On Mon, Nov 06, 2017 at 09:54:30AM +0100 I heard the voice of Frank Steiner, and lo! it spake thus: > > If there is another xterm "shell" overlapping this console xterm, > the focus should change when I move the mouse to "shell", but it > doesn't. [...]
> So in 3.8.2 f.warpto doesn't call f.focus, but 4.0.1 seems to do > that. Well, that in particular isn't the case. And if it were, you wouldn't get freed up by moving into the root; f.focus locks it in all cases. And I can't imagine it's at all related to the fix for chromium/etc focus, since that was purely synthesizing hint on windows that don't have them. I'm unable to reproduce anything like that with an xterm named "CONSOLE", even with/out SloppyFocus and NoTitleFocus which seem the mostly likely contributor to changed activity. What app do you have behind that "CONSOLE" name? Maybe it's doing something twisty with us... At any rate, I'd say the most likely contributor would be the SetFocus() addition in the warping process we added in r500[0]. You could test that by pulling it out; e.g., patch under [1] (that code moved since r500). But, I'm having a hard time imagining how THAT could be causing anything like that behavior either. Maybe if we were getting a garbage past timestamp on the Enter events? But I'd expect that to be breaking the fudge out of all sorts of other things too, so that doesn't seem likely. Are you using any unusual X server or library implementation? [0] http://bazaar.launchpad.net/~ctwm/ctwm/trunk/revision/500 [1] === modified file 'win_utils.c' --- win_utils.c 2017-01-27 07:04:42 +0000 +++ win_utils.c 2017-11-08 02:35:42 +0000 @@ -812,7 +812,7 @@ } XWarpPointer(dpy, None, Scr->Root, 0, 0, 0, 0, x + t->frame_x, y + t->frame_y); - SetFocus(t, EventTime); + //SetFocus(t, EventTime); #ifdef DEBUG { -- Matthew Fuller (MF4839) | [email protected] Systems/Network Administrator | http://www.over-yonder.net/~fullermd/ On the Internet, nobody can hear you scream.
