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

Diff of /code/trunk/ChangeLog

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

revision 31 by nigel, Sat Feb 24 21:38:57 2007 UTC revision 637 by ph10, Sun Jul 24 17:44:12 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    
182    Version 8.12 15-Jan-2011
183    ------------------------
184    
185    1.  Fixed some typos in the markup of the man pages, and wrote a script that
186        checks for such things as part of the documentation building process.
187    
188    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
189        --match-limit and --recursion-limit options (added for 8.11). In
190        particular, this made one of the standard tests fail. (The integer value
191        went into the wrong half of a long int.)
192    
193    3.  If the --colour option was given to pcregrep with -v (invert match), it
194        did strange things, either producing crazy output, or crashing. It should,
195        of course, ignore a request for colour when reporting lines that do not
196        match.
197    
198    4.  Another pcregrep bug caused similar problems if --colour was specified with
199        -M (multiline) and the pattern match finished with a line ending.
200    
201    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
202        matched in multiline mode, the following line was shown as part of the
203        match. This seems wrong, so I have changed it.
204    
205    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
206        the check for further matches in the same line (so they could be coloured)
207        to overrun the end of the current line. If another match was found, it was
208        incorrectly shown (and then shown again when found in the next line).
209    
210    7.  If pcregrep was compiled under Windows, there was a reference to the
211        function pcregrep_exit() before it was defined. I am assuming this was
212        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
213        reported by a user. I've moved the definition above the reference.
214    
215    
216    Version 8.11 10-Dec-2010
217    ------------------------
218    
219    1.  (*THEN) was not working properly if there were untried alternatives prior
220        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
221        backtracked to try for "b" instead of moving to the next alternative branch
222        at the same level (in this case, to look for "c"). The Perl documentation
223        is clear that when (*THEN) is backtracked onto, it goes to the "next
224        alternative in the innermost enclosing group".
225    
226    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
227        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
228        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
229        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
230        (*THEN).
231    
232    3.  If \s appeared in a character class, it removed the VT character from
233        the class, even if it had been included by some previous item, for example
234        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
235        of \s, but is part of the POSIX "space" class.)
236    
237    4.  A partial match never returns an empty string (because you can always
238        match an empty string at the end of the subject); however the checking for
239        an empty string was starting at the "start of match" point. This has been
240        changed to the "earliest inspected character" point, because the returned
241        data for a partial match starts at this character. This means that, for
242        example, /(?<=abc)def/ gives a partial match for the subject "abc"
243        (previously it gave "no match").
244    
245    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
246        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
247        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
248        has an implication that the given string is incomplete (because a partial
249        match is preferred over a full match). For this reason, these items now
250        give a partial match in this situation. [Aside: previously, the one case
251        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
252        match rather than a full match, which was wrong by the old rules, but is
253        now correct.]
254    
255    6.  There was a bug in the handling of #-introduced comments, recognized when
256        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
257        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
258        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
259        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
260        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
261        places in pcre_compile().
262    
263    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
264        comments when looking ahead for named forward references to subpatterns,
265        the only newline sequence it recognized was NL. It now handles newlines
266        according to the set newline convention.
267    
268    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
269        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
270        cater for a lack of strerror(). These oversights have been fixed.
271    
272    9.  Added --match-limit and --recursion-limit to pcregrep.
273    
274    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
275    
276    11. When the -o option was used, pcregrep was setting a return code of 1, even
277        when matches were found, and --line-buffered was not being honoured.
278    
279    12. Added an optional parentheses number to the -o and --only-matching options
280        of pcregrep.
281    
282    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
283        can match an empty string. The code to do it in pcretest and pcredemo
284        needed fixing:
285    
286        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
287            only one byte after an empty string match just before CRLF (this case
288            just got forgotten; "any" and "anycrlf" were OK).
289    
290        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
291            mode when an empty string match preceded an ASCII character followed by
292            a non-ASCII character. (The code for advancing by one character rather
293            than one byte was nonsense.)
294    
295        (c) The pcredemo.c sample program did not have any code at all to handle
296            the cases when CRLF is a valid newline sequence.
297    
298    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
299        as a starting offset was within the subject string. There is now a new
300        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
301        negative or greater than the length of the string. In order to test this,
302        pcretest is extended to allow the setting of negative starting offsets.
303    
304    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
305        starting offset points to the beginning of a UTF-8 character was
306        unnecessarily clumsy. I tidied it up.
307    
308    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
309        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
310    
311    17. Nobody had reported that the --include_dir option, which was added in
312        release 7.7 should have been called --include-dir (hyphen, not underscore)
313        for compatibility with GNU grep. I have changed it to --include-dir, but
314        left --include_dir as an undocumented synonym, and the same for
315        --exclude-dir, though that is not available in GNU grep, at least as of
316        release 2.5.4.
317    
318    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
319        characters from a string of bytes) have been redefined so as not to use
320        loops, in order to improve performance in some environments. At the same
321        time, I abstracted some of the common code into auxiliary macros to save
322        repetition (this should not affect the compiled code).
323    
324    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
325        compile-time error is now given if \c is not followed by an ASCII
326        character, that is, a byte less than 128. (In EBCDIC mode, the code is
327        different, and any byte value is allowed.)
328    
329    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
330        START_OPTIMIZE option, which is now allowed at compile time - but just
331        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
332        to pcregrep and other applications that have no direct access to PCRE
333        options. The new /Y option in pcretest sets this option when calling
334        pcre_compile().
335    
336    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
337        back references. Groups containing recursive back references were forced to
338        be atomic by that change, but in the case of named groups, the amount of
339        memory required was incorrectly computed, leading to "Failed: internal
340        error: code overflow". This has been fixed.
341    
342    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
343        pcretest.c, to avoid build problems in some Borland environments.
344    
345    
346    Version 8.10 25-Jun-2010
347    ------------------------
348    
349    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
350        THEN.
351    
352    2.  (*ACCEPT) was not working when inside an atomic group.
353    
354    3.  Inside a character class, \B is treated as a literal by default, but
355        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
356        causes the error). The code is unchanged, but I tidied the documentation.
357    
358    4.  Inside a character class, PCRE always treated \R and \X as literals,
359        whereas Perl faults them if its -w option is set. I have changed PCRE so
360        that it faults them when PCRE_EXTRA is set.
361    
362    5.  Added support for \N, which always matches any character other than
363        newline. (It is the same as "." when PCRE_DOTALL is not set.)
364    
365    6.  When compiling pcregrep with newer versions of gcc which may have
366        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
367        declared with attribute warn_unused_result" were given. Just casting the
368        result to (void) does not stop the warnings; a more elaborate fudge is
369        needed. I've used a macro to implement this.
370    
371    7.  Minor change to pcretest.c to avoid a compiler warning.
372    
373    8.  Added four artifical Unicode properties to help with an option to make
374        \s etc use properties (see next item). The new properties are: Xan
375        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
376    
377    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
378        use Unicode properties. (*UCP) at the start of a pattern can be used to set
379        this option. Modified pcretest to add /W to test this facility. Added
380        REG_UCP to make it available via the POSIX interface.
381    
382    10. Added --line-buffered to pcregrep.
383    
384    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
385        studied, and the match started with a letter with a code point greater than
386        127 whose first byte was different to the first byte of the other case of
387        the letter, the other case of this starting letter was not recognized
388        (#976).
389    
390    12. If a pattern that was studied started with a repeated Unicode property
391        test, for example, \p{Nd}+, there was the theoretical possibility of
392        setting up an incorrect bitmap of starting bytes, but fortunately it could
393        not have actually happened in practice until change 8 above was made (it
394        added property types that matched character-matching opcodes).
395    
396    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
397        possible starting bytes for non-anchored patterns.
398    
399    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
400        \R, and also a number of cases that involve Unicode properties, both
401        explicit and implicit when PCRE_UCP is set.
402    
403    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
404        input, it could crash or give wrong results if characters with values
405        greater than 0xc0 were present in the subject string. (Detail: it assumed
406        UTF-8 input when processing these items.)
407    
408    16. Added a lot of (int) casts to avoid compiler warnings in systems where
409        size_t is 64-bit (#991).
410    
411    17. Added a check for running out of memory when PCRE is compiled with
412        --disable-stack-for-recursion (#990).
413    
414    18. If the last data line in a file for pcretest does not have a newline on
415        the end, a newline was missing in the output.
416    
417    19. The default pcre_chartables.c file recognizes only ASCII characters (values
418        less than 128) in its various bitmaps. However, there is a facility for
419        generating tables according to the current locale when PCRE is compiled. It
420        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
421        characters, are recognized by isspace() and therefore were getting set in
422        these tables, and indeed these tables seem to approximate to ISO 8859. This
423        caused a problem in UTF-8 mode when pcre_study() was used to create a list
424        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
425        which of course cannot start UTF-8 characters. I have changed the code so
426        that only real ASCII characters (less than 128) and the correct starting
427        bytes for UTF-8 encodings are set for characters greater than 127 when in
428        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
429        altogether.)
430    
431    20. Added the /T option to pcretest so as to be able to run tests with non-
432        standard character tables, thus making it possible to include the tests
433        used for 19 above in the standard set of tests.
434    
435    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
436        reference to a subpattern the other side of a comment that contains an
437        opening parenthesis caused either an internal compiling error, or a
438        reference to the wrong subpattern.
439    
440    
441    Version 8.02 19-Mar-2010
442    ------------------------
443    
444    1.  The Unicode data tables have been updated to Unicode 5.2.0.
445    
446    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
447        configured.
448    
449    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
450        original author of that file, following a query about its status.
451    
452    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
453        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
454    
455    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
456        quantifier applied to a forward-referencing subroutine call, could compile
457        incorrect code or give the error "internal error: previously-checked
458        referenced subpattern not found".
459    
460    6.  Both MS Visual Studio and Symbian OS have problems with initializing
461        variables to point to external functions. For these systems, therefore,
462        pcre_malloc etc. are now initialized to local functions that call the
463        relevant global functions.
464    
465    7.  There were two entries missing in the vectors called coptable and poptable
466        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
467        I've fixed the data, and added a kludgy way of testing at compile time that
468        the lengths are correct (equal to the number of opcodes).
469    
470    8.  Following on from 7, I added a similar kludge to check the length of the
471        eint vector in pcreposix.c.
472    
473    9.  Error texts for pcre_compile() are held as one long string to avoid too
474        much relocation at load time. To find a text, the string is searched,
475        counting zeros. There was no check for running off the end of the string,
476        which could happen if a new error number was added without updating the
477        string.
478    
479    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
480    
481    11. \K was not working if it appeared in an atomic group or in a group that
482        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
483        \K is "not well defined" if used in an assertion. PCRE now accepts it if
484        the assertion is positive, but not if it is negative.
485    
486    12. Change 11 fortuitously reduced the size of the stack frame used in the
487        "match()" function of pcre_exec.c by one pointer. Forthcoming
488        implementation of support for (*MARK) will need an extra pointer on the
489        stack; I have reserved it now, so that the stack frame size does not
490        decrease.
491    
492    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
493        item in branch that calls a recursion is a subroutine call - as in the
494        second branch in the above example - was incorrectly given the compile-
495        time error "recursive call could loop indefinitely" because pcre_compile()
496        was not correctly checking the subroutine for matching a non-empty string.
497    
498    14. The checks for overrunning compiling workspace could trigger after an
499        overrun had occurred. This is a "should never occur" error, but it can be
500        triggered by pathological patterns such as hundreds of nested parentheses.
501        The checks now trigger 100 bytes before the end of the workspace.
502    
503    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
504    
505    
506    Version 8.01 19-Jan-2010
507    ------------------------
508    
509    1.  If a pattern contained a conditional subpattern with only one branch (in
510        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
511        computed the wrong minimum data length (which is of course zero for such
512        subpatterns). This could cause incorrect "no match" results.
513    
514    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
515        the pattern is reset in the first branch, pcre_compile() failed with
516        "internal error: code overflow at offset...". This happened only when
517        the reset was to the original external option setting. (An optimization
518        abstracts leading options settings into an external setting, which was the
519        cause of this.)
520    
521    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
522        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
523        assertion pattern did not match (meaning that the assertion was true), it
524        was incorrectly treated as false if the SKIP had been reached during the
525        matching. This also applied to assertions used as conditions.
526    
527    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
528        assertion subpattern, including such a pattern used as a condition,
529        unpredictable results occurred, instead of the error return
530        PCRE_ERROR_DFA_UITEM.
531    
532    5.  The C++ GlobalReplace function was not working like Perl for the special
533        situation when an empty string is matched. It now does the fancy magic
534        stuff that is necessary.
535    
536    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
537        removed. (These were left over from very, very early versions of PCRE.)
538    
539    7.  Some cosmetic changes to the code to make life easier when compiling it
540        as part of something else:
541    
542        (a) Change DEBUG to PCRE_DEBUG.
543    
544        (b) In pcre_compile(), rename the member of the "branch_chain" structure
545            called "current" as "current_branch", to prevent a collision with the
546            Linux macro when compiled as a kernel module.
547    
548        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
549            prevent a collision with the Linux macro when compiled as a kernel
550            module.
551    
552    8.  In pcre_compile() there are some checks for integer overflows that used to
553        cast potentially large values to (double). This has been changed to that
554        when building, a check for int64_t is made, and if it is found, it is used
555        instead, thus avoiding the use of floating point arithmetic. (There is no
556        other use of FP in PCRE.) If int64_t is not found, the fallback is to
557        double.
558    
559    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
560        2005 (difference between two addresses compared to an unsigned value).
561    
562    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
563        custom one, because of the following reported problem in Windows:
564    
565          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
566              under Win32.
567          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
568              therefore missing the function definition.
569          - The compiler thus generates a "C" signature for the test function.
570          - The linker fails to find the "C" function.
571          - PCRE fails to configure if asked to do so against libbz2.
572    
573    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
574        messages were output:
575    
576          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
577          rerunning libtoolize, to keep the correct libtool macros in-tree.
578          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
579    
580        I have done both of these things.
581    
582    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
583        most of the time, it *can* run out if it is given a pattern that contains a
584        runaway infinite recursion. I updated the discussion in the pcrestack man
585        page.
586    
587    13. Now that we have gone to the x.xx style of version numbers, the minor
588        version may start with zero. Using 08 or 09 is a bad idea because users
589        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
590        interpreted as invalid octal numbers. I've updated the previous comment in
591        configure.ac, and also added a check that gives an error if 08 or 09 are
592        used.
593    
594    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
595        causing partial matching to fail when the end of the subject matched \W
596        in a UTF-8 pattern where \W was quantified with a minimum of 3.
597    
598    15. There were some discrepancies between the declarations in pcre_internal.h
599        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
600        their definitions. The declarations used "const uschar *" and the
601        definitions used USPTR. Even though USPTR is normally defined as "const
602        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
603        reported that: "This difference in casting confuses some C++ compilers, for
604        example, SunCC recognizes above declarations as different functions and
605        generates broken code for hbpcre." I have changed the declarations to use
606        USPTR.
607    
608    16. GNU libtool is named differently on some systems. The autogen.sh script now
609        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
610        (FreeBSD).
611    
612    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
613        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
614        comment: "Figure out how to create a longlong from a string: strtoll and
615        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
616        instance, but it only takes 2 args instead of 3!"
617    
618    18. A subtle bug concerned with back references has been fixed by a change of
619        specification, with a corresponding code fix. A pattern such as
620        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
621        refers, was giving matches when it shouldn't. For example, xa=xaaa would
622        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
623        same bug. Such groups have to be quantified to be useful, or contained
624        inside another quantified group. (If there's no repetition, the reference
625        can never match.) The problem arises because, having left the group and
626        moved on to the rest of the pattern, a later failure that backtracks into
627        the group uses the captured value from the final iteration of the group
628        rather than the correct earlier one. I have fixed this in PCRE by forcing
629        any group that contains a reference to itself to be an atomic group; that
630        is, there cannot be any backtracking into it once it has completed. This is
631        similar to recursive and subroutine calls.
632    
633    
634    Version 8.00 19-Oct-09
635    ----------------------
636    
637    1.  The table for translating pcre_compile() error codes into POSIX error codes
638        was out-of-date, and there was no check on the pcre_compile() error code
639        being within the table. This could lead to an OK return being given in
640        error.
641    
642    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
643        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
644        in a Windows environment.
645    
646    3.  The pcregrep --count option prints the count for each file even when it is
647        zero, as does GNU grep. However, pcregrep was also printing all files when
648        --files-with-matches was added. Now, when both options are given, it prints
649        counts only for those files that have at least one match. (GNU grep just
650        prints the file name in this circumstance, but including the count seems
651        more useful - otherwise, why use --count?) Also ensured that the
652        combination -clh just lists non-zero counts, with no names.
653    
654    4.  The long form of the pcregrep -F option was incorrectly implemented as
655        --fixed_strings instead of --fixed-strings. This is an incompatible change,
656        but it seems right to fix it, and I didn't think it was worth preserving
657        the old behaviour.
