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

Diff of /code/trunk/ChangeLog

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

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