/[pcre]/code/branches/pcre16/ChangeLog
ViewVC logotype

Diff of /code/branches/pcre16/ChangeLog

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

revision 5 by nigel, Sat Feb 24 21:38:05 2007 UTC revision 640 by ph10, Mon Jul 25 10:50:28 2011 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    Version 8.13 30-Apr-2011
5    ------------------------
6    
7    1.  The Unicode data tables have been updated to Unicode 6.0.0.
8    
9    2.  Two minor typos in pcre_internal.h have been fixed.
10    
11    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
12        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
13        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
14    
15    4.  There were a number of related bugs in the code for matching backrefences
16        caselessly in UTF-8 mode when codes for the characters concerned were
17        different numbers of bytes. For example, U+023A and U+2C65 are an upper
18        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
19        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
20        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
21        2-byte code at the end of the subject (it thought there wasn't enough data
22        left).
23    
24    5.  Comprehensive information about what went wrong is now returned by
25        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
26        as the output vector has at least 2 elements. The offset of the start of
27        the failing character and a reason code are placed in the vector.
28    
29    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
30        now returned is for the first byte of the failing character, instead of the
31        last byte inspected. This is an incompatible change, but I hope it is small
32        enough not to be a problem. It makes the returned offset consistent with
33        pcre_exec() and pcre_dfa_exec().
34    
35    7.  pcretest now gives a text phrase as well as the error number when
36        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
37        failure, the offset and reason code are output.
38    
39    8.  When \R was used with a maximizing quantifier it failed to skip backwards
40        over a \r\n pair if the subsequent match failed. Instead, it just skipped
41        back over a single character (\n). This seems wrong (because it treated the
42        two characters as a single entity when going forwards), conflicts with the
43        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
44        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
45        has been changed.
46    
47    9.  Some internal refactoring has changed the processing so that the handling
48        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
49        time (the PCRE_DOTALL option was changed this way some time ago: version
50        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
51        which was always a bit of a fudge. It also means that there is one less
52        argument for the match() function, which reduces its stack requirements
53        slightly. This change also fixes an incompatibility with Perl: the pattern
54        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
55    
56    10. More internal refactoring has drastically reduced the number of recursive
57        calls to match() for possessively repeated groups such as (abc)++ when
58        using pcre_exec().
59    
60    11. While implementing 10, a number of bugs in the handling of groups were
61        discovered and fixed:
62    
63        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
64        (a|)*(?1) gave a compile-time internal error.
65        ((a|)+)+  did not notice that the outer group could match an empty string.
66        (^a|^)+   was not marked as anchored.
67        (.*a|.*)+ was not marked as matching at start or after a newline.
68    
69    12. Yet more internal refactoring has removed another argument from the match()
70        function. Special calls to this function are now indicated by setting a
71        value in a variable in the "match data" data block.
72    
73    13. Be more explicit in pcre_study() instead of relying on "default" for
74        opcodes that mean there is no starting character; this means that when new
75        ones are added and accidentally left out of pcre_study(), testing should
76        pick them up.
77    
78    14. The -s option of pcretest has been documented for ages as being an old
79        synonym of -m (show memory usage). I have changed it to mean "force study
80        for every regex", that is, assume /S for every regex. This is similar to -i
81        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
82        using it. It makes it easier to run collections of tests with and without
83        study enabled, and thereby test pcre_study() more easily. All the standard
84        tests are now run with and without -s (but some patterns can be marked as
85        "never study" - see 20 below).
86    
87    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
88        restoration of the capturing data to the outer values was not happening
89        correctly.
90    
91    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
92        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
93        pattern had matched an empty string, and so incorrectly returned a no
94        match.
95    
96    17. There was optimizing code for the last branch of non-capturing parentheses,
97        and also for the obeyed branch of a conditional subexpression, which used
98        tail recursion to cut down on stack usage. Unfortunately, not that there is
99        the possibility of (*THEN) occurring in these branches, tail recursion is
100        no longer possible because the return has to be checked for (*THEN). These
101        two optimizations have therefore been removed.
102    
103    18. If a pattern containing \R was studied, it was assumed that \R always
104        matched two bytes, thus causing the minimum subject length to be
105        incorrectly computed because \R can also match just one byte.
106    
107    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
108        was incorrectly computed.
109    
110    20. If /S is present twice on a test pattern in pcretest input, it *disables*
111        studying, thereby overriding the use of -s on the command line. This is
112        necessary for one or two tests to keep the output identical in both cases.
113    
114    21. When (*ACCEPT) was used in an assertion that matched an empty string and
115        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
116    
117    22. When an atomic group that contained a capturing parenthesis was
118        successfully matched, but the branch in which it appeared failed, the
119        capturing was not being forgotten if a higher numbered group was later
120        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
121        group 1 to "a", when in fact it should be unset. This applied to multi-
122        branched capturing and non-capturing groups, repeated or not, and also to
123        positive assertions (capturing in negative assertions is not well defined
124        in PCRE) and also to nested atomic groups.
125    
126    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
127        subject after a captured substring (to make it easier to tell which of a
128        number of identical substrings has been captured).
129    
130    24. The way atomic groups are processed by pcre_exec() has been changed so that
131        if they are repeated, backtracking one repetition now resets captured
132        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
133        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
134        "aaa". Previously, it would have been "a". As part of this code
135        refactoring, the way recursive calls are handled has also been changed.
136    
137    24. If an assertion condition captured any substrings, they were not passed
138        back unless some other capturing happened later. For example, if
139        (?(?=(a))a) was matched against "a", no capturing was returned.
140    
141    25. When studying a pattern that contained subroutine calls or assertions,
142        the code for finding the minimum length of a possible match was handling
143        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
144        group 1 called group 2 while simultaneously a separate group 2 called group
145        1). A stack overflow occurred in this case. I have fixed this by limiting
146        the recursion depth to 10.
147    
148    26. Updated RunTest.bat in the distribution to the version supplied by Tom
149        Fortmann. This supports explicit test numbers on the command line, and has
150        argument validation and error reporting.
151    
152    27. An instance of \X with an unlimited repeat could fail if at any point the
153        first character it looked at was a mark character.
154    
155    28. Some minor code refactoring concerning Unicode properties and scripts
156        should reduce the stack requirement of match() slightly.
157    
158    29. Added the '=' option to pcretest to check the setting of unused capturing
159        slots at the end of the pattern, which are documented as being -1, but are
160        not included in the return count.
161    
162    30. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
163        compiled something random. Now it gives a compile-time error (as does
164        Perl).
165    
166    31. A *MARK encountered during the processing of a positive assertion is now
167        recorded and passed back (compatible with Perl).
168    
169    32. If --only-matching or --colour was set on a pcregrep call whose pattern
170        had alternative anchored branches, the search for a second match in a line
171        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
172        matched the line "0102" twice. The same bug affected patterns that started
173        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
174        twice.
175    
176    33. Previously, PCRE did not allow quantification of assertions. However, Perl
177        does, and because of capturing effects, quantifying parenthesized
178        assertions may at times be useful. Quantifiers are now allowed for
179        parenthesized assertions.
180    
181    34. A minor code tidy in pcre_compile() when checking options for \R usage.
182    
183    35. \g was being checked for fancy things in a character class, when it should
184        just be a literal "g".
185    
186    36. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
187        appearance of a nested POSIX class supersedes an apparent external class.
188        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
189        unescaped square brackets may also appear as part of class names. For
190        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
191        more like Perl.
192    
193    37. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
194        was because it thought it was \N{name}, which is not supported).
195    
196    
197    Version 8.12 15-Jan-2011
198    ------------------------
199    
200    1.  Fixed some typos in the markup of the man pages, and wrote a script that
201        checks for such things as part of the documentation building process.
202    
203    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
204        --match-limit and --recursion-limit options (added for 8.11). In
205        particular, this made one of the standard tests fail. (The integer value
206        went into the wrong half of a long int.)
207    
208    3.  If the --colour option was given to pcregrep with -v (invert match), it
209        did strange things, either producing crazy output, or crashing. It should,
210        of course, ignore a request for colour when reporting lines that do not
211        match.
212    
213    4.  Another pcregrep bug caused similar problems if --colour was specified with
214        -M (multiline) and the pattern match finished with a line ending.
215    
216    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
217        matched in multiline mode, the following line was shown as part of the
218        match. This seems wrong, so I have changed it.
219    
220    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
221        the check for further matches in the same line (so they could be coloured)
222        to overrun the end of the current line. If another match was found, it was
223        incorrectly shown (and then shown again when found in the next line).
224    
225    7.  If pcregrep was compiled under Windows, there was a reference to the
226        function pcregrep_exit() before it was defined. I am assuming this was
227        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
228        reported by a user. I've moved the definition above the reference.
229    
230    
231    Version 8.11 10-Dec-2010
232    ------------------------
233    
234    1.  (*THEN) was not working properly if there were untried alternatives prior
235        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
236        backtracked to try for "b" instead of moving to the next alternative branch
237        at the same level (in this case, to look for "c"). The Perl documentation
238        is clear that when (*THEN) is backtracked onto, it goes to the "next
239        alternative in the innermost enclosing group".
240    
241    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
242        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
243        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
244        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
245        (*THEN).
246    
247    3.  If \s appeared in a character class, it removed the VT character from
248        the class, even if it had been included by some previous item, for example
249        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
250        of \s, but is part of the POSIX "space" class.)
251    
252    4.  A partial match never returns an empty string (because you can always
253        match an empty string at the end of the subject); however the checking for
254        an empty string was starting at the "start of match" point. This has been
255        changed to the "earliest inspected character" point, because the returned
256        data for a partial match starts at this character. This means that, for
257        example, /(?<=abc)def/ gives a partial match for the subject "abc"
258        (previously it gave "no match").
259    
260    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
261        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
262        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
263        has an implication that the given string is incomplete (because a partial
264        match is preferred over a full match). For this reason, these items now
265        give a partial match in this situation. [Aside: previously, the one case
266        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
267        match rather than a full match, which was wrong by the old rules, but is
268        now correct.]
269    
270    6.  There was a bug in the handling of #-introduced comments, recognized when
271        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
272        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
273        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
274        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
275        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
276        places in pcre_compile().
277    
278    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
279        comments when looking ahead for named forward references to subpatterns,
280        the only newline sequence it recognized was NL. It now handles newlines
281        according to the set newline convention.
282    
283    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
284        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
285        cater for a lack of strerror(). These oversights have been fixed.
286    
287    9.  Added --match-limit and --recursion-limit to pcregrep.
288    
289    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
290    
291    11. When the -o option was used, pcregrep was setting a return code of 1, even
292        when matches were found, and --line-buffered was not being honoured.
293    
294    12. Added an optional parentheses number to the -o and --only-matching options
295        of pcregrep.
296    
297    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
298        can match an empty string. The code to do it in pcretest and pcredemo
299        needed fixing:
300    
301        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
302            only one byte after an empty string match just before CRLF (this case
303            just got forgotten; "any" and "anycrlf" were OK).
304    
305        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
306            mode when an empty string match preceded an ASCII character followed by
307            a non-ASCII character. (The code for advancing by one character rather
308            than one byte was nonsense.)
309    
310        (c) The pcredemo.c sample program did not have any code at all to handle
311            the cases when CRLF is a valid newline sequence.
312    
313    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
314        as a starting offset was within the subject string. There is now a new
315        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
316        negative or greater than the length of the string. In order to test this,
317        pcretest is extended to allow the setting of negative starting offsets.
318    
319    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
320        starting offset points to the beginning of a UTF-8 character was
321        unnecessarily clumsy. I tidied it up.
322    
323    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
324        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
325    
326    17. Nobody had reported that the --include_dir option, which was added in
327        release 7.7 should have been called --include-dir (hyphen, not underscore)
328        for compatibility with GNU grep. I have changed it to --include-dir, but
329        left --include_dir as an undocumented synonym, and the same for
330        --exclude-dir, though that is not available in GNU grep, at least as of
331        release 2.5.4.
332    
333    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
334        characters from a string of bytes) have been redefined so as not to use
335        loops, in order to improve performance in some environments. At the same
336        time, I abstracted some of the common code into auxiliary macros to save
337        repetition (this should not affect the compiled code).
338    
339    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
340        compile-time error is now given if \c is not followed by an ASCII
341        character, that is, a byte less than 128. (In EBCDIC mode, the code is
342        different, and any byte value is allowed.)
343    
344    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
345        START_OPTIMIZE option, which is now allowed at compile time - but just
346        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
347        to pcregrep and other applications that have no direct access to PCRE
348        options. The new /Y option in pcretest sets this option when calling
349        pcre_compile().
350    
351    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
352        back references. Groups containing recursive back references were forced to
353        be atomic by that change, but in the case of named groups, the amount of
354        memory required was incorrectly computed, leading to "Failed: internal
355        error: code overflow". This has been fixed.
356    
357    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
358        pcretest.c, to avoid build problems in some Borland environments.
359    
360    
361    Version 8.10 25-Jun-2010
362    ------------------------
363    
364    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
365        THEN.
366    
367    2.  (*ACCEPT) was not working when inside an atomic group.
368    
369    3.  Inside a character class, \B is treated as a literal by default, but
370        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
371        causes the error). The code is unchanged, but I tidied the documentation.
372    
373    4.  Inside a character class, PCRE always treated \R and \X as literals,
374        whereas Perl faults them if its -w option is set. I have changed PCRE so
375        that it faults them when PCRE_EXTRA is set.
376    
377    5.  Added support for \N, which always matches any character other than
378        newline. (It is the same as "." when PCRE_DOTALL is not set.)
379    
380    6.  When compiling pcregrep with newer versions of gcc which may have
381        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
382        declared with attribute warn_unused_result" were given. Just casting the
383        result to (void) does not stop the warnings; a more elaborate fudge is
384        needed. I've used a macro to implement this.
385    
386    7.  Minor change to pcretest.c to avoid a compiler warning.
387    
388    8.  Added four artifical Unicode properties to help with an option to make
389        \s etc use properties (see next item). The new properties are: Xan
390        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
391    
392    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
393        use Unicode properties. (*UCP) at the start of a pattern can be used to set
394        this option. Modified pcretest to add /W to test this facility. Added
395        REG_UCP to make it available via the POSIX interface.
396    
397    10. Added --line-buffered to pcregrep.
398    
399    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
400        studied, and the match started with a letter with a code point greater than
401        127 whose first byte was different to the first byte of the other case of
402        the letter, the other case of this starting letter was not recognized
403        (#976).
404    
405    12. If a pattern that was studied started with a repeated Unicode property
406        test, for example, \p{Nd}+, there was the theoretical possibility of
407        setting up an incorrect bitmap of starting bytes, but fortunately it could
408        not have actually happened in practice until change 8 above was made (it
409        added property types that matched character-matching opcodes).
410    
411    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
412        possible starting bytes for non-anchored patterns.
413    
414    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
415        \R, and also a number of cases that involve Unicode properties, both
416        explicit and implicit when PCRE_UCP is set.
417    
418    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
419        input, it could crash or give wrong results if characters with values
420        greater than 0xc0 were present in the subject string. (Detail: it assumed
421        UTF-8 input when processing these items.)
422    
423    16. Added a lot of (int) casts to avoid compiler warnings in systems where
424        size_t is 64-bit (#991).
425    
426    17. Added a check for running out of memory when PCRE is compiled with
427        --disable-stack-for-recursion (#990).
428    
429    18. If the last data line in a file for pcretest does not have a newline on
430        the end, a newline was missing in the output.
431    
432    19. The default pcre_chartables.c file recognizes only ASCII characters (values
433        less than 128) in its various bitmaps. However, there is a facility for
434        generating tables according to the current locale when PCRE is compiled. It
435        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
436        characters, are recognized by isspace() and therefore were getting set in
437        these tables, and indeed these tables seem to approximate to ISO 8859. This
438        caused a problem in UTF-8 mode when pcre_study() was used to create a list
439        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
440        which of course cannot start UTF-8 characters. I have changed the code so
441        that only real ASCII characters (less than 128) and the correct starting
442        bytes for UTF-8 encodings are set for characters greater than 127 when in
443        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
444        altogether.)
445    
446    20. Added the /T option to pcretest so as to be able to run tests with non-
447        standard character tables, thus making it possible to include the tests
448        used for 19 above in the standard set of tests.
449    
450    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
451        reference to a subpattern the other side of a comment that contains an
452        opening parenthesis caused either an internal compiling error, or a
453        reference to the wrong subpattern.
454    
455    
456    Version 8.02 19-Mar-2010
457    ------------------------
458    
459    1.  The Unicode data tables have been updated to Unicode 5.2.0.
460    
461    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
462        configured.
463    
464    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
465        original author of that file, following a query about its status.
466    
467    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
468        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
469    
470    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
471        quantifier applied to a forward-referencing subroutine call, could compile
472        incorrect code or give the error "internal error: previously-checked
473        referenced subpattern not found".
474    
475    6.  Both MS Visual Studio and Symbian OS have problems with initializing
476        variables to point to external functions. For these systems, therefore,
477        pcre_malloc etc. are now initialized to local functions that call the
478        relevant global functions.
479    
480    7.  There were two entries missing in the vectors called coptable and poptable
481        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
482        I've fixed the data, and added a kludgy way of testing at compile time that
483        the lengths are correct (equal to the number of opcodes).
484    
485    8.  Following on from 7, I added a similar kludge to check the length of the
486        eint vector in pcreposix.c.
487    
488    9.  Error texts for pcre_compile() are held as one long string to avoid too
489        much relocation at load time. To find a text, the string is searched,
490        counting zeros. There was no check for running off the end of the string,
491        which could happen if a new error number was added without updating the
492        string.
493    
494    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
495    
496    11. \K was not working if it appeared in an atomic group or in a group that
497        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
498        \K is "not well defined" if used in an assertion. PCRE now accepts it if
499        the assertion is positive, but not if it is negative.
