rename_internals and rename_internals_fsp consolidated

[Jeremy Allison]

On Fri, Jun 15, 2007 at 12:07:51AM +0200, Volker Lendecke wrote:
> Hi, Jeremy!
> 
> Attached find a patch with lots of "-" inside: It changes
> rename_internals to open the file and then calls
> rename_internals_fsp. This then does not open the file a
> second time via can_delete, we already checked the share
> modes in the first open.
> 
> It does not survive the wildcard part of RAW-RENAME yet and
> something in RAW-NOTIFY is broken, but I think that can be
> fixed.

Just one not of caution...

Currently it doesn't map NT_STATUS_SHARING_VIOLATION into
NT_STATUS_ACCESS_DENIED as the old code used to do. Do we
have a torture test for this case ?

I don't think we have to handle deferred opens (I don't
think the old code does) but do we have enough test
coverage to be sure ?

Jeremy