On Thu, 2007-04-19 at 09:43 +0200, Anselm R. Garbe wrote: > Hi there, > > I created new releases: > > http://www.suckless.org/download/dwm-4.0.tar.gz > http://www.suckless.org/download/dmenu-3.0.tar.gz > > Both contain some minor changes (actually I jumped from > dmenu-2.8 to dmenu-3.0, because I didn't checked carefully that > there was no dmenu-2.9, nevermind ;)) > > dwm-4.0 works with clients which set the TRANSIENT_FOR hint to > some junk value (see the mailing list archives for such broken > apps), and it also restores the client border at unmanage() time > to the initial value. This fixes a nasty bug (border remains in > north-west edges) when toggling the mplayer window between > normal and fullscreen state. > > However, I wasn't able to workaround/fix the firefox issue when > toggling firefox into fullscreen mode. Actually firefox is > broken, it does not conform to ConfigureRequest responses > specified in ICCCM. If firefox is a tiled window, dwm ignores > any configurerequest and sends back a synthetic configure notify > event which sets the border_width to the value firefox requested > - this notifies firefox (conforiming to ICCCM) that the WM > ignores the request of the client. But firefox resizes its > internal pixmap buffer anyways and assumes a wrong window size > afterwards. Not to mention that firefox behaves similiar worse > in either floating layout or floating mode as well. > > I recommend you: DO NOT use the fullscreen-feature of firefox if > the side-effects with dwm annoy you. Instead view it in a > separate tag. > > Regards,
Thanks arg! bstack patch should be out this weekend. -RPM
