D15302: [effects/coverswitch] Port to TimeLine

davidedmundson added inline comments.

INLINE COMMENTS

> coverswitch.cpp:406
>          if (stop)  // Fade out windows not on the current desktop
> -            data.setOpacity((1.0 - timeLine.currentValue()));
> +            data.setOpacity((1.0 - timeLine.value()));
>          else // Fade in Windows from other desktops when animation is started

one set of parentheses seems redundant

> coverswitch.cpp:545
>                  stop = true;
> -                timeLine.setCurrentTime(timeLine.duration() - timeLine.currentValue());
> +                timeLine.setElapsed(timeLine.duration() - timeLine.elapsed());
>              } else {

it'll be a bigger refactor, but would it be more in line with Timeline to do .setDirection()

(maybe afterwards)

then we can kill all the

  data.setOpacity(1.0 - timeLine.currentValue());
  if (stop)
      data.setOpacity(timeLine.currentValue());

everywhere.

> coverswitch.cpp:550
>          } else
>              effects->setActiveFullScreenEffect(0);
>          mActivated = false;

we should reset the timeline here

If animateStart is on and animateStop is off and a user cancels halfway through the load, we become inactive straight away and won't hit the main cleanup.

> coverswitch.cpp:740
>      }
>      if (specialHandlingForward) {
> +        data.multiplyOpacity((1.0 - timeLine.value() * 2.0));

I know you've not touched this, but

should these two lines be wrapped in

&& timeLine.value < 0.5

> coverswitch.cpp:918
>      }
>      effects->setActiveFullScreenEffect(0);
>      mActivated = false;

should we not reset the timeline here?

REPOSITORY
  R108 KWin

REVISION DETAIL
  https://phabricator.kde.org/D15302

To: zzag, #kwin
Cc: davidedmundson, kwin, mkulinski, ragreen, jackyalcine, Pitel, iodelay, bwowk, ZrenBot, lesliezhai, ali-mohamed, hardening, jensreuterberg, abetts, sebas, apol, mart