On Thu, Feb 19, 2009 at 10:36 PM, Antoine Martin <[email protected]> wrote: >> The other option is to just leave out the "choose a different base >> port" option. The port ends up being "a + b", where "b" is fully under >> the user's control. So giving them control over "a" really doesn't >> give them any extra power, but skipping it lets us avoid nasty and >> confusing UI possibilities. > > I'm not sure I am following you... Can you provide a sample command line > usage?
Err, all I'm really saying is, what if we just use the same fixed offset always? Don't provide a command line switch to change it, just leave it hard-coded? Because the only reason you would need to change the offset was if some port was in use, right? But you can also solve that problem by just choosing a different xpra display number, right? And in fact this would be a strictly simpler solution for users anyway? > Cool. > What happens if 2 separate clients connect to both sockets. Does one kick > the other out? (no biggie) Yes, just like what happens now if 2 separate clients connect on the one socket. (Well, the behavior is slightly more complicated -- we allow a few simple requests without kicking out the existing client, so that e.g. 'xpra list' can ping a server and find out if it's alive. And that bit of the protocol is a bit ad hoc and should probably be cleaned up. But basically, yes, each new 'xpra attach' wins, and anyway having multiple listen sockets doesn't introduce any new complexities here.) -- Nathaniel _______________________________________________ Parti-discuss mailing list [email protected] http://lists.partiwm.org/cgi-bin/mailman/listinfo/parti-discuss