500    
501    12. Change 11 fortuitously reduced the size of the stack frame used in the
502        "match()" function of pcre_exec.c by one pointer. Forthcoming
503        implementation of support for (*MARK) will need an extra pointer on the
504        stack; I have reserved it now, so that the stack frame size does not
505        decrease.
506    
507    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
508        item in branch that calls a recursion is a subroutine call - as in the
509        second branch in the above example - was incorrectly given the compile-
510        time error "recursive call could loop indefinitely" because pcre_compile()
511        was not correctly checking the subroutine for matching a non-empty string.
512    
513    14. The checks for overrunning compiling workspace could trigger after an
514        overrun had occurred. This is a "should never occur" error, but it can be
515        triggered by pathological patterns such as hundreds of nested parentheses.
516        The checks now trigger 100 bytes before the end of the workspace.
517    
518    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
519    
520    
521    Version 8.01 19-Jan-2010
522    ------------------------
523    
524    1.  If a pattern contained a conditional subpattern with only one branch (in
525        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
526        computed the wrong minimum data length (which is of course zero for such
527        subpatterns). This could cause incorrect "no match" results.
528    
529    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
530        the pattern is reset in the first branch, pcre_compile() failed with
531        "internal error: code overflow at offset...". This happened only when
532        the reset was to the original external option setting. (An optimization
533        abstracts leading options settings into an external setting, which was the
534        cause of this.)
535    
536    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
537        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
538        assertion pattern did not match (meaning that the assertion was true), it
539        was incorrectly treated as false if the SKIP had been reached during the
540        matching. This also applied to assertions used as conditions.
541    
542    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
543        assertion subpattern, including such a pattern used as a condition,
544        unpredictable results occurred, instead of the error return
545        PCRE_ERROR_DFA_UITEM.
546    
547    5.  The C++ GlobalReplace function was not working like Perl for the special
548        situation when an empty string is matched. It now does the fancy magic
549        stuff that is necessary.
550    
551    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
552        removed. (These were left over from very, very early versions of PCRE.)
553    
554    7.  Some cosmetic changes to the code to make life easier when compiling it
555        as part of something else:
556    
557        (a) Change DEBUG to PCRE_DEBUG.
558    
559        (b) In pcre_compile(), rename the member of the "branch_chain" structure
560            called "current" as "current_branch", to prevent a collision with the
561            Linux macro when compiled as a kernel module.
562    
563        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
564            prevent a collision with the Linux macro when compiled as a kernel
565            module.
566    
567    8.  In pcre_compile() there are some checks for integer overflows that used to
568        cast potentially large values to (double). This has been changed to that
569        when building, a check for int64_t is made, and if it is found, it is used
570        instead, thus avoiding the use of floating point arithmetic. (There is no
571        other use of FP in PCRE.) If int64_t is not found, the fallback is to
572        double.
573    
574    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
575        2005 (difference between two addresses compared to an unsigned value).
576    
577    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
578        custom one, because of the following reported problem in Windows:
579    
580          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
581              under Win32.
582          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
583              therefore missing the function definition.
584          - The compiler thus generates a "C" signature for the test function.
585          - The linker fails to find the "C" function.
586          - PCRE fails to configure if asked to do so against libbz2.
587    
588    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
589        messages were output:
590    
591          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
592          rerunning libtoolize, to keep the correct libtool macros in-tree.
593          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
594    
595        I have done both of these things.
596    
597    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
598        most of the time, it *can* run out if it is given a pattern that contains a
599        runaway infinite recursion. I updated the discussion in the pcrestack man
600        page.
601    
602    13. Now that we have gone to the x.xx style of version numbers, the minor
603        version may start with zero. Using 08 or 09 is a bad idea because users
604        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
605        interpreted as invalid octal numbers. I've updated the previous comment in
606        configure.ac, and also added a check that gives an error if 08 or 09 are
607        used.
608    
609    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
610        causing partial matching to fail when the end of the subject matched \W
611        in a UTF-8 pattern where \W was quantified with a minimum of 3.
612    
613    15. There were some discrepancies between the declarations in pcre_internal.h
614        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
615        their definitions. The declarations used "const uschar *" and the
616        definitions used USPTR. Even though USPTR is normally defined as "const
617        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
618        reported that: "This difference in casting confuses some C++ compilers, for
619        example, SunCC recognizes above declarations as different functions and
620        generates broken code for hbpcre." I have changed the declarations to use
621        USPTR.
622    
623    16. GNU libtool is named differently on some systems. The autogen.sh script now
624        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
625        (FreeBSD).
626    
627    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
628        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
629        comment: "Figure out how to create a longlong from a string: strtoll and
630        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
631        instance, but it only takes 2 args instead of 3!"
632    
633    18. A subtle bug concerned with back references has been fixed by a change of
634        specification, with a corresponding code fix. A pattern such as
635        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
636        refers, was giving matches when it shouldn't. For example, xa=xaaa would
637        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
638        same bug. Such groups have to be quantified to be useful, or contained
639        inside another quantified group. (If there's no repetition, the reference
640        can never match.) The problem arises because, having left the group and
641        moved on to the rest of the pattern, a later failure that backtracks into
642        the group uses the captured value from the final iteration of the group
643        rather than the correct earlier one. I have fixed this in PCRE by forcing
644        any group that contains a reference to itself to be an atomic group; that
645        is, there cannot be any backtracking into it once it has completed. This is
646        similar to recursive and subroutine calls.
647    
648    
649    Version 8.00 19-Oct-09
650    ----------------------
651    
652    1.  The table for translating pcre_compile() error codes into POSIX error codes
653        was out-of-date, and there was no check on the pcre_compile() error code
654        being within the table. This could lead to an OK return being given in
655        error.
656    
657    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
658        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
659        in a Windows environment.
660    
661    3.  The pcregrep --count option prints the count for each file even when it is
662        zero, as does GNU grep. However, pcregrep was also printing all files when
663        --files-with-matches was added. Now, when both options are given, it prints
664        counts only for those files that have at least one match. (GNU grep just
665        prints the file name in this circumstance, but including the count seems
666        more useful - otherwise, why use --count?) Also ensured that the
667        combination -clh just lists non-zero counts, with no names.
668    
669    4.  The long form of the pcregrep -F option was incorrectly implemented as
670        --fixed_strings instead of --fixed-strings. This is an incompatible change,
671        but it seems right to fix it, and I didn't think it was worth preserving
672        the old behaviour.
673    
674    5.  The command line items --regex=pattern and --regexp=pattern were not
675        recognized by pcregrep, which required --regex pattern or --regexp pattern
676        (with a space rather than an '='). The man page documented the '=' forms,
677        which are compatible with GNU grep; these now work.
678    
679    6.  No libpcreposix.pc file was created for pkg-config; there was just
680        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
681    
682    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
683        when UCP support is not needed, by modifying the Python script that
684        generates it from Unicode data files. This should not matter if the module
685        is correctly used as a library, but I received one complaint about 50K of
686        unwanted data. My guess is that the person linked everything into his
687        program rather than using a library. Anyway, it does no harm.
688    
689    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
690        was a minimum greater than 1 for a wide character in a possessive
691        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
692        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
693        character. Chaos in the form of incorrect output or a compiling loop could
694        result.
695    
696    9.  The restrictions on what a pattern can contain when partial matching is
697        requested for pcre_exec() have been removed. All patterns can now be
698        partially matched by this function. In addition, if there are at least two
699        slots in the offset vector, the offset of the earliest inspected character
700        for the match and the offset of the end of the subject are set in them when
701        PCRE_ERROR_PARTIAL is returned.
702    
703    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
704        synonymous with PCRE_PARTIAL, for backwards compatibility, and
705        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
706        and may be more useful for multi-segment matching.
707    
708    11. Partial matching with pcre_exec() is now more intuitive. A partial match
709        used to be given if ever the end of the subject was reached; now it is
710        given only if matching could not proceed because another character was
711        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
712        string "Z", which now yields "no match" instead of "partial match". In the
713        case of pcre_dfa_exec(), "no match" is given if every matching path for the
714        final character ended with (*FAIL).
715    
716    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
717        if the pattern had a "must contain" character that was already found in the
718        earlier partial match, unless partial matching was again requested. For
719        example, with the pattern /dog.(body)?/, the "must contain" character is
720        "g". If the first part-match was for the string "dog", restarting with
721        "sbody" failed. This bug has been fixed.
722    
723    13. The string returned by pcre_dfa_exec() after a partial match has been
724        changed so that it starts at the first inspected character rather than the
725        first character of the match. This makes a difference only if the pattern
726        starts with a lookbehind assertion or \b or \B (\K is not supported by
727        pcre_dfa_exec()). It's an incompatible change, but it makes the two
728        matching functions compatible, and I think it's the right thing to do.
729    
730    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
731        so that the demonstration program is easily available in environments where
732        PCRE has not been installed from source.
733    
734    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
735        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
736        library.
737    
738    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
739        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
740        is not the first non-POSIX option to be added. Clearly some people find
741        these options useful.
742    
743    17. If a caller to the POSIX matching function regexec() passes a non-zero
744        value for nmatch with a NULL value for pmatch, the value of
745        nmatch is forced to zero.
746    
747    18. RunGrepTest did not have a test for the availability of the -u option of
748        the diff command, as RunTest does. It now checks in the same way as
749        RunTest, and also checks for the -b option.
750    
751    19. If an odd number of negated classes containing just a single character
752        interposed, within parentheses, between a forward reference to a named
753        subpattern and the definition of the subpattern, compilation crashed with
754        an internal error, complaining that it could not find the referenced
755        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
756        [The bug was that it was starting one character too far in when skipping
757        over the character class, thus treating the ] as data rather than
758        terminating the class. This meant it could skip too much.]
759    
760    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
761        /g option in pcretest when the pattern contains \K, which makes it possible
762        to have an empty string match not at the start, even when the pattern is
763        anchored. Updated pcretest and pcredemo to use this option.
764    
765    21. If the maximum number of capturing subpatterns in a recursion was greater
766        than the maximum at the outer level, the higher number was returned, but
767        with unset values at the outer level. The correct (outer level) value is
768        now given.
769    
770    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
771        PCRE did not set those parentheses (unlike Perl). I have now found a way to
772        make it do so. The string so far is captured, making this feature
773        compatible with Perl.
774    
775    23. The tests have been re-organized, adding tests 11 and 12, to make it
776        possible to check the Perl 5.10 features against Perl 5.10.
777    
778    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
779        pattern matches a fixed length string. PCRE did not allow this; now it
780        does. Neither allows recursion.
781    
782    25. I finally figured out how to implement a request to provide the minimum
783        length of subject string that was needed in order to match a given pattern.
784        (It was back references and recursion that I had previously got hung up
785        on.) This code has now been added to pcre_study(); it finds a lower bound
786        to the length of subject needed. It is not necessarily the greatest lower
787        bound, but using it to avoid searching strings that are too short does give
788        some useful speed-ups. The value is available to calling programs via
789        pcre_fullinfo().
790    
791    26. While implementing 25, I discovered to my embarrassment that pcretest had
792        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
793        study optimizations had never been tested with that matching function.
794        Oops. What is worse, even when it was passed study data, there was a bug in
795        pcre_dfa_exec() that meant it never actually used it. Double oops. There
796        were also very few tests of studied patterns with pcre_dfa_exec().
797    
798    27. If (?| is used to create subpatterns with duplicate numbers, they are now
799        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
800        on the other side of the coin, they are no longer allowed to have different
801        names, because these cannot be distinguished in PCRE, and this has caused
802        confusion. (This is a difference from Perl.)
803    
804    28. When duplicate subpattern names are present (necessarily with different
805        numbers, as required by 27 above), and a test is made by name in a
806        conditional pattern, either for a subpattern having been matched, or for
807        recursion in such a pattern, all the associated numbered subpatterns are
808        tested, and the overall condition is true if the condition is true for any
809        one of them. This is the way Perl works, and is also more like the way
810        testing by number works.
811    
812    
813    Version 7.9 11-Apr-09
814    ---------------------
815    
816    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
817        (pcretest), all targets were linked against these libraries. This included
818        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
819        libraries. This caused unwanted dependencies to be created. This problem
820        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
821        pcretest is linked with readline.
822    
823    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
824        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
825        moved outside it again, because FALSE and TRUE are already defined in AIX,
826        but BOOL is not.
827    
828    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
829        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
830    
831    4.  The pcregrep documentation said spaces were inserted as well as colons (or
832        hyphens) following file names and line numbers when outputting matching
833        lines. This is not true; no spaces are inserted. I have also clarified the
834        wording for the --colour (or --color) option.
835    
836    5.  In pcregrep, when --colour was used with -o, the list of matching strings
837        was not coloured; this is different to GNU grep, so I have changed it to be
838        the same.
839    
840    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
841        each matching line was coloured. Now it goes on to look for further matches
842        of any of the test patterns, which is the same behaviour as GNU grep.
843    
844    7.  A pattern that could match an empty string could cause pcregrep to loop; it
845        doesn't make sense to accept an empty string match in pcregrep, so I have
846        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
847        seems to be how GNU grep behaves.
848    
849    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
850        start or after a newline", because the conditional assertion was not being
851        correctly handled. The rule now is that both the assertion and what follows
852        in the first alternative must satisfy the test.
853    
854    9.  If auto-callout was enabled in a pattern with a conditional group whose
855        condition was an assertion, PCRE could crash during matching, both with
856        pcre_exec() and pcre_dfa_exec().
857    
858    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
859        used for matching.
860    
861    11. Unicode property support in character classes was not working for
862        characters (bytes) greater than 127 when not in UTF-8 mode.
863    
864    12. Added the -M command line option to pcretest.
865    
866    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
867    
868    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
869    
870    16. Added comments and documentation about mis-use of no_arg in the C++
871        wrapper.
872    
873    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
874        from Martin Jerabek that uses macro names for all relevant character and
875        string constants.
876    
877    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
878        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
879        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
880        these, but not everybody uses configure.
881    
882    19. A conditional group that had only one branch was not being correctly
883        recognized as an item that could match an empty string. This meant that an
884        enclosing group might also not be so recognized, causing infinite looping
885        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
886        with the subject "ab", where knowledge that the repeated group can match
887        nothing is needed in order to break the loop.
888    
889    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
890        exec(), but without supplying a callout function, matching went wrong.
891    
892    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
893        leak if the size of the offset vector was greater than 30. When the vector
894        is smaller, the saved offsets during recursion go onto a local stack
895        vector, but for larger vectors malloc() is used. It was failing to free
896        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
897        error, in fact).
898    
899    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
900        heapframe that is used only when UTF-8 support is enabled. This caused no
901        problem, but was untidy.
902    
903    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
904        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
905        included within another project.
906    
907    24. Steven Van Ingelgem's patches to add more options to the CMake support,
908        slightly modified by me:
909    
910          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
911              not building pcregrep.
912    
913          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
914              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
915    
916    25. Forward references, both numeric and by name, in patterns that made use of
917        duplicate group numbers, could behave incorrectly or give incorrect errors,
918        because when scanning forward to find the reference group, PCRE was not
919        taking into account the duplicate group numbers. A pattern such as
920        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
921    
922    26. Changed a few more instances of "const unsigned char *" to USPTR, making
923        the feature of a custom pointer more persuasive (as requested by a user).
924    
925    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
926        pcretest.c, inside #ifndefs, because it seems they are sometimes already
927        pre-defined.
928    
929    28. Added support for (*UTF8) at the start of a pattern.
930    
931    29. Arrange for flags added by the "release type" setting in CMake to be shown
932        in the configuration summary.
933    
934    
935    Version 7.8 05-Sep-08
936    ---------------------
937    
938    1.  Replaced UCP searching code with optimized version as implemented for Ad
939        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
940        stage table and inline lookup instead of a function, giving speed ups of 2
941        to 5 times on some simple patterns that I tested. Permission was given to
942        distribute the MultiStage2.py script that generates the tables (it's not in
943        the tarball, but is in the Subversion repository).
944    
945    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
946        scripts.
947    
948    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
949        a group with a zero qualifier. The result of the study could be incorrect,
950        or the function might crash, depending on the pattern.
951    
952    4.  Caseless matching was not working for non-ASCII characters in back
953        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
954        It now works when Unicode Property Support is available.
955    
956    5.  In pcretest, an escape such as \x{de} in the data was always generating
957        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
958        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
959        truncation.
960    
961    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
962    
963    7.  Added two (int) casts to pcregrep when printing the difference of two
964        pointers, in case they are 64-bit values.
965    
966    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
967        test 2 if it fails.
968    
969    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
970        and a #define of that name to empty if it is not externally set. This is to
971        allow users of MSVC to set it if necessary.
972    
973    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
974        the convenience functions in the pcre_get.c source file.
975    
976    11. An option change at the start of a pattern that had top-level alternatives
977        could cause overwriting and/or a crash. This command provoked a crash in
978        some environments:
979    
980          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
981    
982        This potential security problem was recorded as CVE-2008-2371.
983    
984    12. For a pattern where the match had to start at the beginning or immediately
985        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
986        pcre_dfa_exec() could read past the end of the passed subject if there was
987        no match. To help with detecting such bugs (e.g. with valgrind), I modified
988        pcretest so that it places the subject at the end of its malloc-ed buffer.
989    
990    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
991        exec() might read past the end of the data buffer in UTF-8 mode.
992    
993    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
994        the data contained the byte 0x85 as part of a UTF-8 character within its
995        first line. This applied both to normal and DFA matching.
996    
997    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
998        /^[^d]*?$/8 failed to match "abc".
999    
1000    16. Added a missing copyright notice to pcrecpp_internal.h.
1001    
1002    17. Make it more clear in the documentation that values returned from
1003        pcre_exec() in ovector are byte offsets, not character counts.
