Hi,
It's why we provide the bwild tool... If you are using something
special, you can validate your expression with this tool.
I have run the testfnm regress tool from the glibc with the old bacula
version, and more than one tests are failing...
[ 3] a[/]b matches a/b -> FAIL
[ 5] * matches a/b -> FAIL
[ 6] *[/]b matches a/b -> FAIL
[ 7] *[b] matches a/b -> FAIL
[33] */* matches a/.b -> FAIL
[36] *[[:alpha:]]/*[[:alnum:]] does not match a/b -> FAIL
[37] *[![:digit:]]*/[![:d-d] does not match a/b -> FAIL
[38] *[![:digit:]]*/[[:d-d] does not match a/[ -> FAIL
[48] **/? does not match /b -> FAIL
...
On Monday 03 September 2007 13:45:48 Kern Sibbald wrote:
> Hello,
>
> You would think that wild cards (fnmatch) are well known and that they work
> the same on all systems. Apparently not.
>
> On GNU clib systems,
>
> fnmatch("a*b/*", "abbb/.x", FNM_PATHNAME|FNM_PERIOD) returns fail
> (i.e. FN_NOMATCH).
>
> and on my version of the BSD fnmatch.c it returns success. I could have
> messed up the code in porting it into Bacula, but I consider that *highly*
> unlikely.
>
> In reading the GNU documentation on fnmatch, it is not clear which is
> correct -- in fact, depending on nuances of precedences of the rules, which
> is not documented, both interpretations seem to be correct.
>
> Does anyone have any opinions? Am I missing something?
>
> I must admit: this is somewhat a tricky case. :-)
>
> Regards,
>
> Kern
>
-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems? Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
