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

Diff of /code/trunk/ChangeLog

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

revision 195 by ph10, Mon Jul 30 13:23:28 2007 UTC revision 594 by ph10, Sun May 1 08:22:12 2011 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4  Version 7.3 05-Jul-07  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    
16    Version 8.12 15-Jan-2011
17    ------------------------
18    
19    1.  Fixed some typos in the markup of the man pages, and wrote a script that
20        checks for such things as part of the documentation building process.
21    
22    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
23        --match-limit and --recursion-limit options (added for 8.11). In
24        particular, this made one of the standard tests fail. (The integer value
25        went into the wrong half of a long int.)
26    
27    3.  If the --colour option was given to pcregrep with -v (invert match), it
28        did strange things, either producing crazy output, or crashing. It should,
29        of course, ignore a request for colour when reporting lines that do not
30        match.
31    
32    4.  Another pcregrep bug caused similar problems if --colour was specified with
33        -M (multiline) and the pattern match finished with a line ending.
34    
35    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
36        matched in multiline mode, the following line was shown as part of the
37        match. This seems wrong, so I have changed it.
38    
39    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
40        the check for further matches in the same line (so they could be coloured)
41        to overrun the end of the current line. If another match was found, it was
42        incorrectly shown (and then shown again when found in the next line).
43    
44    7.  If pcregrep was compiled under Windows, there was a reference to the
45        function pcregrep_exit() before it was defined. I am assuming this was
46        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
47        reported by a user. I've moved the definition above the reference.
48    
49    
50    Version 8.11 10-Dec-2010
51    ------------------------
52    
53    1.  (*THEN) was not working properly if there were untried alternatives prior
54        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
55        backtracked to try for "b" instead of moving to the next alternative branch
56        at the same level (in this case, to look for "c"). The Perl documentation
57        is clear that when (*THEN) is backtracked onto, it goes to the "next
58        alternative in the innermost enclosing group".
59    
60    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
61        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
62        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
63        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
64        (*THEN).
65    
66    3.  If \s appeared in a character class, it removed the VT character from
67        the class, even if it had been included by some previous item, for example
68        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
69        of \s, but is part of the POSIX "space" class.)
70    
71    4.  A partial match never returns an empty string (because you can always
72        match an empty string at the end of the subject); however the checking for
73        an empty string was starting at the "start of match" point. This has been
74        changed to the "earliest inspected character" point, because the returned
75        data for a partial match starts at this character. This means that, for
76        example, /(?<=abc)def/ gives a partial match for the subject "abc"
77        (previously it gave "no match").
78    
79    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
80        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
81        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
82        has an implication that the given string is incomplete (because a partial
83        match is preferred over a full match). For this reason, these items now
84        give a partial match in this situation. [Aside: previously, the one case
85        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
86        match rather than a full match, which was wrong by the old rules, but is
87        now correct.]
88    
89    6.  There was a bug in the handling of #-introduced comments, recognized when
90        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
91        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
92        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
93        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
94        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
95        places in pcre_compile().
96    
97    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
98        comments when looking ahead for named forward references to subpatterns,
99        the only newline sequence it recognized was NL. It now handles newlines
100        according to the set newline convention.
101    
102    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
103        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
104        cater for a lack of strerror(). These oversights have been fixed.
105    
106    9.  Added --match-limit and --recursion-limit to pcregrep.
107    
108    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
109    
110    11. When the -o option was used, pcregrep was setting a return code of 1, even
111        when matches were found, and --line-buffered was not being honoured.
112    
113    12. Added an optional parentheses number to the -o and --only-matching options
114        of pcregrep.
115    
116    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
117        can match an empty string. The code to do it in pcretest and pcredemo
118        needed fixing:
119    
120        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
121            only one byte after an empty string match just before CRLF (this case
122            just got forgotten; "any" and "anycrlf" were OK).
123    
124        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
125            mode when an empty string match preceded an ASCII character followed by
126            a non-ASCII character. (The code for advancing by one character rather
127            than one byte was nonsense.)
128    
129        (c) The pcredemo.c sample program did not have any code at all to handle
130            the cases when CRLF is a valid newline sequence.
131    
132    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
133        as a starting offset was within the subject string. There is now a new
134        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
135        negative or greater than the length of the string. In order to test this,
136        pcretest is extended to allow the setting of negative starting offsets.
137    
138    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
139        starting offset points to the beginning of a UTF-8 character was
140        unnecessarily clumsy. I tidied it up.
141    
142    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
143        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
144    
145    17. Nobody had reported that the --include_dir option, which was added in
146        release 7.7 should have been called --include-dir (hyphen, not underscore)
147        for compatibility with GNU grep. I have changed it to --include-dir, but
148        left --include_dir as an undocumented synonym, and the same for
149        --exclude-dir, though that is not available in GNU grep, at least as of
150        release 2.5.4.
151    
152    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
153        characters from a string of bytes) have been redefined so as not to use
154        loops, in order to improve performance in some environments. At the same
155        time, I abstracted some of the common code into auxiliary macros to save
156        repetition (this should not affect the compiled code).
157    
158    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
159        compile-time error is now given if \c is not followed by an ASCII
160        character, that is, a byte less than 128. (In EBCDIC mode, the code is
161        different, and any byte value is allowed.)
162    
163    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
164        START_OPTIMIZE option, which is now allowed at compile time - but just
165        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
166        to pcregrep and other applications that have no direct access to PCRE
167        options. The new /Y option in pcretest sets this option when calling
168        pcre_compile().
169    
170    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
171        back references. Groups containing recursive back references were forced to
172        be atomic by that change, but in the case of named groups, the amount of
173        memory required was incorrectly computed, leading to "Failed: internal
174        error: code overflow". This has been fixed.
175    
176    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
177        pcretest.c, to avoid build problems in some Borland environments.
178    
179    
180    Version 8.10 25-Jun-2010
181    ------------------------
182    
183    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
184        THEN.
185    
186    2.  (*ACCEPT) was not working when inside an atomic group.
187    
188    3.  Inside a character class, \B is treated as a literal by default, but
189        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
190        causes the error). The code is unchanged, but I tidied the documentation.