1004    
1005    18. Tidied a few places to stop certain compilers from issuing warnings.
1006    
1007    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
1008        supplied by Stefan Weber. I made a further small update for 7.8 because
1009        there is a change of source arrangements: the pcre_searchfuncs.c module is
1010        replaced by pcre_ucd.c.
1011    
1012    
1013    Version 7.7 07-May-08
1014    ---------------------
1015    
1016    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
1017        a string to a long long, pretend we don't even have a long long." This is
1018        done by checking for the strtoq, strtoll, and _strtoi64 functions.
1019    
1020    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
1021        pre-7.6 versions, which defined a global no_arg variable instead of putting
1022        it in the RE class. (See also #8 below.)
1023    
1024    3.  Remove a line of dead code, identified by coverity and reported by Nuno
1025        Lopes.
1026    
1027    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
1028    
1029        (1) The include/exclude patterns were being applied to the whole pathnames
1030            of files, instead of just to the final components.
1031    
1032        (2) If there was more than one level of directory, the subdirectories were
1033            skipped unless they satisfied the include/exclude conditions. This is
1034            inconsistent with GNU grep (and could even be seen as contrary to the
1035            pcregrep specification - which I improved to make it absolutely clear).
1036            The action now is always to scan all levels of directory, and just
1037            apply the include/exclude patterns to regular files.
1038    
1039    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
1040        --exclude_dir in the tests to avoid scanning .svn directories.
1041    
1042    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
1043        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
1044        doesn't support NULs in patterns.
1045    
1046    7.  Added some missing "const"s to declarations of static tables in
1047        pcre_compile.c and pcre_dfa_exec.c.
1048    
1049    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
1050        caused by fix #2  above. (Subsequently also a second patch to fix the
1051        first patch. And a third patch - this was a messy problem.)
1052    
1053    9.  Applied Craig's patch to remove the use of push_back().
1054    
1055    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
1056        matching function regexec().
1057    
1058    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
1059        which, however, unlike Perl's \g{...}, are subroutine calls, not back
1060        references. PCRE supports relative numbers with this syntax (I don't think
1061        Oniguruma does).
1062    
1063    12. Previously, a group with a zero repeat such as (...){0} was completely
1064        omitted from the compiled regex. However, this means that if the group
1065        was called as a subroutine from elsewhere in the pattern, things went wrong
1066        (an internal error was given). Such groups are now left in the compiled
1067        pattern, with a new opcode that causes them to be skipped at execution
1068        time.
1069    
1070    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
1071        to the way PCRE behaves:
1072    
1073        (a) A lone ] character is dis-allowed (Perl treats it as data).
1074    
1075        (b) A back reference to an unmatched subpattern matches an empty string
1076            (Perl fails the current match path).
1077    
1078        (c) A data ] in a character class must be notated as \] because if the
1079            first data character in a class is ], it defines an empty class. (In
1080            Perl it is not possible to have an empty class.) The empty class []
1081            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
1082            The negative empty class [^] matches any one character, independently
1083            of the DOTALL setting.
1084    
1085    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
1086        non-existent subpattern following a character class starting with ']' and
1087        containing () gave an internal compiling error instead of "reference to
1088        non-existent subpattern". Fortunately, when the pattern did exist, the
1089        compiled code was correct. (When scanning forwards to check for the
1090        existencd of the subpattern, it was treating the data ']' as terminating
1091        the class, so got the count wrong. When actually compiling, the reference
1092        was subsequently set up correctly.)
1093    
1094    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
1095        it was being rejected as not supported by pcre_dfa_exec(), even though
1096        other assertions are supported. I have made pcre_dfa_exec() support
1097        (*FAIL).
1098    
1099    16. The implementation of 13c above involved the invention of a new opcode,
1100        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
1101        cannot be changed at match time, I realized I could make a small
1102        improvement to matching performance by compiling OP_ALLANY instead of
1103        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
1104        on the OP_ANY path.
1105    
1106    17. Compiling pcretest on Windows with readline support failed without the
1107        following two fixes: (1) Make the unistd.h include conditional on
1108        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
1109    
1110    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
1111        ncurses library to be included for pcretest when ReadLine support is
1112        requested, but also to allow for it to be overridden. This patch came from
1113        Daniel Bergström.
1114    
1115    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
1116        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
1117        any errors with the current Unicode tables. Thanks to Peter Kankowski for
1118        spotting this.
1119    
1120    
1121    Version 7.6 28-Jan-08
1122    ---------------------
1123    
1124    1.  A character class containing a very large number of characters with
1125        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
1126        overflow.
1127    
1128    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
1129        HAVE_LONG_LONG is not defined.
1130    
1131    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
1132        bring them up to date and include new features. This patch includes:
1133    
1134        - Fixed PH's badly added libz and libbz2 support.
1135        - Fixed a problem with static linking.
1136        - Added pcredemo. [But later removed - see 7 below.]
1137        - Fixed dftables problem and added an option.
1138        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
1139            HAVE_LONG_LONG.
1140        - Added readline support for pcretest.
1141        - Added an listing of the option settings after cmake has run.
1142    
1143    4.  A user submitted a patch to Makefile that makes it easy to create
1144        "pcre.dll" under mingw when using Configure/Make. I added stuff to
1145        Makefile.am that cause it to include this special target, without
1146        affecting anything else. Note that the same mingw target plus all
1147        the other distribution libraries and programs are now supported
1148        when configuring with CMake (see 6 below) instead of with
1149        Configure/Make.
1150    
1151    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
1152        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
1153        exported in the Windows port". It has not yet been confirmed that the patch
1154        solves the problem, but it does no harm.
1155    
1156    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
1157        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
1158        with CMake, and also correct the comment about stack recursion.
1159    
1160    7.  Remove the automatic building of pcredemo from the ./configure system and
1161        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
1162        of a program that users should build themselves after PCRE is installed, so
1163        building it automatically is not really right. What is more, it gave
1164        trouble in some build environments.
1165    
1166    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
1167    
1168    
1169    Version 7.5 10-Jan-08
1170    ---------------------
1171    
1172    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
1173        values in parens when parsing an RE using the C++ wrapper."
1174    
1175    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
1176        Characters greater than 255 were excluded from the class instead of being
1177        included.
1178    
1179    3.  The same bug as (2) above applied to negated POSIX classes such as
1180        [:^space:].
1181    
1182    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
1183        defined or documented. It seems to have been a typo for PCRE_STATIC, so
1184        I have changed it.
1185    
1186    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
1187        first named subpattern) and a construct such as (?&a) would reference the
1188        first named subpattern whose name started with "a" (in other words, the
1189        length check was missing). Both these problems are fixed. "Subpattern name
1190        expected" is now given for (?&) (a zero-length name), and this patch also
1191        makes it give the same error for \k'' (previously it complained that that
1192        was a reference to a non-existent subpattern).
1193    
1194    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
1195        this is right because (?- can be followed by option settings as well as by
1196        digits. I have, however, made the messages clearer.
1197    
1198    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
1199        than the number used in the conditional) now cause a compile-time error.
1200        This is actually not compatible with Perl, which accepts such patterns, but
1201        treats the conditional as always being FALSE (as PCRE used to), but it
1202        seems to me that giving a diagnostic is better.
1203    
1204    8.  Change "alphameric" to the more common word "alphanumeric" in comments
1205        and messages.
1206    
1207    9.  Fix two occurrences of "backslash" in comments that should have been
1208        "backspace".
1209    
1210    10. Remove two redundant lines of code that can never be obeyed (their function
1211        was moved elsewhere).
1212    
1213    11. The program that makes PCRE's Unicode character property table had a bug
1214        which caused it to generate incorrect table entries for sequences of
1215        characters that have the same character type, but are in different scripts.
1216        It amalgamated them into a single range, with the script of the first of
1217        them. In other words, some characters were in the wrong script. There were
1218        thirteen such cases, affecting characters in the following ranges:
1219    
1220          U+002b0 - U+002c1
1221          U+0060c - U+0060d
1222          U+0061e - U+00612
1223          U+0064b - U+0065e
1224          U+0074d - U+0076d
1225          U+01800 - U+01805
1226          U+01d00 - U+01d77
1227          U+01d9b - U+01dbf
1228          U+0200b - U+0200f
1229          U+030fc - U+030fe
1230          U+03260 - U+0327f
1231          U+0fb46 - U+0fbb1
1232          U+10450 - U+1049d
1233    
1234    12. The -o option (show only the matching part of a line) for pcregrep was not
1235        compatible with GNU grep in that, if there was more than one match in a
1236        line, it showed only the first of them. It now behaves in the same way as
1237        GNU grep.
1238    
1239    13. If the -o and -v options were combined for pcregrep, it printed a blank
1240        line for every non-matching line. GNU grep prints nothing, and pcregrep now
1241        does the same. The return code can be used to tell if there were any
1242        non-matching lines.
1243    
1244    14. Added --file-offsets and --line-offsets to pcregrep.
1245    
1246    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
1247        infinitely looping recursion. The bug was that positive lookaheads were not
1248        being skipped when checking for a possible empty match (negative lookaheads
1249        and both kinds of lookbehind were skipped).
1250    
1251    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
1252        inclusion of <windows.h> to before rather than after the definition of
1253        INVALID_FILE_ATTRIBUTES (patch from David Byron).
1254    
1255    17. Specifying a possessive quantifier with a specific limit for a Unicode
1256        character property caused pcre_compile() to compile bad code, which led at
1257        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
1258        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
1259        caused the error; without that there was no problem.
1260    
1261    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
1262    
1263    19. Added --enable-pcretest-libreadline.
1264    
1265    20. In pcrecpp.cc, the variable 'count' was incremented twice in
1266        RE::GlobalReplace(). As a result, the number of replacements returned was
1267        double what it should be. I removed one of the increments, but Craig sent a
1268        later patch that removed the other one (the right fix) and added unit tests
1269        that check the return values (which was not done before).
1270    
1271    21. Several CMake things:
1272    
1273        (1) Arranged that, when cmake is used on Unix, the libraries end up with
1274            the names libpcre and libpcreposix, not just pcre and pcreposix.
1275    
1276        (2) The above change means that pcretest and pcregrep are now correctly
1277            linked with the newly-built libraries, not previously installed ones.
1278    
1279        (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
1280    
1281    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
1282        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
1283        UTF-8 newline character). The key issue is that the pattern starts .*;
1284        this means that the match must be either at the beginning, or after a
1285        newline. The bug was in the code for advancing after a failed match and
1286        checking that the new position followed a newline. It was not taking
1287        account of UTF-8 characters correctly.
1288    
1289    23. PCRE was behaving differently from Perl in the way it recognized POSIX
1290        character classes. PCRE was not treating the sequence [:...:] as a
1291        character class unless the ... were all letters. Perl, however, seems to
1292        allow any characters between [: and :], though of course it rejects as
1293        unknown any "names" that contain non-letters, because all the known class
1294        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
1295        for example, whereas PCRE did not - it did not recognize a POSIX character
1296        class. This seemed a bit dangerous, so the code has been changed to be
1297        closer to Perl. The behaviour is not identical to Perl, because PCRE will
1298        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
1299        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
1300        Perl does, and where it didn't before.
1301    
1302    24. Rewrite so as to remove the single use of %n from pcregrep because in some
1303        Windows environments %n is disabled by default.
1304    
1305    
1306    Version 7.4 21-Sep-07
1307    ---------------------
1308    
1309    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
1310        means that a class such as [\s] counted as "explicit reference to CR or
1311        LF". That isn't really right - the whole point of the change was to try to
1312        help when there was an actual mention of one of the two characters. So now
1313        the change happens only if \r or \n (or a literal CR or LF) character is
1314        encountered.
1315    
1316    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
1317        of both had grown to the point where there were only 3 bits left.
1318        Fortunately, there was spare space in the data structure, and so I have
1319        moved the internal flags into a new 16-bit field to free up more option
1320        bits.
1321    
1322    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
1323        but did not set the internal JCHANGED flag - either of these is enough to
1324        control the way the "get" function works - but the PCRE_INFO_JCHANGED
1325        facility is supposed to tell if (?J) was ever used, so now (?J) at the
1326        start sets both bits.
1327    
1328    4.  Added options (at build time, compile time, exec time) to change \R from
1329        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
1330    
1331    5.  doc/pcresyntax.html was missing from the distribution.
1332    
1333    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
1334        compatibility, even though it is no longer used.
1335    
1336    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
1337        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
1338        windows.h file is present (where different names are used). [This was
1339        reversed later after testing - see 16 below.]
1340    
1341    8.  Changed all #include <config.h> to #include "config.h". There were also
1342        some further <pcre.h> cases that I changed to "pcre.h".
1343    
1344    9.  When pcregrep was used with the --colour option, it missed the line ending
1345        sequence off the lines that it output.
1346    
1347    10. It was pointed out to me that arrays of string pointers cause lots of
1348        relocations when a shared library is dynamically loaded. A technique of
1349        using a single long string with a table of offsets can drastically reduce
1350        these. I have refactored PCRE in four places to do this. The result is
1351        dramatic:
1352    
1353          Originally:                          290
1354          After changing UCP table:            187
1355          After changing error message table:   43
1356          After changing table of "verbs"       36
1357          After changing table of Posix names   22
1358    
1359        Thanks to the folks working on Gregex for glib for this insight.
1360    
1361    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
1362        unicode-properties was also set.
1363    
1364    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
1365    
1366    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
1367        checked only for CRLF.
1368    
1369    14. Added casts to pcretest.c to avoid compiler warnings.
1370    
1371    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
1372    
1373    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
1374        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
1375        entirely. This removes changes made in 7 above.
1376    
1377    17. The CMake files have been updated, and there is now more information about
1378        building with CMake in the NON-UNIX-USE document.
1379    
1380    
1381    Version 7.3 28-Aug-07
1382    ---------------------
1383    
1384     1. In the rejigging of the build system that eventually resulted in 7.1, the
1385        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
1386        brackets there is not right, since it causes compilers to look for an
1387        installed pcre.h, not the version that is in the source that is being
1388        compiled (which of course may be different). I have changed it back to:
1389    
1390          #include "pcre.h"
1391    
1392        I have a vague recollection that the change was concerned with compiling in
1393        different directories, but in the new build system, that is taken care of
1394        by the VPATH setting the Makefile.
1395    
1396     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
1397        when the subject happened to end in the byte 0x85 (e.g. if the last
1398        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
1399        characters but of course it shouldn't be taken as a newline when it is part
1400        of another character. The bug was that, for an unlimited repeat of . in
1401        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
1402        characters when looking for a newline.
1403    
1404     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
1405    
1406     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
1407        in debug output.
1408    
1409     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
1410        long printing in the pcrecpp unittest when running under MinGW.
1411    
1412     6. ESC_K was left out of the EBCDIC table.
1413    
1414     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
1415        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
1416        limit also applies to "virtual nesting" when a pattern is recursive, and in
1417        this case 1000 isn't so big. I have been able to remove this limit at the
1418        expense of backing off one optimization in certain circumstances. Normally,
1419        when pcre_exec() would call its internal match() function recursively and
1420        immediately return the result unconditionally, it uses a "tail recursion"
1421        feature to save stack. However, when a subpattern that can match an empty
1422        string has an unlimited repetition quantifier, it no longer makes this
1423        optimization. That gives it a stack frame in which to save the data for
1424        checking that an empty string has been matched. Previously this was taken
1425        from the 1000-entry workspace that had been reserved. So now there is no
1426        explicit limit, but more stack is used.
1427    
1428     8. Applied Daniel's patches to solve problems with the import/export magic
1429        syntax that is required for Windows, and which was going wrong for the
1430        pcreposix and pcrecpp parts of the library. These were overlooked when this
1431        problem was solved for the main library.
1432    
1433     9. There were some crude static tests to avoid integer overflow when computing
1434        the size of patterns that contain repeated groups with explicit upper
1435        limits. As the maximum quantifier is 65535, the maximum group length was
1436        set at 30,000 so that the product of these two numbers did not overflow a
1437        32-bit integer. However, it turns out that people want to use groups that
1438        are longer than 30,000 bytes (though not repeat them that many times).
1439        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1440        made it possible to implement the integer overflow checks in a much more
1441        dynamic way, which I have now done. The artificial limitation on group
1442        length has been removed - we now have only the limit on the total length of
1443        the compiled pattern, which depends on the LINK_SIZE setting.
1444    
1445    10. Fixed a bug in the documentation for get/copy named substring when
1446        duplicate names are permitted. If none of the named substrings are set, the
1447        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1448        empty string.
1449    
1450    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1451        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1452        because the ] is interpreted as the first data character and the
1453        terminating ] is not found. PCRE has been made compatible with Perl in this
1454        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1455        cause memory overwriting.
1456    
1457    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1458        string has been matched (to stop an infinite loop). It was not recognizing
1459        a conditional subpattern that could match an empty string if that
1460        subpattern was within another subpattern. For example, it looped when
1461        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
1462        condition was not nested. This bug has been fixed.
1463    
1464    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1465        past the start of the subject in the presence of bytes with the top bit
1466        set, for example "\x8aBCD".
1467    
1468    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1469        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1470    
1471    14. Optimized (?!) to (*FAIL).
1472    
1473    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1474        This restricts code points to be within the range 0 to 0x10FFFF, excluding
1475        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1476        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1477        does: it's just the validity check that is more restrictive.
1478    
1479    16. Inserted checks for integer overflows during escape sequence (backslash)
1480        processing, and also fixed erroneous offset values for syntax errors during
1481        backslash processing.
1482    
1483    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1484        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1485    
1486    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1487        caused an overrun.
1488    
1489    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1490        something other than just ASCII characters) inside a group that had an
1491        unlimited repeat caused a loop at compile time (while checking to see
1492        whether the group could match an empty string).
1493    
1494    20. Debugging a pattern containing \p or \P could cause a crash. For example,
1495        [\P{Any}] did so. (Error in the code for printing property names.)
