been a
3242    few cases of people hitting this limit. The code now uses macros to handle the
3243    storing of links as offsets within the compiled pattern. It defaults to 2-byte
3244    links, but this can be changed to 3 or 4 bytes by --with-link-size when
3245    configuring. Tests 2 and 5 work only with 2-byte links because they output
3246    debugging information about compiled patterns.
3247    
3248    33. Internal code re-arrangements:
3249    
3250    (a) Moved the debugging function for printing out a compiled regex into
3251        its own source file (printint.c) and used #include to pull it into
3252        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
3253        separate copies.
3254    
3255    (b) Defined the list of op-code names for debugging as a macro in
3256        internal.h so that it is next to the definition of the opcodes.
3257    
3258    (c) Defined a table of op-code lengths for simpler skipping along compiled
3259        code. This is again a macro in internal.h so that it is next to the
3260        definition of the opcodes.
3261    
3262    34. Added support for recursive calls to individual subpatterns, along the
3263    lines of Robin Houston's patch (but implemented somewhat differently).
3264    
3265    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
3266    allow it to read and process whole directories in Win32. This code was
3267    contributed by Lionel Fourquaux; it has not been tested by me.
3268    
3269    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
3270    used to name a group. Names consist of alphanumerics and underscores, and must
3271    be unique. Back references use the syntax (?P=name) and recursive calls use
3272    (?P>name) which is a PCRE extension to the Python extension. Groups still have
3273    numbers. The function pcre_fullinfo() can be used after compilation to extract
3274    a name/number map. There are three relevant calls:
3275    
3276      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
3277      PCRE_INFO_NAMECOUNT            yields the number of entries
3278      PCRE_INFO_NAMETABLE            yields a pointer to the map.
3279    
3280    The map is a vector of fixed-size entries. The size of each entry depends on
3281    the length of the longest name used. The first two bytes of each entry are the
3282    group number, most significant byte first. There follows the corresponding
3283    name, zero terminated. The names are in alphabetical order.
3284    
3285    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
3286    case instead of 255. Making it the same both with and without UTF-8 support
3287    means that the same test output works with both.
3288    
3289    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
3290    calling malloc() with a zero argument.
3291    
3292    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
3293    optimization. I've improved things by keeping a bitmap of backreferences with
3294    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
3295    fact referenced, the optimization can be applied. It is unlikely that a
3296    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
3297    the match to follow \n) will appear inside brackets with a number greater than
3298    31, but if it does, any back reference > 31 suppresses the optimization.
3299    
3300    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
3301    of disabling numbered capturing parentheses. Any opening parenthesis that is
3302    not followed by ? behaves as if it were followed by ?: but named parentheses
3303    can still be used for capturing (and they will acquire numbers in the usual
3304    way).
3305    
3306    41. Redesigned the return codes from the match() function into yes/no/error so
3307    that errors can be passed back from deep inside the nested calls. A malloc
3308    failure while inside a recursive subpattern call now causes the
3309    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
3310    
3311    42. It is now possible to set a limit on the number of times the match()
3312    function is called in a call to pcre_exec(). This facility makes it possible to
3313    limit the amount of recursion and backtracking, though not in a directly
3314    obvious way, because the match() function is used in a number of different
3315    circumstances. The count starts from zero for each position in the subject
3316    string (for non-anchored patterns). The default limit is, for compatibility, a
3317    large number, namely 10 000 000. You can change this in two ways:
3318    
3319    (a) When configuring PCRE before making, you can use --with-match-limit=n
3320        to set a default value for the compiled library.
3321    
3322    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
3323        a different value is set. See 45 below.
3324    
3325    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
3326    
3327    43. Added a new function pcre_config(int, void *) to enable run-time extraction
3328    of things that can be changed at compile time. The first argument specifies
3329    what is wanted and the second points to where the information is to be placed.
3330    The current list of available information is:
3331    
3332      PCRE_CONFIG_UTF8
3333    
3334    The output is an integer that is set to one if UTF-8 support is available;
3335    otherwise it is set to zero.
3336    
3337      PCRE_CONFIG_NEWLINE
3338    
3339    The output is an integer that it set to the value of the code that is used for
3340    newline. It is either LF (10) or CR (13).
3341    
3342      PCRE_CONFIG_LINK_SIZE
3343    
3344    The output is an integer that contains the number of bytes used for internal
3345    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
3346    
3347      PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
3348    
3349    The output is an integer that contains the threshold above which the POSIX
3350    interface uses malloc() for output vectors. See item 31 above.