191    
192    4.  Inside a character class, PCRE always treated \R and \X as literals,
193        whereas Perl faults them if its -w option is set. I have changed PCRE so
194        that it faults them when PCRE_EXTRA is set.
195    
196    5.  Added support for \N, which always matches any character other than
197        newline. (It is the same as "." when PCRE_DOTALL is not set.)
198    
199    6.  When compiling pcregrep with newer versions of gcc which may have
200        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
201        declared with attribute warn_unused_result" were given. Just casting the
202        result to (void) does not stop the warnings; a more elaborate fudge is
203        needed. I've used a macro to implement this.
204    
205    7.  Minor change to pcretest.c to avoid a compiler warning.
206    
207    8.  Added four artifical Unicode properties to help with an option to make
208        \s etc use properties (see next item). The new properties are: Xan
209        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
210    
211    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
212        use Unicode properties. (*UCP) at the start of a pattern can be used to set
213        this option. Modified pcretest to add /W to test this facility. Added
214        REG_UCP to make it available via the POSIX interface.
215    
216    10. Added --line-buffered to pcregrep.
217    
218    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
219        studied, and the match started with a letter with a code point greater than
220        127 whose first byte was different to the first byte of the other case of
221        the letter, the other case of this starting letter was not recognized
222        (#976).
223    
224    12. If a pattern that was studied started with a repeated Unicode property
225        test, for example, \p{Nd}+, there was the theoretical possibility of
226        setting up an incorrect bitmap of starting bytes, but fortunately it could
227        not have actually happened in practice until change 8 above was made (it
228        added property types that matched character-matching opcodes).
229    
230    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
231        possible starting bytes for non-anchored patterns.
232    
233    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
234        \R, and also a number of cases that involve Unicode properties, both
235        explicit and implicit when PCRE_UCP is set.
236    
237    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
238        input, it could crash or give wrong results if characters with values
239        greater than 0xc0 were present in the subject string. (Detail: it assumed
240        UTF-8 input when processing these items.)
241    
242    16. Added a lot of (int) casts to avoid compiler warnings in systems where
243        size_t is 64-bit (#991).
244    
245    17. Added a check for running out of memory when PCRE is compiled with
246        --disable-stack-for-recursion (#990).
247    
248    18. If the last data line in a file for pcretest does not have a newline on
249        the end, a newline was missing in the output.
250    
251    19. The default pcre_chartables.c file recognizes only ASCII characters (values
252        less than 128) in its various bitmaps. However, there is a facility for
253        generating tables according to the current locale when PCRE is compiled. It
254        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
255        characters, are recognized by isspace() and therefore were getting set in
256        these tables, and indeed these tables seem to approximate to ISO 8859. This
257        caused a problem in UTF-8 mode when pcre_study() was used to create a list
258        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
259        which of course cannot start UTF-8 characters. I have changed the code so
260        that only real ASCII characters (less than 128) and the correct starting
261        bytes for UTF-8 encodings are set for characters greater than 127 when in
262        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
263        altogether.)
264    
265    20. Added the /T option to pcretest so as to be able to run tests with non-
266        standard character tables, thus making it possible to include the tests
267        used for 19 above in the standard set of tests.
268    
269    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
270        reference to a subpattern the other side of a comment that contains an
271        opening parenthesis caused either an internal compiling error, or a
272        reference to the wrong subpattern.
273    
274    
275    Version 8.02 19-Mar-2010
276    ------------------------
277    
278    1.  The Unicode data tables have been updated to Unicode 5.2.0.
279    
280    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
281        configured.
282    
283    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
284        original author of that file, following a query about its status.
285    
286    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
287        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
288    
289    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
290        quantifier applied to a forward-referencing subroutine call, could compile
291        incorrect code or give the error "internal error: previously-checked
292        referenced subpattern not found".
293    
294    6.  Both MS Visual Studio and Symbian OS have problems with initializing
295        variables to point to external functions. For these systems, therefore,
296        pcre_malloc etc. are now initialized to local functions that call the
297        relevant global functions.
298    
299    7.  There were two entries missing in the vectors called coptable and poptable
300        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
301        I've fixed the data, and added a kludgy way of testing at compile time that
302        the lengths are correct (equal to the number of opcodes).
303    
304    8.  Following on from 7, I added a similar kludge to check the length of the
305        eint vector in pcreposix.c.
306    
307    9.  Error texts for pcre_compile() are held as one long string to avoid too
308        much relocation at load time. To find a text, the string is searched,
309        counting zeros. There was no check for running off the end of the string,
310        which could happen if a new error number was added without updating the
311        string.
312    
313    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
314    
315    11. \K was not working if it appeared in an atomic group or in a group that
316        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
317        \K is "not well defined" if used in an assertion. PCRE now accepts it if
318        the assertion is positive, but not if it is negative.
319    
320    12. Change 11 fortuitously reduced the size of the stack frame used in the
321        "match()" function of pcre_exec.c by one pointer. Forthcoming
322        implementation of support for (*MARK) will need an extra pointer on the
323        stack; I have reserved it now, so that the stack frame size does not
324        decrease.
325    
326    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
327        item in branch that calls a recursion is a subroutine call - as in the
328        second branch in the above example - was incorrectly given the compile-
329        time error "recursive call could loop indefinitely" because pcre_compile()
330        was not correctly checking the subroutine for matching a non-empty string.
331    
332    14. The checks for overrunning compiling workspace could trigger after an
333        overrun had occurred. This is a "should never occur" error, but it can be
334        triggered by pathological patterns such as hundreds of nested parentheses.
335        The checks now trigger 100 bytes before the end of the workspace.
336    
337    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
338    
339    
340    Version 8.01 19-Jan-2010
341    ------------------------
342    
343    1.  If a pattern contained a conditional subpattern with only one branch (in
344        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
345        computed the wrong minimum data length (which is of course zero for such
346        subpatterns). This could cause incorrect "no match" results.
347    
348    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
349        the pattern is reset in the first branch, pcre_compile() failed with
350        "internal error: code overflow at offset...". This happened only when
351        the reset was to the original external option setting. (An optimization
352        abstracts leading options settings into an external setting, which was the
353        cause of this.)
354    
355    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
356        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
357        assertion pattern did not match (meaning that the assertion was true), it
358        was incorrectly treated as false if the SKIP had been reached during the
359        matching. This also applied to assertions used as conditions.
