Re: GNOME Summary, June 7-14

[Tom Tromey <tromey cygnus com>]

>>>>> "Tony" == Tony Lill <Tony.Lill@ajlc.waterloo.on.ca> writes:

>> Speaking of which, gnome-session needs a lot of work, if someone is
>> so inclined. The SMlib.PS document gives you an idea how powerful a
>> session manager could really be, in conjunction with SM-compliant
>> applications.

Tony> Here's a question, how does one go about changeing the SM interface?
Tony> The problem that lead me into looking at the session manager is that
Tony> the session manager has no idea what host the client is running on.

Normally you don't extend the SM interface.  However, we occasionally
have to because it is so lame.

For this particular case, the xsm session manager already defines an
extension that can be used to do what you want.  I think we ought to
adopt the same convention.  This is already on the gnome-session to-do
list.

We could change libgnomeui to automatically set this property when
required.

Tony> Then there's the UserId property. The spec also says "On
Tony> POSIX-based systems this will contain the the user's name (the
Tony> pw_name field of struct passwd)." In reality, on a local linux
Tony> box it works, but on a remote one, you get the uid number,
Tony> making it useless for doing something like "rsh -l username host
Tony> cmd".

I don't understand why this would happen.
The code to set UserId looks correct to me.
Can you find out why this fails for you?

Tom