On Mon, Dec 09, 2002 at 04:46:14PM +0100, Ivan Popov wrote:
> On Mon, 9 Dec 2002, Jan Harkes wrote:
> 
> > > exported by Coda. No offence meant. Coda offers very special properties
> > > that are hardly synchronizable with other "access paths", i.e if you'd
> > > insist on accessing your files via the local filesystem.
> >
> > Not only that, but the server wouldn't necessarily notice when a user
> > modifies a file on the 'exported' filesystem. As a result it would not
> > break callbacks to inform client that a file was updated. And because
> 
> It *could* be possible to rewrite the server (and probably have to
> implement a local file system suitable for putting the files on it), so
> that the server *would* be able to notice the changes. That's why I wrote
> "hardly" instead of "absolutely not".
> 
> While the meaning is very much the same :-)
> 
> Such a project would be far away from being a reasonable one, wouldn't it?

Not really possible. The current semantics is that a modification on the
server is an atomic change from the old version to the new. When someone
is 'editing' a file from some magical exported tree on the local
machine, when should we break the callback?

- When the file is opened for writing? Clients will refetching the file
  before any of the changes are saved, and get an bad copy.

- When the file is closed? Client that 'accidentally' fetch the file
  between the open and the close will get a bad copy.

Jan