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

Diff of /code/trunk/ChangeLog

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

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