360    
361    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
362        assertion subpattern, including such a pattern used as a condition,
363        unpredictable results occurred, instead of the error return
364        PCRE_ERROR_DFA_UITEM.
365    
366    5.  The C++ GlobalReplace function was not working like Perl for the special
367        situation when an empty string is matched. It now does the fancy magic
368        stuff that is necessary.
369    
370    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
371        removed. (These were left over from very, very early versions of PCRE.)
372    
373    7.  Some cosmetic changes to the code to make life easier when compiling it
374        as part of something else:
375    
376        (a) Change DEBUG to PCRE_DEBUG.
377    
378        (b) In pcre_compile(), rename the member of the "branch_chain" structure
379            called "current" as "current_branch", to prevent a collision with the
380            Linux macro when compiled as a kernel module.
381    
382        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
383            prevent a collision with the Linux macro when compiled as a kernel
384            module.
385    
386    8.  In pcre_compile() there are some checks for integer overflows that used to
387        cast potentially large values to (double). This has been changed to that
388        when building, a check for int64_t is made, and if it is found, it is used
389        instead, thus avoiding the use of floating point arithmetic. (There is no
390        other use of FP in PCRE.) If int64_t is not found, the fallback is to
391        double.
392    
393    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
394        2005 (difference between two addresses compared to an unsigned value).
395    
396    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
397        custom one, because of the following reported problem in Windows:
398    
399          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
400              under Win32.
401          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
402              therefore missing the function definition.
403          - The compiler thus generates a "C" signature for the test function.
404          - The linker fails to find the "C" function.
405          - PCRE fails to configure if asked to do so against libbz2.
406    
407    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
408        messages were output:
409    
410          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
411          rerunning libtoolize, to keep the correct libtool macros in-tree.
412          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
413    
414        I have done both of these things.
415    
416    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
417        most of the time, it *can* run out if it is given a pattern that contains a
418        runaway infinite recursion. I updated the discussion in the pcrestack man
419        page.
420    
421    13. Now that we have gone to the x.xx style of version numbers, the minor
422        version may start with zero. Using 08 or 09 is a bad idea because users
423        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
424        interpreted as invalid octal numbers. I've updated the previous comment in
425        configure.ac, and also added a check that gives an error if 08 or 09 are
426        used.
427    
428    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
429        causing partial matching to fail when the end of the subject matched \W
430        in a UTF-8 pattern where \W was quantified with a minimum of 3.
431    
432    15. There were some discrepancies between the declarations in pcre_internal.h
433        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
434        their definitions. The declarations used "const uschar *" and the
435        definitions used USPTR. Even though USPTR is normally defined as "const
436        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
437        reported that: "This difference in casting confuses some C++ compilers, for
438        example, SunCC recognizes above declarations as different functions and
439        generates broken code for hbpcre." I have changed the declarations to use
440        USPTR.
441    
442    16. GNU libtool is named differently on some systems. The autogen.sh script now
443        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
444        (FreeBSD).
445    
446    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
447        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
448        comment: "Figure out how to create a longlong from a string: strtoll and
449        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
450        instance, but it only takes 2 args instead of 3!"
451    
452    18. A subtle bug concerned with back references has been fixed by a change of
453        specification, with a corresponding code fix. A pattern such as
454        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
455        refers, was giving matches when it shouldn't. For example, xa=xaaa would
456        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
457        same bug. Such groups have to be quantified to be useful, or contained
458        inside another quantified group. (If there's no repetition, the reference
459        can never match.) The problem arises because, having left the group and
460        moved on to the rest of the pattern, a later failure that backtracks into
461        the group uses the captured value from the final iteration of the group
462        rather than the correct earlier one. I have fixed this in PCRE by forcing
463        any group that contains a reference to itself to be an atomic group; that
464        is, there cannot be any backtracking into it once it has completed. This is
465        similar to recursive and subroutine calls.
466    
467    
468    Version 8.00 19-Oct-09
469    ----------------------
470    
471    1.  The table for translating pcre_compile() error codes into POSIX error codes
472        was out-of-date, and there was no check on the pcre_compile() error code
473        being within the table. This could lead to an OK return being given in
474        error.
475    
476    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
477        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
478        in a Windows environment.
479    
480    3.  The pcregrep --count option prints the count for each file even when it is
481        zero, as does GNU grep. However, pcregrep was also printing all files when
482        --files-with-matches was added. Now, when both options are given, it prints
483        counts only for those files that have at least one match. (GNU grep just
484        prints the file name in this circumstance, but including the count seems
485        more useful - otherwise, why use --count?) Also ensured that the
486        combination -clh just lists non-zero counts, with no names.
487    
488    4.  The long form of the pcregrep -F option was incorrectly implemented as
489        --fixed_strings instead of --fixed-strings. This is an incompatible change,
490        but it seems right to fix it, and I didn't think it was worth preserving
491        the old behaviour.
492    
493    5.  The command line items --regex=pattern and --regexp=pattern were not
494        recognized by pcregrep, which required --regex pattern or --regexp pattern
495        (with a space rather than an '='). The man page documented the '=' forms,
496        which are compatible with GNU grep; these now work.
497    
498    6.  No libpcreposix.pc file was created for pkg-config; there was just
499        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
500    
501    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
502        when UCP support is not needed, by modifying the Python script that
503        generates it from Unicode data files. This should not matter if the module
504        is correctly used as a library, but I received one complaint about 50K of
505        unwanted data. My guess is that the person linked everything into his
506        program rather than using a library. Anyway, it does no harm.
507    
508    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
509        was a minimum greater than 1 for a wide character in a possessive
510        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
511        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
512        character. Chaos in the form of incorrect output or a compiling loop could
513        result.
514    
515    9.  The restrictions on what a pattern can contain when partial matching is
516        requested for pcre_exec() have been removed. All patterns can now be
517        partially matched by this function. In addition, if there are at least two
518        slots in the offset vector, the offset of the earliest inspected character
519        for the match and the offset of the end of the subject are set in them when
520        PCRE_ERROR_PARTIAL is returned.
521    
522    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
523        synonymous with PCRE_PARTIAL, for backwards compatibility, and
524        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
525        and may be more useful for multi-segment matching.