658    
659    5.  The command line items --regex=pattern and --regexp=pattern were not
660        recognized by pcregrep, which required --regex pattern or --regexp pattern
661        (with a space rather than an '='). The man page documented the '=' forms,
662        which are compatible with GNU grep; these now work.
663    
664    6.  No libpcreposix.pc file was created for pkg-config; there was just
665        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
666    
667    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
668        when UCP support is not needed, by modifying the Python script that
669        generates it from Unicode data files. This should not matter if the module
670        is correctly used as a library, but I received one complaint about 50K of
671        unwanted data. My guess is that the person linked everything into his
672        program rather than using a library. Anyway, it does no harm.
673    
674    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
675        was a minimum greater than 1 for a wide character in a possessive
676        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
677        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
678        character. Chaos in the form of incorrect output or a compiling loop could
679        result.
680    
681    9.  The restrictions on what a pattern can contain when partial matching is
682        requested for pcre_exec() have been removed. All patterns can now be
683        partially matched by this function. In addition, if there are at least two
684        slots in the offset vector, the offset of the earliest inspected character
685        for the match and the offset of the end of the subject are set in them when
686        PCRE_ERROR_PARTIAL is returned.
687    
688    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
689        synonymous with PCRE_PARTIAL, for backwards compatibility, and
690        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
691        and may be more useful for multi-segment matching.
692    
693    11. Partial matching with pcre_exec() is now more intuitive. A partial match
694        used to be given if ever the end of the subject was reached; now it is
695        given only if matching could not proceed because another character was
696        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
697        string "Z", which now yields "no match" instead of "partial match". In the
698        case of pcre_dfa_exec(), "no match" is given if every matching path for the
699        final character ended with (*FAIL).
700    
701    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
702        if the pattern had a "must contain" character that was already found in the
703        earlier partial match, unless partial matching was again requested. For
704        example, with the pattern /dog.(body)?/, the "must contain" character is
705        "g". If the first part-match was for the string "dog", restarting with
706        "sbody" failed. This bug has been fixed.
707    
708    13. The string returned by pcre_dfa_exec() after a partial match has been
709        changed so that it starts at the first inspected character rather than the
710        first character of the match. This makes a difference only if the pattern
711        starts with a lookbehind assertion or \b or \B (\K is not supported by
712        pcre_dfa_exec()). It's an incompatible change, but it makes the two
713        matching functions compatible, and I think it's the right thing to do.
714    
715    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
716        so that the demonstration program is easily available in environments where
717        PCRE has not been installed from source.
718    
719    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
720        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
721        library.
722    
723    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
724        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
725        is not the first non-POSIX option to be added. Clearly some people find
726        these options useful.
727    
728    17. If a caller to the POSIX matching function regexec() passes a non-zero
729        value for nmatch with a NULL value for pmatch, the value of
730        nmatch is forced to zero.
731    
732    18. RunGrepTest did not have a test for the availability of the -u option of
733        the diff command, as RunTest does. It now checks in the same way as
734        RunTest, and also checks for the -b option.
735    
736    19. If an odd number of negated classes containing just a single character
737        interposed, within parentheses, between a forward reference to a named
738        subpattern and the definition of the subpattern, compilation crashed with
739        an internal error, complaining that it could not find the referenced
740        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
741        [The bug was that it was starting one character too far in when skipping
742        over the character class, thus treating the ] as data rather than
743        terminating the class. This meant it could skip too much.]
744    
745    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
746        /g option in pcretest when the pattern contains \K, which makes it possible
747        to have an empty string match not at the start, even when the pattern is
748        anchored. Updated pcretest and pcredemo to use this option.
749    
750    21. If the maximum number of capturing subpatterns in a recursion was greater
751        than the maximum at the outer level, the higher number was returned, but
752        with unset values at the outer level. The correct (outer level) value is
753        now given.
754    
755    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
756        PCRE did not set those parentheses (unlike Perl). I have now found a way to
757        make it do so. The string so far is captured, making this feature
758        compatible with Perl.
759    
760    23. The tests have been re-organized, adding tests 11 and 12, to make it
761        possible to check the Perl 5.10 features against Perl 5.10.
762    
763    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
764        pattern matches a fixed length string. PCRE did not allow this; now it
765        does. Neither allows recursion.
766    
767    25. I finally figured out how to implement a request to provide the minimum
768        length of subject string that was needed in order to match a given pattern.
769        (It was back references and recursion that I had previously got hung up
770        on.) This code has now been added to pcre_study(); it finds a lower bound
771        to the length of subject needed. It is not necessarily the greatest lower
772        bound, but using it to avoid searching strings that are too short does give
773        some useful speed-ups. The value is available to calling programs via
774        pcre_fullinfo().
775    
776    26. While implementing 25, I discovered to my embarrassment that pcretest had
777        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
778        study optimizations had never been tested with that matching function.
779        Oops. What is worse, even when it was passed study data, there was a bug in
780        pcre_dfa_exec() that meant it never actually used it. Double oops. There
781        were also very few tests of studied patterns with pcre_dfa_exec().
782    
783    27. If (?| is used to create subpatterns with duplicate numbers, they are now
784        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
785        on the other side of the coin, they are no longer allowed to have different
786        names, because these cannot be distinguished in PCRE, and this has caused
787        confusion. (This is a difference from Perl.)
788    
789    28. When duplicate subpattern names are present (necessarily with different
790        numbers, as required by 27 above), and a test is made by name in a
791        conditional pattern, either for a subpattern having been matched, or for
792        recursion in such a pattern, all the associated numbered subpatterns are
793        tested, and the overall condition is true if the condition is true for any
794        one of them. This is the way Perl works, and is also more like the way
795        testing by number works.
796    
797    
798    Version 7.9 11-Apr-09
799    ---------------------
800    
801    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
802        (pcretest), all targets were linked against these libraries. This included
803        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
804        libraries. This caused unwanted dependencies to be created. This problem
805        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
806        pcretest is linked with readline.
807    
808    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
809        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
810        moved outside it again, because FALSE and TRUE are already defined in AIX,
811        but BOOL is not.
812    
813    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
814        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
815    
816    4.  The pcregrep documentation said spaces were inserted as well as colons (or
817        hyphens) following file names and line numbers when outputting matching
818        lines. This is not true; no spaces are inserted. I have also clarified the
819        wording for the --colour (or --color) option.
820    
821    5.  In pcregrep, when --colour was used with -o, the list of matching strings
822        was not coloured; this is different to GNU grep, so I have changed it to be
823        the same.
824    
825    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
826        each matching line was coloured. Now it goes on to look for further matches
827        of any of the test patterns, which is the same behaviour as GNU grep.
828    
829    7.  A pattern that could match an empty string could cause pcregrep to loop; it
830        doesn't make sense to accept an empty string match in pcregrep, so I have
831        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
832        seems to be how GNU grep behaves.
833    
834    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
835        start or after a newline", because the conditional assertion was not being
836        correctly handled. The rule now is that both the assertion and what follows
837        in the first alternative must satisfy the test.
838    
839    9.  If auto-callout was enabled in a pattern with a conditional group whose
840        condition was an assertion, PCRE could crash during matching, both with
841        pcre_exec() and pcre_dfa_exec().
842    
843    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
844        used for matching.
845    
846    11. Unicode property support in character classes was not working for
847        characters (bytes) greater than 127 when not in UTF-8 mode.
848    
849    12. Added the -M command line option to pcretest.
850    
851    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
852    
853    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
854    
855    16. Added comments and documentation about mis-use of no_arg in the C++
856        wrapper.
857    
858    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
859        from Martin Jerabek that uses macro names for all relevant character and
860        string constants.
861    
862    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
863        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
864        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
865        these, but not everybody uses configure.
866    
867    19. A conditional group that had only one branch was not being correctly
868        recognized as an item that could match an empty string. This meant that an
869        enclosing group might also not be so recognized, causing infinite looping
870        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
871        with the subject "ab", where knowledge that the repeated group can match
872        nothing is needed in order to break the loop.
873    
874    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
875        exec(), but without supplying a callout function, matching went wrong.
876    
877    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
878        leak if the size of the offset vector was greater than 30. When the vector
879        is smaller, the saved offsets during recursion go onto a local stack
880        vector, but for larger vectors malloc() is used. It was failing to free
881        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
882        error, in fact).
883    
884    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
885        heapframe that is used only when UTF-8 support is enabled. This caused no
886        problem, but was untidy.
887    
888    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
889        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
890        included within another project.
891    
892    24. Steven Van Ingelgem's patches to add more options to the CMake support,
893        slightly modified by me:
894    
895          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
896              not building pcregrep.
897    
898          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
899              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
900    
901    25. Forward references, both numeric and by name, in patterns that made use of
902        duplicate group numbers, could behave incorrectly or give incorrect errors,
903        because when scanning forward to find the reference group, PCRE was not
904        taking into account the duplicate group numbers. A pattern such as
905        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
906    
907    26. Changed a few more instances of "const unsigned char *" to USPTR, making
908        the feature of a custom pointer more persuasive (as requested by a user).
909    
910    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
911        pcretest.c, inside #ifndefs, because it seems they are sometimes already
912        pre-defined.
913    
914    28. Added support for (*UTF8) at the start of a pattern.
915    
916    29. Arrange for flags added by the "release type" setting in CMake to be shown
917        in the configuration summary.
918    
919    
920    Version 7.8 05-Sep-08
921    ---------------------
922    
923    1.  Replaced UCP searching code with optimized version as implemented for Ad
924        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
925        stage table and inline lookup instead of a function, giving speed ups of 2
926        to 5 times on some simple patterns that I tested. Permission was given to
927        distribute the MultiStage2.py script that generates the tables (it's not in
928        the tarball, but is in the Subversion repository).
929    
930    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
931        scripts.
932    
933    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
934        a group with a zero qualifier. The result of the study could be incorrect,
935        or the function might crash, depending on the pattern.
936    
937    4.  Caseless matching was not working for non-ASCII characters in back
938        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
939        It now works when Unicode Property Support is available.
940    
941    5.  In pcretest, an escape such as \x{de} in the data was always generating
942        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
943        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
944        truncation.
945    
946    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
947    
948    7.  Added two (int) casts to pcregrep when printing the difference of two
949        pointers, in case they are 64-bit values.
950    
951    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
952        test 2 if it fails.
953    
954    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
955        and a #define of that name to empty if it is not externally set. This is to
956        allow users of MSVC to set it if necessary.
957    
958    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
959        the convenience functions in the pcre_get.c source file.
960    
961    11. An option change at the start of a pattern that had top-level alternatives
962        could cause overwriting and/or a crash. This command provoked a crash in
963        some environments:
964    
965          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
966    
967        This potential security problem was recorded as CVE-2008-2371.
968    
969    12. For a pattern where the match had to start at the beginning or immediately
970        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
971        pcre_dfa_exec() could read past the end of the passed subject if there was
972        no match. To help with detecting such bugs (e.g. with valgrind), I modified
973        pcretest so that it places the subject at the end of its malloc-ed buffer.
974    
975    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
976        exec() might read past the end of the data buffer in UTF-8 mode.
977    
978    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
979        the data contained the byte 0x85 as part of a UTF-8 character within its
980        first line. This applied both to normal and DFA matching.
981    
982    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
983        /^[^d]*?$/8 failed to match "abc".
984    
985    16. Added a missing copyright notice to pcrecpp_internal.h.
986    
987    17. Make it more clear in the documentation that values returned from
988        pcre_exec() in ovector are byte offsets, not character counts.
989    
990    18. Tidied a few places to stop certain compilers from issuing warnings.
991    
992    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
993        supplied by Stefan Weber. I made a further small update for 7.8 because
994        there is a change of source arrangements: the pcre_searchfuncs.c module is
995        replaced by pcre_ucd.c.
996    
997    
998    Version 7.7 07-May-08
999    ---------------------
1000    
1001    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
1002        a string to a long long, pretend we don't even have a long long." This is
1003        done by checking for the strtoq, strtoll, and _strtoi64 functions.
1004    
1005    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
1006        pre-7.6 versions, which defined a global no_arg variable instead of putting
1007        it in the RE class. (See also #8 below.)
1008    
1009    3.  Remove a line of dead code, identified by coverity and reported by Nuno
1010        Lopes.
1011    
1012    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
1013    
1014        (1) The include/exclude patterns were being applied to the whole pathnames
1015            of files, instead of just to the final components.
1016    
1017        (2) If there was more than one level of directory, the subdirectories were
1018            skipped unless they satisfied the include/exclude conditions. This is
1019            inconsistent with GNU grep (and could even be seen as contrary to the
1020            pcregrep specification - which I improved to make it absolutely clear).
1021            The action now is always to scan all levels of directory, and just
1022            apply the include/exclude patterns to regular files.
1023    
1024    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
1025        --exclude_dir in the tests to avoid scanning .svn directories.
1026    
1027    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
1028        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
1029        doesn't support NULs in patterns.
1030    
1031    7.  Added some missing "const"s to declarations of static tables in
1032        pcre_compile.c and pcre_dfa_exec.c.
1033    
1034    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
1035        caused by fix #2  above. (Subsequently also a second patch to fix the
1036        first patch. And a third patch - this was a messy problem.)
1037    
1038    9.  Applied Craig's patch to remove the use of push_back().
1039    
1040    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
1041        matching function regexec().
1042    
1043    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
1044        which, however, unlike Perl's \g{...}, are subroutine calls, not back
1045        references. PCRE supports relative numbers with this syntax (I don't think
1046        Oniguruma does).
1047    
1048    12. Previously, a group with a zero repeat such as (...){0} was completely
1049        omitted from the compiled regex. However, this means that if the group
1050        was called as a subroutine from elsewhere in the pattern, things went wrong
1051        (an internal error was given). Such groups are now left in the compiled
1052        pattern, with a new opcode that causes them to be skipped at execution
1053        time.
1054    
1055    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
1056        to the way PCRE behaves:
1057    
1058        (a) A lone ] character is dis-allowed (Perl treats it as data).
1059    
1060        (b) A back reference to an unmatched subpattern matches an empty string
1061            (Perl fails the current match path).
1062    
1063        (c) A data ] in a character class must be notated as \] because if the
1064            first data character in a class is ], it defines an empty class. (In
1065            Perl it is not possible to have an empty class.) The empty class []
1066            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
1067            The negative empty class [^] matches any one character, independently
1068            of the DOTALL setting.
1069    
1070    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
1071        non-existent subpattern following a character class starting with ']' and
1072        containing () gave an internal compiling error instead of "reference to
1073        non-existent subpattern". Fortunately, when the pattern did exist, the
1074        compiled code was correct. (When scanning forwards to check for the
1075        existencd of the subpattern, it was treating the data ']' as terminating
1076        the class, so got the count wrong. When actually compiling, the reference
1077        was subsequently set up correctly.)
1078    
1079    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
1080        it was being rejected as not supported by pcre_dfa_exec(), even though
1081        other assertions are supported. I have made pcre_dfa_exec() support
1082        (*FAIL).
1083    
1084    16. The implementation of 13c above involved the invention of a new opcode,
1085        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
1086        cannot be changed at match time, I realized I could make a small
1087        improvement to matching performance by compiling OP_ALLANY instead of
1088        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
1089        on the OP_ANY path.
1090    
1091    17. Compiling pcretest on Windows with readline support failed without the
1092        following two fixes: (1) Make the unistd.h include conditional on
1093        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
1094    
1095    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
1096        ncurses library to be included for pcretest when ReadLine support is
1097        requested, but also to allow for it to be overridden. This patch came from
1098        Daniel Bergström.
1099    
1100    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
1101        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
1102        any errors with the current Unicode tables. Thanks to Peter Kankowski for
1103        spotting this.
1104    
1105    
1106    Version 7.6 28-Jan-08
1107    ---------------------
1108    
1109    1.  A character class containing a very large number of characters with
1110        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
1111        overflow.
1112    
1113    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
1114        HAVE_LONG_LONG is not defined.
1115    
1116    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
1117        bring them up to date and include new features. This patch includes:
1118    
1119        - Fixed PH's badly added libz and libbz2 support.
1120        - Fixed a problem with static linking.
1121        - Added pcredemo. [But later removed - see 7 below.]
1122        - Fixed dftables problem and added an option.
1123        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
1124            HAVE_LONG_LONG.
1125        - Added readline support for pcretest.
1126        - Added an listing of the option settings after cmake has run.
1127    
1128    4.  A user submitted a patch to Makefile that makes it easy to create
1129        "pcre.dll" under mingw when using Configure/Make. I added stuff to
1130        Makefile.am that cause it to include this special target, without
1131        affecting anything else. Note that the same mingw target plus all
1132        the other distribution libraries and programs are now supported
1133        when configuring with CMake (see 6 below) instead of with
1134        Configure/Make.
1135    
1136    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
1137        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
1138        exported in the Windows port". It has not yet been confirmed that the patch
1139        solves the problem, but it does no harm.
1140    
1141    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
1142        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
1143        with CMake, and also correct the comment about stack recursion.
