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

Diff of /code/trunk/ChangeLog

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

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