526    
527    11. Partial matching with pcre_exec() is now more intuitive. A partial match
528        used to be given if ever the end of the subject was reached; now it is
529        given only if matching could not proceed because another character was
530        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
531        string "Z", which now yields "no match" instead of "partial match". In the
532        case of pcre_dfa_exec(), "no match" is given if every matching path for the
533        final character ended with (*FAIL).
534    
535    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
536        if the pattern had a "must contain" character that was already found in the
537        earlier partial match, unless partial matching was again requested. For
538        example, with the pattern /dog.(body)?/, the "must contain" character is
539        "g". If the first part-match was for the string "dog", restarting with
540        "sbody" failed. This bug has been fixed.
541    
542    13. The string returned by pcre_dfa_exec() after a partial match has been
543        changed so that it starts at the first inspected character rather than the
544        first character of the match. This makes a difference only if the pattern
545        starts with a lookbehind assertion or \b or \B (\K is not supported by
546        pcre_dfa_exec()). It's an incompatible change, but it makes the two
547        matching functions compatible, and I think it's the right thing to do.
548    
549    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
550        so that the demonstration program is easily available in environments where
551        PCRE has not been installed from source.
552    
553    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
554        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
555        library.
556    
557    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
558        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
559        is not the first non-POSIX option to be added. Clearly some people find
560        these options useful.
561    
562    17. If a caller to the POSIX matching function regexec() passes a non-zero
563        value for nmatch with a NULL value for pmatch, the value of
564        nmatch is forced to zero.
565    
566    18. RunGrepTest did not have a test for the availability of the -u option of
567        the diff command, as RunTest does. It now checks in the same way as
568        RunTest, and also checks for the -b option.
569    
570    19. If an odd number of negated classes containing just a single character
571        interposed, within parentheses, between a forward reference to a named
572        subpattern and the definition of the subpattern, compilation crashed with
573        an internal error, complaining that it could not find the referenced
574        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
575        [The bug was that it was starting one character too far in when skipping
576        over the character class, thus treating the ] as data rather than
577        terminating the class. This meant it could skip too much.]
578    
579    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
580        /g option in pcretest when the pattern contains \K, which makes it possible
581        to have an empty string match not at the start, even when the pattern is
582        anchored. Updated pcretest and pcredemo to use this option.
583    
584    21. If the maximum number of capturing subpatterns in a recursion was greater
585        than the maximum at the outer level, the higher number was returned, but
586        with unset values at the outer level. The correct (outer level) value is
587        now given.
588    
589    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
590        PCRE did not set those parentheses (unlike Perl). I have now found a way to
591        make it do so. The string so far is captured, making this feature
592        compatible with Perl.
593    
594    23. The tests have been re-organized, adding tests 11 and 12, to make it
595        possible to check the Perl 5.10 features against Perl 5.10.
596    
597    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
598        pattern matches a fixed length string. PCRE did not allow this; now it
599        does. Neither allows recursion.
600    
601    25. I finally figured out how to implement a request to provide the minimum
602        length of subject string that was needed in order to match a given pattern.
603        (It was back references and recursion that I had previously got hung up
604        on.) This code has now been added to pcre_study(); it finds a lower bound
605        to the length of subject needed. It is not necessarily the greatest lower
606        bound, but using it to avoid searching strings that are too short does give
607        some useful speed-ups. The value is available to calling programs via
608        pcre_fullinfo().
609    
610    26. While implementing 25, I discovered to my embarrassment that pcretest had
611        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
612        study optimizations had never been tested with that matching function.
613        Oops. What is worse, even when it was passed study data, there was a bug in
614        pcre_dfa_exec() that meant it never actually used it. Double oops. There
615        were also very few tests of studied patterns with pcre_dfa_exec().
616    
617    27. If (?| is used to create subpatterns with duplicate numbers, they are now
618        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
619        on the other side of the coin, they are no longer allowed to have different
620        names, because these cannot be distinguished in PCRE, and this has caused
621        confusion. (This is a difference from Perl.)
622    
623    28. When duplicate subpattern names are present (necessarily with different
624        numbers, as required by 27 above), and a test is made by name in a
625        conditional pattern, either for a subpattern having been matched, or for
626        recursion in such a pattern, all the associated numbered subpatterns are
627        tested, and the overall condition is true if the condition is true for any
628        one of them. This is the way Perl works, and is also more like the way
629        testing by number works.
630    
631    
632    Version 7.9 11-Apr-09
633    ---------------------
634    
635    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
636        (pcretest), all targets were linked against these libraries. This included
637        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
638        libraries. This caused unwanted dependencies to be created. This problem
639        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
640        pcretest is linked with readline.
641    
642    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
643        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
644        moved outside it again, because FALSE and TRUE are already defined in AIX,
645        but BOOL is not.
646    
647    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
648        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
649    
650    4.  The pcregrep documentation said spaces were inserted as well as colons (or
651        hyphens) following file names and line numbers when outputting matching
652        lines. This is not true; no spaces are inserted. I have also clarified the
653        wording for the --colour (or --color) option.
654    
655    5.  In pcregrep, when --colour was used with -o, the list of matching strings
656        was not coloured; this is different to GNU grep, so I have changed it to be
657        the same.
658    
659    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
660        each matching line was coloured. Now it goes on to look for further matches
661        of any of the test patterns, which is the same behaviour as GNU grep.
662    
663    7.  A pattern that could match an empty string could cause pcregrep to loop; it
664        doesn't make sense to accept an empty string match in pcregrep, so I have
665        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
666        seems to be how GNU grep behaves.
667    
668    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
669        start or after a newline", because the conditional assertion was not being
670        correctly handled. The rule now is that both the assertion and what follows
671        in the first alternative must satisfy the test.
672    
673    9.  If auto-callout was enabled in a pattern with a conditional group whose
674        condition was an assertion, PCRE could crash during matching, both with
675        pcre_exec() and pcre_dfa_exec().
676    
677    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
678        used for matching.
679    
680    11. Unicode property support in character classes was not working for
681        characters (bytes) greater than 127 when not in UTF-8 mode.
682    
683    12. Added the -M command line option to pcretest.
684    
685    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
686    
687    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
688    
689    16. Added comments and documentation about mis-use of no_arg in the C++
690        wrapper.