1144    
1145    7.  Remove the automatic building of pcredemo from the ./configure system and
1146        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
1147        of a program that users should build themselves after PCRE is installed, so
1148        building it automatically is not really right. What is more, it gave
1149        trouble in some build environments.
1150    
1151    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
1152    
1153    
1154    Version 7.5 10-Jan-08
1155    ---------------------
1156    
1157    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
1158        values in parens when parsing an RE using the C++ wrapper."
1159    
1160    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
1161        Characters greater than 255 were excluded from the class instead of being
1162        included.
1163    
1164    3.  The same bug as (2) above applied to negated POSIX classes such as
1165        [:^space:].
1166    
1167    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
1168        defined or documented. It seems to have been a typo for PCRE_STATIC, so
1169        I have changed it.
1170    
1171    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
1172        first named subpattern) and a construct such as (?&a) would reference the
1173        first named subpattern whose name started with "a" (in other words, the
1174        length check was missing). Both these problems are fixed. "Subpattern name
1175        expected" is now given for (?&) (a zero-length name), and this patch also
1176        makes it give the same error for \k'' (previously it complained that that
1177        was a reference to a non-existent subpattern).
1178    
1179    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
1180        this is right because (?- can be followed by option settings as well as by
1181        digits. I have, however, made the messages clearer.
1182    
1183    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
1184        than the number used in the conditional) now cause a compile-time error.
1185        This is actually not compatible with Perl, which accepts such patterns, but
1186        treats the conditional as always being FALSE (as PCRE used to), but it
1187        seems to me that giving a diagnostic is better.
1188    
1189    8.  Change "alphameric" to the more common word "alphanumeric" in comments
1190        and messages.
1191    
1192    9.  Fix two occurrences of "backslash" in comments that should have been
1193        "backspace".
1194    
1195    10. Remove two redundant lines of code that can never be obeyed (their function
1196        was moved elsewhere).
1197    
1198    11. The program that makes PCRE's Unicode character property table had a bug
1199        which caused it to generate incorrect table entries for sequences of
1200        characters that have the same character type, but are in different scripts.
1201        It amalgamated them into a single range, with the script of the first of
1202        them. In other words, some characters were in the wrong script. There were
1203        thirteen such cases, affecting characters in the following ranges:
1204    
1205          U+002b0 - U+002c1
1206          U+0060c - U+0060d
1207          U+0061e - U+00612
1208          U+0064b - U+0065e
1209          U+0074d - U+0076d
1210          U+01800 - U+01805
1211          U+01d00 - U+01d77
1212          U+01d9b - U+01dbf
1213          U+0200b - U+0200f
1214          U+030fc - U+030fe
1215          U+03260 - U+0327f
1216          U+0fb46 - U+0fbb1
1217          U+10450 - U+1049d
1218    
1219    12. The -o option (show only the matching part of a line) for pcregrep was not
1220        compatible with GNU grep in that, if there was more than one match in a
1221        line, it showed only the first of them. It now behaves in the same way as
1222        GNU grep.
1223    
1224    13. If the -o and -v options were combined for pcregrep, it printed a blank
1225        line for every non-matching line. GNU grep prints nothing, and pcregrep now
1226        does the same. The return code can be used to tell if there were any
1227        non-matching lines.
1228    
1229    14. Added --file-offsets and --line-offsets to pcregrep.
1230    
1231    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
1232        infinitely looping recursion. The bug was that positive lookaheads were not
1233        being skipped when checking for a possible empty match (negative lookaheads
1234        and both kinds of lookbehind were skipped).
1235    
1236    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
1237        inclusion of <windows.h> to before rather than after the definition of
1238        INVALID_FILE_ATTRIBUTES (patch from David Byron).
1239    
1240    17. Specifying a possessive quantifier with a specific limit for a Unicode
1241        character property caused pcre_compile() to compile bad code, which led at
1242        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
1243        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
1244        caused the error; without that there was no problem.
1245    
1246    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
1247    
1248    19. Added --enable-pcretest-libreadline.
1249    
1250    20. In pcrecpp.cc, the variable 'count' was incremented twice in
1251        RE::GlobalReplace(). As a result, the number of replacements returned was
1252        double what it should be. I removed one of the increments, but Craig sent a
1253        later patch that removed the other one (the right fix) and added unit tests
1254        that check the return values (which was not done before).
1255    
1256    21. Several CMake things:
1257    
1258        (1) Arranged that, when cmake is used on Unix, the libraries end up with
1259            the names libpcre and libpcreposix, not just pcre and pcreposix.
1260    
1261        (2) The above change means that pcretest and pcregrep are now correctly
1262            linked with the newly-built libraries, not previously installed ones.
1263    
1264        (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
1265    
1266    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
1267        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
1268        UTF-8 newline character). The key issue is that the pattern starts .*;
1269        this means that the match must be either at the beginning, or after a
1270        newline. The bug was in the code for advancing after a failed match and
1271        checking that the new position followed a newline. It was not taking
1272        account of UTF-8 characters correctly.
1273    
1274    23. PCRE was behaving differently from Perl in the way it recognized POSIX
1275        character classes. PCRE was not treating the sequence [:...:] as a
1276        character class unless the ... were all letters. Perl, however, seems to
1277        allow any characters between [: and :], though of course it rejects as
1278        unknown any "names" that contain non-letters, because all the known class
1279        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
1280        for example, whereas PCRE did not - it did not recognize a POSIX character
1281        class. This seemed a bit dangerous, so the code has been changed to be
1282        closer to Perl. The behaviour is not identical to Perl, because PCRE will
1283        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
1284        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
1285        Perl does, and where it didn't before.
1286    
1287    24. Rewrite so as to remove the single use of %n from pcregrep because in some
1288        Windows environments %n is disabled by default.
1289    
1290    
1291    Version 7.4 21-Sep-07
1292    ---------------------
1293    
1294    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
1295        means that a class such as [\s] counted as "explicit reference to CR or
1296        LF". That isn't really right - the whole point of the change was to try to
1297        help when there was an actual mention of one of the two characters. So now
1298        the change happens only if \r or \n (or a literal CR or LF) character is
1299        encountered.
1300    
1301    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
1302        of both had grown to the point where there were only 3 bits left.
1303        Fortunately, there was spare space in the data structure, and so I have
1304        moved the internal flags into a new 16-bit field to free up more option
1305        bits.
1306    
1307    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
1308        but did not set the internal JCHANGED flag - either of these is enough to
1309        control the way the "get" function works - but the PCRE_INFO_JCHANGED
1310        facility is supposed to tell if (?J) was ever used, so now (?J) at the
1311        start sets both bits.
1312    
1313    4.  Added options (at build time, compile time, exec time) to change \R from
1314        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
1315    
1316    5.  doc/pcresyntax.html was missing from the distribution.
1317    
1318    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
1319        compatibility, even though it is no longer used.
1320    
1321    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
1322        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
1323        windows.h file is present (where different names are used). [This was
1324        reversed later after testing - see 16 below.]
1325    
1326    8.  Changed all #include <config.h> to #include "config.h". There were also
1327        some further <pcre.h> cases that I changed to "pcre.h".
1328    
1329    9.  When pcregrep was used with the --colour option, it missed the line ending
1330        sequence off the lines that it output.
1331    
1332    10. It was pointed out to me that arrays of string pointers cause lots of
1333        relocations when a shared library is dynamically loaded. A technique of
1334        using a single long string with a table of offsets can drastically reduce
1335        these. I have refactored PCRE in four places to do this. The result is
1336        dramatic:
1337    
1338          Originally:                          290
1339          After changing UCP table:            187
1340          After changing error message table:   43
1341          After changing table of "verbs"       36
1342          After changing table of Posix names   22
1343    
1344        Thanks to the folks working on Gregex for glib for this insight.
1345    
1346    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
1347        unicode-properties was also set.
1348    
1349    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
1350    
1351    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
1352        checked only for CRLF.
1353    
1354    14. Added casts to pcretest.c to avoid compiler warnings.
1355    
1356    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
1357    
1358    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
1359        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
1360        entirely. This removes changes made in 7 above.
1361    
1362    17. The CMake files have been updated, and there is now more information about
1363        building with CMake in the NON-UNIX-USE document.
1364    
1365    
1366    Version 7.3 28-Aug-07
1367    ---------------------
1368    
1369     1. In the rejigging of the build system that eventually resulted in 7.1, the
1370        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
1371        brackets there is not right, since it causes compilers to look for an
1372        installed pcre.h, not the version that is in the source that is being
1373        compiled (which of course may be different). I have changed it back to:
1374    
1375          #include "pcre.h"
1376    
1377        I have a vague recollection that the change was concerned with compiling in
1378        different directories, but in the new build system, that is taken care of
1379        by the VPATH setting the Makefile.
1380    
1381     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
1382        when the subject happened to end in the byte 0x85 (e.g. if the last
1383        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
1384        characters but of course it shouldn't be taken as a newline when it is part
1385        of another character. The bug was that, for an unlimited repeat of . in
1386        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
1387        characters when looking for a newline.
1388    
1389     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
1390    
1391     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
1392        in debug output.
1393    
1394     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
1395        long printing in the pcrecpp unittest when running under MinGW.
1396    
1397     6. ESC_K was left out of the EBCDIC table.
1398    
1399     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
1400        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
1401        limit also applies to "virtual nesting" when a pattern is recursive, and in
1402        this case 1000 isn't so big. I have been able to remove this limit at the
1403        expense of backing off one optimization in certain circumstances. Normally,
1404        when pcre_exec() would call its internal match() function recursively and
1405        immediately return the result unconditionally, it uses a "tail recursion"
1406        feature to save stack. However, when a subpattern that can match an empty
1407        string has an unlimited repetition quantifier, it no longer makes this
1408        optimization. That gives it a stack frame in which to save the data for
1409        checking that an empty string has been matched. Previously this was taken
1410        from the 1000-entry workspace that had been reserved. So now there is no
1411        explicit limit, but more stack is used.
1412    
1413     8. Applied Daniel's patches to solve problems with the import/export magic
1414        syntax that is required for Windows, and which was going wrong for the
1415        pcreposix and pcrecpp parts of the library. These were overlooked when this
1416        problem was solved for the main library.
1417    
1418     9. There were some crude static tests to avoid integer overflow when computing
1419        the size of patterns that contain repeated groups with explicit upper
1420        limits. As the maximum quantifier is 65535, the maximum group length was
1421        set at 30,000 so that the product of these two numbers did not overflow a
1422        32-bit integer. However, it turns out that people want to use groups that
1423        are longer than 30,000 bytes (though not repeat them that many times).
1424        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1425        made it possible to implement the integer overflow checks in a much more
1426        dynamic way, which I have now done. The artificial limitation on group
1427        length has been removed - we now have only the limit on the total length of
1428        the compiled pattern, which depends on the LINK_SIZE setting.
1429    
1430    10. Fixed a bug in the documentation for get/copy named substring when
1431        duplicate names are permitted. If none of the named substrings are set, the
1432        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1433        empty string.
1434    
1435    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1436        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1437        because the ] is interpreted as the first data character and the
1438        terminating ] is not found. PCRE has been made compatible with Perl in this
1439        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1440        cause memory overwriting.
1441    
1442    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1443        string has been matched (to stop an infinite loop). It was not recognizing
1444        a conditional subpattern that could match an empty string if that
1445        subpattern was within another subpattern. For example, it looped when
1446        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
1447        condition was not nested. This bug has been fixed.
1448    
1449    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1450        past the start of the subject in the presence of bytes with the top bit
1451        set, for example "\x8aBCD".
1452    
1453    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1454        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1455    
1456    14. Optimized (?!) to (*FAIL).
1457    
1458    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1459        This restricts code points to be within the range 0 to 0x10FFFF, excluding
1460        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1461        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1462        does: it's just the validity check that is more restrictive.
1463    
1464    16. Inserted checks for integer overflows during escape sequence (backslash)
1465        processing, and also fixed erroneous offset values for syntax errors during
1466        backslash processing.
1467    
1468    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1469        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1470    
1471    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1472        caused an overrun.
1473    
1474    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1475        something other than just ASCII characters) inside a group that had an
1476        unlimited repeat caused a loop at compile time (while checking to see
1477        whether the group could match an empty string).
1478    
1479    20. Debugging a pattern containing \p or \P could cause a crash. For example,
1480        [\P{Any}] did so. (Error in the code for printing property names.)
1481    
1482    21. An orphan \E inside a character class could cause a crash.
1483    
1484    22. A repeated capturing bracket such as (A)? could cause a wild memory
1485        reference during compilation.
1486    
1487    23. There are several functions in pcre_compile() that scan along a compiled
1488        expression for various reasons (e.g. to see if it's fixed length for look
1489        behind). There were bugs in these functions when a repeated \p or \P was
1490        present in the pattern. These operators have additional parameters compared
1491        with \d, etc, and these were not being taken into account when moving along
1492        the compiled data. Specifically:
1493    
1494        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1495            length.
1496    
1497        (b) An item such as \pL+ within a repeated group could cause crashes or
1498            loops.
1499    
1500        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1501            "reference to non-existent subpattern" error.
1502    
1503        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1504    
1505    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1506        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1507    
1508    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1509    
1510    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1511        character were causing crashes (broken optimization).
1512    
1513    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1514        \p or \P) caused a compile-time loop.
1515    
1516    28. More problems have arisen in unanchored patterns when CRLF is a valid line
1517        break. For example, the unstudied pattern [\r\n]A does not match the string
1518        "\r\nA" because change 7.0/46 below moves the current point on by two
1519        characters after failing to match at the start. However, the pattern \nA
1520        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
1521        the same is true. There doesn't seem any very clean way out of this, but
1522        what I have chosen to do makes the common cases work: PCRE now takes note
1523        of whether there can be an explicit match for \r or \n anywhere in the
1524        pattern, and if so, 7.0/46 no longer applies. As part of this change,
1525        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1526        pattern has explicit CR or LF references.
1527    
1528    29. Added (*CR) etc for changing newline setting at start of pattern.
1529    
1530    
1531    Version 7.2 19-Jun-07
1532    ---------------------
1533    
1534     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
1535        which is apparently normally available under Windows.
1536    
1537     2. Re-jig the pcregrep tests with different newline settings in an attempt
1538        to make them independent of the local environment's newline setting.
1539    
1540     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
1541    
1542     4. Some of the "internals" tests were previously cut out when the link size
1543        was not 2, because the output contained actual offsets. The recent new
1544        "Z" feature of pcretest means that these can be cut out, making the tests
1545        usable with all link sizes.
1546    
1547     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
1548        stack recursion. This gives a massive performance boost under BSD, but just
1549        a small improvement under Linux. However, it saves one field in the frame
1550        in all cases.
1551    
1552     6. Added more features from the forthcoming Perl 5.10:
1553    
1554        (a) (?-n) (where n is a string of digits) is a relative subroutine or
1555            recursion call. It refers to the nth most recently opened parentheses.
1556    
1557        (b) (?+n) is also a relative subroutine call; it refers to the nth next
1558            to be opened parentheses.
1559    
1560        (c) Conditions that refer to capturing parentheses can be specified
1561            relatively, for example, (?(-2)... or (?(+3)...
1562    
1563        (d) \K resets the start of the current match so that everything before
1564            is not part of it.
1565    
1566        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1567    
1568        (f) \g{name} is another synonym - part of Perl 5.10's unification of
1569            reference syntax.
1570    
1571        (g) (?| introduces a group in which the numbering of parentheses in each
1572            alternative starts with the same number.
1573    
1574        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
1575    
1576     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
1577        PCRE_INFO_JCHANGED.
1578    
1579     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
1580        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
1581        for detecting groups that can match an empty string.
1582    
1583     9. A pattern with a very large number of alternatives (more than several
1584        hundred) was running out of internal workspace during the pre-compile
1585        phase, where pcre_compile() figures out how much memory will be needed. A
1586        bit of new cunning has reduced the workspace needed for groups with
1587        alternatives. The 1000-alternative test pattern now uses 12 bytes of
1588        workspace instead of running out of the 4096 that are available.
1589    
1590    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
1591    
1592    11. Applied patch from Google to remove an optimization that didn't quite work.
1593        The report of the bug said:
1594    
1595          pcrecpp::RE("a*").FullMatch("aaa") matches, while
1596          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
1597          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
1598    
1599    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
1600        it matched the wrong number of bytes.
1601    
1602    
1603    Version 7.1 24-Apr-07
1604    ---------------------
1605    
1606     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
1607        that is more "standard", making use of automake and other Autotools. There
1608        is some re-arrangement of the files and adjustment of comments consequent
1609        on this.
1610    
1611     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
1612        for recursive directory scanning broke on some systems because the files
1613        are not scanned in any specific order and on different systems the order
1614        was different. A call to "sort" has been inserted into RunGrepTest for the
1615        approprate test as a short-term fix. In the longer term there may be an
1616        alternative.