1496    
1497    21. An orphan \E inside a character class could cause a crash.
1498    
1499    22. A repeated capturing bracket such as (A)? could cause a wild memory
1500        reference during compilation.
1501    
1502    23. There are several functions in pcre_compile() that scan along a compiled
1503        expression for various reasons (e.g. to see if it's fixed length for look
1504        behind). There were bugs in these functions when a repeated \p or \P was
1505        present in the pattern. These operators have additional parameters compared
1506        with \d, etc, and these were not being taken into account when moving along
1507        the compiled data. Specifically:
1508    
1509        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1510            length.
1511    
1512        (b) An item such as \pL+ within a repeated group could cause crashes or
1513            loops.
1514    
1515        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1516            "reference to non-existent subpattern" error.
1517    
1518        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1519    
1520    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1521        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1522    
1523    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1524    
1525    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1526        character were causing crashes (broken optimization).
1527    
1528    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1529        \p or \P) caused a compile-time loop.
1530    
1531    28. More problems have arisen in unanchored patterns when CRLF is a valid line
1532        break. For example, the unstudied pattern [\r\n]A does not match the string
1533        "\r\nA" because change 7.0/46 below moves the current point on by two
1534        characters after failing to match at the start. However, the pattern \nA
1535        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
1536        the same is true. There doesn't seem any very clean way out of this, but
1537        what I have chosen to do makes the common cases work: PCRE now takes note
1538        of whether there can be an explicit match for \r or \n anywhere in the
1539        pattern, and if so, 7.0/46 no longer applies. As part of this change,
1540        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1541        pattern has explicit CR or LF references.
1542    
1543    29. Added (*CR) etc for changing newline setting at start of pattern.
1544    
1545    
1546    Version 7.2 19-Jun-07
1547    ---------------------
1548    
1549     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
1550        which is apparently normally available under Windows.
1551    
1552     2. Re-jig the pcregrep tests with different newline settings in an attempt
1553        to make them independent of the local environment's newline setting.
1554    
1555     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
1556    
1557     4. Some of the "internals" tests were previously cut out when the link size
1558        was not 2, because the output contained actual offsets. The recent new
1559        "Z" feature of pcretest means that these can be cut out, making the tests
1560        usable with all link sizes.
1561    
1562     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
1563        stack recursion. This gives a massive performance boost under BSD, but just
1564        a small improvement under Linux. However, it saves one field in the frame
1565        in all cases.
1566    
1567     6. Added more features from the forthcoming Perl 5.10:
1568    
1569        (a) (?-n) (where n is a string of digits) is a relative subroutine or
1570            recursion call. It refers to the nth most recently opened parentheses.
1571    
1572        (b) (?+n) is also a relative subroutine call; it refers to the nth next
1573            to be opened parentheses.
1574    
1575        (c) Conditions that refer to capturing parentheses can be specified
1576            relatively, for example, (?(-2)... or (?(+3)...
1577    
1578        (d) \K resets the start of the current match so that everything before
1579            is not part of it.
1580    
1581        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1582    
1583        (f) \g{name} is another synonym - part of Perl 5.10's unification of
1584            reference syntax.
1585    
1586        (g) (?| introduces a group in which the numbering of parentheses in each
1587            alternative starts with the same number.
1588    
1589        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
1590    
1591     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
1592        PCRE_INFO_JCHANGED.
1593    
1594     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
1595        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
1596        for detecting groups that can match an empty string.
1597    
1598     9. A pattern with a very large number of alternatives (more than several
1599        hundred) was running out of internal workspace during the pre-compile
1600        phase, where pcre_compile() figures out how much memory will be needed. A
1601        bit of new cunning has reduced the workspace needed for groups with
1602        alternatives. The 1000-alternative test pattern now uses 12 bytes of
1603        workspace instead of running out of the 4096 that are available.
1604    
1605    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
1606    
1607    11. Applied patch from Google to remove an optimization that didn't quite work.
1608        The report of the bug said:
1609    
1610          pcrecpp::RE("a*").FullMatch("aaa") matches, while
1611          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
1612          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
1613    
1614    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
1615        it matched the wrong number of bytes.
1616    
1617    
1618    Version 7.1 24-Apr-07
1619    ---------------------
1620    
1621     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
1622        that is more "standard", making use of automake and other Autotools. There
1623        is some re-arrangement of the files and adjustment of comments consequent
1624        on this.
1625    
1626     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
1627        for recursive directory scanning broke on some systems because the files
1628        are not scanned in any specific order and on different systems the order
1629        was different. A call to "sort" has been inserted into RunGrepTest for the
1630        approprate test as a short-term fix. In the longer term there may be an
1631        alternative.
1632    
1633     3. I had an email from Eric Raymond about problems translating some of PCRE's
1634        man pages to HTML (despite the fact that I distribute HTML pages, some
1635        people do their own conversions for various reasons). The problems
1636        concerned the use of low-level troff macros .br and .in. I have therefore
1637        removed all such uses from the man pages (some were redundant, some could
1638        be replaced by .nf/.fi pairs). The 132html script that I use to generate
1639        HTML has been updated to handle .nf/.fi and to complain if it encounters
1640        .br or .in.
1641    
1642     4. Updated comments in configure.ac that get placed in config.h.in and also
1643        arranged for config.h to be included in the distribution, with the name
1644        config.h.generic, for the benefit of those who have to compile without
1645        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
1646    
1647     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
1648        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
1649        makevp.bat for the current PCRE, using the additional files
1650        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
1651    
1652     6. A Windows user reported a minor discrepancy with test 2, which turned out
1653        to be caused by a trailing space on an input line that had got lost in his
1654        copy. The trailing space was an accident, so I've just removed it.
1655    
1656     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
1657        that is needed.
1658    
1659     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
1660        as "const" (a) because they are and (b) because it helps the PHP
1661        maintainers who have recently made a script to detect big data structures
1662        in the php code that should be moved to the .rodata section. I remembered
1663        to update Builducptable as well, so it won't revert if ucptable.h is ever
1664        re-created.
1665    
1666     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
1667        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
1668        order to be able to cut out the UTF-8 tables in the latter when UTF-8
1669        support is not required. This saves 1.5-2K of code, which is important in
1670        some applications.
1671    
1672        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
1673        so as not to refer to the tables, even though these functions will never be
1674        called when UTF-8 support is disabled. Otherwise there are problems with a
1675        shared library.
1676    
1677    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
1678    
1679        (a) It was defining its arguments as char * instead of void *.
1680    
1681        (b) It was assuming that all moves were upwards in memory; this was true
1682            a long time ago when I wrote it, but is no longer the case.
1683    
1684        The emulated memove() is provided for those environments that have neither
1685        memmove() nor bcopy(). I didn't think anyone used it these days, but that
1686        is clearly not the case, as these two bugs were recently reported.
1687    
1688    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
1689        and Detrail to create the HTML documentation, the .txt form of the man
1690        pages, and it removes trailing spaces from listed files. It also creates
1691        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
1692        case, it wraps all the #defines with #ifndefs. This script should be run
1693        before "make dist".
1694    
1695    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1696        with Unicode property support.
1697    
1698        (a) For a maximizing quantifier, if the two different cases of the
1699            character were of different lengths in their UTF-8 codings (there are
1700            some cases like this - I found 11), and the matching function had to
1701            back up over a mixture of the two cases, it incorrectly assumed they
1702            were both the same length.
1703    
1704        (b) When PCRE was configured to use the heap rather than the stack for
1705            recursion during matching, it was not correctly preserving the data for
1706            the other case of a UTF-8 character when checking ahead for a match
1707            while processing a minimizing repeat. If the check also involved
1708            matching a wide character, but failed, corruption could cause an
1709            erroneous result when trying to check for a repeat of the original
1710            character.
1711    
1712    13. Some tidying changes to the testing mechanism:
1713    
1714        (a) The RunTest script now detects the internal link size and whether there
1715            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1716            values substituted by "configure". (The RunGrepTest script already did
1717            this for UTF-8.) The configure.ac script no longer substitutes the
1718            relevant variables.
1719    
1720        (b) The debugging options /B and /D in pcretest show the compiled bytecode
1721            with length and offset values. This means that the output is different
1722            for different internal link sizes. Test 2 is skipped for link sizes
1723            other than 2 because of this, bypassing the problem. Unfortunately,
1724            there was also a test in test 3 (the locale tests) that used /B and
1725            failed for link sizes other than 2. Rather than cut the whole test out,
1726            I have added a new /Z option to pcretest that replaces the length and
1727            offset values with spaces. This is now used to make test 3 independent
1728            of link size. (Test 2 will be tidied up later.)
1729    
1730    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1731        segmentation fault instead of returning the appropriate error message.
1732    
1733    15. In multiline mode when the newline sequence was set to "any", the pattern
1734        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1735        This doesn't seem right; it now treats the CRLF combination as the line
1736        ending, and so does not match in that case. It's only a pattern such as ^$
1737        that would hit this one: something like ^ABC$ would have failed after \r
1738        and then tried again after \r\n.
1739    
1740    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1741        in an attempt to make files that differ only in their line terminators
1742        compare equal. This works on Linux.
1743    
1744    17. Under certain error circumstances pcregrep might try to free random memory
1745        as it exited. This is now fixed, thanks to valgrind.
1746    
1747    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1748        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1749        was because its rules for how to advance for /g after matching an empty
1750        string at the end of a line did not allow for this case. They now check for
1751        it specially.
1752    
1753    20. pcretest is supposed to handle patterns and data of any length, by
1754        extending its buffers when necessary. It was getting this wrong when the
1755        buffer for a data line had to be extended.
1756    
1757    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1758        CRLF as a newline sequence.
1759    
1760    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1761        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1762        I have nevertheless tidied it up.
1763    
1764    23. Added some casts to kill warnings from HP-UX ia64 compiler.
1765    
1766    24. Added a man page for pcre-config.
1767    
1768    
1769    Version 7.0 19-Dec-06
1770    ---------------------
1771    
1772     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1773        moving to gcc 4.1.1.
1774    
1775     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1776        sys/time.h, which is documented as needed for this function. It doesn't
1777        seem to matter on Linux, but it showed up on some releases of OS X.
1778    
1779     3. It seems that there are systems where bytes whose values are greater than
1780        127 match isprint() in the "C" locale. The "C" locale should be the
1781        default when a C program starts up. In most systems, only ASCII printing
1782        characters match isprint(). This difference caused the output from pcretest
1783        to vary, making some of the tests fail. I have changed pcretest so that:
1784    
1785        (a) When it is outputting text in the compiled version of a pattern, bytes
1786            other than 32-126 are always shown as hex escapes.
1787    
1788        (b) When it is outputting text that is a matched part of a subject string,
1789            it does the same, unless a different locale has been set for the match
1790            (using the /L modifier). In this case, it uses isprint() to decide.
1791    
1792     4. Fixed a major bug that caused incorrect computation of the amount of memory
1793        required for a compiled pattern when options that changed within the
1794        pattern affected the logic of the preliminary scan that determines the
1795        length. The relevant options are -x, and -i in UTF-8 mode. The result was
1796        that the computed length was too small. The symptoms of this bug were
1797        either the PCRE error "internal error: code overflow" from pcre_compile(),
1798        or a glibc crash with a message such as "pcretest: free(): invalid next
1799        size (fast)". Examples of patterns that provoked this bug (shown in
1800        pcretest format) are:
1801    
1802          /(?-x: )/x
1803          /(?x)(?-x: \s*#\s*)/
1804          /((?i)[\x{c0}])/8
1805          /(?i:[\x{c0}])/8
1806    
1807        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1808        is now done differently.
1809    
1810     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1811        wrapper classes; (b) implement a new function in the C++ scanner that is
1812        more efficient than the old way of doing things because it avoids levels of
1813        recursion in the regex matching; (c) add a paragraph to the documentation
1814        for the FullMatch() function.
1815    
1816     6. The escape sequence \n was being treated as whatever was defined as
1817        "newline". Not only was this contrary to the documentation, which states
1818        that \n is character 10 (hex 0A), but it also went horribly wrong when
1819        "newline" was defined as CRLF. This has been fixed.
1820    
1821     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1822        was being set to -1 for the "end of line" case (supposedly a value that no
1823        character can have). Though this value is never used (the check for end of
1824        line is "zero bytes in current character"), it caused compiler complaints.
1825        I've changed it to 0xffffffff.
1826    
1827     8. In pcre_version.c, the version string was being built by a sequence of
1828        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1829        string (as it is for production releases) called a macro with an empty
1830        argument. The C standard says the result of this is undefined. The gcc
1831        compiler treats it as an empty string (which was what was wanted) but it is
1832        reported that Visual C gives an error. The source has been hacked around to
1833        avoid this problem.
1834    
1835     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1836        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1837        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1838        of them did).
1839    
1840    10. Originally, pcretest opened its input and output without "b"; then I was
1841        told that "b" was needed in some environments, so it was added for release
1842        5.0 to both the input and output. (It makes no difference on Unix-like
1843        systems.) Later I was told that it is wrong for the input on Windows. I've
1844        now abstracted the modes into two macros, to make it easier to fiddle with
1845        them, and removed "b" from the input mode under Windows.
1846    
1847    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1848    
1849    12. Added -help and --help to pcretest as an official way of being reminded
1850        of the options.
1851    
1852    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1853        and pcrecpp.cc because they annoy compilers at high warning levels.
1854    
1855    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1856    
1857    15. Fixed an occurrence of == in configure.ac that should have been = (shell
1858        scripts are not C programs :-) and which was not noticed because it works
1859        on Linux.
1860    
1861    16. pcretest is supposed to handle any length of pattern and data line (as one
1862        line or as a continued sequence of lines) by extending its input buffer if
1863        necessary. This feature was broken for very long pattern lines, leading to
1864        a string of junk being passed to pcre_compile() if the pattern was longer
1865        than about 50K.
1866    
1867    17. I have done a major re-factoring of the way pcre_compile() computes the
1868        amount of memory needed for a compiled pattern. Previously, there was code
1869        that made a preliminary scan of the pattern in order to do this. That was
1870        OK when PCRE was new, but as the facilities have expanded, it has become
1871        harder and harder to keep it in step with the real compile phase, and there
1872        have been a number of bugs (see for example, 4 above). I have now found a
1873        cunning way of running the real compile function in a "fake" mode that
1874        enables it to compute how much memory it would need, while actually only
1875        ever using a few hundred bytes of working memory and without too many
1876        tests of the mode. This should make future maintenance and development
1877        easier. A side effect of this work is that the limit of 200 on the nesting
1878        depth of parentheses has been removed (though this was never a serious
1879        limitation, I suspect). However, there is a downside: pcre_compile() now
1880        runs more slowly than before (30% or more, depending on the pattern). I
1881        hope this isn't a big issue. There is no effect on runtime performance.
1882    
1883    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1884        newline (only possible for the last line of a file) and it was a
1885        pattern that set a locale (followed by /Lsomething), pcretest crashed.
1886    
1887    19. Added additional timing features to pcretest. (1) The -tm option now times
1888        matching only, not compiling. (2) Both -t and -tm can be followed, as a
1889        separate command line item, by a number that specifies the number of
1890        repeats to use when timing. The default is 50000; this gives better
1891        precision, but takes uncomfortably long for very large patterns.
1892    
1893    20. Extended pcre_study() to be more clever in cases where a branch of a
1894        subpattern has no definite first character. For example, (a*|b*)[cd] would
1895        previously give no result from pcre_study(). Now it recognizes that the
1896        first character must be a, b, c, or d.
1897    
1898    21. There was an incorrect error "recursive call could loop indefinitely" if
1899        a subpattern (or the entire pattern) that was being tested for matching an
1900        empty string contained only one non-empty item after a nested subpattern.
1901        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1902        incorrectly, because the \d was being skipped in the check.
1903    
1904    22. The pcretest program now has a new pattern option /B and a command line
1905        option -b, which is equivalent to adding /B to every pattern. This causes
1906        it to show the compiled bytecode, without the additional information that
1907        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1908        is the same as /B/I).
1909    
1910    23. A new optimization is now able automatically to treat some sequences such
1911        as a*b as a*+b. More specifically, if something simple (such as a character
1912        or a simple class like \d) has an unlimited quantifier, and is followed by
1913        something that cannot possibly match the quantified thing, the quantifier
1914        is automatically "possessified".
1915    
1916    24. A recursive reference to a subpattern whose number was greater than 39
1917        went wrong under certain circumstances in UTF-8 mode. This bug could also
1918        have affected the operation of pcre_study().
1919    
1920    25. Realized that a little bit of performance could be had by replacing
1921        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1922    
1923    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1924    
1925    27. Possessive quantifiers such as a++ were previously implemented by turning
1926        them into atomic groups such as ($>a+). Now they have their own opcodes,
1927        which improves performance. This includes the automatically created ones
1928        from 23 above.
1929    
1930    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1931        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1932        the first matched character to be a colon. This applied both to named and
1933        numbered groups.
1934    
1935    29. The ucpinternal.h header file was missing its idempotency #ifdef.
1936    
1937    30. I was sent a "project" file called libpcre.a.dev which I understand makes
1938        building PCRE on Windows easier, so I have included it in the distribution.
1939    
1940    31. There is now a check in pcretest against a ridiculously large number being
1941        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1942        loop, the loop is abandoned.
1943    
1944    32. Forward references to subpatterns in conditions such as (?(2)...) where
1945        subpattern 2 is defined later cause pcre_compile() to search forwards in
1946        the pattern for the relevant set of parentheses. This search went wrong
1947        when there were unescaped parentheses in a character class, parentheses
1948        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1949    
1950    33. "Subroutine" calls and backreferences were previously restricted to
1951        referencing subpatterns earlier in the regex. This restriction has now
1952        been removed.