691    
692    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
693        from Martin Jerabek that uses macro names for all relevant character and
694        string constants.
695    
696    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
697        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
698        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
699        these, but not everybody uses configure.
700    
701    19. A conditional group that had only one branch was not being correctly
702        recognized as an item that could match an empty string. This meant that an
703        enclosing group might also not be so recognized, causing infinite looping
704        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
705        with the subject "ab", where knowledge that the repeated group can match
706        nothing is needed in order to break the loop.
707    
708    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
709        exec(), but without supplying a callout function, matching went wrong.
710    
711    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
712        leak if the size of the offset vector was greater than 30. When the vector
713        is smaller, the saved offsets during recursion go onto a local stack
714        vector, but for larger vectors malloc() is used. It was failing to free
715        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
716        error, in fact).
717    
718    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
719        heapframe that is used only when UTF-8 support is enabled. This caused no
720        problem, but was untidy.
721    
722    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
723        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
724        included within another project.
725    
726    24. Steven Van Ingelgem's patches to add more options to the CMake support,
727        slightly modified by me:
728    
729          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
730              not building pcregrep.
731    
732          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
733              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
734    
735    25. Forward references, both numeric and by name, in patterns that made use of
736        duplicate group numbers, could behave incorrectly or give incorrect errors,
737        because when scanning forward to find the reference group, PCRE was not
738        taking into account the duplicate group numbers. A pattern such as
739        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
740    
741    26. Changed a few more instances of "const unsigned char *" to USPTR, making
742        the feature of a custom pointer more persuasive (as requested by a user).
743    
744    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
745        pcretest.c, inside #ifndefs, because it seems they are sometimes already
746        pre-defined.
747    
748    28. Added support for (*UTF8) at the start of a pattern.
749    
750    29. Arrange for flags added by the "release type" setting in CMake to be shown
751        in the configuration summary.
752    
753    
754    Version 7.8 05-Sep-08
755    ---------------------
756    
757    1.  Replaced UCP searching code with optimized version as implemented for Ad
758        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
759        stage table and inline lookup instead of a function, giving speed ups of 2
760        to 5 times on some simple patterns that I tested. Permission was given to
761        distribute the MultiStage2.py script that generates the tables (it's not in
762        the tarball, but is in the Subversion repository).
763    
764    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
765        scripts.
766    
767    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
768        a group with a zero qualifier. The result of the study could be incorrect,
769        or the function might crash, depending on the pattern.
770    
771    4.  Caseless matching was not working for non-ASCII characters in back
772        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
773        It now works when Unicode Property Support is available.
774    
775    5.  In pcretest, an escape such as \x{de} in the data was always generating
776        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
777        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
778        truncation.
779    
780    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
781    
782    7.  Added two (int) casts to pcregrep when printing the difference of two
783        pointers, in case they are 64-bit values.
784    
785    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
786        test 2 if it fails.
787    
788    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
789        and a #define of that name to empty if it is not externally set. This is to
790        allow users of MSVC to set it if necessary.
791    
792    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
793        the convenience functions in the pcre_get.c source file.
794    
795    11. An option change at the start of a pattern that had top-level alternatives
796        could cause overwriting and/or a crash. This command provoked a crash in
797        some environments:
798    
799          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
800    
801        This potential security problem was recorded as CVE-2008-2371.
802    
803    12. For a pattern where the match had to start at the beginning or immediately
804        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
805        pcre_dfa_exec() could read past the end of the passed subject if there was
806        no match. To help with detecting such bugs (e.g. with valgrind), I modified
807        pcretest so that it places the subject at the end of its malloc-ed buffer.
808    
809    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
810        exec() might read past the end of the data buffer in UTF-8 mode.
811    
812    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
813        the data contained the byte 0x85 as part of a UTF-8 character within its
814        first line. This applied both to normal and DFA matching.
815    
816    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
817        /^[^d]*?$/8 failed to match "abc".
818    
819    16. Added a missing copyright notice to pcrecpp_internal.h.
820    
821    17. Make it more clear in the documentation that values returned from
822        pcre_exec() in ovector are byte offsets, not character counts.
823    
824    18. Tidied a few places to stop certain compilers from issuing warnings.
825    
826    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
827        supplied by Stefan Weber. I made a further small update for 7.8 because
828        there is a change of source arrangements: the pcre_searchfuncs.c module is
829        replaced by pcre_ucd.c.
830    
831    
832    Version 7.7 07-May-08
833    ---------------------
834    
835    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
836        a string to a long long, pretend we don't even have a long long." This is
837        done by checking for the strtoq, strtoll, and _strtoi64 functions.
838    
839    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
840        pre-7.6 versions, which defined a global no_arg variable instead of putting
841        it in the RE class. (See also #8 below.)
842    
843    3.  Remove a line of dead code, identified by coverity and reported by Nuno
844        Lopes.
845    
846    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
847    
848        (1) The include/exclude patterns were being applied to the whole pathnames
849            of files, instead of just to the final components.
850    
851        (2) If there was more than one level of directory, the subdirectories were
852            skipped unless they satisfied the include/exclude conditions. This is
853            inconsistent with GNU grep (and could even be seen as contrary to the
854            pcregrep specification - which I improved to make it absolutely clear).
855            The action now is always to scan all levels of directory, and just
856            apply the include/exclude patterns to regular files.
857    
858    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
859        --exclude_dir in the tests to avoid scanning .svn directories.
860    
861    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
862        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
863        doesn't support NULs in patterns.
864    
865    7.  Added some missing "const"s to declarations of static tables in
866        pcre_compile.c and pcre_dfa_exec.c.
867    
868    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
869        caused by fix #2  above. (Subsequently also a second patch to fix the
870        first patch. And a third patch - this was a messy problem.)
871    
872    9.  Applied Craig's patch to remove the use of push_back().
873    
874    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
875        matching function regexec().
876    
877    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
878        which, however, unlike Perl's \g{...}, are subroutine calls, not back
879        references. PCRE supports relative numbers with this syntax (I don't think
880        Oniguruma does).
881    
882    12. Previously, a group with a zero repeat such as (...){0} was completely
883        omitted from the compiled regex. However, this means that if the group
884        was called as a subroutine from elsewhere in the pattern, things went wrong
885        (an internal error was given). Such groups are now left in the compiled
886        pattern, with a new opcode that causes them to be skipped at execution
887        time.
