Il giorno ven, 19/01/2007 alle 15.55 +0100, Mario Torre ha scritto:

> I have done some tests and seems ok. What do you think (I need some help
> to see if on other platforms it broke things)?

Too few, it seems....

+  
+  return ((statbuf.st_mode & perms) == perms) ? 0 : 1;
+}

Indeed, this test do the job only in the common cases.
Fix this is not hard, but I was wondering if we can use the "access"
instead of "stat" and go with it.

This could also be used in the other tests, and I think is way faster
that opening a file for read/write and check if it fails...

The only drawback of access I can think of is that it does the checks
basing on the Real IDs, but I don't think a java program will be ever
have the sticky bit, so I guess it is safe... Access is not reliable on
NFS filesystems too, though, so this maybe a stopper.

What do you think?

Mario
-- 
Lima Software, SO.PR.IND. s.r.l.
http://www.limasoftware.net/
pgp key: http://subkeys.pgp.net/

Please, support open standards:
http://opendocumentfellowship.org/petition/
http://www.nosoftwarepatents.com/

Attachment: signature.asc
Description: Questa รจ una parte del messaggio firmata digitalmente

Reply via email to