1953    
1954    34. Added a number of extra features that are going to be in Perl 5.10. On the
1955        whole, these are just syntactic alternatives for features that PCRE had
1956        previously implemented using the Python syntax or my own invention. The
1957        other formats are all retained for compatibility.
1958    
1959        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1960            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1961            also .NET compatible.
1962    
1963        (b) A recursion or subroutine call to a named group can now be defined as
1964            (?&name) as well as (?P>name).
1965    
1966        (c) A backreference to a named group can now be defined as \k<name> or
1967            \k'name' as well as (?P=name). The new forms, as well as being in Perl
1968            5.10, are also .NET compatible.
1969    
1970        (d) A conditional reference to a named group can now use the syntax
1971            (?(<name>) or (?('name') as well as (?(name).
1972    
1973        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1974            groups (named and numbered) that are never evaluated inline, but can be
1975            called as "subroutines" from elsewhere. In effect, the DEFINE condition
1976            is always false. There may be only one alternative in such a group.
1977    
1978        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1979            as the simple (?(R). The condition is true only if the most recent
1980            recursion is that of the given number or name. It does not search out
1981            through the entire recursion stack.
1982    
1983        (g) The escape \gN or \g{N} has been added, where N is a positive or
1984            negative number, specifying an absolute or relative reference.
1985    
1986    35. Tidied to get rid of some further signed/unsigned compiler warnings and
1987        some "unreachable code" warnings.
1988    
1989    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1990        things, this adds five new scripts.
1991    
1992    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1993        There were also incompatibilities regarding the handling of \Q..\E inside
1994        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1995        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1996    
1997    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1998        matches an empty string, and forcibly breaks the loop. There were bugs in
1999        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
2000        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
2001        separate and independent bugs (that affected different cases) have been
2002        fixed.
2003    
2004    39. Refactored the code to abolish the use of different opcodes for small
2005        capturing bracket numbers. This is a tidy that I avoided doing when I
2006        removed the limit on the number of capturing brackets for 3.5 back in 2001.
2007        The new approach is not only tidier, it makes it possible to reduce the
2008        memory needed to fix the previous bug (38).
2009    
2010    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
2011        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
2012        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
2013        mode.
2014    
2015    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
2016        report.
2017    
2018    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
2019        copy construction and assignment in the C++ wrapper.
2020    
2021    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
2022        couple of bugs that could have given wrong results in the "--newline=crlf"
2023        case.
2024    
2025    44. Added a number of casts and did some reorganization of signed/unsigned int
2026        variables following suggestions from Dair Grant. Also renamed the variable
2027        "this" as "item" because it is a C++ keyword.
2028    
2029    45. Arranged for dftables to add
2030    
2031          #include "pcre_internal.h"
2032    
2033        to pcre_chartables.c because without it, gcc 4.x may remove the array
2034        definition from the final binary if PCRE is built into a static library and
2035        dead code stripping is activated.
2036    
2037    46. For an unanchored pattern, if a match attempt fails at the start of a
2038        newline sequence, and the newline setting is CRLF or ANY, and the next two
2039        characters are CRLF, advance by two characters instead of one.
2040    
2041    
2042    Version 6.7 04-Jul-06
2043    ---------------------
2044    
2045     1. In order to handle tests when input lines are enormously long, pcretest has
2046        been re-factored so that it automatically extends its buffers when
2047        necessary. The code is crude, but this _is_ just a test program. The
2048        default size has been increased from 32K to 50K.
2049    
2050     2. The code in pcre_study() was using the value of the re argument before
2051        testing it for NULL. (Of course, in any sensible call of the function, it
2052        won't be NULL.)
2053    
2054     3. The memmove() emulation function in pcre_internal.h, which is used on
2055        systems that lack both memmove() and bcopy() - that is, hardly ever -
2056        was missing a "static" storage class specifier.
2057    
2058     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
2059        containing an extended class (one that cannot be represented by a bitmap
2060        because it contains high-valued characters or Unicode property items, e.g.
2061        [\pZ]). Almost always one would set UTF-8 mode when processing such a
2062        pattern, but PCRE should not loop if you do not (it no longer does).
2063        [Detail: two cases were found: (a) a repeated subpattern containing an
2064        extended class; (b) a recursive reference to a subpattern that followed a
2065        previous extended class. It wasn't skipping over the extended class
2066        correctly when UTF-8 mode was not set.]
2067    
2068     5. A negated single-character class was not being recognized as fixed-length
2069        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
2070        compile error "lookbehind assertion is not fixed length".
2071    
2072     6. The RunPerlTest auxiliary script was showing an unexpected difference
2073        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
2074        write a Perl script that can interpret lines of an input file either as
2075        byte characters or as UTF-8, which is what "perltest" was being required to
2076        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
2077        can't do is switch easily at run time between having the "use utf8;" pragma
2078        or not. In the end, I fudged it by using the RunPerlTest script to insert
2079        "use utf8;" explicitly for the UTF-8 tests.
2080    
2081     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
2082        the end of the subject string, contrary to the documentation and to what
2083        Perl does. This was true of both matching functions. Now it matches only at
2084        the start of the subject and immediately after *internal* newlines.
2085    
2086     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
2087        a pointer to an int instead of a pointer to an unsigned long int. This
2088        caused problems on 64-bit systems.
2089    
2090     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
2091        instance of the 'standard' template library not being so standard".
2092    
2093    10. There was no check on the number of named subpatterns nor the maximum
2094        length of a subpattern name. The product of these values is used to compute
2095        the size of the memory block for a compiled pattern. By supplying a very
2096        long subpattern name and a large number of named subpatterns, the size
2097        computation could be caused to overflow. This is now prevented by limiting
2098        the length of names to 32 characters, and the number of named subpatterns
2099        to 10,000.
2100    
2101    11. Subpatterns that are repeated with specific counts have to be replicated in
2102        the compiled pattern. The size of memory for this was computed from the
2103        length of the subpattern and the repeat count. The latter is limited to
2104        65535, but there was no limit on the former, meaning that integer overflow
2105        could in principle occur. The compiled length of a repeated subpattern is
2106        now limited to 30,000 bytes in order to prevent this.
2107    
2108    12. Added the optional facility to have named substrings with the same name.
2109    
2110    13. Added the ability to use a named substring as a condition, using the
2111        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
2112        are numbers (not recommended). Forward references are permitted.
2113    
2114    14. Added forward references in named backreferences (if you see what I mean).
2115    
2116    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
2117        pattern could run off the end of the subject. For example, the pattern
2118        "(?s)(.{1,5})"8 did this with the subject "ab".
2119    
2120    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
2121        PCRE_CASELESS was set when matching characters that were quantified with ?
2122        or *.
2123    
2124    17. A character class other than a single negated character that had a minimum
2125        but no maximum quantifier - for example [ab]{6,} - was not handled
2126        correctly by pce_dfa_exec(). It would match only one character.
2127    
2128    18. A valid (though odd) pattern that looked like a POSIX character
2129        class but used an invalid character after [ (for example [[,abc,]]) caused
2130        pcre_compile() to give the error "Failed: internal error: code overflow" or
2131        in some cases to crash with a glibc free() error. This could even happen if
2132        the pattern terminated after [[ but there just happened to be a sequence of
2133        letters, a binary zero, and a closing ] in the memory that followed.
2134    
2135    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
2136        over the years. Originally (before any Unicode support), just the bottom 8
2137        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
2138        output from "man perlunicode" includes this:
2139    
2140          The regular expression compiler produces polymorphic opcodes.  That
2141          is, the pattern adapts to the data and automatically switches to
2142          the Unicode character scheme when presented with Unicode data--or
2143          instead uses a traditional byte scheme when presented with byte
2144          data.
2145    
2146        Sadly, a wide octal escape does not cause a switch, and in a string with
2147        no other multibyte characters, these octal escapes are treated as before.
2148        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
2149        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
2150        Unicode string.
2151    
2152        I have not perpetrated such confusion in PCRE. Up till now, it took just
2153        the bottom 8 bits, as in old Perl. I have now made octal escapes with
2154        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
2155        translate to the appropriate multibyte character.
2156    
2157    29. Applied some refactoring to reduce the number of warnings from Microsoft
2158        and Borland compilers. This has included removing the fudge introduced
2159        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
2160        a warning about an unused variable.
2161    
2162    21. PCRE has not included VT (character 0x0b) in the set of whitespace
2163        characters since release 4.0, because Perl (from release 5.004) does not.
2164        [Or at least, is documented not to: some releases seem to be in conflict
2165        with the documentation.] However, when a pattern was studied with
2166        pcre_study() and all its branches started with \s, PCRE still included VT
2167        as a possible starting character. Of course, this did no harm; it just
2168        caused an unnecessary match attempt.
2169    
2170    22. Removed a now-redundant internal flag bit that recorded the fact that case
2171        dependency changed within the pattern. This was once needed for "required
2172        byte" processing, but is no longer used. This recovers a now-scarce options
2173        bit. Also moved the least significant internal flag bit to the most-
2174        significant bit of the word, which was not previously used (hangover from
2175        the days when it was an int rather than a uint) to free up another bit for
2176        the future.
2177    
2178    23. Added support for CRLF line endings as well as CR and LF. As well as the
2179        default being selectable at build time, it can now be changed at runtime
2180        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
2181        specify that it is scanning data with non-default line endings.
2182    
2183    24. Changed the definition of CXXLINK to make it agree with the definition of
2184        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
2185    
2186    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
2187        recursions. This makes a big different to stack usage for some patterns.
2188    
2189    26. If a subpattern containing a named recursion or subroutine reference such
2190        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
2191        the space required for the compiled pattern went wrong and gave too small a
2192        value. Depending on the environment, this could lead to "Failed: internal
2193        error: code overflow at offset 49" or "glibc detected double free or
2194        corruption" errors.
2195    
2196    27. Applied patches from Google (a) to support the new newline modes and (b) to
2197        advance over multibyte UTF-8 characters in GlobalReplace.
2198    
2199    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
2200        difference for some implementation of PCRE in some Windows version.
2201    
2202    29. Added some extra testing facilities to pcretest:
2203    
2204        \q<number>   in a data line sets the "match limit" value
2205        \Q<number>   in a data line sets the "match recursion limt" value
2206        -S <number>  sets the stack size, where <number> is in megabytes
2207    
2208        The -S option isn't available for Windows.
2209    
2210    
2211    Version 6.6 06-Feb-06
2212    ---------------------
2213    
2214     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
2215        in pcreposix.h. I have copied the definition from pcre.h.
2216    
2217     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
2218        because pcre.h is no longer a built file.
2219    
2220     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
2221        not normally included in the compiled code.
2222    
2223    
2224    Version 6.5 01-Feb-06
2225    ---------------------
2226    
2227     1. When using the partial match feature with pcre_dfa_exec(), it was not
2228        anchoring the second and subsequent partial matches at the new starting
2229        point. This could lead to incorrect results. For example, with the pattern
2230        /1234/, partially matching against "123" and then "a4" gave a match.
2231    
2232     2. Changes to pcregrep:
2233    
2234        (a) All non-match returns from pcre_exec() were being treated as failures
2235            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
2236            error message is output. Some extra information is given for the
2237            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
2238            probably the only errors that are likely to be caused by users (by
2239            specifying a regex that has nested indefinite repeats, for instance).
2240            If there are more than 20 of these errors, pcregrep is abandoned.
2241    
2242        (b) A binary zero was treated as data while matching, but terminated the
2243            output line if it was written out. This has been fixed: binary zeroes
2244            are now no different to any other data bytes.
2245    
2246        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
2247            used to set a locale for matching. The --locale=xxxx long option has
2248            been added (no short equivalent) to specify a locale explicitly on the
2249            pcregrep command, overriding the environment variables.
2250    
2251        (d) When -B was used with -n, some line numbers in the output were one less
2252            than they should have been.
2253    
2254        (e) Added the -o (--only-matching) option.
2255    
2256        (f) If -A or -C was used with -c (count only), some lines of context were
2257            accidentally printed for the final match.
2258    
2259        (g) Added the -H (--with-filename) option.
2260    
2261        (h) The combination of options -rh failed to suppress file names for files
2262            that were found from directory arguments.
2263    
2264        (i) Added the -D (--devices) and -d (--directories) options.
2265    
2266        (j) Added the -F (--fixed-strings) option.
2267    
2268        (k) Allow "-" to be used as a file name for -f as well as for a data file.
2269    
2270        (l) Added the --colo(u)r option.
2271    
2272        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
2273            is not present by default.
2274    
2275     3. A nasty bug was discovered in the handling of recursive patterns, that is,
2276        items such as (?R) or (?1), when the recursion could match a number of
2277        alternatives. If it matched one of the alternatives, but subsequently,
2278        outside the recursion, there was a failure, the code tried to back up into
2279        the recursion. However, because of the way PCRE is implemented, this is not
2280        possible, and the result was an incorrect result from the match.
2281    
2282        In order to prevent this happening, the specification of recursion has
2283        been changed so that all such subpatterns are automatically treated as
2284        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
2285    
2286     4. I had overlooked the fact that, in some locales, there are characters for
2287        which isalpha() is true but neither isupper() nor islower() are true. In
2288        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
2289        and ordfeminine) are like this. This affected the treatment of \w and \W
2290        when they appeared in character classes, but not when they appeared outside
2291        a character class. The bit map for "word" characters is now created
2292        separately from the results of isalnum() instead of just taking it from the
2293        upper, lower, and digit maps. (Plus the underscore character, of course.)
2294    
2295     5. The above bug also affected the handling of POSIX character classes such as
2296        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
2297        permanent tables. Instead, the bit maps for such a class were previously
2298        created as the appropriate unions of the upper, lower, and digit bitmaps.
2299        Now they are created by subtraction from the [[:word:]] class, which has
2300        its own bitmap.
2301    
2302     6. The [[:blank:]] character class matches horizontal, but not vertical space.
2303        It is created by subtracting the vertical space characters (\x09, \x0a,
2304        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
2305        subtraction was done in the overall bitmap for a character class, meaning
2306        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
2307        be recognized. This bug has been fixed.
2308    
2309     7. Patches from the folks at Google:
2310    
2311          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
2312          real life, but is still worth protecting against".
2313    
2314          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
2315          regular expressions".
2316    
2317          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
2318          have it.
2319    
2320          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
2321          "configure" and the latter not, in order to fix a problem somebody had
2322          with compiling the Arg class on HP-UX.
2323    
2324          (e) Improve the error-handling of the C++ wrapper a little bit.
2325    
2326          (f) New tests for checking recursion limiting.
2327    
2328     8. The pcre_memmove() function, which is used only if the environment does not
2329        have a standard memmove() function (and is therefore rarely compiled),
2330        contained two bugs: (a) use of int instead of size_t, and (b) it was not
2331        returning a result (though PCRE never actually uses the result).
2332    
2333     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
2334        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
2335        returned instead of calling malloc() with an overflowing number that would
2336        most likely cause subsequent chaos.
2337    
2338    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
2339    
2340    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
2341        with this option is matched, the nmatch and pmatch options of regexec() are
2342        ignored.
2343    
2344    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
2345        provided in case anyone wants to the the POSIX interface with UTF-8
2346        strings.
2347    
2348    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
2349        C++ linking (needed for some HP-UX environments).
2350    
2351    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
2352        (unused parameter) and in the pcre_printint() function (omitted "default"
2353        switch label when the default is to do nothing).
2354    
2355    15. Added some code to make it possible, when PCRE is compiled as a C++
2356        library, to replace subject pointers for pcre_exec() with a smart pointer
2357        class, thus making it possible to process discontinuous strings.
2358    
2359    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
2360        much the same function. They were added by different people who were trying
2361        to make PCRE easy to compile on non-Unix systems. It has been suggested
2362        that PCRE_EXPORT be abolished now that there is more automatic apparatus
2363        for compiling on Windows systems. I have therefore replaced it with
2364        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
2365        defaults to "extern" for C or "extern C" for C++, which works fine on
2366        Unix-like systems. It is now possible to override the value of PCRE_DATA_
2367        SCOPE with something explicit in config.h. In addition:
2368    
2369        (a) pcreposix.h still had just "extern" instead of either of these macros;
2370            I have replaced it with PCRE_DATA_SCOPE.
2371    
2372        (b) Functions such as _pcre_xclass(), which are internal to the library,
2373            but external in the C sense, all had PCRE_EXPORT in their definitions.
2374            This is apparently wrong for the Windows case, so I have removed it.
2375            (It makes no difference on Unix-like systems.)
2376    
2377    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
2378        of recursive calls to match(). This is different to MATCH_LIMIT because
2379        that limits the total number of calls to match(), not all of which increase
2380        the depth of recursion. Limiting the recursion depth limits the amount of
2381        stack (or heap if NO_RECURSE is set) that is used. The default can be set
2382        when PCRE is compiled, and changed at run time. A patch from Google adds
2383        this functionality to the C++ interface.
2384    
2385    18. Changes to the handling of Unicode character properties:
2386    
2387        (a) Updated the table to Unicode 4.1.0.
2388    
2389        (b) Recognize characters that are not in the table as "Cn" (undefined).
2390    
2391        (c) I revised the way the table is implemented to a much improved format
2392            which includes recognition of ranges. It now supports the ranges that
2393            are defined in UnicodeData.txt, and it also amalgamates other
2394            characters into ranges. This has reduced the number of entries in the
2395            table from around 16,000 to around 3,000, thus reducing its size
2396            considerably. I realized I did not need to use a tree structure after
2397            all - a binary chop search is just as efficient. Having reduced the
2398            number of entries, I extended their size from 6 bytes to 8 bytes to
2399            allow for more data.
2400    
2401        (d) Added support for Unicode script names via properties such as \p{Han}.
2402    
2403    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
2404        matching that character.
