bzr and ntfsmount (fuse) issues?

[Aaron Bentley]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

John Arbash Meinel wrote:
> Aaron Bentley wrote:
>> I don't think we should swallow the exception.  I think we should behave
>> like we're running under win32.  I.e. when we're checking/setting the
>> execute bit, we should do it via the inventory.

> How would you detect it?

One option would be to call os.chmod on '.bzr' or similar, and see if it
raises an exception.

>  On win32 we do call 'os.chmod()'. It just
> doesn't *do* anything.

I thought win32 supported some bits, just not all of them.

> It seems to complicate things a lot to have to
> start doing mount-point detection.
> 
> It is Fuse which is setting EPERM when we try to call os.chmod(). Since
> it doesn't permit chmod() on an ntfs mount.

It's always been a bit bogus to assume that chmod is supported on all
unix filesystems.  We should be testing on a per-path basis, I think.

Aaron
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2.2 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFE13fI0F+nu1YWqI0RArneAJ9rhf8Qa/i+20zXTdv6xS8fHrLBbQCdH1Vk
td8a7Vneu461pnjb69/6yeo=
=ineN
-----END PGP SIGNATURE-----