888    
889    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
890        to the way PCRE behaves:
891    
892        (a) A lone ] character is dis-allowed (Perl treats it as data).
893    
894        (b) A back reference to an unmatched subpattern matches an empty string
895            (Perl fails the current match path).
896    
897        (c) A data ] in a character class must be notated as \] because if the
898            first data character in a class is ], it defines an empty class. (In
899            Perl it is not possible to have an empty class.) The empty class []
900            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
901            The negative empty class [^] matches any one character, independently
902            of the DOTALL setting.
903    
904    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
905        non-existent subpattern following a character class starting with ']' and
906        containing () gave an internal compiling error instead of "reference to
907        non-existent subpattern". Fortunately, when the pattern did exist, the
908        compiled code was correct. (When scanning forwards to check for the
909        existencd of the subpattern, it was treating the data ']' as terminating
910        the class, so got the count wrong. When actually compiling, the reference
911        was subsequently set up correctly.)
912    
913    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
914        it was being rejected as not supported by pcre_dfa_exec(), even though
915        other assertions are supported. I have made pcre_dfa_exec() support
916        (*FAIL).
917    
918    16. The implementation of 13c above involved the invention of a new opcode,
919        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
920        cannot be changed at match time, I realized I could make a small
921        improvement to matching performance by compiling OP_ALLANY instead of
922        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
923        on the OP_ANY path.
924    
925    17. Compiling pcretest on Windows with readline support failed without the
926        following two fixes: (1) Make the unistd.h include conditional on
927        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
928    
929    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
930        ncurses library to be included for pcretest when ReadLine support is
931        requested, but also to allow for it to be overridden. This patch came from
932        Daniel Bergström.
933    
934    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
935        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
936        any errors with the current Unicode tables. Thanks to Peter Kankowski for
937        spotting this.
938    
939    
940    Version 7.6 28-Jan-08
941    ---------------------
942    
943    1.  A character class containing a very large number of characters with
944        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
945        overflow.
946    
947    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
948        HAVE_LONG_LONG is not defined.
949    
950    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
951        bring them up to date and include new features. This patch includes:
952    
953        - Fixed PH's badly added libz and libbz2 support.
954        - Fixed a problem with static linking.
955        - Added pcredemo. [But later removed - see 7 below.]
956        - Fixed dftables problem and added an option.
957        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
958            HAVE_LONG_LONG.
959        - Added readline support for pcretest.
960        - Added an listing of the option settings after cmake has run.
961    
962    4.  A user submitted a patch to Makefile that makes it easy to create
963        "pcre.dll" under mingw when using Configure/Make. I added stuff to
964        Makefile.am that cause it to include this special target, without
965        affecting anything else. Note that the same mingw target plus all
966        the other distribution libraries and programs are now supported
967        when configuring with CMake (see 6 below) instead of with
968        Configure/Make.
969    
970    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
971        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
972        exported in the Windows port". It has not yet been confirmed that the patch
973        solves the problem, but it does no harm.
974    
975    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
976        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
977        with CMake, and also correct the comment about stack recursion.
978    
979    7.  Remove the automatic building of pcredemo from the ./configure system and
980        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
981        of a program that users should build themselves after PCRE is installed, so
982        building it automatically is not really right. What is more, it gave
983        trouble in some build environments.
984    
985    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
986    
987    
988    Version 7.5 10-Jan-08
989    ---------------------
990    
991    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
992        values in parens when parsing an RE using the C++ wrapper."
993    
994    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
995        Characters greater than 255 were excluded from the class instead of being
996        included.
997    
998    3.  The same bug as (2) above applied to negated POSIX classes such as
999        [:^space:].
1000    
1001    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
1002        defined or documented. It seems to have been a typo for PCRE_STATIC, so
1003        I have changed it.
1004    
1005    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
1006        first named subpattern) and a construct such as (?&a) would reference the
1007        first named subpattern whose name started with "a" (in other words, the
1008        length check was missing). Both these problems are fixed. "Subpattern name
1009        expected" is now given for (?&) (a zero-length name), and this patch also
1010        makes it give the same error for \k'' (previously it complained that that
1011        was a reference to a non-existent subpattern).
1012    
1013    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
1014        this is right because (?- can be followed by option settings as well as by
1015        digits. I have, however, made the messages clearer.
1016    
1017    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
1018        than the number used in the conditional) now cause a compile-time error.
1019        This is actually not compatible with Perl, which accepts such patterns, but
1020        treats the conditional as always being FALSE (as PCRE used to), but it
1021        seems to me that giving a diagnostic is better.
1022    
1023    8.  Change "alphameric" to the more common word "alphanumeric" in comments
1024        and messages.
1025    
1026    9.  Fix two occurrences of "backslash" in comments that should have been
1027        "backspace".
1028    
1029    10. Remove two redundant lines of code that can never be obeyed (their function
1030        was moved elsewhere).
1031    
1032    11. The program that makes PCRE's Unicode character property table had a bug
1033        which caused it to generate incorrect table entries for sequences of
1034        characters that have the same character type, but are in different scripts.
1035        It amalgamated them into a single range, with the script of the first of
1036        them. In other words, some characters were in the wrong script. There were
1037        thirteen such cases, affecting characters in the following ranges:
1038    
1039          U+002b0 - U+002c1
1040          U+0060c - U+0060d
1041          U+0061e - U+00612
1042          U+0064b - U+0065e
1043          U+0074d - U+0076d
1044          U+01800 - U+01805
1045          U+01d00 - U+01d77
1046          U+01d9b - U+01dbf
1047          U+0200b - U+0200f
1048          U+030fc - U+030fe
1049          U+03260 - U+0327f
1050          U+0fb46 - U+0fbb1
1051          U+10450 - U+1049d
1052    
1053    12. The -o option (show only the matching part of a line) for pcregrep was not
1054        compatible with GNU grep in that, if there was more than one match in a
1055        line, it showed only the first of them. It now behaves in the same way as
1056        GNU grep.
1057    
1058    13. If the -o and -v options were combined for pcregrep, it printed a blank
1059        line for every non-matching line. GNU grep prints nothing, and pcregrep now
1060        does the same. The return code can be used to tell if there were any
1061        non-matching lines.