2405    
2406    20. When matching a repeated Unicode property with a minimum greater than zero,
2407        (for example \pL{2,}), PCRE could look past the end of the subject if it
2408        reached it while seeking the minimum number of characters. This could
2409        happen only if some of the characters were more than one byte long, because
2410        there is a check for at least the minimum number of bytes.
2411    
2412    21. Refactored the implementation of \p and \P so as to be more general, to
2413        allow for more different types of property in future. This has changed the
2414        compiled form incompatibly. Anybody with saved compiled patterns that use
2415        \p or \P will have to recompile them.
2416    
2417    22. Added "Any" and "L&" to the supported property types.
2418    
2419    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
2420        but give a compile time error if the value is greater than 0xff.
2421    
2422    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
2423        accidentally not being installed or uninstalled.
2424    
2425    25. The pcre.h file was built from pcre.h.in, but the only changes that were
2426        made were to insert the current release number. This seemed silly, because
2427        it made things harder for people building PCRE on systems that don't run
2428        "configure". I have turned pcre.h into a distributed file, no longer built
2429        by "configure", with the version identification directly included. There is
2430        no longer a pcre.h.in file.
2431    
2432        However, this change necessitated a change to the pcre-config script as
2433        well. It is built from pcre-config.in, and one of the substitutions was the
2434        release number. I have updated configure.ac so that ./configure now finds
2435        the release number by grepping pcre.h.
2436    
2437    26. Added the ability to run the tests under valgrind.
2438    
2439    
2440    Version 6.4 05-Sep-05
2441    ---------------------
2442    
2443     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
2444        "--" to be printed when multiple files were scanned, even when none of the
2445        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
2446        consider it to be a bug, and have restored the previous behaviour.
2447    
2448     2. A couple of code tidies to get rid of compiler warnings.
2449    
2450     3. The pcretest program used to cheat by referring to symbols in the library
2451        whose names begin with _pcre_. These are internal symbols that are not
2452        really supposed to be visible externally, and in some environments it is
2453        possible to suppress them. The cheating is now confined to including
2454        certain files from the library's source, which is a bit cleaner.
2455    
2456     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
2457        file's purpose clearer.
2458    
2459     5. Reorganized pcre_ucp_findchar().
2460    
2461    
2462    Version 6.3 15-Aug-05
2463    ---------------------
2464    
2465     1. The file libpcre.pc.in did not have general read permission in the tarball.
2466    
2467     2. There were some problems when building without C++ support:
2468    
2469        (a) If C++ support was not built, "make install" and "make test" still
2470            tried to test it.
2471    
2472        (b) There were problems when the value of CXX was explicitly set. Some
2473            changes have been made to try to fix these, and ...
2474    
2475        (c) --disable-cpp can now be used to explicitly disable C++ support.
2476    
2477        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
2478            backslash in a target when C++ was disabled. This confuses some
2479            versions of "make", apparently. Using an intermediate variable solves
2480            this. (Same for CPP_LOBJ.)
2481    
2482     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
2483        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
2484        necessary on certain architectures.
2485    
2486     4. Added a setting of -export-symbols-regex to the link command to remove
2487        those symbols that are exported in the C sense, but actually are local
2488        within the library, and not documented. Their names all begin with
2489        "_pcre_". This is not a perfect job, because (a) we have to except some
2490        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
2491        available (and never for static libraries). I have made a note to try to
2492        find a way round (a) in the future.
2493    
2494    
2495    Version 6.2 01-Aug-05
2496    ---------------------
2497    
2498     1. There was no test for integer overflow of quantifier values. A construction
2499        such as {1111111111111111} would give undefined results. What is worse, if
2500        a minimum quantifier for a parenthesized subpattern overflowed and became
2501        negative, the calculation of the memory size went wrong. This could have
2502        led to memory overwriting.
2503    
2504     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
2505    
2506     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
2507        operating environments where this matters.
2508    
2509     4. Applied Giuseppe Maxia's patch to add additional features for controlling
2510        PCRE options from within the C++ wrapper.
2511    
2512     5. Named capturing subpatterns were not being correctly counted when a pattern
2513        was compiled. This caused two problems: (a) If there were more than 100
2514        such subpatterns, the calculation of the memory needed for the whole
2515        compiled pattern went wrong, leading to an overflow error. (b) Numerical
2516        back references of the form \12, where the number was greater than 9, were
2517        not recognized as back references, even though there were sufficient
2518        previous subpatterns.
2519    
2520     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
2521        versions of gcc, e.g. 2.95.4.
2522    
2523    
2524    Version 6.1 21-Jun-05
2525    ---------------------
2526    
2527     1. There was one reference to the variable "posix" in pcretest.c that was not
2528        surrounded by "#if !defined NOPOSIX".
2529    
2530     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
2531        the cross-check on the old pcre_info() function, for the benefit of the
2532        cut-down version of PCRE that is currently imported into Exim.
2533    
2534     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
2535        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
2536        patterns that start (?i) but I don't think that matters. The use of (?i) is
2537        just an example; this all applies to the other options as well.
2538    
2539     4. Since libtool seems to echo the compile commands it is issuing, the output
2540        from "make" can be reduced a bit by putting "@" in front of each libtool
2541        compile command.
2542    
2543     5. Patch from the folks at Google for configure.in to be a bit more thorough
2544        in checking for a suitable C++ installation before trying to compile the
2545        C++ stuff. This should fix a reported problem when a compiler was present,
2546        but no suitable headers.
2547    
2548     6. The man pages all had just "PCRE" as their title. I have changed them to
2549        be the relevant file name. I have also arranged that these names are
2550        retained in the file doc/pcre.txt, which is a concatenation in text format
2551        of all the man pages except the little individual ones for each function.
2552    
2553     7. The NON-UNIX-USE file had not been updated for the different set of source
2554        files that come with release 6. I also added a few comments about the C++
2555        wrapper.
2556    
2557    
2558    Version 6.0 07-Jun-05
2559    ---------------------
2560    
2561     1. Some minor internal re-organization to help with my DFA experiments.
2562    
2563     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
2564        didn't matter for the library itself when fully configured, but did matter
2565        when compiling without UCP support, or within Exim, where the ucp files are
2566        not imported.
2567    
2568     3. Refactoring of the library code to split up the various functions into
2569        different source modules. The addition of the new DFA matching code (see
2570        below) to a single monolithic source would have made it really too
2571        unwieldy, quite apart from causing all the code to be include in a
2572        statically linked application, when only some functions are used. This is
2573        relevant even without the DFA addition now that patterns can be compiled in
2574        one application and matched in another.
2575    
2576        The downside of splitting up is that there have to be some external
2577        functions and data tables that are used internally in different modules of
2578        the library but which are not part of the API. These have all had their
2579        names changed to start with "_pcre_" so that they are unlikely to clash
2580        with other external names.
2581    
2582     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
2583        a different (DFA) algorithm. Although it is slower than the original
2584        function, it does have some advantages for certain types of matching
2585        problem.
2586    
2587     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
2588        including restarting after a partial match.
2589    
2590     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
2591        defined when compiling for Windows was sent to me. I have put it into the
2592        code, though I have no means of testing or verifying it.
2593    
2594     7. Added the pcre_refcount() auxiliary function.
2595    
2596     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
2597        match before or at the first newline in the subject string. In pcretest,
2598        the /f option on a pattern can be used to set this.
2599    
2600     9. A repeated \w when used in UTF-8 mode with characters greater than 256
2601        would behave wrongly. This has been present in PCRE since release 4.0.
2602    
2603    10. A number of changes to the pcregrep command:
2604    
2605        (a) Refactored how -x works; insert ^(...)$ instead of setting
2606            PCRE_ANCHORED and checking the length, in preparation for adding
2607            something similar for -w.
2608    
2609        (b) Added the -w (match as a word) option.
2610    
2611        (c) Refactored the way lines are read and buffered so as to have more
2612            than one at a time available.
2613    
2614        (d) Implemented a pcregrep test script.
2615    
2616        (e) Added the -M (multiline match) option. This allows patterns to match
2617            over several lines of the subject. The buffering ensures that at least
2618            8K, or the rest of the document (whichever is the shorter) is available
2619            for matching (and similarly the previous 8K for lookbehind assertions).
2620    
2621        (f) Changed the --help output so that it now says
2622    
2623              -w, --word-regex(p)
2624    
2625            instead of two lines, one with "regex" and the other with "regexp"
2626            because that confused at least one person since the short forms are the
2627            same. (This required a bit of code, as the output is generated
2628            automatically from a table. It wasn't just a text change.)
2629    
2630        (g) -- can be used to terminate pcregrep options if the next thing isn't an
2631            option but starts with a hyphen. Could be a pattern or a path name
2632            starting with a hyphen, for instance.
2633    
2634        (h) "-" can be given as a file name to represent stdin.
2635    
2636        (i) When file names are being printed, "(standard input)" is used for
2637            the standard input, for compatibility with GNU grep. Previously
2638            "<stdin>" was used.
2639    
2640        (j) The option --label=xxx can be used to supply a name to be used for
2641            stdin when file names are being printed. There is no short form.
2642    
2643        (k) Re-factored the options decoding logic because we are going to add
2644            two more options that take data. Such options can now be given in four
2645            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
2646    
2647        (l) Added the -A, -B, and -C options for requesting that lines of context
2648            around matches be printed.
2649    
2650        (m) Added the -L option to print the names of files that do not contain
2651            any matching lines, that is, the complement of -l.
2652    
2653        (n) The return code is 2 if any file cannot be opened, but pcregrep does
2654            continue to scan other files.
2655    
2656        (o) The -s option was incorrectly implemented. For compatibility with other
2657            greps, it now suppresses the error message for a non-existent or non-
2658            accessible file (but not the return code). There is a new option called
2659            -q that suppresses the output of matching lines, which was what -s was
2660            previously doing.
2661    
2662        (p) Added --include and --exclude options to specify files for inclusion
2663            and exclusion when recursing.
2664    
2665    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
2666        Hopefully, it now does.
2667    
2668    12. Missing cast in pcre_study().
2669    
2670    13. Added an "uninstall" target to the makefile.
2671    
2672    14. Replaced "extern" in the function prototypes in Makefile.in with
2673        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
2674        world, but is set differently for Windows.
2675    
2676    15. Added a second compiling function called pcre_compile2(). The only
2677        difference is that it has an extra argument, which is a pointer to an
2678        integer error code. When there is a compile-time failure, this is set
2679        non-zero, in addition to the error test pointer being set to point to an
2680        error message. The new argument may be NULL if no error number is required
2681        (but then you may as well call pcre_compile(), which is now just a
2682        wrapper). This facility is provided because some applications need a
2683        numeric error indication, but it has also enabled me to tidy up the way
2684        compile-time errors are handled in the POSIX wrapper.
2685    
2686    16. Added VPATH=.libs to the makefile; this should help when building with one
2687        prefix path and installing with another. (Or so I'm told by someone who
2688        knows more about this stuff than I do.)
2689    
2690    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
2691        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
2692        match everything, including newlines. This is not POSIX-compatible, but
2693        somebody wanted the feature. From pcretest it can be activated by using
2694        both the P and the s flags.
2695    
2696    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
2697    
2698    19. libpcre.pc was being incorrectly installed as executable.
2699    
2700    20. A couple of places in pcretest check for end-of-line by looking for '\n';
2701        it now also looks for '\r' so that it will work unmodified on Windows.
2702    
2703    21. Added Google's contributed C++ wrapper to the distribution.
2704    
2705    22. Added some untidy missing memory free() calls in pcretest, to keep
2706        Electric Fence happy when testing.
2707    
2708    
2709    
2710    Version 5.0 13-Sep-04
2711    ---------------------
2712    
2713     1. Internal change: literal characters are no longer packed up into items
2714        containing multiple characters in a single byte-string. Each character
2715        is now matched using a separate opcode. However, there may be more than one
2716        byte in the character in UTF-8 mode.
2717    
2718     2. The pcre_callout_block structure has two new fields: pattern_position and
2719        next_item_length. These contain the offset in the pattern to the next match
2720        item, and its length, respectively.
2721    
2722     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
2723        insertion of callouts before each pattern item. Added the /C option to
2724        pcretest to make use of this.
2725    
2726     4. On the advice of a Windows user, the lines
2727    
2728          #if defined(_WIN32) || defined(WIN32)
2729          _setmode( _fileno( stdout ), 0x8000 );
2730          #endif  /* defined(_WIN32) || defined(WIN32) */
2731    
2732        have been added to the source of pcretest. This apparently does useful
2733        magic in relation to line terminators.
2734    
2735     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
2736        for the benefit of those environments where the "b" makes a difference.
2737    
2738     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
2739        to know about it. I have put a hack into configure.in that adds in code
2740        to set GCC=yes if CC=icc. This seems to end up at a point in the
2741        generated configure script that is early enough to affect the setting of
2742        compiler options, which is what is needed, but I have no means of testing
2743        whether it really works. (The user who reported this had patched the
2744        generated configure script, which of course I cannot do.)
2745    
2746        LATER: After change 22 below (new libtool files), the configure script
2747        seems to know about icc (and also ecc). Therefore, I have commented out
2748        this hack in configure.in.
2749    
2750     7. Added support for pkg-config (2 patches were sent in).
2751    
2752     8. Negated POSIX character classes that used a combination of internal tables
2753        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2754        [[:^ascii]]. Typically, they would match almost any characters. The other
2755        POSIX classes were not broken in this way.
2756    
2757     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2758        to find the match, as PCRE was deluded into thinking that the match had to
2759        start at the start point or following a newline. The same bug applied to
2760        patterns with negative forward assertions or any backward assertions
2761        preceding ".*" at the start, unless the pattern required a fixed first
2762        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2763    
2764    10. In UTF-8 mode, when moving forwards in the subject after a failed match
2765        starting at the last subject character, bytes beyond the end of the subject
2766        string were read.
2767    
2768    11. Renamed the variable "class" as "classbits" to make life easier for C++
2769        users. (Previously there was a macro definition, but it apparently wasn't
2770        enough.)
2771    
2772    12. Added the new field "tables" to the extra data so that tables can be passed
2773        in at exec time, or the internal tables can be re-selected. This allows
2774        a compiled regex to be saved and re-used at a later time by a different
2775        program that might have everything at different addresses.
2776    
2777    13. Modified the pcre-config script so that, when run on Solaris, it shows a
2778        -R library as well as a -L library.
2779    
2780    14. The debugging options of pcretest (-d on the command line or D on a
2781        pattern) showed incorrect output for anything following an extended class
2782        that contained multibyte characters and which was followed by a quantifier.
2783    
2784    15. Added optional support for general category Unicode character properties
2785        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2786        support. It adds about 90K to the size of the library. The meanings of the
2787        inbuilt class escapes such as \d and \s have NOT been changed.
2788    
2789    16. Updated pcredemo.c to include calls to free() to release the memory for the
2790        compiled pattern.
2791    
2792    17. The generated file chartables.c was being created in the source directory
2793        instead of in the building directory. This caused the build to fail if the
2794        source directory was different from the building directory, and was
2795        read-only.
2796    
2797    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2798        file. No doubt somebody will tell me if they don't make sense... Also added
2799        Dan Mooney's comments about building on OpenVMS.
2800    
2801    19. Added support for partial matching via the PCRE_PARTIAL option for
2802        pcre_exec() and the \P data escape in pcretest.
2803    
2804    20. Extended pcretest with 3 new pattern features:
2805    
2806        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
2807              write the compiled pattern to the file whose name is "rest-of-line".
2808              This is a straight binary dump of the data, with the saved pointer to
2809              the character tables forced to be NULL. The study data, if any, is
2810              written too. After writing, pcretest reads a new pattern.
2811    
2812        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2813              compiled pattern from the given file. There must not be any
2814              occurrences of "<" in the file name (pretty unlikely); if there are,
2815              pcretest will instead treat the initial "<" as a pattern delimiter.
2816              After reading in the pattern, pcretest goes on to read data lines as
2817              usual.
2818    
2819        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2820              and 16-bit fields in a compiled pattern, to simulate a pattern that
2821              was compiled on a host of opposite endianness.
2822    
2823    21. The pcre-exec() function can now cope with patterns that were compiled on
2824        hosts of opposite endianness, with this restriction:
2825    
2826          As for any compiled expression that is saved and used later, the tables
2827          pointer field cannot be preserved; the extra_data field in the arguments
2828          to pcre_exec() should be used to pass in a tables address if a value
2829          other than the default internal tables were used at compile time.
2830    
2831    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2832        now diagnosed as an error. Previously, most of the time, a negative number
2833        would have been treated as zero, but if in addition "ovector" was passed as
2834        NULL, a crash could occur.
2835    
2836    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2837        new versions from the libtool 1.5 distribution (the last one is a copy of
2838        a file called libtool.m4). This seems to have fixed the need to patch
2839        "configure" to support Darwin 1.3 (which I used to do). However, I still
2840        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2841        workstation).
2842    
2843    24. Changed the PCRE licence to be the more standard "BSD" licence.
2844    
2845    
2846    Version 4.5 01-Dec-03
2847    ---------------------
2848    
2849     1. There has been some re-arrangement of the code for the match() function so
2850        that it can be compiled in a version that does not call itself recursively.
2851        Instead, it keeps those local variables that need separate instances for
2852        each "recursion" in a frame on the heap, and gets/frees frames whenever it
2853        needs to "recurse". Keeping track of where control must go is done by means
2854        of setjmp/longjmp. The whole thing is implemented by a set of macros that
2855        hide most of the details from the main code, and operates only if
2856        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2857        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2858        operating.
2859    
2860        To make it easier for callers to provide specially tailored get/free
2861        functions for this usage, two new functions, pcre_stack_malloc, and
2862        pcre_stack_free, are used. They are always called in strict stacking order,
2863        and the size of block requested is always the same.
2864    
2865        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2866        PCRE has been compiled to use the stack or the heap for recursion. The
2867        -C option of pcretest uses this to show which version is compiled.
