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

Diff of /code/trunk/ChangeLog

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

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

Legend:
Removed from v.57  
changed lines
  Added in v.608

  ViewVC Help
Powered by ViewVC 1.1.5