1617    
1618     3. I had an email from Eric Raymond about problems translating some of PCRE's
1619        man pages to HTML (despite the fact that I distribute HTML pages, some
1620        people do their own conversions for various reasons). The problems
1621        concerned the use of low-level troff macros .br and .in. I have therefore
1622        removed all such uses from the man pages (some were redundant, some could
1623        be replaced by .nf/.fi pairs). The 132html script that I use to generate
1624        HTML has been updated to handle .nf/.fi and to complain if it encounters
1625        .br or .in.
1626    
1627     4. Updated comments in configure.ac that get placed in config.h.in and also
1628        arranged for config.h to be included in the distribution, with the name
1629        config.h.generic, for the benefit of those who have to compile without
1630        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
1631    
1632     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
1633        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
1634        makevp.bat for the current PCRE, using the additional files
1635        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
1636    
1637     6. A Windows user reported a minor discrepancy with test 2, which turned out
1638        to be caused by a trailing space on an input line that had got lost in his
1639        copy. The trailing space was an accident, so I've just removed it.
1640    
1641     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
1642        that is needed.
1643    
1644     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
1645        as "const" (a) because they are and (b) because it helps the PHP
1646        maintainers who have recently made a script to detect big data structures
1647        in the php code that should be moved to the .rodata section. I remembered
1648        to update Builducptable as well, so it won't revert if ucptable.h is ever
1649        re-created.
1650    
1651     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
1652        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
1653        order to be able to cut out the UTF-8 tables in the latter when UTF-8
1654        support is not required. This saves 1.5-2K of code, which is important in
1655        some applications.
1656    
1657        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
1658        so as not to refer to the tables, even though these functions will never be
1659        called when UTF-8 support is disabled. Otherwise there are problems with a
1660        shared library.
1661    
1662    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
1663    
1664        (a) It was defining its arguments as char * instead of void *.
1665    
1666        (b) It was assuming that all moves were upwards in memory; this was true
1667            a long time ago when I wrote it, but is no longer the case.
1668    
1669        The emulated memove() is provided for those environments that have neither
1670        memmove() nor bcopy(). I didn't think anyone used it these days, but that
1671        is clearly not the case, as these two bugs were recently reported.
1672    
1673    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
1674        and Detrail to create the HTML documentation, the .txt form of the man
1675        pages, and it removes trailing spaces from listed files. It also creates
1676        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
1677        case, it wraps all the #defines with #ifndefs. This script should be run
1678        before "make dist".
1679    
1680    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1681        with Unicode property support.
1682    
1683        (a) For a maximizing quantifier, if the two different cases of the
1684            character were of different lengths in their UTF-8 codings (there are
1685            some cases like this - I found 11), and the matching function had to
1686            back up over a mixture of the two cases, it incorrectly assumed they
1687            were both the same length.
1688    
1689        (b) When PCRE was configured to use the heap rather than the stack for
1690            recursion during matching, it was not correctly preserving the data for
1691            the other case of a UTF-8 character when checking ahead for a match
1692            while processing a minimizing repeat. If the check also involved
1693            matching a wide character, but failed, corruption could cause an
1694            erroneous result when trying to check for a repeat of the original
1695            character.
1696    
1697    13. Some tidying changes to the testing mechanism:
1698    
1699        (a) The RunTest script now detects the internal link size and whether there
1700            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1701            values substituted by "configure". (The RunGrepTest script already did
1702            this for UTF-8.) The configure.ac script no longer substitutes the
1703            relevant variables.
1704    
1705        (b) The debugging options /B and /D in pcretest show the compiled bytecode
1706            with length and offset values. This means that the output is different
1707            for different internal link sizes. Test 2 is skipped for link sizes
1708            other than 2 because of this, bypassing the problem. Unfortunately,
1709            there was also a test in test 3 (the locale tests) that used /B and
1710            failed for link sizes other than 2. Rather than cut the whole test out,
1711            I have added a new /Z option to pcretest that replaces the length and
1712            offset values with spaces. This is now used to make test 3 independent
1713            of link size. (Test 2 will be tidied up later.)
1714    
1715    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1716        segmentation fault instead of returning the appropriate error message.
1717    
1718    15. In multiline mode when the newline sequence was set to "any", the pattern
1719        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1720        This doesn't seem right; it now treats the CRLF combination as the line
1721        ending, and so does not match in that case. It's only a pattern such as ^$
1722        that would hit this one: something like ^ABC$ would have failed after \r
1723        and then tried again after \r\n.
1724    
1725    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1726        in an attempt to make files that differ only in their line terminators
1727        compare equal. This works on Linux.
1728    
1729    17. Under certain error circumstances pcregrep might try to free random memory
1730        as it exited. This is now fixed, thanks to valgrind.
1731    
1732    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1733        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1734        was because its rules for how to advance for /g after matching an empty
1735        string at the end of a line did not allow for this case. They now check for
1736        it specially.
1737    
1738    20. pcretest is supposed to handle patterns and data of any length, by
1739        extending its buffers when necessary. It was getting this wrong when the
1740        buffer for a data line had to be extended.
1741    
1742    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1743        CRLF as a newline sequence.
1744    
1745    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1746        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1747        I have nevertheless tidied it up.
1748    
1749    23. Added some casts to kill warnings from HP-UX ia64 compiler.
1750    
1751    24. Added a man page for pcre-config.
1752    
1753    
1754    Version 7.0 19-Dec-06
1755    ---------------------
1756    
1757     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1758        moving to gcc 4.1.1.
1759    
1760     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1761        sys/time.h, which is documented as needed for this function. It doesn't
1762        seem to matter on Linux, but it showed up on some releases of OS X.
1763    
1764     3. It seems that there are systems where bytes whose values are greater than
1765        127 match isprint() in the "C" locale. The "C" locale should be the
1766        default when a C program starts up. In most systems, only ASCII printing
1767        characters match isprint(). This difference caused the output from pcretest
1768        to vary, making some of the tests fail. I have changed pcretest so that:
1769    
1770        (a) When it is outputting text in the compiled version of a pattern, bytes
1771            other than 32-126 are always shown as hex escapes.
1772    
1773        (b) When it is outputting text that is a matched part of a subject string,
1774            it does the same, unless a different locale has been set for the match
1775            (using the /L modifier). In this case, it uses isprint() to decide.
1776    
1777     4. Fixed a major bug that caused incorrect computation of the amount of memory
1778        required for a compiled pattern when options that changed within the
1779        pattern affected the logic of the preliminary scan that determines the
1780        length. The relevant options are -x, and -i in UTF-8 mode. The result was
1781        that the computed length was too small. The symptoms of this bug were
1782        either the PCRE error "internal error: code overflow" from pcre_compile(),
1783        or a glibc crash with a message such as "pcretest: free(): invalid next
1784        size (fast)". Examples of patterns that provoked this bug (shown in
1785        pcretest format) are:
1786    
1787          /(?-x: )/x
1788          /(?x)(?-x: \s*#\s*)/
1789          /((?i)[\x{c0}])/8
1790          /(?i:[\x{c0}])/8
1791    
1792        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1793        is now done differently.
1794    
1795     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1796        wrapper classes; (b) implement a new function in the C++ scanner that is
1797        more efficient than the old way of doing things because it avoids levels of
1798        recursion in the regex matching; (c) add a paragraph to the documentation
1799        for the FullMatch() function.
1800    
1801     6. The escape sequence \n was being treated as whatever was defined as
1802        "newline". Not only was this contrary to the documentation, which states
1803        that \n is character 10 (hex 0A), but it also went horribly wrong when
1804        "newline" was defined as CRLF. This has been fixed.
1805    
1806     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1807        was being set to -1 for the "end of line" case (supposedly a value that no
1808        character can have). Though this value is never used (the check for end of
1809        line is "zero bytes in current character"), it caused compiler complaints.
1810        I've changed it to 0xffffffff.
1811    
1812     8. In pcre_version.c, the version string was being built by a sequence of
1813        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1814        string (as it is for production releases) called a macro with an empty
1815        argument. The C standard says the result of this is undefined. The gcc
1816        compiler treats it as an empty string (which was what was wanted) but it is
1817        reported that Visual C gives an error. The source has been hacked around to
1818        avoid this problem.
1819    
1820     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1821        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1822        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1823        of them did).
1824    
1825    10. Originally, pcretest opened its input and output without "b"; then I was
1826        told that "b" was needed in some environments, so it was added for release
1827        5.0 to both the input and output. (It makes no difference on Unix-like
1828        systems.) Later I was told that it is wrong for the input on Windows. I've
1829        now abstracted the modes into two macros, to make it easier to fiddle with
1830        them, and removed "b" from the input mode under Windows.
1831    
1832    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1833    
1834    12. Added -help and --help to pcretest as an official way of being reminded
1835        of the options.
1836    
1837    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1838        and pcrecpp.cc because they annoy compilers at high warning levels.
1839    
1840    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1841    
1842    15. Fixed an occurrence of == in configure.ac that should have been = (shell
1843        scripts are not C programs :-) and which was not noticed because it works
1844        on Linux.
1845    
1846    16. pcretest is supposed to handle any length of pattern and data line (as one
1847        line or as a continued sequence of lines) by extending its input buffer if
1848        necessary. This feature was broken for very long pattern lines, leading to
1849        a string of junk being passed to pcre_compile() if the pattern was longer
1850        than about 50K.
1851    
1852    17. I have done a major re-factoring of the way pcre_compile() computes the
1853        amount of memory needed for a compiled pattern. Previously, there was code
1854        that made a preliminary scan of the pattern in order to do this. That was
1855        OK when PCRE was new, but as the facilities have expanded, it has become
1856        harder and harder to keep it in step with the real compile phase, and there
1857        have been a number of bugs (see for example, 4 above). I have now found a
1858        cunning way of running the real compile function in a "fake" mode that
1859        enables it to compute how much memory it would need, while actually only
1860        ever using a few hundred bytes of working memory and without too many
1861        tests of the mode. This should make future maintenance and development
1862        easier. A side effect of this work is that the limit of 200 on the nesting
1863        depth of parentheses has been removed (though this was never a serious
1864        limitation, I suspect). However, there is a downside: pcre_compile() now
1865        runs more slowly than before (30% or more, depending on the pattern). I
1866        hope this isn't a big issue. There is no effect on runtime performance.
1867    
1868    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1869        newline (only possible for the last line of a file) and it was a
1870        pattern that set a locale (followed by /Lsomething), pcretest crashed.
1871    
1872    19. Added additional timing features to pcretest. (1) The -tm option now times
1873        matching only, not compiling. (2) Both -t and -tm can be followed, as a
1874        separate command line item, by a number that specifies the number of
1875        repeats to use when timing. The default is 50000; this gives better
1876        precision, but takes uncomfortably long for very large patterns.
1877    
1878    20. Extended pcre_study() to be more clever in cases where a branch of a
1879        subpattern has no definite first character. For example, (a*|b*)[cd] would
1880        previously give no result from pcre_study(). Now it recognizes that the
1881        first character must be a, b, c, or d.
1882    
1883    21. There was an incorrect error "recursive call could loop indefinitely" if
1884        a subpattern (or the entire pattern) that was being tested for matching an
1885        empty string contained only one non-empty item after a nested subpattern.
1886        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1887        incorrectly, because the \d was being skipped in the check.
1888    
1889    22. The pcretest program now has a new pattern option /B and a command line
1890        option -b, which is equivalent to adding /B to every pattern. This causes
1891        it to show the compiled bytecode, without the additional information that
1892        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1893        is the same as /B/I).
1894    
1895    23. A new optimization is now able automatically to treat some sequences such
1896        as a*b as a*+b. More specifically, if something simple (such as a character
1897        or a simple class like \d) has an unlimited quantifier, and is followed by
1898        something that cannot possibly match the quantified thing, the quantifier
1899        is automatically "possessified".
1900    
1901    24. A recursive reference to a subpattern whose number was greater than 39
1902        went wrong under certain circumstances in UTF-8 mode. This bug could also
1903        have affected the operation of pcre_study().
1904    
1905    25. Realized that a little bit of performance could be had by replacing
1906        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1907    
1908    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1909    
1910    27. Possessive quantifiers such as a++ were previously implemented by turning
1911        them into atomic groups such as ($>a+). Now they have their own opcodes,
1912        which improves performance. This includes the automatically created ones
1913        from 23 above.
1914    
1915    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1916        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1917        the first matched character to be a colon. This applied both to named and
1918        numbered groups.
1919    
1920    29. The ucpinternal.h header file was missing its idempotency #ifdef.
1921    
1922    30. I was sent a "project" file called libpcre.a.dev which I understand makes
1923        building PCRE on Windows easier, so I have included it in the distribution.
1924    
1925    31. There is now a check in pcretest against a ridiculously large number being
1926        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1927        loop, the loop is abandoned.
1928    
1929    32. Forward references to subpatterns in conditions such as (?(2)...) where
1930        subpattern 2 is defined later cause pcre_compile() to search forwards in
1931        the pattern for the relevant set of parentheses. This search went wrong
1932        when there were unescaped parentheses in a character class, parentheses
1933        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1934    
1935    33. "Subroutine" calls and backreferences were previously restricted to
1936        referencing subpatterns earlier in the regex. This restriction has now
1937        been removed.
1938    
1939    34. Added a number of extra features that are going to be in Perl 5.10. On the
1940        whole, these are just syntactic alternatives for features that PCRE had
1941        previously implemented using the Python syntax or my own invention. The
1942        other formats are all retained for compatibility.
1943    
1944        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1945            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1946            also .NET compatible.
1947    
1948        (b) A recursion or subroutine call to a named group can now be defined as
1949            (?&name) as well as (?P>name).
1950    
1951        (c) A backreference to a named group can now be defined as \k<name> or
1952            \k'name' as well as (?P=name). The new forms, as well as being in Perl
1953            5.10, are also .NET compatible.
1954    
1955        (d) A conditional reference to a named group can now use the syntax
1956            (?(<name>) or (?('name') as well as (?(name).
1957    
1958        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1959            groups (named and numbered) that are never evaluated inline, but can be
1960            called as "subroutines" from elsewhere. In effect, the DEFINE condition
1961            is always false. There may be only one alternative in such a group.
1962    
1963        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1964            as the simple (?(R). The condition is true only if the most recent
1965            recursion is that of the given number or name. It does not search out
1966            through the entire recursion stack.
1967    
1968        (g) The escape \gN or \g{N} has been added, where N is a positive or
1969            negative number, specifying an absolute or relative reference.
1970    
1971    35. Tidied to get rid of some further signed/unsigned compiler warnings and
1972        some "unreachable code" warnings.
1973    
1974    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1975        things, this adds five new scripts.
1976    
1977    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1978        There were also incompatibilities regarding the handling of \Q..\E inside
1979        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1980        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1981    
1982    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1983        matches an empty string, and forcibly breaks the loop. There were bugs in
1984        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
1985        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
1986        separate and independent bugs (that affected different cases) have been
1987        fixed.
1988    
1989    39. Refactored the code to abolish the use of different opcodes for small
1990        capturing bracket numbers. This is a tidy that I avoided doing when I
1991        removed the limit on the number of capturing brackets for 3.5 back in 2001.
1992        The new approach is not only tidier, it makes it possible to reduce the
1993        memory needed to fix the previous bug (38).
1994    
1995    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1996        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1997        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1998        mode.
1999    
2000    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
2001        report.
2002    
2003    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
2004        copy construction and assignment in the C++ wrapper.
2005    
2006    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
2007        couple of bugs that could have given wrong results in the "--newline=crlf"
2008        case.
2009    
2010    44. Added a number of casts and did some reorganization of signed/unsigned int
2011        variables following suggestions from Dair Grant. Also renamed the variable
2012        "this" as "item" because it is a C++ keyword.
2013    
2014    45. Arranged for dftables to add
2015    
2016          #include "pcre_internal.h"
2017    
2018        to pcre_chartables.c because without it, gcc 4.x may remove the array
2019        definition from the final binary if PCRE is built into a static library and
2020        dead code stripping is activated.
2021    
2022    46. For an unanchored pattern, if a match attempt fails at the start of a
2023        newline sequence, and the newline setting is CRLF or ANY, and the next two
2024        characters are CRLF, advance by two characters instead of one.
2025    
2026    
2027    Version 6.7 04-Jul-06
2028    ---------------------
2029    
2030     1. In order to handle tests when input lines are enormously long, pcretest has
2031        been re-factored so that it automatically extends its buffers when
2032        necessary. The code is crude, but this _is_ just a test program. The
2033        default size has been increased from 32K to 50K.
2034    
2035     2. The code in pcre_study() was using the value of the re argument before
2036        testing it for NULL. (Of course, in any sensible call of the function, it
2037        won't be NULL.)
2038    
2039     3. The memmove() emulation function in pcre_internal.h, which is used on
2040        systems that lack both memmove() and bcopy() - that is, hardly ever -
2041        was missing a "static" storage class specifier.
2042    
2043     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
2044        containing an extended class (one that cannot be represented by a bitmap
2045        because it contains high-valued characters or Unicode property items, e.g.
2046        [\pZ]). Almost always one would set UTF-8 mode when processing such a
2047        pattern, but PCRE should not loop if you do not (it no longer does).
