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

Diff of /code/trunk/ChangeLog

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

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

Legend:
Removed from v.69  
changed lines
  Added in v.638

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12