On Wed, 2011-09-14 at 22:22 +0100, Colin Guthrie wrote: > 'Twas brillig, and David Henningsson at 14/09/11 15:10 did gyre and gimble: > > On 09/14/2011 03:45 PM, Maarten Bosmans wrote: > >> 2011/9/14 David Henningsson<david.henningsson at canonical.com>: > >>> Module-dbus-protocol gets the default sink, which sets the default sink > >>> if not already set. This is turn makes module-default-device-restore do > >>> nothing. > >>> To solve the problem, make sure module-default-device-restore is loaded > >>> before module-dbus-protocol and not the other way around. > >> > >> Wouldn't this gloss over the fact the one could reasonably expectation > >> that loading a protocol module does not affect the state of the daemon > >> other than that (obviously) a module has been added? > >> > >> I don't believe that module-dbus-protocol should be special in this > >> regard. > > > > Well, I don't like that asking for the default sink turns out actually > > saving a default sink to disk, if nobody has set the default sink > > earlier. But fixing that is more difficult, and once we have the > > priority lists, this will be handled differently and hopefully much > > better. This is more of a quick fix/workaround for v1.0. > > > > Ok to push? > > While I agree with Maarten, I think this will likely have to do for v1.0. Interesting catch. There's callers of pa_namereg_get_default_sink() in a *lot* of places, and we really should get rid of the side-effect of this call. Even if we do this, David's fix is required, so I'll push this shortly and file a bug to get rid of side effect so we don't forget post-1.0. > I'm too tired just now, so I'll push this in the morning (unless Arun's > jetlag-based productivity beats me to it) and push out a tarball in the > morning. Still waking up at absurd hours. :) > Failing anything else horrible happening, that'll be 1.0 (I know Arun > may have a couple fixes to push but I don't envisage anything else > major). It's so close I can taste it :D ++ :D Cheers, Arun