2048        [Detail: two cases were found: (a) a repeated subpattern containing an
2049        extended class; (b) a recursive reference to a subpattern that followed a
2050        previous extended class. It wasn't skipping over the extended class
2051        correctly when UTF-8 mode was not set.]
2052    
2053     5. A negated single-character class was not being recognized as fixed-length
2054        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
2055        compile error "lookbehind assertion is not fixed length".
2056    
2057     6. The RunPerlTest auxiliary script was showing an unexpected difference
2058        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
2059        write a Perl script that can interpret lines of an input file either as
2060        byte characters or as UTF-8, which is what "perltest" was being required to
2061        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
2062        can't do is switch easily at run time between having the "use utf8;" pragma
2063        or not. In the end, I fudged it by using the RunPerlTest script to insert
2064        "use utf8;" explicitly for the UTF-8 tests.
2065    
2066     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
2067        the end of the subject string, contrary to the documentation and to what
2068        Perl does. This was true of both matching functions. Now it matches only at
2069        the start of the subject and immediately after *internal* newlines.
2070    
2071     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
2072        a pointer to an int instead of a pointer to an unsigned long int. This
2073        caused problems on 64-bit systems.
2074    
2075     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
2076        instance of the 'standard' template library not being so standard".
2077    
2078    10. There was no check on the number of named subpatterns nor the maximum
2079        length of a subpattern name. The product of these values is used to compute
2080        the size of the memory block for a compiled pattern. By supplying a very
2081        long subpattern name and a large number of named subpatterns, the size
2082        computation could be caused to overflow. This is now prevented by limiting
2083        the length of names to 32 characters, and the number of named subpatterns
2084        to 10,000.
2085    
2086    11. Subpatterns that are repeated with specific counts have to be replicated in
2087        the compiled pattern. The size of memory for this was computed from the
2088        length of the subpattern and the repeat count. The latter is limited to
2089        65535, but there was no limit on the former, meaning that integer overflow
2090        could in principle occur. The compiled length of a repeated subpattern is
2091        now limited to 30,000 bytes in order to prevent this.
2092    
2093    12. Added the optional facility to have named substrings with the same name.
2094    
2095    13. Added the ability to use a named substring as a condition, using the
2096        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
2097        are numbers (not recommended). Forward references are permitted.
2098    
2099    14. Added forward references in named backreferences (if you see what I mean).
2100    
2101    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
2102        pattern could run off the end of the subject. For example, the pattern
2103        "(?s)(.{1,5})"8 did this with the subject "ab".
2104    
2105    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
2106        PCRE_CASELESS was set when matching characters that were quantified with ?
2107        or *.
2108    
2109    17. A character class other than a single negated character that had a minimum
2110        but no maximum quantifier - for example [ab]{6,} - was not handled
2111        correctly by pce_dfa_exec(). It would match only one character.
2112    
2113    18. A valid (though odd) pattern that looked like a POSIX character
2114        class but used an invalid character after [ (for example [[,abc,]]) caused
2115        pcre_compile() to give the error "Failed: internal error: code overflow" or
2116        in some cases to crash with a glibc free() error. This could even happen if
2117        the pattern terminated after [[ but there just happened to be a sequence of
2118        letters, a binary zero, and a closing ] in the memory that followed.
2119    
2120    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
2121        over the years. Originally (before any Unicode support), just the bottom 8
2122        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
2123        output from "man perlunicode" includes this:
2124    
2125          The regular expression compiler produces polymorphic opcodes.  That
2126          is, the pattern adapts to the data and automatically switches to
2127          the Unicode character scheme when presented with Unicode data--or
2128          instead uses a traditional byte scheme when presented with byte
2129          data.
2130    
2131        Sadly, a wide octal escape does not cause a switch, and in a string with
2132        no other multibyte characters, these octal escapes are treated as before.
2133        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
2134        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
2135        Unicode string.
2136    
2137        I have not perpetrated such confusion in PCRE. Up till now, it took just
2138        the bottom 8 bits, as in old Perl. I have now made octal escapes with
2139        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
2140        translate to the appropriate multibyte character.
2141    
2142    29. Applied some refactoring to reduce the number of warnings from Microsoft
2143        and Borland compilers. This has included removing the fudge introduced
2144        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
2145        a warning about an unused variable.
2146    
2147    21. PCRE has not included VT (character 0x0b) in the set of whitespace
2148        characters since release 4.0, because Perl (from release 5.004) does not.
2149        [Or at least, is documented not to: some releases seem to be in conflict
2150        with the documentation.] However, when a pattern was studied with
2151        pcre_study() and all its branches started with \s, PCRE still included VT
2152        as a possible starting character. Of course, this did no harm; it just
2153        caused an unnecessary match attempt.
2154    
2155    22. Removed a now-redundant internal flag bit that recorded the fact that case
2156        dependency changed within the pattern. This was once needed for "required
2157        byte" processing, but is no longer used. This recovers a now-scarce options
2158        bit. Also moved the least significant internal flag bit to the most-
2159        significant bit of the word, which was not previously used (hangover from
2160        the days when it was an int rather than a uint) to free up another bit for
2161        the future.
2162    
2163    23. Added support for CRLF line endings as well as CR and LF. As well as the
2164        default being selectable at build time, it can now be changed at runtime
2165        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
2166        specify that it is scanning data with non-default line endings.
2167    
2168    24. Changed the definition of CXXLINK to make it agree with the definition of
2169        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
2170    
2171    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
2172        recursions. This makes a big different to stack usage for some patterns.
2173    
2174    26. If a subpattern containing a named recursion or subroutine reference such
2175        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
2176        the space required for the compiled pattern went wrong and gave too small a
2177        value. Depending on the environment, this could lead to "Failed: internal
2178        error: code overflow at offset 49" or "glibc detected double free or
2179        corruption" errors.
2180    
2181    27. Applied patches from Google (a) to support the new newline modes and (b) to
2182        advance over multibyte UTF-8 characters in GlobalReplace.
2183    
2184    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
2185        difference for some implementation of PCRE in some Windows version.
2186    
2187    29. Added some extra testing facilities to pcretest:
2188    
2189        \q<number>   in a data line sets the "match limit" value
2190        \Q<number>   in a data line sets the "match recursion limt" value
2191        -S <number>  sets the stack size, where <number> is in megabytes
2192    
2193        The -S option isn't available for Windows.
2194    
2195    
2196    Version 6.6 06-Feb-06
2197    ---------------------
2198    
2199     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
2200        in pcreposix.h. I have copied the definition from pcre.h.
2201    
2202     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
2203        because pcre.h is no longer a built file.
2204    
2205     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
2206        not normally included in the compiled code.
2207    
2208    
2209    Version 6.5 01-Feb-06
2210    ---------------------
2211    
2212     1. When using the partial match feature with pcre_dfa_exec(), it was not
2213        anchoring the second and subsequent partial matches at the new starting
2214        point. This could lead to incorrect results. For example, with the pattern
2215        /1234/, partially matching against "123" and then "a4" gave a match.
2216    
2217     2. Changes to pcregrep:
2218    
2219        (a) All non-match returns from pcre_exec() were being treated as failures
2220            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
2221            error message is output. Some extra information is given for the
2222            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
2223            probably the only errors that are likely to be caused by users (by
2224            specifying a regex that has nested indefinite repeats, for instance).
2225            If there are more than 20 of these errors, pcregrep is abandoned.
2226    
2227        (b) A binary zero was treated as data while matching, but terminated the
2228            output line if it was written out. This has been fixed: binary zeroes
2229            are now no different to any other data bytes.
2230    
2231        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
2232            used to set a locale for matching. The --locale=xxxx long option has
2233            been added (no short equivalent) to specify a locale explicitly on the
2234            pcregrep command, overriding the environment variables.
2235    
2236        (d) When -B was used with -n, some line numbers in the output were one less
2237            than they should have been.
2238    
2239        (e) Added the -o (--only-matching) option.
2240    
2241        (f) If -A or -C was used with -c (count only), some lines of context were
2242            accidentally printed for the final match.
2243    
2244        (g) Added the -H (--with-filename) option.
2245    
2246        (h) The combination of options -rh failed to suppress file names for files
2247            that were found from directory arguments.
2248    
2249        (i) Added the -D (--devices) and -d (--directories) options.
2250    
2251        (j) Added the -F (--fixed-strings) option.
2252    
2253        (k) Allow "-" to be used as a file name for -f as well as for a data file.
2254    
2255        (l) Added the --colo(u)r option.
2256    
2257        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
2258            is not present by default.
2259    
2260     3. A nasty bug was discovered in the handling of recursive patterns, that is,
2261        items such as (?R) or (?1), when the recursion could match a number of
2262        alternatives. If it matched one of the alternatives, but subsequently,
2263        outside the recursion, there was a failure, the code tried to back up into
2264        the recursion. However, because of the way PCRE is implemented, this is not
2265        possible, and the result was an incorrect result from the match.
2266    
2267        In order to prevent this happening, the specification of recursion has
2268        been changed so that all such subpatterns are automatically treated as
2269        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
2270    
2271     4. I had overlooked the fact that, in some locales, there are characters for
2272        which isalpha() is true but neither isupper() nor islower() are true. In
2273        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
2274        and ordfeminine) are like this. This affected the treatment of \w and \W
2275        when they appeared in character classes, but not when they appeared outside
2276        a character class. The bit map for "word" characters is now created
2277        separately from the results of isalnum() instead of just taking it from the
2278        upper, lower, and digit maps. (Plus the underscore character, of course.)
2279    
2280     5. The above bug also affected the handling of POSIX character classes such as
2281        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
2282        permanent tables. Instead, the bit maps for such a class were previously
2283        created as the appropriate unions of the upper, lower, and digit bitmaps.
2284        Now they are created by subtraction from the [[:word:]] class, which has
2285        its own bitmap.
2286    
2287     6. The [[:blank:]] character class matches horizontal, but not vertical space.
2288        It is created by subtracting the vertical space characters (\x09, \x0a,
2289        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
2290        subtraction was done in the overall bitmap for a character class, meaning
2291        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
2292        be recognized. This bug has been fixed.
2293    
2294     7. Patches from the folks at Google:
2295    
2296          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
2297          real life, but is still worth protecting against".
2298    
2299          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
2300          regular expressions".
2301    
2302          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
2303          have it.
2304    
2305          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
2306          "configure" and the latter not, in order to fix a problem somebody had
2307          with compiling the Arg class on HP-UX.
2308    
2309          (e) Improve the error-handling of the C++ wrapper a little bit.
2310    
2311          (f) New tests for checking recursion limiting.
2312    
2313     8. The pcre_memmove() function, which is used only if the environment does not
2314        have a standard memmove() function (and is therefore rarely compiled),
2315        contained two bugs: (a) use of int instead of size_t, and (b) it was not
2316        returning a result (though PCRE never actually uses the result).
2317    
2318     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
2319        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
2320        returned instead of calling malloc() with an overflowing number that would
2321        most likely cause subsequent chaos.
2322    
2323    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
2324    
2325    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
2326        with this option is matched, the nmatch and pmatch options of regexec() are
2327        ignored.
2328    
2329    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
2330        provided in case anyone wants to the the POSIX interface with UTF-8
2331        strings.
2332    
2333    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
2334        C++ linking (needed for some HP-UX environments).
2335    
2336    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
2337        (unused parameter) and in the pcre_printint() function (omitted "default"
2338        switch label when the default is to do nothing).
2339    
2340    15. Added some code to make it possible, when PCRE is compiled as a C++
2341        library, to replace subject pointers for pcre_exec() with a smart pointer
2342        class, thus making it possible to process discontinuous strings.
2343    
2344    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
2345        much the same function. They were added by different people who were trying
2346        to make PCRE easy to compile on non-Unix systems. It has been suggested
2347        that PCRE_EXPORT be abolished now that there is more automatic apparatus
2348        for compiling on Windows systems. I have therefore replaced it with
2349        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
2350        defaults to "extern" for C or "extern C" for C++, which works fine on
2351        Unix-like systems. It is now possible to override the value of PCRE_DATA_
2352        SCOPE with something explicit in config.h. In addition:
2353    
2354        (a) pcreposix.h still had just "extern" instead of either of these macros;
2355            I have replaced it with PCRE_DATA_SCOPE.
2356    
2357        (b) Functions such as _pcre_xclass(), which are internal to the library,
2358            but external in the C sense, all had PCRE_EXPORT in their definitions.
2359            This is apparently wrong for the Windows case, so I have removed it.
2360            (It makes no difference on Unix-like systems.)
2361    
2362    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
2363        of recursive calls to match(). This is different to MATCH_LIMIT because
2364        that limits the total number of calls to match(), not all of which increase
2365        the depth of recursion. Limiting the recursion depth limits the amount of
2366        stack (or heap if NO_RECURSE is set) that is used. The default can be set
2367        when PCRE is compiled, and changed at run time. A patch from Google adds
2368        this functionality to the C++ interface.
2369    
2370    18. Changes to the handling of Unicode character properties:
2371    
2372        (a) Updated the table to Unicode 4.1.0.
2373    
2374        (b) Recognize characters that are not in the table as "Cn" (undefined).
2375    
2376        (c) I revised the way the table is implemented to a much improved format
2377            which includes recognition of ranges. It now supports the ranges that
2378            are defined in UnicodeData.txt, and it also amalgamates other
2379            characters into ranges. This has reduced the number of entries in the
2380            table from around 16,000 to around 3,000, thus reducing its size
2381            considerably. I realized I did not need to use a tree structure after
2382            all - a binary chop search is just as efficient. Having reduced the
2383            number of entries, I extended their size from 6 bytes to 8 bytes to
2384            allow for more data.
2385    
2386        (d) Added support for Unicode script names via properties such as \p{Han}.
2387    
2388    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
2389        matching that character.
2390    
2391    20. When matching a repeated Unicode property with a minimum greater than zero,
2392        (for example \pL{2,}), PCRE could look past the end of the subject if it
2393        reached it while seeking the minimum number of characters. This could
2394        happen only if some of the characters were more than one byte long, because
2395        there is a check for at least the minimum number of bytes.
2396    
2397    21. Refactored the implementation of \p and \P so as to be more general, to
2398        allow for more different types of property in future. This has changed the
2399        compiled form incompatibly. Anybody with saved compiled patterns that use
2400        \p or \P will have to recompile them.
2401    
2402    22. Added "Any" and "L&" to the supported property types.
2403    
2404    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
2405        but give a compile time error if the value is greater than 0xff.
2406    
2407    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
2408        accidentally not being installed or uninstalled.
2409    
2410    25. The pcre.h file was built from pcre.h.in, but the only changes that were
2411        made were to insert the current release number. This seemed silly, because
2412        it made things harder for people building PCRE on systems that don't run
2413        "configure". I have turned pcre.h into a distributed file, no longer built
2414        by "configure", with the version identification directly included. There is
2415        no longer a pcre.h.in file.
2416    
2417        However, this change necessitated a change to the pcre-config script as
2418        well. It is built from pcre-config.in, and one of the substitutions was the
2419        release number. I have updated configure.ac so that ./configure now finds
2420        the release number by grepping pcre.h.
2421    
2422    26. Added the ability to run the tests under valgrind.
2423    
2424    
2425    Version 6.4 05-Sep-05
2426    ---------------------
2427    
2428     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
2429        "--" to be printed when multiple files were scanned, even when none of the
2430        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
2431        consider it to be a bug, and have restored the previous behaviour.
2432    
2433     2. A couple of code tidies to get rid of compiler warnings.
2434    
2435     3. The pcretest program used to cheat by referring to symbols in the library
2436        whose names begin with _pcre_. These are internal symbols that are not
2437        really supposed to be visible externally, and in some environments it is
2438        possible to suppress them. The cheating is now confined to including
2439        certain files from the library's source, which is a bit cleaner.
2440    
2441     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
2442        file's purpose clearer.
2443    
2444     5. Reorganized pcre_ucp_findchar().
2445    
2446    
2447    Version 6.3 15-Aug-05
2448    ---------------------
2449    
2450     1. The file libpcre.pc.in did not have general read permission in the tarball.
2451    
2452     2. There were some problems when building without C++ support:
2453    
2454        (a) If C++ support was not built, "make install" and "make test" still
2455            tried to test it.
2456    
2457        (b) There were problems when the value of CXX was explicitly set. Some
2458            changes have been made to try to fix these, and ...
2459    
2460        (c) --disable-cpp can now be used to explicitly disable C++ support.
2461    
2462        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
2463            backslash in a target when C++ was disabled. This confuses some
2464            versions of "make", apparently. Using an intermediate variable solves
2465            this. (Same for CPP_LOBJ.)
2466    
2467     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
2468        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
2469        necessary on certain architectures.
2470    
2471     4. Added a setting of -export-symbols-regex to the link command to remove
2472        those symbols that are exported in the C sense, but actually are local
2473        within the library, and not documented. Their names all begin with
2474        "_pcre_". This is not a perfect job, because (a) we have to except some
2475        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
2476        available (and never for static libraries). I have made a note to try to
2477        find a way round (a) in the future.
