Notice that there is a remote chance to break
DnD between clients (but virtually everything speaks v3 of the
protocol, also Qt5 is badly broken reg. DnD in toto)
Sorry for the build failure - i simply collected the patches but didn't try to build them w/o other ones.
About DnD, try eg. xfe or any Qt application - gtk+ seems extremely picky where it sends client messages.
Not yet sure about the details - this could fail for window hovering and require internal dispatching for the iconbar :-(
I'll have another look.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Afaics, gtk+ does not send events to override redirect windows (the toolbar...) and otherwise only to the highest child with the XdndAware property set (ie. the client - dragging onto the titlebar works, though)
There isn't much we can do here, the entire thing won't really work with gtk+ DnD's :-(
Boils down to "is a half-working feature better than none?" (where gtk+ may be so important to the fluxbox audience that it's not exactly "half")
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Let's stall this.
I noticed broken DnD between two different toolkits with this and also gtk+ doesn't work.
We might ditch the entire approach and investigate on tracking the DnD selection buffer to poll (uahhhhhh; xinput2 may come to the rescue) the pointer position ;-)
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
i tested the patch and it did nothing: i tried to dnd things from thunar, chrome etc onto the iconbar: no raise.
after merging to the iconbar-branch i got a compile error in https://github.com/luebking/fluxbox/commit/19327d0119e83027665ab7a27e13483f096c8150#diff-d930a6eae074ce5ede372cd087f736dcR102 (line 102) because the forwarded class FluxboxWindow is used to find the FluxboxWindow::raise() function. i solved it by adding
#include "Window.hh"
in the top.but the main issue stays: how can i test the patch?
Sorry for the build failure - i simply collected the patches but didn't try to build them w/o other ones.
About DnD, try eg. xfe or any Qt application - gtk+ seems extremely picky where it sends client messages.
Not yet sure about the details - this could fail for window hovering and require internal dispatching for the iconbar :-(
I'll have another look.
GRUMMELBRUMMELGRUMMELBRUMMELGRUMMELBRUMMELGRUMMELBRUMMEL.
Afaics, gtk+ does not send events to override redirect windows (the toolbar...) and otherwise only to the highest child with the XdndAware property set (ie. the client - dragging onto the titlebar works, though)
There isn't much we can do here, the entire thing won't really work with gtk+ DnD's :-(
Boils down to "is a half-working feature better than none?" (where gtk+ may be so important to the fluxbox audience that it's not exactly "half")
yep :)
Let's stall this.
I noticed broken DnD between two different toolkits with this and also gtk+ doesn't work.
We might ditch the entire approach and investigate on tracking the DnD selection buffer to poll (uahhhhhh; xinput2 may come to the rescue) the pointer position ;-)
Should we have the dnd raise over the iconbar and deactivate the direct raise on window entering for the time being?