3351    
3352      PCRE_CONFIG_MATCH_LIMIT
3353    
3354    The output is an unsigned integer that contains the default limit of the number
3355    of match() calls in a pcre_exec() execution. See 42 above.
3356    
3357    44. pcretest has been upgraded by the addition of the -C option. This causes it
3358    to extract all the available output from the new pcre_config() function, and to
3359    output it. The program then exits immediately.
3360    
3361    45. A need has arisen to pass over additional data with calls to pcre_exec() in
3362    order to support additional features. One way would have been to define
3363    pcre_exec2() (for example) with extra arguments, but this would not have been
3364    extensible, and would also have required all calls to the original function to
3365    be mapped to the new one. Instead, I have chosen to extend the mechanism that
3366    is used for passing in "extra" data from pcre_study().
3367    
3368    The pcre_extra structure is now exposed and defined in pcre.h. It currently
3369    contains the following fields:
3370    
3371      flags         a bitmap indicating which of the following fields are set
3372      study_data    opaque data from pcre_study()
3373      match_limit   a way of specifying a limit on match() calls for a specific
3374                      call to pcre_exec()
3375      callout_data  data for callouts (see 49 below)
3376    
3377    The flag bits are also defined in pcre.h, and are
3378    
3379      PCRE_EXTRA_STUDY_DATA
3380      PCRE_EXTRA_MATCH_LIMIT
3381      PCRE_EXTRA_CALLOUT_DATA
3382    
3383    The pcre_study() function now returns one of these new pcre_extra blocks, with
3384    the actual study data pointed to by the study_data field, and the
3385    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
3386    before. That is, this change is entirely upwards-compatible and requires no
3387    change to existing code.
3388    
3389    If you want to pass in additional data to pcre_exec(), you can either place it
3390    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
3391    block.
3392    
3393    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
3394    data string contains the escape sequence \M, pcretest calls pcre_exec() several
3395    times with different match limits, until it finds the minimum value needed for
3396    pcre_exec() to complete. The value is then output. This can be instructive; for
3397    most simple matches the number is quite small, but for pathological cases it
3398    gets very large very quickly.
3399    
3400    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
3401    returns the size of the data block pointed to by the study_data field in a
3402    pcre_extra block, that is, the value that was passed as the argument to
3403    pcre_malloc() when PCRE was getting memory in which to place the information
3404    created by pcre_study(). The fourth argument should point to a size_t variable.
3405    pcretest has been extended so that this information is shown after a successful
3406    pcre_study() call when information about the compiled regex is being displayed.
3407    
3408    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
3409    because what follows is always an absolute path. (Later: it turns out that this
3410    is more than cosmetic for MinGW, because it doesn't like empty path
3411    components.)
3412    
3413    49. Some changes have been made to the callout feature (see 28 above):
3414    
3415    (i)  A callout function now has three choices for what it returns:
3416    
3417           0  =>  success, carry on matching
3418         > 0  =>  failure at this point, but backtrack if possible
3419         < 0  =>  serious error, return this value from pcre_exec()
3420    
3421         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
3422         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
3423         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
3424         use by callout functions. It will never be used by PCRE itself.
3425    
3426    (ii) The pcre_extra structure (see 45 above) has a void * field called
3427         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
3428         pcre_callout_block structure has a field of the same name. The contents of
3429         the field passed in the pcre_extra structure are passed to the callout
3430         function in the corresponding field in the callout block. This makes it
3431         easier to use the same callout-containing regex from multiple threads. For
3432         testing, the pcretest program has a new data escape
3433    
3434           \C*n        pass the number n (may be negative) as callout_data
3435    
3436         If the callout function in pcretest receives a non-zero value as
3437         callout_data, it returns that value.
3438    
3439    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
3440    there were some redundant $(CFLAGS) in commands that are now specified as
3441    $(LINK), which already includes $(CFLAGS).
3442    
3443    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
3444    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
3445    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
3446    one-byte characters throughout. Note that case-insensitive matching applies
3447    only to characters whose values are less than 256. PCRE doesn't support the
3448    notion of cases for higher-valued characters.
3449    
3450    (i)   A character class whose characters are all within 0-255 is handled as
3451          a bit map, and the map is inverted for negative classes. Previously, a
3452          character > 255 always failed to match such a class; however it should
3453          match if the class was a negative one (e.g. [^ab]). This has been fixed.