2478    
2479    
2480    Version 6.2 01-Aug-05
2481    ---------------------
2482    
2483     1. There was no test for integer overflow of quantifier values. A construction
2484        such as {1111111111111111} would give undefined results. What is worse, if
2485        a minimum quantifier for a parenthesized subpattern overflowed and became
2486        negative, the calculation of the memory size went wrong. This could have
2487        led to memory overwriting.
2488    
2489     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
2490    
2491     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
2492        operating environments where this matters.
2493    
2494     4. Applied Giuseppe Maxia's patch to add additional features for controlling
2495        PCRE options from within the C++ wrapper.
2496    
2497     5. Named capturing subpatterns were not being correctly counted when a pattern
2498        was compiled. This caused two problems: (a) If there were more than 100
2499        such subpatterns, the calculation of the memory needed for the whole
2500        compiled pattern went wrong, leading to an overflow error. (b) Numerical
2501        back references of the form \12, where the number was greater than 9, were
2502        not recognized as back references, even though there were sufficient
2503        previous subpatterns.
2504    
2505     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
2506        versions of gcc, e.g. 2.95.4.
2507    
2508    
2509    Version 6.1 21-Jun-05
2510    ---------------------
2511    
2512     1. There was one reference to the variable "posix" in pcretest.c that was not
2513        surrounded by "#if !defined NOPOSIX".
2514    
2515     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
2516        the cross-check on the old pcre_info() function, for the benefit of the
2517        cut-down version of PCRE that is currently imported into Exim.
2518    
2519     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
2520        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
2521        patterns that start (?i) but I don't think that matters. The use of (?i) is
2522        just an example; this all applies to the other options as well.
2523    
2524     4. Since libtool seems to echo the compile commands it is issuing, the output
2525        from "make" can be reduced a bit by putting "@" in front of each libtool
2526        compile command.
2527    
2528     5. Patch from the folks at Google for configure.in to be a bit more thorough
2529        in checking for a suitable C++ installation before trying to compile the
2530        C++ stuff. This should fix a reported problem when a compiler was present,
2531        but no suitable headers.
2532    
2533     6. The man pages all had just "PCRE" as their title. I have changed them to
2534        be the relevant file name. I have also arranged that these names are
2535        retained in the file doc/pcre.txt, which is a concatenation in text format
2536        of all the man pages except the little individual ones for each function.
2537    
2538     7. The NON-UNIX-USE file had not been updated for the different set of source
2539        files that come with release 6. I also added a few comments about the C++
2540        wrapper.
2541    
2542    
2543    Version 6.0 07-Jun-05
2544    ---------------------
2545    
2546     1. Some minor internal re-organization to help with my DFA experiments.
2547    
2548     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
2549        didn't matter for the library itself when fully configured, but did matter
2550        when compiling without UCP support, or within Exim, where the ucp files are
2551        not imported.
2552    
2553     3. Refactoring of the library code to split up the various functions into
2554        different source modules. The addition of the new DFA matching code (see
2555        below) to a single monolithic source would have made it really too
2556        unwieldy, quite apart from causing all the code to be include in a
2557        statically linked application, when only some functions are used. This is
2558        relevant even without the DFA addition now that patterns can be compiled in
2559        one application and matched in another.
2560    
2561        The downside of splitting up is that there have to be some external
2562        functions and data tables that are used internally in different modules of
2563        the library but which are not part of the API. These have all had their
2564        names changed to start with "_pcre_" so that they are unlikely to clash
2565        with other external names.
2566    
2567     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
2568        a different (DFA) algorithm. Although it is slower than the original
2569        function, it does have some advantages for certain types of matching
2570        problem.
2571    
2572     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
2573        including restarting after a partial match.
2574    
2575     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
2576        defined when compiling for Windows was sent to me. I have put it into the
2577        code, though I have no means of testing or verifying it.
2578    
2579     7. Added the pcre_refcount() auxiliary function.
2580    
2581     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
2582        match before or at the first newline in the subject string. In pcretest,
2583        the /f option on a pattern can be used to set this.
2584    
2585     9. A repeated \w when used in UTF-8 mode with characters greater than 256
2586        would behave wrongly. This has been present in PCRE since release 4.0.
2587    
2588    10. A number of changes to the pcregrep command:
2589    
2590        (a) Refactored how -x works; insert ^(...)$ instead of setting
2591            PCRE_ANCHORED and checking the length, in preparation for adding
2592            something similar for -w.
2593    
2594        (b) Added the -w (match as a word) option.
2595    
2596        (c) Refactored the way lines are read and buffered so as to have more
2597            than one at a time available.
2598    
2599        (d) Implemented a pcregrep test script.
2600    
2601        (e) Added the -M (multiline match) option. This allows patterns to match
2602            over several lines of the subject. The buffering ensures that at least
2603            8K, or the rest of the document (whichever is the shorter) is available
2604            for matching (and similarly the previous 8K for lookbehind assertions).
2605    
2606        (f) Changed the --help output so that it now says
2607    
2608              -w, --word-regex(p)
2609    
2610            instead of two lines, one with "regex" and the other with "regexp"
2611            because that confused at least one person since the short forms are the
2612            same. (This required a bit of code, as the output is generated
2613            automatically from a table. It wasn't just a text change.)
2614    
2615        (g) -- can be used to terminate pcregrep options if the next thing isn't an
2616            option but starts with a hyphen. Could be a pattern or a path name
2617            starting with a hyphen, for instance.
2618    
2619        (h) "-" can be given as a file name to represent stdin.
2620    
2621        (i) When file names are being printed, "(standard input)" is used for
2622            the standard input, for compatibility with GNU grep. Previously
2623            "<stdin>" was used.
2624    
2625        (j) The option --label=xxx can be used to supply a name to be used for
2626            stdin when file names are being printed. There is no short form.
2627    
2628        (k) Re-factored the options decoding logic because we are going to add
2629            two more options that take data. Such options can now be given in four
2630            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
2631    
2632        (l) Added the -A, -B, and -C options for requesting that lines of context
2633            around matches be printed.
2634    
2635        (m) Added the -L option to print the names of files that do not contain
2636            any matching lines, that is, the complement of -l.
2637    
2638        (n) The return code is 2 if any file cannot be opened, but pcregrep does
2639            continue to scan other files.
2640    
2641        (o) The -s option was incorrectly implemented. For compatibility with other
2642            greps, it now suppresses the error message for a non-existent or non-
2643            accessible file (but not the return code). There is a new option called
2644            -q that suppresses the output of matching lines, which was what -s was
2645            previously doing.
2646    
2647        (p) Added --include and --exclude options to specify files for inclusion
2648            and exclusion when recursing.
2649    
2650    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
2651        Hopefully, it now does.
2652    
2653    12. Missing cast in pcre_study().
2654    
2655    13. Added an "uninstall" target to the makefile.
2656    
2657    14. Replaced "extern" in the function prototypes in Makefile.in with
2658        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
2659        world, but is set differently for Windows.
2660    
2661    15. Added a second compiling function called pcre_compile2(). The only
2662        difference is that it has an extra argument, which is a pointer to an
2663        integer error code. When there is a compile-time failure, this is set
2664        non-zero, in addition to the error test pointer being set to point to an
2665        error message. The new argument may be NULL if no error number is required
2666        (but then you may as well call pcre_compile(), which is now just a
2667        wrapper). This facility is provided because some applications need a
2668        numeric error indication, but it has also enabled me to tidy up the way
2669        compile-time errors are handled in the POSIX wrapper.
2670    
2671    16. Added VPATH=.libs to the makefile; this should help when building with one
2672        prefix path and installing with another. (Or so I'm told by someone who
2673        knows more about this stuff than I do.)
2674    
2675    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
2676        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
2677        match everything, including newlines. This is not POSIX-compatible, but
2678        somebody wanted the feature. From pcretest it can be activated by using
2679        both the P and the s flags.
2680    
2681    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
2682    
2683    19. libpcre.pc was being incorrectly installed as executable.
2684    
2685    20. A couple of places in pcretest check for end-of-line by looking for '\n';
2686        it now also looks for '\r' so that it will work unmodified on Windows.
2687    
2688    21. Added Google's contributed C++ wrapper to the distribution.
2689    
2690    22. Added some untidy missing memory free() calls in pcretest, to keep
2691        Electric Fence happy when testing.
2692    
2693    
2694    
2695    Version 5.0 13-Sep-04
2696    ---------------------
2697    
2698     1. Internal change: literal characters are no longer packed up into items
2699        containing multiple characters in a single byte-string. Each character
2700        is now matched using a separate opcode. However, there may be more than one
2701        byte in the character in UTF-8 mode.
2702    
2703     2. The pcre_callout_block structure has two new fields: pattern_position and
2704        next_item_length. These contain the offset in the pattern to the next match
2705        item, and its length, respectively.
2706    
2707     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
2708        insertion of callouts before each pattern item. Added the /C option to
2709        pcretest to make use of this.
2710    
2711     4. On the advice of a Windows user, the lines
2712    
2713          #if defined(_WIN32) || defined(WIN32)
2714          _setmode( _fileno( stdout ), 0x8000 );
2715          #endif  /* defined(_WIN32) || defined(WIN32) */
2716    
2717        have been added to the source of pcretest. This apparently does useful
2718        magic in relation to line terminators.
2719    
2720     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
2721        for the benefit of those environments where the "b" makes a difference.
2722    
2723     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
2724        to know about it. I have put a hack into configure.in that adds in code
2725        to set GCC=yes if CC=icc. This seems to end up at a point in the
2726        generated configure script that is early enough to affect the setting of
2727        compiler options, which is what is needed, but I have no means of testing
2728        whether it really works. (The user who reported this had patched the
2729        generated configure script, which of course I cannot do.)
2730    
2731        LATER: After change 22 below (new libtool files), the configure script
2732        seems to know about icc (and also ecc). Therefore, I have commented out
2733        this hack in configure.in.
2734    
2735     7. Added support for pkg-config (2 patches were sent in).
2736    
2737     8. Negated POSIX character classes that used a combination of internal tables
2738        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2739        [[:^ascii]]. Typically, they would match almost any characters. The other
2740        POSIX classes were not broken in this way.
2741    
2742     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2743        to find the match, as PCRE was deluded into thinking that the match had to
2744        start at the start point or following a newline. The same bug applied to
2745        patterns with negative forward assertions or any backward assertions
2746        preceding ".*" at the start, unless the pattern required a fixed first
2747        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2748    
2749    10. In UTF-8 mode, when moving forwards in the subject after a failed match
2750        starting at the last subject character, bytes beyond the end of the subject
2751        string were read.
2752    
2753    11. Renamed the variable "class" as "classbits" to make life easier for C++
2754        users. (Previously there was a macro definition, but it apparently wasn't
2755        enough.)
2756    
2757    12. Added the new field "tables" to the extra data so that tables can be passed
2758        in at exec time, or the internal tables can be re-selected. This allows
2759        a compiled regex to be saved and re-used at a later time by a different
2760        program that might have everything at different addresses.
2761    
2762    13. Modified the pcre-config script so that, when run on Solaris, it shows a
2763        -R library as well as a -L library.
2764    
2765    14. The debugging options of pcretest (-d on the command line or D on a
2766        pattern) showed incorrect output for anything following an extended class
2767        that contained multibyte characters and which was followed by a quantifier.
2768    
2769    15. Added optional support for general category Unicode character properties
2770        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2771        support. It adds about 90K to the size of the library. The meanings of the
2772        inbuilt class escapes such as \d and \s have NOT been changed.
2773    
2774    16. Updated pcredemo.c to include calls to free() to release the memory for the
2775        compiled pattern.
2776    
2777    17. The generated file chartables.c was being created in the source directory
2778        instead of in the building directory. This caused the build to fail if the
2779        source directory was different from the building directory, and was
2780        read-only.
2781    
2782    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2783        file. No doubt somebody will tell me if they don't make sense... Also added
2784        Dan Mooney's comments about building on OpenVMS.
2785    
2786    19. Added support for partial matching via the PCRE_PARTIAL option for
2787        pcre_exec() and the \P data escape in pcretest.
2788    
2789    20. Extended pcretest with 3 new pattern features:
2790    
2791        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
2792              write the compiled pattern to the file whose name is "rest-of-line".
2793              This is a straight binary dump of the data, with the saved pointer to
2794              the character tables forced to be NULL. The study data, if any, is
2795              written too. After writing, pcretest reads a new pattern.
2796    
2797        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2798              compiled pattern from the given file. There must not be any
2799              occurrences of "<" in the file name (pretty unlikely); if there are,
2800              pcretest will instead treat the initial "<" as a pattern delimiter.
2801              After reading in the pattern, pcretest goes on to read data lines as
2802              usual.
2803    
2804        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2805              and 16-bit fields in a compiled pattern, to simulate a pattern that
2806              was compiled on a host of opposite endianness.
2807    
2808    21. The pcre-exec() function can now cope with patterns that were compiled on
2809        hosts of opposite endianness, with this restriction:
2810    
2811          As for any compiled expression that is saved and used later, the tables
2812          pointer field cannot be preserved; the extra_data field in the arguments
2813          to pcre_exec() should be used to pass in a tables address if a value
2814          other than the default internal tables were used at compile time.
2815    
2816    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2817        now diagnosed as an error. Previously, most of the time, a negative number
2818        would have been treated as zero, but if in addition "ovector" was passed as
2819        NULL, a crash could occur.
2820    
2821    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2822        new versions from the libtool 1.5 distribution (the last one is a copy of
2823        a file called libtool.m4). This seems to have fixed the need to patch
2824        "configure" to support Darwin 1.3 (which I used to do). However, I still
2825        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2826        workstation).
2827    
2828    24. Changed the PCRE licence to be the more standard "BSD" licence.
2829    
2830    
2831    Version 4.5 01-Dec-03
2832    ---------------------
2833    
2834     1. There has been some re-arrangement of the code for the match() function so
2835        that it can be compiled in a version that does not call itself recursively.
2836        Instead, it keeps those local variables that need separate instances for
2837        each "recursion" in a frame on the heap, and gets/frees frames whenever it
2838        needs to "recurse". Keeping track of where control must go is done by means
2839        of setjmp/longjmp. The whole thing is implemented by a set of macros that
2840        hide most of the details from the main code, and operates only if
2841        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2842        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2843        operating.
2844    
2845        To make it easier for callers to provide specially tailored get/free
2846        functions for this usage, two new functions, pcre_stack_malloc, and
2847        pcre_stack_free, are used. They are always called in strict stacking order,
2848        and the size of block requested is always the same.
2849    
2850        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2851        PCRE has been compiled to use the stack or the heap for recursion. The
2852        -C option of pcretest uses this to show which version is compiled.
2853    
2854        A new data escape \S, is added to pcretest; it causes the amounts of store
2855        obtained and freed by both kinds of malloc/free at match time to be added
2856        to the output.
2857    
2858     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2859        what's available on my current Linux desktop machine.
2860    
2861     3. When matching a UTF-8 string, the test for a valid string at the start has
2862        been extended. If start_offset is not zero, PCRE now checks that it points
2863        to a byte that is the start of a UTF-8 character. If not, it returns
2864        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2865        this is necessary because there may be backward assertions in the pattern.
2866        When matching the same subject several times, it may save resources to use
2867        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2868    
2869     4. The code for checking the validity of UTF-8 strings has been tightened so
2870        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2871        containing "overlong sequences".
2872    
2873     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2874        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2875        should have been "&", but it just so happened that all the cases this let
2876        through by mistake were picked up later in the function.
2877    
2878     6. I had used a variable called "isblank" - this is a C99 function, causing
2879        some compilers to warn. To avoid this, I renamed it (as "blankclass").
2880    
2881     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2882        prompting; (b) run "./pcretest /dev/null" at the start of the test script
2883        so the version is shown; (c) stop "make test" echoing "./RunTest".
2884    
2885     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2886    
2887     9. The prototype for memmove() for systems that don't have it was using
2888        size_t, but the inclusion of the header that defines size_t was later. I've
2889        moved the #includes for the C headers earlier to avoid this.
2890    
2891    10. Added some adjustments to the code to make it easier to compiler on certain
2892        special systems:
2893    
2894          (a) Some "const" qualifiers were missing.
2895          (b) Added the macro EXPORT before all exported functions; by default this
2896              is defined to be empty.
2897          (c) Changed the dftables auxiliary program (that builds chartables.c) so
2898              that it reads its output file name as an argument instead of writing
2899              to the standard output and assuming this can be redirected.
2900    
2901    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2902        class containing characters with values greater than 255, PCRE compilation
2903        went into a loop.
2904    
2905    12. A recursive reference to a subpattern that was within another subpattern
2906        that had a minimum quantifier of zero caused PCRE to crash. For example,
2907        (x(y(?2))z)? provoked this bug with a subject that got as far as the
2908        recursion. If the recursively-called subpattern itself had a zero repeat,
2909        that was OK.
