On Sun, Jun 03, 2001 at 03:48:32PM +0200, Tomas Ogren wrote: > On 03 June, 2001 - Mikhael Goikhman sent me these 1.9K bytes: > > > On 03 Jun 2001 05:21:36 +0200, Tomas Ogren wrote: > > > > > > There's still (cvs as of now) problems with focus in Nautilus.. I've > > > been busy+++ lately, but I will have spare time to spend now.. It > > > doesn't get any focus.. > > > > I though Dominik fixed this... > > He tried..
Well, it's hard to fix without any feedback ;-) I fixed the problem with the "Globally Active" focus model. I can't test Nautilus since I don't have it, though. > > > There is also a problem when an app sets it's (GNOME) WM layer before > > > the window is shown. It then sets a property on the window which level > > > it's supposed to be on, but it doesn't get read by fvwm it seems... > > > I see the code in gnome.c and style.c to read it, but I don't know which > > > order stuff is happening/supposed to happen, so it's a bit hard... > > > > > > Attached is a test program written in gtk/xlib... > > > create some windows on layer 4 and then start the program.. one of the > > > windows is supposed to land on layer 3 (below the others).. then you can > > > raise/lower it with the up/down buttons.. > > > > > > If given any argument to the program, it will show (map) the window > > > before trying to set the layer which makes it end up on the correct > > > layer directly (almost.. you can see it starting infront of the windows, > > > then jumping back) > > > > 1) There is a typo in the program, so it can't be compiled without a fix, > > the variable "level" should be replaced with "layer" globally. > > Sorry.. I was quite tired when I sent this and I made some last second > changes without compiling.. > > > 2) To compile it: gcc -o winhints `gtk-config --cflags --libs` winhints.c > > > > 3) I can't reproduce any problem with this program, it works just well > > with or without arguments. Of course it is not consistent because it > > first requests to place itself to "layer-1" and then to "layer". > > But if you patch the program to replace "layer-1" with "layer" it will > > become consistent. > > > > A note, it is intended that a window is lowered in FVWM when it changes a > > layer, say from 5 to 4, and raised - when changes a layer from 3 to 4. > > Ok, I made some more changes to make it consistent now. The point is: It > doesn't matter what layer I request if I do it _before_ mapping the > window. Fvwm will still put it on the default layer. If you want to put > it behind everything (like the nautilus background thingie), you don't > want it to first cover the entire screen then pop back below everything. > > There seems to be code to read that stuff when the window is mapped, but > I suppose it's called in the wrong order or something. > > How to reproduce stuff: > 1) compile with $CC -o winhints `gtk-config --cflags --libs` winhints.c > 2) make a window cover most of the screen so there will be overlapping > 3) run winhints .. if your normal layer is 4 (which is default), then the > larger of the windows from 'winhints' should be below the window from 2) > but that doesn't happen. > 4) Click on "Same" which will set the layer to layer 2 again but by > sending an event this time, now it pops down to layer 2. > 5) Clicking on up/down will raise/lower the window as expected. > > If you give any arguments to winhints, it will first map the window, > then set the layer.. so it will appear at the default layer, then pop > back to layer 2. Fvwm does not honour GNOME hints, only GNOME client messages. No need to test code that is never compiled. Bye Dominik ^_^ ^_^ -- Dominik Vogt, [EMAIL PROTECTED] Reply-To: [EMAIL PROTECTED] -- Visit the official FVWM web page at <URL:http://www.fvwm.org/>. To unsubscribe from the list, send "unsubscribe fvwm-workers" in the body of a message to [EMAIL PROTECTED] To report problems, send mail to [EMAIL PROTECTED]