3454    
3455    (ii)  A negated character class with a single character < 255 is coded as
3456          "not this character" (OP_NOT). This wasn't working properly when the test
3457          character was multibyte, either singly or repeated.
3458    
3459    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
3460          mode, for example: \x{100}{2,3}.
3461    
3462    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
3463          singly or repeated) now correctly test multibyte characters. However,
3464          PCRE doesn't recognize any characters with values greater than 255 as
3465          digits, spaces, or word characters. Such characters always match \D, \S,
3466          and \W, and never match \d, \s, or \w.
3467    
3468    (v)   Classes may now contain characters and character ranges with values
3469          greater than 255. For example: [ab\x{100}-\x{400}].
3470    
3471    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
3472          PCRE in UTF-8 mode.
3473    
3474    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
3475    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
3476    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
3477    value.)
3478    
3479    53. The single man page has become too large. I have therefore split it up into
3480    a number of separate man pages. These also give rise to individual HTML pages;
3481    these are now put in a separate directory, and there is an index.html page that
3482    lists them all. Some hyperlinking between the pages has been installed.
3483    
3484    54. Added convenience functions for handling named capturing parentheses.
3485    
3486    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
3487    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
3488    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
3489    are faulted.
3490    
3491    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
3492    calling configure. These values are used when compiling the dftables.c program
3493    which is run to generate the source of the default character tables. They
3494    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
3495    you will need to set these values.
3496    
3497    57. Updated the building process for Windows DLL, as provided by Fred Cox.
3498    
3499    
3500    Version 3.9 02-Jan-02
3501    ---------------------
3502    
3503    1. A bit of extraneous text had somehow crept into the pcregrep documentation.
3504    
3505    2. If --disable-static was given, the building process failed when trying to
3506    build pcretest and pcregrep. (For some reason it was using libtool to compile
3507    them, which is not right, as they aren't part of the library.)
3508    
3509    
3510    Version 3.8 18-Dec-01
3511    ---------------------
3512    
3513    1. The experimental UTF-8 code was completely screwed up. It was packing the
3514    bytes in the wrong order. How dumb can you get?
3515    
3516    
3517    Version 3.7 29-Oct-01
3518    ---------------------
3519    
3520    1. In updating pcretest to check change 1 of version 3.6, I screwed up.
3521    This caused pcretest, when used on the test data, to segfault. Unfortunately,
3522    this didn't happen under Solaris 8, where I normally test things.
3523    
3524    2. The Makefile had to be changed to make it work on BSD systems, where 'make'
3525    doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
3526    isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
3527    this fix an hour or so after the initial 3.7 release.)
3528    
3529    
3530    Version 3.6 23-Oct-01
3531    ---------------------
3532    
3533    1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
3534    offsets passed as NULL with zero offset count.
3535    
3536    2. The config.guess and config.sub files had not been updated when I moved to
3537    the latest autoconf.
3538    
3539    
3540    Version 3.5 15-Aug-01
3541    ---------------------
3542    
3543    1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
3544    had been forgotten.
3545    
3546    2. By using declared but undefined structures, we can avoid using "void"
3547    definitions in pcre.h while keeping the internal definitions of the structures
3548    private.
3549    
3550    3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
3551    user point of view, this means that both static and shared libraries are built
3552    by default, but this can be individually controlled. More of the work of
3553    handling this static/shared cases is now inside libtool instead of PCRE's make
3554    file.
3555    
3556    4. The pcretest utility is now installed along with pcregrep because it is
3557    useful for users (to test regexs) and by doing this, it automatically gets
3558    relinked by libtool. The documentation has been turned into a man page, so
3559    there are now .1, .txt, and .html versions in /doc.
3560    
3561    5. Upgrades to pcregrep:
3562       (i)   Added long-form option names like gnu grep.
3563       (ii)  Added --help to list all options with an explanatory phrase.
3564       (iii) Added -r, --recursive to recurse into sub-directories.
3565       (iv)  Added -f, --file to read patterns from a file.
3566    
3567    6. pcre_exec() was referring to its "code" argument before testing that
3568    argument for NULL (and giving an error if it was NULL).
3569    
3570    7. Upgraded Makefile.in to allow for compiling in a different directory from
3571    the source directory.
3572    
3573    8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
3574    options bits, the pointer it was passed was to an int instead of to an unsigned
3575    long int. This mattered only on 64-bit systems.
3576    
3577    9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
3578    generated) instead of pcre.in, which it its source. Also made the same change
3579    in several of the .c files.
3580    
3581    10. A new release of gcc defines printf() as a macro, which broke pcretest