2868    
2869        A new data escape \S, is added to pcretest; it causes the amounts of store
2870        obtained and freed by both kinds of malloc/free at match time to be added
2871        to the output.
2872    
2873     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2874        what's available on my current Linux desktop machine.
2875    
2876     3. When matching a UTF-8 string, the test for a valid string at the start has
2877        been extended. If start_offset is not zero, PCRE now checks that it points
2878        to a byte that is the start of a UTF-8 character. If not, it returns
2879        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2880        this is necessary because there may be backward assertions in the pattern.
2881        When matching the same subject several times, it may save resources to use
2882        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2883    
2884     4. The code for checking the validity of UTF-8 strings has been tightened so
2885        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2886        containing "overlong sequences".
2887    
2888     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2889        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2890        should have been "&", but it just so happened that all the cases this let
2891        through by mistake were picked up later in the function.
2892    
2893     6. I had used a variable called "isblank" - this is a C99 function, causing
2894        some compilers to warn. To avoid this, I renamed it (as "blankclass").
2895    
2896     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2897        prompting; (b) run "./pcretest /dev/null" at the start of the test script
2898        so the version is shown; (c) stop "make test" echoing "./RunTest".
2899    
2900     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2901    
2902     9. The prototype for memmove() for systems that don't have it was using
2903        size_t, but the inclusion of the header that defines size_t was later. I've
2904        moved the #includes for the C headers earlier to avoid this.
2905    
2906    10. Added some adjustments to the code to make it easier to compiler on certain
2907        special systems:
2908    
2909          (a) Some "const" qualifiers were missing.
2910          (b) Added the macro EXPORT before all exported functions; by default this
2911              is defined to be empty.
2912          (c) Changed the dftables auxiliary program (that builds chartables.c) so
2913              that it reads its output file name as an argument instead of writing
2914              to the standard output and assuming this can be redirected.
2915    
2916    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2917        class containing characters with values greater than 255, PCRE compilation
2918        went into a loop.
2919    
2920    12. A recursive reference to a subpattern that was within another subpattern
2921        that had a minimum quantifier of zero caused PCRE to crash. For example,
2922        (x(y(?2))z)? provoked this bug with a subject that got as far as the
2923        recursion. If the recursively-called subpattern itself had a zero repeat,
2924        that was OK.
2925    
2926    13. In pcretest, the buffer for reading a data line was set at 30K, but the
2927        buffer into which it was copied (for escape processing) was still set at
2928        1024, so long lines caused crashes.
2929    
2930    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2931        "internal error: code overflow...". This applied to any character class
2932        that was followed by a possessive quantifier.
2933    
2934    15. Modified the Makefile to add libpcre.la as a prerequisite for
2935        libpcreposix.la because I was told this is needed for a parallel build to
2936        work.
2937    
2938    16. If a pattern that contained .* following optional items at the start was
2939        studied, the wrong optimizing data was generated, leading to matching
2940        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2941        matching string must start with a or b or c. The correct conclusion for
2942        this pattern is that a match can start with any character.
2943    
2944    
2945    Version 4.4 13-Aug-03
2946    ---------------------
2947    
2948     1. In UTF-8 mode, a character class containing characters with values between
2949        127 and 255 was not handled correctly if the compiled pattern was studied.
2950        In fixing this, I have also improved the studying algorithm for such
2951        classes (slightly).
2952    
2953     2. Three internal functions had redundant arguments passed to them. Removal
2954        might give a very teeny performance improvement.
2955    
2956     3. Documentation bug: the value of the capture_top field in a callout is *one
2957        more than* the number of the hightest numbered captured substring.
2958    
2959     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2960        in incorrectly linking with a previously installed version. They now link
2961        explicitly with libpcre.la.
2962    
2963     5. configure.in no longer needs to recognize Cygwin specially.
2964    
2965     6. A problem in pcre.in for Windows platforms is fixed.
2966    
2967     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2968        pcretest, it used to include the size of the study block as part of its
2969        output. Unfortunately, the structure contains a field that has a different
2970        size on different hardware architectures. This meant that the tests that
2971        showed this size failed. As the block is currently always of a fixed size,
2972        this information isn't actually particularly useful in pcretest output, so
2973        I have just removed it.
2974    
2975     8. Three pre-processor statements accidentally did not start in column 1.
2976        Sadly, there are *still* compilers around that complain, even though
2977        standard C has not required this for well over a decade. Sigh.
2978    
2979     9. In pcretest, the code for checking callouts passed small integers in the
2980        callout_data field, which is a void * field. However, some picky compilers
2981        complained about the casts involved for this on 64-bit systems. Now
2982        pcretest passes the address of the small integer instead, which should get
2983        rid of the warnings.
2984    
2985    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2986        both compile and run time, and gives an error if an invalid UTF-8 sequence
2987        is found. There is a option for disabling this check in cases where the
2988        string is known to be correct and/or the maximum performance is wanted.
2989    
2990    11. In response to a bug report, I changed one line in Makefile.in from
2991    
2992            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2993        to
2994            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2995    
2996        to look similar to other lines, but I have no way of telling whether this
2997        is the right thing to do, as I do not use Windows. No doubt I'll get told
2998        if it's wrong...
2999    
3000    
3001    Version 4.3 21-May-03
3002    ---------------------
3003    
3004    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
3005       Makefile.
3006    
3007    2. Some refactoring to improve the quality of the code:
3008    
3009       (i)   The utf8_table... variables are now declared "const".
3010    
3011       (ii)  The code for \cx, which used the "case flipping" table to upper case
3012             lower case letters, now just substracts 32. This is ASCII-specific,
3013             but the whole concept of \cx is ASCII-specific, so it seems
3014             reasonable.
3015    
3016       (iii) PCRE was using its character types table to recognize decimal and
3017             hexadecimal digits in the pattern. This is silly, because it handles
3018             only 0-9, a-f, and A-F, but the character types table is locale-
3019             specific, which means strange things might happen. A private
3020             table is now used for this - though it costs 256 bytes, a table is
3021             much faster than multiple explicit tests. Of course, the standard
3022             character types table is still used for matching digits in subject
3023             strings against \d.
3024    
3025       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
3026             ending in _t are). So I've renamed it as ESC_tee.
3027    
3028    3. The first argument for regexec() in the POSIX wrapper should have been
3029       defined as "const".
3030    
3031    4. Changed pcretest to use malloc() for its buffers so that they can be
3032       Electric Fenced for debugging.
3033    
3034    5. There were several places in the code where, in UTF-8 mode, PCRE would try
3035       to read one or more bytes before the start of the subject string. Often this
3036       had no effect on PCRE's behaviour, but in some circumstances it could
3037       provoke a segmentation fault.
3038    
3039    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
3040       to try to read one or more bytes before the start of the subject string.
3041    
3042    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
3043       UTF-8 support could misbehave in various ways if the subject string
3044       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
3045       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
3046       back over UTF-8 characters.)
3047    
3048    
3049    Version 4.2 14-Apr-03
3050    ---------------------
3051    
3052    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
3053    
3054    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
3055         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
3056         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
3057         [WIN_PREFIX]: new variable, "cyg" for Cygwin
3058         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
3059           and BUILD_EXEEXT
3060         Note: automatic setting of the BUILD variables is not yet working
3061         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
3062           compile-time but not at link-time
3063         [LINK]: use for linking executables only
3064         make different versions for Windows and non-Windows
3065         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
3066           libraries
3067         [LINK_FOR_BUILD]: new variable
3068         [OBJEXT]: use throughout
3069         [EXEEXT]: use throughout
3070         <winshared>: new target
3071         <wininstall>: new target
3072         <dftables.o>: use native compiler
3073         <dftables>: use native linker
3074         <install>: handle Windows platform correctly
3075         <clean>: ditto
3076         <check>: ditto
3077         copy DLL to top builddir before testing
3078    
3079       As part of these changes, -no-undefined was removed again. This was reported
3080       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
3081       in any case.
3082    
3083    3. Some tidies to get rid of compiler warnings:
3084    
3085       . In the match_data structure, match_limit was an unsigned long int, whereas
3086         match_call_count was an int. I've made them both unsigned long ints.
3087    
3088       . In pcretest the fact that a const uschar * doesn't automatically cast to
3089         a void * provoked a warning.
3090    
3091       . Turning on some more compiler warnings threw up some "shadow" variables
3092         and a few more missing casts.
3093    
3094    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
3095       option, a class that contained a single character with a value between 128
3096       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
3097    
3098    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
3099       option, a class that contained several characters, but with at least one
3100       whose value was between 128 and 255 caused PCRE to crash.
3101    
3102    
3103    Version 4.1 12-Mar-03
3104    ---------------------
3105    
3106    1. Compiling with gcc -pedantic found a couple of places where casts were
3107    needed, and a string in dftables.c that was longer than standard compilers are
3108    required to support.
3109    
3110    2. Compiling with Sun's compiler found a few more places where the code could
3111    be tidied up in order to avoid warnings.
3112    
3113    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
3114    first of these names is deprecated in the latest Autoconf in favour of the name
3115    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
3116    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
3117    analogy I have changed it to CFLAGS_FOR_BUILD.
3118    
3119    4. Added -no-undefined to the linking command in the Makefile, because this is
3120    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
3121    linking step for the pcreposix library.
3122    
3123    5. PCRE was failing to diagnose the case of two named groups with the same
3124    name.
3125    
3126    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
3127    literal character that is needed in the subject for a match, and scans along to
3128    ensure that it is present before embarking on the full matching process. This
3129    saves time in cases of nested unlimited repeats that are never going to match.
3130    Problem: the scan can take a lot of time if the subject is very long (e.g.
3131    megabytes), thus penalizing straightforward matches. It is now done only if the
3132    amount of subject to be scanned is less than 1000 bytes.
3133    
3134    7. A lesser problem with the same optimization is that it was recording the
3135    first character of an anchored pattern as "needed", thus provoking a search
3136    right along the subject, even when the first match of the pattern was going to
3137    fail. The "needed" character is now not set for anchored patterns, unless it
3138    follows something in the pattern that is of non-fixed length. Thus, it still
3139    fulfils its original purpose of finding quick non-matches in cases of nested
3140    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
3141    
3142    
3143    Version 4.0 17-Feb-03
3144    ---------------------
3145    
3146    1. If a comment in an extended regex that started immediately after a meta-item
3147    extended to the end of string, PCRE compiled incorrect data. This could lead to
3148    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
3149    
3150    2. Moved to autoconf 2.53 and libtool 1.4.2.
3151    
3152    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
3153    the special perltest8 script is no longer needed - all the tests can be run
3154    from a single perltest script.
3155    
3156    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
3157    by \s. It has now been removed in PCRE. This means it isn't recognized as
3158    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
3159    class [:space:] *does* include VT, thereby creating a mess.
3160    
3161    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
3162    space and tab.
3163    
3164    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
3165    its new features into the main test script, reducing the number of scripts.
3166    
3167    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
3168    were backward compatible, and made the (?i) apply to the whole pattern, as if
3169    /i were given. Now it behaves more logically, and applies the option setting
3170    only to what follows. PCRE has been changed to follow suit. However, if it
3171    finds options settings right at the start of the pattern, it extracts them into
3172    the global options, as before. Thus, they show up in the info data.
3173    
3174    8. Added support for the \Q...\E escape sequence. Characters in between are
3175    treated as literals. This is slightly different from Perl in that $ and @ are
3176    also handled as literals inside the quotes. In Perl, they will cause variable
3177    interpolation. Note the following examples:
3178    
3179        Pattern            PCRE matches      Perl matches
3180    
3181        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
3182        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
3183        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
3184    
3185    For compatibility with Perl, \Q...\E sequences are recognized inside character
3186    classes as well as outside them.
3187    
3188    9. Re-organized 3 code statements in pcretest to avoid "overflow in
3189    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
3190    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
3191    signed/unsigned warnings.
3192    
3193    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
3194    option for pcretest, so I've replaced it by a simple function that does just
3195    that job.
3196    
3197    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
3198    "pcregrep -".
3199    
3200    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
3201    Java package. This provides some syntactic sugar for simple cases of what my
3202    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
3203    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
3204    item, you can use this simplified notation. Note that only makes sense with
3205    greedy quantifiers. Consequently, the use of the possessive quantifier forces
3206    greediness, whatever the setting of the PCRE_UNGREEDY option.
3207    
3208    13. A change of greediness default within a pattern was not taking effect at
3209    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
3210    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
3211    was abstracted outside.
3212    
3213    14. PCRE now supports the \G assertion. It is true when the current matching
3214    position is at the start point of the match. This differs from \A when the
3215    starting offset is non-zero. Used with the /g option of pcretest (or similar
3216    code), it works in the same way as it does for Perl's /g option. If all
3217    alternatives of a regex begin with \G, the expression is anchored to the start
3218    match position, and the "anchored" flag is set in the compiled expression.
3219    
3220    15. Some bugs concerning the handling of certain option changes within patterns
3221    have been fixed. These applied to options other than (?ims). For example,
3222    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
3223    been the other way round. Some of this was related to change 7 above.
3224    
3225    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
3226    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
3227    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
3228    POSIX classes only within a class (e.g. /[[:alpha:]]/).
3229    
3230    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
3231    mode. Unlike ".", it always matches newline, whatever the setting of
3232    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
3233    assertions. Perl allows it, but it doesn't (in general) work because it can't
3234    calculate the length of the lookbehind. At least, that's the case for Perl
3235    5.8.0 - I've been told they are going to document that it doesn't work in
3236    future.
3237    
3238    18. Added an error diagnosis for escapes that PCRE does not support: these are
3239    \L, \l, \N, \P, \p, \U, \u, and \X.
3240    
3241    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
3242    reading past the end of the pattern in cases such as /[abcd/.
3243    
3244    20. PCRE was getting more memory than necessary for patterns with classes that
3245    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
3246    
3247    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
3248    compiling PCRE for use with Virtual Pascal.
3249    
3250    22. Small fix to the Makefile to make it work properly if the build is done
3251    outside the source tree.
3252    
3253    23. Added a new extension: a condition to go with recursion. If a conditional
3254    subpattern starts with (?(R) the "true" branch is used if recursion has
3255    happened, whereas the "false" branch is used only at the top level.
3256    
3257    24. When there was a very long string of literal characters (over 255 bytes
3258    without UTF support, over 250 bytes with UTF support), the computation of how
3259    much memory was required could be incorrect, leading to segfaults or other
3260    strange effects.
3261    
3262    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
3263    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
3264    there was a subsequent back reference to those brackets. This meant that, for
3265    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
3266    possible to check for precisely this case. All we can do is abandon the
3267    optimization if .* occurs inside capturing brackets when there are any back
3268    references whatsoever. (See below for a better fix that came later.)
3269    
3270    26. The handling of the optimization for finding the first character of a
3271    non-anchored pattern, and for finding a character that is required later in the
3272    match were failing in some cases. This didn't break the matching; it just
3273    failed to optimize when it could. The way this is done has been re-implemented.
3274    
3275    27. Fixed typo in error message for invalid (?R item (it said "(?p").
3276    
3277    28. Added a new feature that provides some of the functionality that Perl
3278    provides with (?{...}). The facility is termed a "callout". The way it is done
3279    in PCRE is for the caller to provide an optional function, by setting
3280    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
3281    global variable. By default it is unset, which disables all calling out. To get
3282    the function called, the regex must include (?C) at appropriate points. This
3283    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
3284    This provides a means of identifying different callout points. When PCRE
3285    reaches such a point in the regex, if pcre_callout has been set, the external
3286    function is called. It is provided with data in a structure called
3287    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
3288    matching continues; if it returns a non-zero value, the match at the current
3289    point fails. However, backtracking will occur if possible. [This was changed
3290    later and other features added - see item 49 below.]
3291    
3292    29. pcretest is upgraded to test the callout functionality. It provides a
3293    callout function that displays information. By default, it shows the start of
3294    the match and the current position in the text. There are some new data escapes
3295    to vary what happens:
3296    
3297        \C+         in addition, show current contents of captured substrings
3298        \C-         do not supply a callout function
3299        \C!n        return 1 when callout number n is reached
3300        \C!n!m      return 1 when callout number n is reached for the mth time
3301    
3302    30. If pcregrep was called with the -l option and just a single file name, it
3303    output "<stdin>" if a match was found, instead of the file name.
3304    
3305    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
3306    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
3307    pcre_exec(). This saves a malloc/free per call. The default value of
3308    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
3309    when configuring.
3310    
3311    32. The default maximum size of a compiled pattern is 64K. There have been a
3312    few cases of people hitting this limit. The code now uses macros to handle the
3313    storing of links as offsets within the compiled pattern. It defaults to 2-byte
3314    links, but this can be changed to 3 or 4 bytes by --with-link-size when
3315    configuring. Tests 2 and 5 work only with 2-byte links because they output
3316    debugging information about compiled patterns.
3317    
3318    33. Internal code re-arrangements:
3319    
3320    (a) Moved the debugging function for printing out a compiled regex into
3321        its own source file (printint.c) and used #include to pull it into
3322        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
3323        separate copies.
3324    
3325    (b) Defined the list of op-code names for debugging as a macro in
3326        internal.h so that it is next to the definition of the opcodes.
3327    
3328    (c) Defined a table of op-code lengths for simpler skipping along compiled
3329        code. This is again a macro in internal.h so that it is next to the
3330        definition of the opcodes.
3331    
3332    34. Added support for recursive calls to individual subpatterns, along the
3333    lines of Robin Houston's patch (but implemented somewhat differently).
3334    
3335    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
3336    allow it to read and process whole directories in Win32. This code was
3337    contributed by Lionel Fourquaux; it has not been tested by me.
