GNOME Bugzilla – Bug 402475
use gdm socket command to initiate login screen with user name
Last modified: 2008-06-10 22:55:43 UTC
...since it is a bit ridiculous to pick a user from the menu, only to be forced to type the user name in the login screen. The bug tracking the necessary gdm addition is bug 335786
Created attachment 81521 [details] [review] use new gdm socket commands The patch is untested, but it compiles, at least. It is missing some form of runtime detection to ensure that the running gdm supports the new socket commands. That shouldn't be hard to add, using the socket command to get the gdm version number.
Isn't this a duplicate of bug 172912?
Looks like it, yes. It would be nice to get some endorsement for the gdm addition required for this from the fusa maintainer side - Brian asked for this on the gdm bug linked below.
I've tested this quite a bit and it works very well. Do you think it is really necessary to have gdm version detection since GDM follows the same release schedule?
it would be good, but I don't see it as a blocker. For gnome-screensaver there is at least a build-time dependency to ensure that g-s is built against a new-enough gdm, but the fusa code duplicates the gdm defines internally. It could be as simple as handling an error return from gdm and falling back to the old command.
(In reply to comment #3) > It would be nice to get some endorsement for the gdm addition required for this > from the fusa maintainer side - Brian asked for this on the gdm bug linked > below. I think this is a marvellous idea (as I think I might have said on bug 172912, but perhaps not). Thanks very much for the patch: I'll try and review it before the end of the week (sorry you've caught me at such a busy time) and if it's good I'll do a special 2.17.x release. From previous experience, checking the GDM version number all the time makes the code at least half as complex again as it needs to be, especially since (as Matthias said) all we need to do is fall back on error.
*** Bug 447457 has been marked as a duplicate of this bug. ***
*** Bug 172912 has been marked as a duplicate of this bug. ***
Exactly what was the point of marking an OLDER bug as a duplicate of this bug? Shouldn't it go the other way around? Marking it as solved at a time when the proposed patch hasn't been merged and the bug that offers the patch is still marked as unconfirmed wasn't particularly smart either.
I typically keep the bug with the patch open, regardless of which one is older. There is no way to mark a bug as duplicate without setting it to resolved, so that is not indicative of any (lack of) smartness.
Hello Mr Maintainer, I'm pinging you, just to review the patches; Thanks a lot -- This message was brought to you by the Bugsquad team ©
The patch has been obsoleted by the gdm rewrite.