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