Reset warp lock after first motion

[acrisci]

Reset has_warp to false (after it has been set after a successful
warp) when the next motion is received.

This should still prevent the cursor "looping" when warping, while
preventing the cursor from getting "stuck" in torus mode.

fixes #35

[acrisci]

Especially, consider this scenario:

       +---+
 +---+ | 3 |
 | 1 +-+-+-+
 +---+ 2 |
     +---+

Imagine hitting the upper right edge of 1 (moving from left to right) without the current fix. You'd instantly end up on 3 because you warp from 1 to 2 as expected, but then also from the top edge of 2 to 3.

This doesn't address this issue, but I think it's more natural to go from 1 to 3 in this case.

[Airblader]

I'm not really happy with this. I get a lot of really weird twitches if I happen to move the cursor in a straight line, even in a standard use case screen setup.

+---+
| 1 |---+
+---+ 2 |
    +---+

Moving from left to right in the upper area in a straight line causes the correct warp to 2, but continuing to move the cursor makes it warp continuously between the top and bottom edge, which is really twitchy.

[Airblader]

What do you think about using a very short "timeout"? If the mouse stays still for a bit, we reset the flag.This would both allow continous movement as in my case and solve your problem if I understand it correctly, right?

[acrisci]

I'm not really happy with this. I get a lot of really weird twitches if I happen to move the cursor in a straight line, even in a standard use case screen setup.

I'm not seeing this on my end.

What do you think about using a very short "timeout"?

I'm not really into this. Maybe I'll think of something else.

[Airblader]

I'm not seeing this on my end.

Reproduce like this:

xedgewarp -t both -o "200x200+0+0,200x200+200+100" &
sleep 1; xdotool mousemove 150 50; sleep 1; for i in {1..150}; do xdotool mousemove_relative 1 0 && sleep 0.05; done
killall xedgewarp

[Airblader]

I'm going to close this PR for now. If we can come up with a different way to overcome this, we can open a new PR, but this solution is not what I want to merge.