2910    
2911    13. In pcretest, the buffer for reading a data line was set at 30K, but the
2912        buffer into which it was copied (for escape processing) was still set at
2913        1024, so long lines caused crashes.
2914    
2915    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2916        "internal error: code overflow...". This applied to any character class
2917        that was followed by a possessive quantifier.
2918    
2919    15. Modified the Makefile to add libpcre.la as a prerequisite for
2920        libpcreposix.la because I was told this is needed for a parallel build to
2921        work.
2922    
2923    16. If a pattern that contained .* following optional items at the start was
2924        studied, the wrong optimizing data was generated, leading to matching
2925        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2926        matching string must start with a or b or c. The correct conclusion for
2927        this pattern is that a match can start with any character.
2928    
2929    
2930    Version 4.4 13-Aug-03
2931    ---------------------
2932    
2933     1. In UTF-8 mode, a character class containing characters with values between
2934        127 and 255 was not handled correctly if the compiled pattern was studied.
2935        In fixing this, I have also improved the studying algorithm for such
2936        classes (slightly).
2937    
2938     2. Three internal functions had redundant arguments passed to them. Removal
2939        might give a very teeny performance improvement.
2940    
2941     3. Documentation bug: the value of the capture_top field in a callout is *one
2942        more than* the number of the hightest numbered captured substring.
2943    
2944     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2945        in incorrectly linking with a previously installed version. They now link
2946        explicitly with libpcre.la.
2947    
2948     5. configure.in no longer needs to recognize Cygwin specially.
2949    
2950     6. A problem in pcre.in for Windows platforms is fixed.
2951    
2952     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2953        pcretest, it used to include the size of the study block as part of its
2954        output. Unfortunately, the structure contains a field that has a different
2955        size on different hardware architectures. This meant that the tests that
2956        showed this size failed. As the block is currently always of a fixed size,
2957        this information isn't actually particularly useful in pcretest output, so
2958        I have just removed it.
2959    
2960     8. Three pre-processor statements accidentally did not start in column 1.
2961        Sadly, there are *still* compilers around that complain, even though
2962        standard C has not required this for well over a decade. Sigh.
2963    
2964     9. In pcretest, the code for checking callouts passed small integers in the
2965        callout_data field, which is a void * field. However, some picky compilers
2966        complained about the casts involved for this on 64-bit systems. Now
2967        pcretest passes the address of the small integer instead, which should get
2968        rid of the warnings.
2969    
2970    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2971        both compile and run time, and gives an error if an invalid UTF-8 sequence
2972        is found. There is a option for disabling this check in cases where the
2973        string is known to be correct and/or the maximum performance is wanted.
2974    
2975    11. In response to a bug report, I changed one line in Makefile.in from
2976    
2977            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2978        to
2979            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2980    
2981        to look similar to other lines, but I have no way of telling whether this
2982        is the right thing to do, as I do not use Windows. No doubt I'll get told
2983        if it's wrong...
2984    
2985    
2986    Version 4.3 21-May-03
2987    ---------------------
2988    
2989    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2990       Makefile.
2991    
2992    2. Some refactoring to improve the quality of the code:
2993    
2994       (i)   The utf8_table... variables are now declared "const".
2995    
2996       (ii)  The code for \cx, which used the "case flipping" table to upper case
2997             lower case letters, now just substracts 32. This is ASCII-specific,
2998             but the whole concept of \cx is ASCII-specific, so it seems
2999             reasonable.
3000    
3001       (iii) PCRE was using its character types table to recognize decimal and
3002             hexadecimal digits in the pattern. This is silly, because it handles
3003             only 0-9, a-f, and A-F, but the character types table is locale-
3004             specific, which means strange things might happen. A private
3005             table is now used for this - though it costs 256 bytes, a table is
3006             much faster than multiple explicit tests. Of course, the standard
3007             character types table is still used for matching digits in subject
3008             strings against \d.
3009    
3010       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
3011             ending in _t are). So I've renamed it as ESC_tee.
3012    
3013    3. The first argument for regexec() in the POSIX wrapper should have been
3014       defined as "const".
3015    
3016    4. Changed pcretest to use malloc() for its buffers so that they can be
3017       Electric Fenced for debugging.
3018    
3019    5. There were several places in the code where, in UTF-8 mode, PCRE would try
3020       to read one or more bytes before the start of the subject string. Often this
3021       had no effect on PCRE's behaviour, but in some circumstances it could
3022       provoke a segmentation fault.
3023    
3024    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
3025       to try to read one or more bytes before the start of the subject string.
3026    
3027    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
3028       UTF-8 support could misbehave in various ways if the subject string
3029       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
3030       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
3031       back over UTF-8 characters.)
3032    
3033    
3034    Version 4.2 14-Apr-03
3035    ---------------------
3036    
3037    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
3038    
3039    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
3040         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
3041         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
3042         [WIN_PREFIX]: new variable, "cyg" for Cygwin
3043         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
3044           and BUILD_EXEEXT
3045         Note: automatic setting of the BUILD variables is not yet working
3046         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
3047           compile-time but not at link-time
3048         [LINK]: use for linking executables only
3049         make different versions for Windows and non-Windows
3050         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
3051           libraries
3052         [LINK_FOR_BUILD]: new variable
3053         [OBJEXT]: use throughout
3054         [EXEEXT]: use throughout
3055         <winshared>: new target
3056         <wininstall>: new target
3057         <dftables.o>: use native compiler
3058         <dftables>: use native linker
3059         <install>: handle Windows platform correctly
3060         <clean>: ditto
3061         <check>: ditto
3062         copy DLL to top builddir before testing
3063    
3064       As part of these changes, -no-undefined was removed again. This was reported
3065       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
3066       in any case.
3067    
3068    3. Some tidies to get rid of compiler warnings:
3069    
3070       . In the match_data structure, match_limit was an unsigned long int, whereas
3071         match_call_count was an int. I've made them both unsigned long ints.
3072    
3073       . In pcretest the fact that a const uschar * doesn't automatically cast to
3074         a void * provoked a warning.
3075    
3076       . Turning on some more compiler warnings threw up some "shadow" variables
3077         and a few more missing casts.
3078    
3079    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
3080       option, a class that contained a single character with a value between 128
3081       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
3082    
3083    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
3084       option, a class that contained several characters, but with at least one
3085       whose value was between 128 and 255 caused PCRE to crash.
3086    
3087    
3088    Version 4.1 12-Mar-03
3089    ---------------------
3090    
3091    1. Compiling with gcc -pedantic found a couple of places where casts were
3092    needed, and a string in dftables.c that was longer than standard compilers are
3093    required to support.
3094    
3095    2. Compiling with Sun's compiler found a few more places where the code could
3096    be tidied up in order to avoid warnings.
3097    
3098    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
3099    first of these names is deprecated in the latest Autoconf in favour of the name
3100    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
3101    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
3102    analogy I have changed it to CFLAGS_FOR_BUILD.
3103    
3104    4. Added -no-undefined to the linking command in the Makefile, because this is
3105    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
3106    linking step for the pcreposix library.
3107    
3108    5. PCRE was failing to diagnose the case of two named groups with the same
3109    name.
3110    
3111    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
3112    literal character that is needed in the subject for a match, and scans along to
3113    ensure that it is present before embarking on the full matching process. This
3114    saves time in cases of nested unlimited repeats that are never going to match.
3115    Problem: the scan can take a lot of time if the subject is very long (e.g.
3116    megabytes), thus penalizing straightforward matches. It is now done only if the
3117    amount of subject to be scanned is less than 1000 bytes.
3118    
3119    7. A lesser problem with the same optimization is that it was recording the
3120    first character of an anchored pattern as "needed", thus provoking a search
3121    right along the subject, even when the first match of the pattern was going to
3122    fail. The "needed" character is now not set for anchored patterns, unless it
3123    follows something in the pattern that is of non-fixed length. Thus, it still
3124    fulfils its original purpose of finding quick non-matches in cases of nested
3125    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
3126    
3127    
3128    Version 4.0 17-Feb-03
3129    ---------------------
3130    
3131    1. If a comment in an extended regex that started immediately after a meta-item
3132    extended to the end of string, PCRE compiled incorrect data. This could lead to
3133    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
3134    
3135    2. Moved to autoconf 2.53 and libtool 1.4.2.
3136    
3137    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
3138    the special perltest8 script is no longer needed - all the tests can be run
3139    from a single perltest script.
3140    
3141    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
3142    by \s. It has now been removed in PCRE. This means it isn't recognized as
3143    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
3144    class [:space:] *does* include VT, thereby creating a mess.
3145    
3146    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
3147    space and tab.
3148    
3149    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
3150    its new features into the main test script, reducing the number of scripts.
3151    
3152    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
3153    were backward compatible, and made the (?i) apply to the whole pattern, as if
3154    /i were given. Now it behaves more logically, and applies the option setting
3155    only to what follows. PCRE has been changed to follow suit. However, if it
3156    finds options settings right at the start of the pattern, it extracts them into
3157    the global options, as before. Thus, they show up in the info data.
3158    
3159    8. Added support for the \Q...\E escape sequence. Characters in between are
3160    treated as literals. This is slightly different from Perl in that $ and @ are
3161    also handled as literals inside the quotes. In Perl, they will cause variable
3162    interpolation. Note the following examples:
3163    
3164        Pattern            PCRE matches      Perl matches
3165    
3166        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
3167        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
3168        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
3169    
3170    For compatibility with Perl, \Q...\E sequences are recognized inside character
3171    classes as well as outside them.
3172    
3173    9. Re-organized 3 code statements in pcretest to avoid "overflow in
3174    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
3175    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
3176    signed/unsigned warnings.
3177    
3178    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
3179    option for pcretest, so I've replaced it by a simple function that does just
3180    that job.
3181    
3182    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
3183    "pcregrep -".
3184    
3185    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
3186    Java package. This provides some syntactic sugar for simple cases of what my
3187    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
3188    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
3189    item, you can use this simplified notation. Note that only makes sense with
3190    greedy quantifiers. Consequently, the use of the possessive quantifier forces
3191    greediness, whatever the setting of the PCRE_UNGREEDY option.
3192    
3193    13. A change of greediness default within a pattern was not taking effect at
3194    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
3195    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
3196    was abstracted outside.
3197    
3198    14. PCRE now supports the \G assertion. It is true when the current matching
3199    position is at the start point of the match. This differs from \A when the
3200    starting offset is non-zero. Used with the /g option of pcretest (or similar
3201    code), it works in the same way as it does for Perl's /g option. If all
3202    alternatives of a regex begin with \G, the expression is anchored to the start
3203    match position, and the "anchored" flag is set in the compiled expression.
3204    
3205    15. Some bugs concerning the handling of certain option changes within patterns
3206    have been fixed. These applied to options other than (?ims). For example,
3207    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
3208    been the other way round. Some of this was related to change 7 above.
3209    
3210    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
3211    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
3212    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
3213    POSIX classes only within a class (e.g. /[[:alpha:]]/).
3214    
3215    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
3216    mode. Unlike ".", it always matches newline, whatever the setting of
3217    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
3218    assertions. Perl allows it, but it doesn't (in general) work because it can't
3219    calculate the length of the lookbehind. At least, that's the case for Perl
3220    5.8.0 - I've been told they are going to document that it doesn't work in
3221    future.
3222    
3223    18. Added an error diagnosis for escapes that PCRE does not support: these are
3224    \L, \l, \N, \P, \p, \U, \u, and \X.
3225    
3226    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
3227    reading past the end of the pattern in cases such as /[abcd/.
3228    
3229    20. PCRE was getting more memory than necessary for patterns with classes that
3230    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
3231    
3232    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
3233    compiling PCRE for use with Virtual Pascal.
3234    
3235    22. Small fix to the Makefile to make it work properly if the build is done
3236    outside the source tree.
3237    
3238    23. Added a new extension: a condition to go with recursion. If a conditional
3239    subpattern starts with (?(R) the "true" branch is used if recursion has
3240    happened, whereas the "false" branch is used only at the top level.
3241    
3242    24. When there was a very long string of literal characters (over 255 bytes
3243    without UTF support, over 250 bytes with UTF support), the computation of how
3244    much memory was required could be incorrect, leading to segfaults or other
3245    strange effects.
3246    
3247    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
3248    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
3249    there was a subsequent back reference to those brackets. This meant that, for
3250    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
3251    possible to check for precisely this case. All we can do is abandon the
3252    optimization if .* occurs inside capturing brackets when there are any back
3253    references whatsoever. (See below for a better fix that came later.)
3254    
3255    26. The handling of the optimization for finding the first character of a
3256    non-anchored pattern, and for finding a character that is required later in the
3257    match were failing in some cases. This didn't break the matching; it just
3258    failed to optimize when it could. The way this is done has been re-implemented.
3259    
3260    27. Fixed typo in error message for invalid (?R item (it said "(?p").
3261    
3262    28. Added a new feature that provides some of the functionality that Perl
3263    provides with (?{...}). The facility is termed a "callout". The way it is done
3264    in PCRE is for the caller to provide an optional function, by setting
3265    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
3266    global variable. By default it is unset, which disables all calling out. To get
3267    the function called, the regex must include (?C) at appropriate points. This
3268    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
3269    This provides a means of identifying different callout points. When PCRE
3270    reaches such a point in the regex, if pcre_callout has been set, the external
3271    function is called. It is provided with data in a structure called
3272    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
3273    matching continues; if it returns a non-zero value, the match at the current
3274    point fails. However, backtracking will occur if possible. [This was changed
3275    later and other features added - see item 49 below.]
3276    
3277    29. pcretest is upgraded to test the callout functionality. It provides a
3278    callout function that displays information. By default, it shows the start of
3279    the match and the current position in the text. There are some new data escapes
3280    to vary what happens:
3281    
3282        \C+         in addition, show current contents of captured substrings
3283        \C-         do not supply a callout function
3284        \C!n        return 1 when callout number n is reached
3285        \C!n!m      return 1 when callout number n is reached for the mth time
3286    
3287    30. If pcregrep was called with the -l option and just a single file name, it
3288    output "<stdin>" if a match was found, instead of the file name.
3289    
3290    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
3291    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
3292    pcre_exec(). This saves a malloc/free per call. The default value of
3293    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
3294    when configuring.
3295    
3296    32. The default maximum size of a compiled pattern is 64K. There have been a
3297    few cases of people hitting this limit. The code now uses macros to handle the
3298    storing of links as offsets within the compiled pattern. It defaults to 2-byte
3299    links, but this can be changed to 3 or 4 bytes by --with-link-size when
3300    configuring. Tests 2 and 5 work only with 2-byte links because they output
3301    debugging information about compiled patterns.
3302    
3303    33. Internal code re-arrangements:
3304    
3305    (a) Moved the debugging function for printing out a compiled regex into
3306        its own source file (printint.c) and used #include to pull it into
3307        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
3308        separate copies.
3309    
3310    (b) Defined the list of op-code names for debugging as a macro in
3311        internal.h so that it is next to the definition of the opcodes.
3312    
3313    (c) Defined a table of op-code lengths for simpler skipping along compiled
3314        code. This is again a macro in internal.h so that it is next to the
3315        definition of the opcodes.
3316    
3317    34. Added support for recursive calls to individual subpatterns, along the
3318    lines of Robin Houston's patch (but implemented somewhat differently).
3319    
3320    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
3321    allow it to read and process whole directories in Win32. This code was
3322    contributed by Lionel Fourquaux; it has not been tested by me.
3323    
3324    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
3325    used to name a group. Names consist of alphanumerics and underscores, and must
3326    be unique. Back references use the syntax (?P=name) and recursive calls use
3327    (?P>name) which is a PCRE extension to the Python extension. Groups still have
3328    numbers. The function pcre_fullinfo() can be used after compilation to extract
3329    a name/number map. There are three relevant calls:
3330    
3331      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
3332      PCRE_INFO_NAMECOUNT            yields the number of entries
3333      PCRE_INFO_NAMETABLE            yields a pointer to the map.
3334    
3335    The map is a vector of fixed-size entries. The size of each entry depends on
3336    the length of the longest name used. The first two bytes of each entry are the
3337    group number, most significant byte first. There follows the corresponding
3338    name, zero terminated. The names are in alphabetical order.
3339    
3340    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
3341    case instead of 255. Making it the same both with and without UTF-8 support
3342    means that the same test output works with both.
3343    
3344    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
3345    calling malloc() with a zero argument.
3346    
3347    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
3348    optimization. I've improved things by keeping a bitmap of backreferences with
3349    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
3350    fact referenced, the optimization can be applied. It is unlikely that a
3351    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
3352    the match to follow \n) will appear inside brackets with a number greater than
3353    31, but if it does, any back reference > 31 suppresses the optimization.
3354    
3355    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
3356    of disabling numbered capturing parentheses. Any opening parenthesis that is
3357    not followed by ? behaves as if it were followed by ?: but named parentheses
3358    can still be used for capturing (and they will acquire numbers in the usual
3359    way).
