On Wed, Jul 21, 2010 at 04:39:26PM -0400, Chris Siebenmann wrote: > I am currently using a somewhat hacked version of fvwm 2.4.20 and I > would like to move to fvwm 2.5.x (for whatever x is current when I get > all of the pieces working). One of the stumbling blocks is duplicating > one piece of icon placement behavior in my current setup. > > I do not have an icon box ('Style * IconBox none'). How I want icon > placement to work is this:
Why? Just let it be the default (which is the whole screen). > - when a window is iconified for the first time (and does not have any > sort of icon placement specified), the icon is put right under the > mouse cursor. You can do this with a state, windowstyle, and FvwmEvent. Here's a rough outline -- oh, and it assumes 2.5.X, since you can't do this at all easily on 2.4.X. Good luck. # All windows, regardless, get state 1 which is removed when iconified. Style * State 1 DestroyModuleConfig A:* *A: iconify SomeFunc *A: deiconify SomeOtherFunc AddToFunc StartFunction I Module FvwmEvent A DestroyFunc A AddToFunc A + I ThisWindow (State 1) MoveIconWindow + I TestRc (NoMatch) Break DestroyFunc MoveIconWindow AddToFunc MoveIconWindow + I Move m+0 m+0 + I ... (other things?) DestroyFunc SomeOtherFunc AddToFunc SomeOtherFunc + I ThisWindow (State 1) WindowStyle 1 False ... which is far from idea -- something would have to set the state of the window to state 1 again if the placement of the icon was to go back to being under the mouse (unless the window was then subsequently closed anyway. YMMV.) > - when a window is iconified after that, the icon reappears wherever > it was last. See above. > I can see how to use a user function to handle the first-time icon > placement (I think; just use a function that does Iconify + Move to > $[pointer.x] $[pointer.y]), but the bit I'm stuck on is how to get icons > to stay where they are after that. Is there a conditional or a command > that would apply for this? See above. > (It looks like $[i.x] et al are only defined if the window is iconified, > so one can't abuse them to determine whether the icon has already been > positioned somewhere before the Iconify happens.) See above. > The other option is that I could add a new conditional that tests on > IS_ICON_MOVED() et al; would people find such a patch generally useful > and interesting for the main line code? No thanks. Not for 2.4.0, and not for 2.5.X, FVWM's in a feature freeze, and 2.4.X receives no new features anyway. See above. :) -- Thomas Adam -- "Deep in my heart I wish I was wrong. But deep in my heart I know I am not." -- Morrissey ("Girl Least Likely To" -- off of Viva Hate.)