On Sat, Nov 20, 2021 at 5:22 PM Dominik Vogt <dominik.v...@gmx.de> wrote: > > > 1) Can we safely assume that the "bool" C type is available in > every relevant compiler?
The bool type is used elsewhere in the fvwm code, so I would assume it is okay. > 2) Please break up the calculations in DoSnapGrid into multiple > lines unsing temp variable. It's too hard to read = hard to > maintain. The previous un-readability of the code shouldn't be > the standard we aim at. ;-) I tried to clean this up a bit, was mostly just cutting/pasting code. > 3) Same for "smap_mon = ...". Ouch. Agreed, tried to simplify it a bit. > 4) I'm not sure whether screen boundaries should have a higher > priority than windows and not equal priorities. That would > require scoring. > > Shouldn't be too difficult to implement; the new sub functions > would simply return a score = the absolute snapping distance > that was used, and the lowest one wins. (Gird is still only > used if there was no other snapping.) > Returning a single score won't work, since vertical and horizontal are treated separately. Instead I now find both an appropriate monitor edge and window edge. Then snap to the closest. If none is found, fallback to using the grid. > 5) Edge resistance is a kind of reverse snapping with a different > snap distance: A window that has been moved past the screen > edge "snaps" back until it has moved more than the edge > resistance past the edge. This is exactly what snapattraction > does, just with the snap distance from a different variable. > > Can that be merged with the new code? This ended up being a bit more complicated, since the resistance isn't the same as snap attraction (since snap attraction snaps both left/right but resistance is only one direction). But I was able to update the monitor function to be able to now deal with both snapping and resistance by passing it a few bools to control its behavior. I also took advantage of the new info about if a monitor edge is an outside or inside edge to distinguish resistance between monitors and between pages. Though the code is a bit more complicated, I hope it is readable. On Sat, Nov 20, 2021 at 9:45 PM Dominik Vogt <dominik.v...@gmx.de> wrote: > > And there's still a bug that had been present in the old code: > > I.e. the top/left/right/bottom border of a window shouldn't snap > to the bottom/right/left/top border of a windows if that latter is > aligned with a screen edge. > I have added a final check in DoSnapWindow that ignores snapping to windows on the monitor edge. I think this fixes the issue. All of the updates have been pushed to js/snapattract. jaimos