1062    
1063    14. Added --file-offsets and --line-offsets to pcregrep.
1064    
1065    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
1066        infinitely looping recursion. The bug was that positive lookaheads were not
1067        being skipped when checking for a possible empty match (negative lookaheads
1068        and both kinds of lookbehind were skipped).
1069    
1070    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
1071        inclusion of <windows.h> to before rather than after the definition of
1072        INVALID_FILE_ATTRIBUTES (patch from David Byron).
1073    
1074    17. Specifying a possessive quantifier with a specific limit for a Unicode
1075        character property caused pcre_compile() to compile bad code, which led at
1076        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
1077        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
1078        caused the error; without that there was no problem.
1079    
1080    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
1081    
1082    19. Added --enable-pcretest-libreadline.
1083    
1084    20. In pcrecpp.cc, the variable 'count' was incremented twice in
1085        RE::GlobalReplace(). As a result, the number of replacements returned was
1086        double what it should be. I removed one of the increments, but Craig sent a
1087        later patch that removed the other one (the right fix) and added unit tests
1088        that check the return values (which was not done before).
1089    
1090    21. Several CMake things:
1091    
1092        (1) Arranged that, when cmake is used on Unix, the libraries end up with
1093            the names libpcre and libpcreposix, not just pcre and pcreposix.
1094    
1095        (2) The above change means that pcretest and pcregrep are now correctly
1096            linked with the newly-built libraries, not previously installed ones.
1097    
1098        (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
1099    
1100    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
1101        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
1102        UTF-8 newline character). The key issue is that the pattern starts .*;
1103        this means that the match must be either at the beginning, or after a
1104        newline. The bug was in the code for advancing after a failed match and
1105        checking that the new position followed a newline. It was not taking
1106        account of UTF-8 characters correctly.
1107    
1108    23. PCRE was behaving differently from Perl in the way it recognized POSIX
1109        character classes. PCRE was not treating the sequence [:...:] as a
1110        character class unless the ... were all letters. Perl, however, seems to
1111        allow any characters between [: and :], though of course it rejects as
1112        unknown any "names" that contain non-letters, because all the known class
1113        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
1114        for example, whereas PCRE did not - it did not recognize a POSIX character
1115        class. This seemed a bit dangerous, so the code has been changed to be
1116        closer to Perl. The behaviour is not identical to Perl, because PCRE will
1117        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
1118        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
1119        Perl does, and where it didn't before.
1120    
1121    24. Rewrite so as to remove the single use of %n from pcregrep because in some
1122        Windows environments %n is disabled by default.
1123    
1124    
1125    Version 7.4 21-Sep-07
1126    ---------------------
1127    
1128    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
1129        means that a class such as [\s] counted as "explicit reference to CR or
1130        LF". That isn't really right - the whole point of the change was to try to
1131        help when there was an actual mention of one of the two characters. So now
1132        the change happens only if \r or \n (or a literal CR or LF) character is
1133        encountered.
1134    
1135    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
1136        of both had grown to the point where there were only 3 bits left.
1137        Fortunately, there was spare space in the data structure, and so I have
1138        moved the internal flags into a new 16-bit field to free up more option
1139        bits.
1140    
1141    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
1142        but did not set the internal JCHANGED flag - either of these is enough to
1143        control the way the "get" function works - but the PCRE_INFO_JCHANGED
1144        facility is supposed to tell if (?J) was ever used, so now (?J) at the
1145        start sets both bits.
1146    
1147    4.  Added options (at build time, compile time, exec time) to change \R from
1148        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
1149    
1150    5.  doc/pcresyntax.html was missing from the distribution.
1151    
1152    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
1153        compatibility, even though it is no longer used.
1154    
1155    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
1156        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
1157        windows.h file is present (where different names are used). [This was
1158        reversed later after testing - see 16 below.]
1159    
1160    8.  Changed all #include <config.h> to #include "config.h". There were also
1161        some further <pcre.h> cases that I changed to "pcre.h".
1162    
1163    9.  When pcregrep was used with the --colour option, it missed the line ending
1164        sequence off the lines that it output.
1165    
1166    10. It was pointed out to me that arrays of string pointers cause lots of
1167        relocations when a shared library is dynamically loaded. A technique of
1168        using a single long string with a table of offsets can drastically reduce
1169        these. I have refactored PCRE in four places to do this. The result is
1170        dramatic:
1171    
1172          Originally:                          290
1173          After changing UCP table:            187
1174          After changing error message table:   43
1175          After changing table of "verbs"       36
1176          After changing table of Posix names   22
1177    
1178        Thanks to the folks working on Gregex for glib for this insight.
1179    
1180    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
1181        unicode-properties was also set.
1182    
1183    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
1184    
1185    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
1186        checked only for CRLF.
1187    
1188    14. Added casts to pcretest.c to avoid compiler warnings.
1189    
1190    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
1191    
1192    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
1193        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
1194        entirely. This removes changes made in 7 above.
1195    
1196    17. The CMake files have been updated, and there is now more information about
1197        building with CMake in the NON-UNIX-USE document.
1198    
1199    
1200    Version 7.3 28-Aug-07
1201  ---------------------  ---------------------
1202    
1203   1. In the rejigging of the build system that eventually resulted in 7.1, the   1. In the rejigging of the build system that eventually resulted in 7.1, the
# Line 9  Version 7.3 05-Jul-07 Line 1205  Version 7.3 05-Jul-07
1205      brackets there is not right, since it causes compilers to look for an      brackets there is not right, since it causes compilers to look for an
1206      installed pcre.h, not the version that is in the source that is being      installed pcre.h, not the version that is in the source that is being
1207      compiled (which of course may be different). I have changed it back to:      compiled (which of course may be different). I have changed it back to:
1208    
1209        #include "pcre.h"        #include "pcre.h"
1210    
1211      I have a vague recollection that the change was concerned with compiling in      I have a vague recollection that the change was concerned with compiling in
1212      different directories, but in the new build system, that is taken care of      different directories, but in the new build system, that is taken care of
1213      by the VPATH setting the Makefile.      by the VPATH setting the Makefile.
