Added _NET_WM_MOVERESIZE handling for the move case

[Jeffrey-P-McAteer]

Added _NET_WM_MOVERESIZE handling for the move case and an empty function wm_resize_node in case anyone wants to let windows resize themselves.

This change adds the ability for programs like MPV to request that they be moved, see #864.

Tested by doing:

make bspwm && cp bspwm /tmp/
# Switch to tty2
startx /tmp/bspwm
# launch mpv and w+m1 drag without the modifier key

[baskerville]

These constants are already defined in xcb_ewmh.h.

[Jeffrey-P-McAteer]

Thank you, I updated the changes in d7a31ad. I don't know if github can do a PR as a squashed merge but I'm fine re-committing the 3 changes as 1 because it makes the history cleaner.

[baskerville]

I get the following warnings using GCC:

src/events.c: In function 'wm_move_resize_node':
src/events.c:355:3: warning: range expressions in switch statements are non-standard [-Wpedantic]
   case XCB_EWMH_WM_MOVERESIZE_SIZE_TOPLEFT ... XCB_EWMH_WM_MOVERESIZE_SIZE_LEFT:
   ^~~~
src/events.c: In function 'wm_resize_node':
src/events.c:387:49: warning: unused parameter 'e' [-Wunused-parameter]
 void wm_resize_node(xcb_client_message_event_t* e, coordinates_t loc)
                     ~~~~~~~~~~~~~~~~~~~~~~~~~~~~^
src/events.c:387:66: warning: unused parameter 'loc' [-Wunused-parameter]
 void wm_resize_node(xcb_client_message_event_t* e, coordinates_t loc)
                                                    ~~~~~~~~~~~~~~^~~

[deeedob]

Hey, what's the current state of this? The patch works for me. I think having the _NET_WM_MOVERESIZE atom enabled is kinda important, at least for me :]

[Jeffrey-P-McAteer]

@deeedob I remember there being community flak about code style when I originally worked on this change, which is why it was never merged in. I have since abandoned the change and moved back to using i3 personally because I like moving client windows which make use of that atom.