3360    
3361    41. Redesigned the return codes from the match() function into yes/no/error so
3362    that errors can be passed back from deep inside the nested calls. A malloc
3363    failure while inside a recursive subpattern call now causes the
3364    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
3365    
3366    42. It is now possible to set a limit on the number of times the match()
3367    function is called in a call to pcre_exec(). This facility makes it possible to
3368    limit the amount of recursion and backtracking, though not in a directly
3369    obvious way, because the match() function is used in a number of different
3370    circumstances. The count starts from zero for each position in the subject
3371    string (for non-anchored patterns). The default limit is, for compatibility, a
3372    large number, namely 10 000 000. You can change this in two ways:
3373    
3374    (a) When configuring PCRE before making, you can use --with-match-limit=n
3375        to set a default value for the compiled library.
3376    
3377    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
3378        a different value is set. See 45 below.
3379    
3380    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
3381    
3382    43. Added a new function pcre_config(int, void *) to enable run-time extraction
3383    of things that can be changed at compile time. The first argument specifies
3384    what is wanted and the second points to where the information is to be placed.
3385    The current list of available information is:
3386    
3387      PCRE_CONFIG_UTF8
3388    
3389    The output is an integer that is set to one if UTF-8 support is available;
3390    otherwise it is set to zero.
3391    
3392      PCRE_CONFIG_NEWLINE
3393    
3394    The output is an integer that it set to the value of the code that is used for
3395    newline. It is either LF (10) or CR (13).
3396    
3397      PCRE_CONFIG_LINK_SIZE
3398    
3399    The output is an integer that contains the number of bytes used for internal
3400    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
3401    
3402      PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
3403    
3404    The output is an integer that contains the threshold above which the POSIX
3405    interface uses malloc() for output vectors. See item 31 above.
3406    
3407      PCRE_CONFIG_MATCH_LIMIT
3408    
3409    The output is an unsigned integer that contains the default limit of the number
3410    of match() calls in a pcre_exec() execution. See 42 above.
3411    
3412    44. pcretest has been upgraded by the addition of the -C option. This causes it
3413    to extract all the available output from the new pcre_config() function, and to
3414    output it. The program then exits immediately.
3415    
3416    45. A need has arisen to pass over additional data with calls to pcre_exec() in
3417    order to support additional features. One way would have been to define
3418    pcre_exec2() (for example) with extra arguments, but this would not have been
3419    extensible, and would also have required all calls to the original function to
3420    be mapped to the new one. Instead, I have chosen to extend the mechanism that
3421    is used for passing in "extra" data from pcre_study().
3422    
3423    The pcre_extra structure is now exposed and defined in pcre.h. It currently
3424    contains the following fields:
3425    
3426      flags         a bitmap indicating which of the following fields are set
3427      study_data    opaque data from pcre_study()
3428      match_limit   a way of specifying a limit on match() calls for a specific
3429                      call to pcre_exec()
3430      callout_data  data for callouts (see 49 below)
3431    
3432    The flag bits are also defined in pcre.h, and are
3433    
3434      PCRE_EXTRA_STUDY_DATA
3435      PCRE_EXTRA_MATCH_LIMIT
3436      PCRE_EXTRA_CALLOUT_DATA
3437    
3438    The pcre_study() function now returns one of these new pcre_extra blocks, with
3439    the actual study data pointed to by the study_data field, and the
3440    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
3441    before. That is, this change is entirely upwards-compatible and requires no
3442    change to existing code.
3443    
3444    If you want to pass in additional data to pcre_exec(), you can either place it
3445    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
3446    block.
3447    
3448    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
3449    data string contains the escape sequence \M, pcretest calls pcre_exec() several
3450    times with different match limits, until it finds the minimum value needed for
3451    pcre_exec() to complete. The value is then output. This can be instructive; for
3452    most simple matches the number is quite small, but for pathological cases it
3453    gets very large very quickly.
3454    
3455    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
3456    returns the size of the data block pointed to by the study_data field in a
3457    pcre_extra block, that is, the value that was passed as the argument to
3458    pcre_malloc() when PCRE was getting memory in which to place the information
3459    created by pcre_study(). The fourth argument should point to a size_t variable.
3460    pcretest has been extended so that this information is shown after a successful
3461    pcre_study() call when information about the compiled regex is being displayed.
3462    
3463    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
3464    because what follows is always an absolute path. (Later: it turns out that this
3465    is more than cosmetic for MinGW, because it doesn't like empty path
3466    components.)
3467    
3468    49. Some changes have been made to the callout feature (see 28 above):
3469    
3470    (i)  A callout function now has three choices for what it returns:
3471    
3472           0  =>  success, carry on matching
3473         > 0  =>  failure at this point, but backtrack if possible
3474         < 0  =>  serious error, return this value from pcre_exec()
3475    
3476         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
3477         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
3478         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
3479         use by callout functions. It will never be used by PCRE itself.
3480    
3481    (ii) The pcre_extra structure (see 45 above) has a void * field called
3482         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
3483         pcre_callout_block structure has a field of the same name. The contents of
3484         the field passed in the pcre_extra structure are passed to the callout
3485         function in the corresponding field in the callout block. This makes it
3486         easier to use the same callout-containing regex from multiple threads. For
3487         testing, the pcretest program has a new data escape
3488    
3489           \C*n        pass the number n (may be negative) as callout_data
3490    
3491         If the callout function in pcretest receives a non-zero value as
3492         callout_data, it returns that value.
3493    
3494    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
3495    there were some redundant $(CFLAGS) in commands that are now specified as
3496    $(LINK), which already includes $(CFLAGS).
3497    
3498    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
3499    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
3500    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
3501    one-byte characters throughout. Note that case-insensitive matching applies
3502    only to characters whose values are less than 256. PCRE doesn't support the
3503    notion of cases for higher-valued characters.
3504    
3505    (i)   A character class whose characters are all within 0-255 is handled as
3506          a bit map, and the map is inverted for negative classes. Previously, a
3507          character > 255 always failed to match such a class; however it should
3508          match if the class was a negative one (e.g. [^ab]). This has been fixed.
3509    
3510    (ii)  A negated character class with a single character < 255 is coded as
3511          "not this character" (OP_NOT). This wasn't working properly when the test
3512          character was multibyte, either singly or repeated.
3513    
3514    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
3515          mode, for example: \x{100}{2,3}.
3516    
3517    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
3518          singly or repeated) now correctly test multibyte characters. However,
3519          PCRE doesn't recognize any characters with values greater than 255 as
3520          digits, spaces, or word characters. Such characters always match \D, \S,
3521          and \W, and never match \d, \s, or \w.
3522    
3523    (v)   Classes may now contain characters and character ranges with values
3524          greater than 255. For example: [ab\x{100}-\x{400}].
3525    
3526    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
3527          PCRE in UTF-8 mode.
3528    
3529    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
3530    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
3531    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
3532    value.)
3533    
3534    53. The single man page has become too large. I have therefore split it up into
3535    a number of separate man pages. These also give rise to individual HTML pages;
3536    these are now put in a separate directory, and there is an index.html page that
3537    lists them all. Some hyperlinking between the pages has been installed.
3538    
3539    54. Added convenience functions for handling named capturing parentheses.
3540    
3541    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
3542    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
3543    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
3544    are faulted.
3545    
3546    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
3547    calling configure. These values are used when compiling the dftables.c program
3548    which is run to generate the source of the default character tables. They
3549    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
3550    you will need to set these values.
3551    
3552    57. Updated the building process for Windows DLL, as provided by Fred Cox.
3553    
3554    
3555    Version 3.9 02-Jan-02
3556    ---------------------
3557    
3558    1. A bit of extraneous text had somehow crept into the pcregrep documentation.
3559    
3560    2. If --disable-static was given, the building process failed when trying to
3561    build pcretest and pcregrep. (For some reason it was using libtool to compile
3562    them, which is not right, as they aren't part of the library.)
3563    
3564    
3565    Version 3.8 18-Dec-01
3566    ---------------------
3567    
3568    1. The experimental UTF-8 code was completely screwed up. It was packing the
3569    bytes in the wrong order. How dumb can you get?
3570    
3571    
3572    Version 3.7 29-Oct-01
3573    ---------------------
3574    
3575    1. In updating pcretest to check change 1 of version 3.6, I screwed up.
3576    This caused pcretest, when used on the test data, to segfault. Unfortunately,
3577    this didn't happen under Solaris 8, where I normally test things.
3578    
3579    2. The Makefile had to be changed to make it work on BSD systems, where 'make'
3580    doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
3581    isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
3582    this fix an hour or so after the initial 3.7 release.)
3583    
3584    
3585    Version 3.6 23-Oct-01
3586    ---------------------
3587    
3588    1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
3589    offsets passed as NULL with zero offset count.
3590    
3591    2. The config.guess and config.sub files had not been updated when I moved to
3592    the latest autoconf.
3593    
3594    
3595    Version 3.5 15-Aug-01
3596    ---------------------
3597    
3598    1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
3599    had been forgotten.
3600    
3601    2. By using declared but undefined structures, we can avoid using "void"
3602    definitions in pcre.h while keeping the internal definitions of the structures
3603    private.
3604    
3605    3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
3606    user point of view, this means that both static and shared libraries are built
3607    by default, but this can be individually controlled. More of the work of
3608    handling this static/shared cases is now inside libtool instead of PCRE's make
3609    file.
3610    
3611    4. The pcretest utility is now installed along with pcregrep because it is
3612    useful for users (to test regexs) and by doing this, it automatically gets
3613    relinked by libtool. The documentation has been turned into a man page, so
3614    there are now .1, .txt, and .html versions in /doc.
3615    
3616    5. Upgrades to pcregrep:
3617       (i)   Added long-form option names like gnu grep.
3618       (ii)  Added --help to list all options with an explanatory phrase.
3619       (iii) Added -r, --recursive to recurse into sub-directories.
3620       (iv)  Added -f, --file to read patterns from a file.
3621    
3622    6. pcre_exec() was referring to its "code" argument before testing that
3623    argument for NULL (and giving an error if it was NULL).
3624    
3625    7. Upgraded Makefile.in to allow for compiling in a different directory from
3626    the source directory.
3627    
3628    8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
3629    options bits, the pointer it was passed was to an int instead of to an unsigned
3630    long int. This mattered only on 64-bit systems.
3631    
3632    9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
3633    generated) instead of pcre.in, which it its source. Also made the same change
3634    in several of the .c files.
3635    
3636    10. A new release of gcc defines printf() as a macro, which broke pcretest
3637    because it had an ifdef in the middle of a string argument for printf(). Fixed
3638    by using separate calls to printf().
3639    
3640    11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
3641    script, to force use of CR or LF instead of \n in the source. On non-Unix
3642    systems, the value can be set in config.h.
3643    
3644    12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
3645    absolute limit. Changed the text of the error message to make this clear, and
3646    likewise updated the man page.
3647    
3648    13. The limit of 99 on the number of capturing subpatterns has been removed.
3649    The new limit is 65535, which I hope will not be a "real" limit.
3650    
3651    
3652    Version 3.4 22-Aug-00
3653    ---------------------
3654    
3655    1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
3656    
3657    2. Diagnose condition (?(0) as an error instead of crashing on matching.
3658    
3659    
3660    Version 3.3 01-Aug-00
3661    ---------------------
3662    
3663    1. If an octal character was given, but the value was greater than \377, it
3664    was not getting masked to the least significant bits, as documented. This could
3665    lead to crashes in some systems.
3666    
3667    2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
3668    the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
3669    
3670    3. Added the functions pcre_free_substring() and pcre_free_substring_list().
3671    These just pass their arguments on to (pcre_free)(), but they are provided
3672    because some uses of PCRE bind it to non-C systems that can call its functions,
3673    but cannot call free() or pcre_free() directly.
3674    
3675    4. Add "make test" as a synonym for "make check". Corrected some comments in
3676    the Makefile.
3677    
3678    5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
3679    Makefile.
3680    
3681    6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
3682    command called pgrep for grepping around the active processes.
3683    
3684    7. Added the beginnings of support for UTF-8 character strings.
3685    
3686    8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
3687    RANLIB to ./ltconfig so that they are used by libtool. I think these are all
3688    the relevant ones. (AR is not passed because ./ltconfig does its own figuring
3689    out for the ar command.)
3690    
3691    
3692    Version 3.2 12-May-00
3693    ---------------------
3694    
3695    This is purely a bug fixing release.
3696    
3697    1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
3698    of ZA. This was just one example of several cases that could provoke this bug,
3699    which was introduced by change 9 of version 2.00. The code for breaking
3700    infinite loops after an iteration that matches an empty string was't working
3701    correctly.
3702    
3703    2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
3704    when matched against abbab (for example). After matching an empty string, it
3705    wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
3706    caused it to match further down the string than it should.
3707    
3708    3. The code contained an inclusion of sys/types.h. It isn't clear why this
3709    was there because it doesn't seem to be needed, and it causes trouble on some
3710    systems, as it is not a Standard C header. It has been removed.
3711    
3712    4. Made 4 silly changes to the source to avoid stupid compiler warnings that
3713    were reported on the Macintosh. The changes were from
3714    
3715      while ((c = *(++ptr)) != 0 && c != '\n');
3716    to
3717      while ((c = *(++ptr)) != 0 && c != '\n') ;
3718    
3719    Totally extraordinary, but if that's what it takes...
3720    
3721    5. PCRE is being used in one environment where neither memmove() nor bcopy() is
3722    available. Added HAVE_BCOPY and an autoconf test for it; if neither
3723    HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
3724    assumes the way PCRE uses memmove() (always moving upwards).
3725    
3726    6. PCRE is being used in one environment where strchr() is not available. There
3727    was only one use in pcre.c, and writing it out to avoid strchr() probably gives
3728    faster code anyway.
3729    
3730    
3731    Version 3.1 09-Feb-00
3732    ---------------------
3733    
3734    The only change in this release is the fixing of some bugs in Makefile.in for
3735    the "install" target:
3736    
3737    (1) It was failing to install pcreposix.h.
3738    
3739    (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
3740    
3741    
3742    Version 3.0 01-Feb-00
3743    ---------------------
3744    
3745    1. Add support for the /+ modifier to perltest (to output $` like it does in
3746    pcretest).
3747    
3748    2. Add support for the /g modifier to perltest.
3749    
3750    3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
3751    matches null strings.
3752    
3753    4. Fix perltest so that it doesn't do unwanted things when fed an empty
3754    pattern. Perl treats empty patterns specially - it reuses the most recent
3755    pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
3756    effect.
3757    
3758    5. The POSIX interface was broken in that it was just handing over the POSIX
3759    captured string vector to pcre_exec(), but (since release 2.00) PCRE has
3760    required a bigger vector, with some working space on the end. This means that
3761    the POSIX wrapper now has to get and free some memory, and copy the results.
3762    
3763    6. Added some simple autoconf support, placing the test data and the
3764    documentation in separate directories, re-organizing some of the
3765    information files, and making it build pcre-config (a GNU standard). Also added
3766    libtool support for building PCRE as a shared library, which is now the
3767    default.
3768    
3769    7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
3770    09 are not valid octal constants. Single digits will be used for minor values
3771    less than 10.
3772    
3773    8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
3774    existing programs that set these in the POSIX interface can use PCRE without
3775    modification.
3776    
3777    9. Added a new function, pcre_fullinfo() with an extensible interface. It can
3778    return all that pcre_info() returns, plus additional data. The pcre_info()
3779    function is retained for compatibility, but is considered to be obsolete.
3780    
3781    10. Added experimental recursion feature (?R) to handle one common case that
3782    Perl 5.6 will be able to do with (?p{...}).
3783    
3784    11. Added support for POSIX character classes like [:alpha:], which Perl is
3785    adopting.
3786    
3787    
3788    Version 2.08 31-Aug-99
3789    ----------------------
3790    
3791    1. When startoffset was not zero and the pattern began with ".*", PCRE was not
3792    trying to match at the startoffset position, but instead was moving forward to
3793    the next newline as if a previous match had failed.
3794    
3795    2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
3796    and could get into a loop if a null string was matched other than at the start
3797    of the subject.
3798    
3799    3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
3800    be distinguished at compile time, and for completeness also added PCRE_DATE.
3801    
3802    5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
3803    in GnuWin32 environments.
3804    
3805    
3806    Version 2.07 29-Jul-99
3807    ----------------------
3808    
3809    1. The documentation is now supplied in plain text form and HTML as well as in
3810    the form of man page sources.
3811    
3812    2. C++ compilers don't like assigning (void *) values to other pointer types.
3813    In particular this affects malloc(). Although there is no problem in Standard
3814    C, I've put in casts to keep C++ compilers happy.
3815    
3816    3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
3817    should be (const char *).
3818    
3819    4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
3820    be useful for non-Unix systems who don't want to bother with the POSIX stuff.
3821    However, I haven't made this a standard facility. The documentation doesn't
3822    mention it, and the Makefile doesn't support it.
3823    
3824    5. The Makefile now contains an "install" target, with editable destinations at
3825    the top of the file. The pcretest program is not installed.
3826    
3827    6. pgrep -V now gives the PCRE version number and date.