3338    
3339    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
3340    used to name a group. Names consist of alphanumerics and underscores, and must
3341    be unique. Back references use the syntax (?P=name) and recursive calls use
3342    (?P>name) which is a PCRE extension to the Python extension. Groups still have
3343    numbers. The function pcre_fullinfo() can be used after compilation to extract
3344    a name/number map. There are three relevant calls:
3345    
3346      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
3347      PCRE_INFO_NAMECOUNT            yields the number of entries
3348      PCRE_INFO_NAMETABLE            yields a pointer to the map.
3349    
3350    The map is a vector of fixed-size entries. The size of each entry depends on
3351    the length of the longest name used. The first two bytes of each entry are the
3352    group number, most significant byte first. There follows the corresponding
3353    name, zero terminated. The names are in alphabetical order.
3354    
3355    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
3356    case instead of 255. Making it the same both with and without UTF-8 support
3357    means that the same test output works with both.
3358    
3359    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
3360    calling malloc() with a zero argument.
3361    
3362    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
3363    optimization. I've improved things by keeping a bitmap of backreferences with
3364    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
3365    fact referenced, the optimization can be applied. It is unlikely that a
3366    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
3367    the match to follow \n) will appear inside brackets with a number greater than
3368    31, but if it does, any back reference > 31 suppresses the optimization.
3369    
3370    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
3371    of disabling numbered capturing parentheses. Any opening parenthesis that is
3372    not followed by ? behaves as if it were followed by ?: but named parentheses
3373    can still be used for capturing (and they will acquire numbers in the usual
3374    way).
3375    
3376    41. Redesigned the return codes from the match() function into yes/no/error so
3377    that errors can be passed back from deep inside the nested calls. A malloc
3378    failure while inside a recursive subpattern call now causes the
3379    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
3380    
3381    42. It is now possible to set a limit on the number of times the match()
3382    function is called in a call to pcre_exec(). This facility makes it possible to
3383    limit the amount of recursion and backtracking, though not in a directly
3384    obvious way, because the match() function is used in a number of different
3385    circumstances. The count starts from zero for each position in the subject
3386    string (for non-anchored patterns). The default limit is, for compatibility, a
3387    large number, namely 10 000 000. You can change this in two ways:
3388    
3389    (a) When configuring PCRE before making, you can use --with-match-limit=n
3390        to set a default value for the compiled library.
3391    
3392    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
3393        a different value is set. See 45 below.
3394    
3395    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
3396    
3397    43. Added a new function pcre_config(int, void *) to enable run-time extraction
3398    of things that can be changed at compile time. The first argument specifies
3399    what is wanted and the second points to where the information is to be placed.
3400    The current list of available information is:
3401    
3402      PCRE_CONFIG_UTF8
3403    
3404    The output is an integer that is set to one if UTF-8 support is available;
3405    otherwise it is set to zero.
3406    
3407      PCRE_CONFIG_NEWLINE
3408    
3409    The output is an integer that it set to the value of the code that is used for
3410    newline. It is either LF (10) or CR (13).
3411    
3412      PCRE_CONFIG_LINK_SIZE
3413    
3414    The output is an integer that contains the number of bytes used for internal
3415    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
3416    
3417      PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
3418    
3419    The output is an integer that contains the threshold above which the POSIX
3420    interface uses malloc() for output vectors. See item 31 above.
3421    
3422      PCRE_CONFIG_MATCH_LIMIT
3423    
3424    The output is an unsigned integer that contains the default limit of the number
3425    of match() calls in a pcre_exec() execution. See 42 above.
3426    
3427    44. pcretest has been upgraded by the addition of the -C option. This causes it
3428    to extract all the available output from the new pcre_config() function, and to
3429    output it. The program then exits immediately.
3430    
3431    45. A need has arisen to pass over additional data with calls to pcre_exec() in
3432    order to support additional features. One way would have been to define
3433    pcre_exec2() (for example) with extra arguments, but this would not have been
3434    extensible, and would also have required all calls to the original function to
3435    be mapped to the new one. Instead, I have chosen to extend the mechanism that
3436    is used for passing in "extra" data from pcre_study().
3437    
3438    The pcre_extra structure is now exposed and defined in pcre.h. It currently
3439    contains the following fields:
3440    
3441      flags         a bitmap indicating which of the following fields are set
3442      study_data    opaque data from pcre_study()
3443      match_limit   a way of specifying a limit on match() calls for a specific
3444                      call to pcre_exec()
3445      callout_data  data for callouts (see 49 below)
3446    
3447    The flag bits are also defined in pcre.h, and are
3448    
3449      PCRE_EXTRA_STUDY_DATA
3450      PCRE_EXTRA_MATCH_LIMIT
3451      PCRE_EXTRA_CALLOUT_DATA
3452    
3453    The pcre_study() function now returns one of these new pcre_extra blocks, with
3454    the actual study data pointed to by the study_data field, and the
3455    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
3456    before. That is, this change is entirely upwards-compatible and requires no
3457    change to existing code.
3458    
3459    If you want to pass in additional data to pcre_exec(), you can either place it
3460    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
3461    block.
3462    
3463    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
3464    data string contains the escape sequence \M, pcretest calls pcre_exec() several
3465    times with different match limits, until it finds the minimum value needed for
3466    pcre_exec() to complete. The value is then output. This can be instructive; for
3467    most simple matches the number is quite small, but for pathological cases it
3468    gets very large very quickly.
3469    
3470    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
3471    returns the size of the data block pointed to by the study_data field in a
3472    pcre_extra block, that is, the value that was passed as the argument to
3473    pcre_malloc() when PCRE was getting memory in which to place the information
3474    created by pcre_study(). The fourth argument should point to a size_t variable.
3475    pcretest has been extended so that this information is shown after a successful
3476    pcre_study() call when information about the compiled regex is being displayed.
3477    
3478    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
3479    because what follows is always an absolute path. (Later: it turns out that this
3480    is more than cosmetic for MinGW, because it doesn't like empty path
3481    components.)
3482    
3483    49. Some changes have been made to the callout feature (see 28 above):
3484    
3485    (i)  A callout function now has three choices for what it returns:
3486    
3487           0  =>  success, carry on matching
3488         > 0  =>  failure at this point, but backtrack if possible
3489         < 0  =>  serious error, return this value from pcre_exec()
3490    
3491         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
3492         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
3493         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
3494         use by callout functions. It will never be used by PCRE itself.
3495    
3496    (ii) The pcre_extra structure (see 45 above) has a void * field called
3497         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
3498         pcre_callout_block structure has a field of the same name. The contents of
3499         the field passed in the pcre_extra structure are passed to the callout
3500         function in the corresponding field in the callout block. This makes it
3501         easier to use the same callout-containing regex from multiple threads. For
3502         testing, the pcretest program has a new data escape
3503    
3504           \C*n        pass the number n (may be negative) as callout_data
3505    
3506         If the callout function in pcretest receives a non-zero value as
3507         callout_data, it returns that value.
3508    
3509    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
3510    there were some redundant $(CFLAGS) in commands that are now specified as
3511    $(LINK), which already includes $(CFLAGS).
3512    
3513    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
3514    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
3515    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
3516    one-byte characters throughout. Note that case-insensitive matching applies
3517    only to characters whose values are less than 256. PCRE doesn't support the
3518    notion of cases for higher-valued characters.
3519    
3520    (i)   A character class whose characters are all within 0-255 is handled as
3521          a bit map, and the map is inverted for negative classes. Previously, a
3522          character > 255 always failed to match such a class; however it should
3523          match if the class was a negative one (e.g. [^ab]). This has been fixed.
3524    
3525    (ii)  A negated character class with a single character < 255 is coded as
3526          "not this character" (OP_NOT). This wasn't working properly when the test
3527          character was multibyte, either singly or repeated.
3528    
3529    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
3530          mode, for example: \x{100}{2,3}.
3531    
3532    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
3533          singly or repeated) now correctly test multibyte characters. However,
3534          PCRE doesn't recognize any characters with values greater than 255 as
3535          digits, spaces, or word characters. Such characters always match \D, \S,
3536          and \W, and never match \d, \s, or \w.
3537    
3538    (v)   Classes may now contain characters and character ranges with values
3539          greater than 255. For example: [ab\x{100}-\x{400}].
3540    
3541    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
3542          PCRE in UTF-8 mode.
3543    
3544    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
3545    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
3546    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
3547    value.)
3548    
3549    53. The single man page has become too large. I have therefore split it up into
3550    a number of separate man pages. These also give rise to individual HTML pages;
3551    these are now put in a separate directory, and there is an index.html page that
3552    lists them all. Some hyperlinking between the pages has been installed.
3553    
3554    54. Added convenience functions for handling named capturing parentheses.
3555    
3556    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
3557    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
3558    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
3559    are faulted.
3560    
3561    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
3562    calling configure. These values are used when compiling the dftables.c program
3563    which is run to generate the source of the default character tables. They
3564    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
3565    you will need to set these values.
3566    
3567    57. Updated the building process for Windows DLL, as provided by Fred Cox.
3568    
3569    
3570    Version 3.9 02-Jan-02
3571    ---------------------
3572    
3573    1. A bit of extraneous text had somehow crept into the pcregrep documentation.
3574    
3575    2. If --disable-static was given, the building process failed when trying to
3576    build pcretest and pcregrep. (For some reason it was using libtool to compile
3577    them, which is not right, as they aren't part of the library.)
3578    
3579    
3580    Version 3.8 18-Dec-01
3581    ---------------------
3582    
3583    1. The experimental UTF-8 code was completely screwed up. It was packing the
3584    bytes in the wrong order. How dumb can you get?
3585    
3586    
3587    Version 3.7 29-Oct-01
3588    ---------------------
3589    
3590    1. In updating pcretest to check change 1 of version 3.6, I screwed up.
3591    This caused pcretest, when used on the test data, to segfault. Unfortunately,
3592    this didn't happen under Solaris 8, where I normally test things.
3593    
3594    2. The Makefile had to be changed to make it work on BSD systems, where 'make'
3595    doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
3596    isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
3597    this fix an hour or so after the initial 3.7 release.)
3598    
3599    
3600    Version 3.6 23-Oct-01
3601    ---------------------
3602    
3603    1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
3604    offsets passed as NULL with zero offset count.
3605    
3606    2. The config.guess and config.sub files had not been updated when I moved to
3607    the latest autoconf.
3608    
3609    
3610    Version 3.5 15-Aug-01
3611    ---------------------
3612    
3613    1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
3614    had been forgotten.
3615    
3616    2. By using declared but undefined structures, we can avoid using "void"
3617    definitions in pcre.h while keeping the internal definitions of the structures
3618    private.
3619    
3620    3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
3621    user point of view, this means that both static and shared libraries are built
3622    by default, but this can be individually controlled. More of the work of
3623    handling this static/shared cases is now inside libtool instead of PCRE's make
3624    file.
3625    
3626    4. The pcretest utility is now installed along with pcregrep because it is
3627    useful for users (to test regexs) and by doing this, it automatically gets
3628    relinked by libtool. The documentation has been turned into a man page, so
3629    there are now .1, .txt, and .html versions in /doc.
3630    
3631    5. Upgrades to pcregrep:
3632       (i)   Added long-form option names like gnu grep.
3633       (ii)  Added --help to list all options with an explanatory phrase.
3634       (iii) Added -r, --recursive to recurse into sub-directories.
3635       (iv)  Added -f, --file to read patterns from a file.
3636    
3637    6. pcre_exec() was referring to its "code" argument before testing that
3638    argument for NULL (and giving an error if it was NULL).
3639    
3640    7. Upgraded Makefile.in to allow for compiling in a different directory from
3641    the source directory.
3642    
3643    8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
3644    options bits, the pointer it was passed was to an int instead of to an unsigned
3645    long int. This mattered only on 64-bit systems.
3646    
3647    9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
3648    generated) instead of pcre.in, which it its source. Also made the same change
3649    in several of the .c files.
3650    
3651    10. A new release of gcc defines printf() as a macro, which broke pcretest
3652    because it had an ifdef in the middle of a string argument for printf(). Fixed
3653    by using separate calls to printf().
3654    
3655    11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
3656    script, to force use of CR or LF instead of \n in the source. On non-Unix
3657    systems, the value can be set in config.h.
3658    
3659    12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
3660    absolute limit. Changed the text of the error message to make this clear, and
3661    likewise updated the man page.
3662    
3663    13. The limit of 99 on the number of capturing subpatterns has been removed.
3664    The new limit is 65535, which I hope will not be a "real" limit.
3665    
3666    
3667    Version 3.4 22-Aug-00
3668    ---------------------
3669    
3670    1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
3671    
3672    2. Diagnose condition (?(0) as an error instead of crashing on matching.
3673    
3674    
3675    Version 3.3 01-Aug-00
3676    ---------------------
3677    
3678    1. If an octal character was given, but the value was greater than \377, it
3679    was not getting masked to the least significant bits, as documented. This could
3680    lead to crashes in some systems.
3681    
3682    2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
3683    the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
3684    
3685    3. Added the functions pcre_free_substring() and pcre_free_substring_list().
3686    These just pass their arguments on to (pcre_free)(), but they are provided
3687    because some uses of PCRE bind it to non-C systems that can call its functions,
3688    but cannot call free() or pcre_free() directly.
3689    
3690    4. Add "make test" as a synonym for "make check". Corrected some comments in
3691    the Makefile.
3692    
3693    5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
3694    Makefile.
3695    
3696    6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
3697    command called pgrep for grepping around the active processes.
3698    
3699    7. Added the beginnings of support for UTF-8 character strings.
3700    
3701    8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
3702    RANLIB to ./ltconfig so that they are used by libtool. I think these are all
3703    the relevant ones. (AR is not passed because ./ltconfig does its own figuring
3704    out for the ar command.)
3705    
3706    
3707    Version 3.2 12-May-00
3708    ---------------------
3709    
3710    This is purely a bug fixing release.
3711    
3712    1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
3713    of ZA. This was just one example of several cases that could provoke this bug,
3714    which was introduced by change 9 of version 2.00. The code for breaking
3715    infinite loops after an iteration that matches an empty string was't working
3716    correctly.
3717    
3718    2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
3719    when matched against abbab (for example). After matching an empty string, it
3720    wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
3721    caused it to match further down the string than it should.
3722    
3723    3. The code contained an inclusion of sys/types.h. It isn't clear why this
3724    was there because it doesn't seem to be needed, and it causes trouble on some
3725    systems, as it is not a Standard C header. It has been removed.
3726    
3727    4. Made 4 silly changes to the source to avoid stupid compiler warnings that
3728    were reported on the Macintosh. The changes were from
3729    
3730      while ((c = *(++ptr)) != 0 && c != '\n');
3731    to
3732      while ((c = *(++ptr)) != 0 && c != '\n') ;
3733    
3734    Totally extraordinary, but if that's what it takes...
3735    
3736    5. PCRE is being used in one environment where neither memmove() nor bcopy() is
3737    available. Added HAVE_BCOPY and an autoconf test for it; if neither
3738    HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
3739    assumes the way PCRE uses memmove() (always moving upwards).
3740    
3741    6. PCRE is being used in one environment where strchr() is not available. There
3742    was only one use in pcre.c, and writing it out to avoid strchr() probably gives
3743    faster code anyway.
3744    
3745    
3746    Version 3.1 09-Feb-00
3747    ---------------------
3748    
3749    The only change in this release is the fixing of some bugs in Makefile.in for
3750    the "install" target:
3751    
3752    (1) It was failing to install pcreposix.h.
3753    
3754    (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
3755    
3756    
3757    Version 3.0 01-Feb-00
3758    ---------------------
3759    
3760    1. Add support for the /+ modifier to perltest (to output $` like it does in
3761    pcretest).
3762    
3763    2. Add support for the /g modifier to perltest.
3764    
3765    3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
3766    matches null strings.
3767    
3768    4. Fix perltest so that it doesn't do unwanted things when fed an empty
3769    pattern. Perl treats empty patterns specially - it reuses the most recent
3770    pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
3771    effect.
3772    
3773    5. The POSIX interface was broken in that it was just handing over the POSIX
3774    captured string vector to pcre_exec(), but (since release 2.00) PCRE has
3775    required a bigger vector, with some working space on the end. This means that
3776    the POSIX wrapper now has to get and free some memory, and copy the results.
3777    
3778    6. Added some simple autoconf support, placing the test data and the
3779    documentation in separate directories, re-organizing some of the
3780    information files, and making it build pcre-config (a GNU standard). Also added
3781    libtool support for building PCRE as a shared library, which is now the
3782    default.
3783    
3784    7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
3785    09 are not valid octal constants. Single digits will be used for minor values
3786    less than 10.
3787    
3788    8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
3789    existing programs that set these in the POSIX interface can use PCRE without
3790    modification.
3791    
3792    9. Added a new function, pcre_fullinfo() with an extensible interface. It can
3793    return all that pcre_info() returns, plus additional data. The pcre_info()
3794    function is retained for compatibility, but is considered to be obsolete.
3795    
3796    10. Added experimental recursion feature (?R) to handle one common case that
3797    Perl 5.6 will be able to do with (?p{...}).
3798    
3799    11. Added support for POSIX character classes like [:alpha:], which Perl is
3800    adopting.
3801    
3802    
3803    Version 2.08 31-Aug-99
3804    ----------------------
3805    
3806    1. When startoffset was not zero and the pattern began with ".*", PCRE was not
3807    trying to match at the startoffset position, but instead was moving forward to
3808    the next newline as if a previous match had failed.
3809    
3810    2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
3811    and could get into a loop if a null string was matched other than at the start
3812    of the subject.
3813    
3814    3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
3815    be distinguished at compile time, and for completeness also added PCRE_DATE.
3816    
3817    5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
3818    in GnuWin32 environments.
3819    
3820    
3821    Version 2.07 29-Jul-99
3822    ----------------------
3823    
3824    1. The documentation is now supplied in plain text form and HTML as well as in
3825    the form of man page sources.
3826