[jcifs] Problems with Wins Resolution

[Michael B. Allen]

On Fri, 27 Sep 2002 13:42:16 -0500
"Christopher R. Hertel" <crh at ubiqx.mn.org> wrote:

> On Thu, Sep 26, 2002 at 02:38:18AM -0400, Michael B. Allen wrote:
> :
> > > ...unless there was a default workgroup.  The default workgroup, by the
> > > way, is not part of the SMB URL.  Samba and Windows have a default
> > > workgroup already, so this would just bring jCIFS up-to-speed in that
> > > sense.
> > 
> > Am I hearing you correctly?  Wouldn't that really screw up the
> > semantics? If you get the servers who are members of your default
> > workgroup using just 'smb://' then how do you get the workgroup list?
> 
> If that's what you're hearing then I'm doing a really lousy job of 
> explaining.  :)
> 
> No, I'm really not suggesting mangling the semantics of the SMB URL.  I'm
> saying that if the client software (jCIFS, whatever) has a default
> workgroup name then that name could be used by the client code to find the
> DMB and grab the workgroup list from that, just as would be done using an
> LMB.  The default workgroup name is part of the application...nothing to 
> do with the SMB URL.
> 
> So, given the string "SMB://", the client might first look for the
> MSBROWSE name.  If that is not found, then the client could look for 
> <default_wg>#1D (likely futile since the MSBROWSE name wasn't found) and 
> the <default_wg>#1B name (last ditch).  If the <default_wg>#1B name is 
> found (via the NBNS), then the client would query for the browse list, and 
> present just the list of workgroups.
> 
> Am I making any more sense?  :)

Yes. Added it to docs/todo.txt.

-- 
A  program should be written to model the concepts of the task it
performs rather than the physical world or a process because this
maximizes  the  potential  for it to be applied to tasks that are
conceptually  similar and more importantly to tasks that have not
yet been conceived.