1214    
1215   2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed   2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
1216      when the subject happened to end in the byte 0x85 (e.g. if the last      when the subject happened to end in the byte 0x85 (e.g. if the last
1217      character was \x{1ec5}). *Character* 0x85 is one of the "any" newline      character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
# Line 23  Version 7.3 05-Jul-07 Line 1219  Version 7.3 05-Jul-07
1219      of another character. The bug was that, for an unlimited repeat of . in      of another character. The bug was that, for an unlimited repeat of . in
1220      not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by      not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
1221      characters when looking for a newline.      characters when looking for a newline.
   
  3. A small performance improvement in the DOTALL UTF-8 mode .* case.  
   
  4. Debugging: adjusted the names of opcodes for different kinds of parentheses  
     in debug output.  
   
  5. Arrange to use "%I64d" instead of "%lld" for long printing in the pcrecpp  
     unittest when running under MinGW.  
   
  6. ESC_K was left out of the EBCDIC table.  
1222    
1223     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
1224    
1225     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
1226        in debug output.
1227    
1228     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
1229        long printing in the pcrecpp unittest when running under MinGW.
1230    
1231     6. ESC_K was left out of the EBCDIC table.
1232    
1233     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
1234        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
1235        limit also applies to "virtual nesting" when a pattern is recursive, and in
1236        this case 1000 isn't so big. I have been able to remove this limit at the
1237        expense of backing off one optimization in certain circumstances. Normally,
1238        when pcre_exec() would call its internal match() function recursively and
1239        immediately return the result unconditionally, it uses a "tail recursion"
1240        feature to save stack. However, when a subpattern that can match an empty
1241        string has an unlimited repetition quantifier, it no longer makes this
1242        optimization. That gives it a stack frame in which to save the data for
1243        checking that an empty string has been matched. Previously this was taken
1244        from the 1000-entry workspace that had been reserved. So now there is no
1245        explicit limit, but more stack is used.
1246    
1247     8. Applied Daniel's patches to solve problems with the import/export magic
1248        syntax that is required for Windows, and which was going wrong for the
1249        pcreposix and pcrecpp parts of the library. These were overlooked when this
1250        problem was solved for the main library.
1251    
1252     9. There were some crude static tests to avoid integer overflow when computing
1253        the size of patterns that contain repeated groups with explicit upper
1254        limits. As the maximum quantifier is 65535, the maximum group length was
1255        set at 30,000 so that the product of these two numbers did not overflow a
1256        32-bit integer. However, it turns out that people want to use groups that
1257        are longer than 30,000 bytes (though not repeat them that many times).
1258        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1259        made it possible to implement the integer overflow checks in a much more
1260        dynamic way, which I have now done. The artificial limitation on group
1261        length has been removed - we now have only the limit on the total length of
1262        the compiled pattern, which depends on the LINK_SIZE setting.
1263    
1264    10. Fixed a bug in the documentation for get/copy named substring when
1265        duplicate names are permitted. If none of the named substrings are set, the
1266        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1267        empty string.
1268    
1269    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1270        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1271        because the ] is interpreted as the first data character and the
1272        terminating ] is not found. PCRE has been made compatible with Perl in this
1273        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1274        cause memory overwriting.
1275    
1276    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1277        string has been matched (to stop an infinite loop). It was not recognizing
1278        a conditional subpattern that could match an empty string if that
1279        subpattern was within another subpattern. For example, it looped when
1280        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
1281        condition was not nested. This bug has been fixed.
1282    
1283    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1284        past the start of the subject in the presence of bytes with the top bit
1285        set, for example "\x8aBCD".
1286    
1287    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1288        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1289    
1290    14. Optimized (?!) to (*FAIL).
1291    
1292    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1293        This restricts code points to be within the range 0 to 0x10FFFF, excluding
1294        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1295        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1296        does: it's just the validity check that is more restrictive.
1297    
1298    16. Inserted checks for integer overflows during escape sequence (backslash)
1299        processing, and also fixed erroneous offset values for syntax errors during
1300        backslash processing.
1301    
1302    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1303        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1304    
1305    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1306        caused an overrun.
1307    
1308    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1309        something other than just ASCII characters) inside a group that had an
1310        unlimited repeat caused a loop at compile time (while checking to see
1311        whether the group could match an empty string).
1312    
1313    20. Debugging a pattern containing \p or \P could cause a crash. For example,
1314        [\P{Any}] did so. (Error in the code for printing property names.)
1315    
1316    21. An orphan \E inside a character class could cause a crash.
1317    
1318    22. A repeated capturing bracket such as (A)? could cause a wild memory
1319        reference during compilation.
1320    
1321    23. There are several functions in pcre_compile() that scan along a compiled
1322        expression for various reasons (e.g. to see if it's fixed length for look
1323        behind). There were bugs in these functions when a repeated \p or \P was
1324        present in the pattern. These operators have additional parameters compared
1325        with \d, etc, and these were not being taken into account when moving along
1326        the compiled data. Specifically:
1327    
1328        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1329            length.
1330    
1331        (b) An item such as \pL+ within a repeated group could cause crashes or
1332            loops.
1333    
1334        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1335            "reference to non-existent subpattern" error.
1336    
1337        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1338    
1339    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1340        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1341    
1342    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1343    
1344    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1345        character were causing crashes (broken optimization).
1346    
1347    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1348        \p or \P) caused a compile-time loop.
1349    
1350    28. More problems have arisen in unanchored patterns when CRLF is a valid line
1351        break. For example, the unstudied pattern [\r\n]A does not match the string
1352        "\r\nA" because change 7.0/46 below moves the current point on by two
1353        characters after failing to match at the start. However, the pattern \nA
1354        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
1355        the same is true. There doesn't seem any very clean way out of this, but
1356        what I have chosen to do makes the common cases work: PCRE now takes note
1357        of whether there can be an explicit match for \r or \n anywhere in the
1358        pattern, and if so, 7.0/46 no longer applies. As part of this change,
1359        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1360        pattern has explicit CR or LF references.
1361    
1362    29. Added (*CR) etc for changing newline setting at start of pattern.
1363    
1364    
1365  Version 7.2 19-Jun-07  Version 7.2 19-Jun-07

Legend:
Removed from v.195  
changed lines
  Added in v.594

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12