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

Diff of /code/trunk/ChangeLog

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

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