> Questions: there are 2 appSplash instances? Is that if user starts 2 apps
> quickly?
Yeah, it mostly catches the case that the screenshot is not available for some reason, for example the one you mentioned.
>
>
> + property bool attachedToView: true
> a comment explaining this would be nice. In fact, the logic in the
> EdgeDragArea could do with the same.
very valid. added.
>
> + if (oneWayFlick && spreadView.shiftedContentX > units.gu(2) &&
> spreadView.shiftedContentX < spreadView.positionMarker1 * spreadView.width) {
> please wrap that line a bit
wrapped all to 120 chars now.
>
> + property real positionMarker3: 0.6
> + property real positionMarker4: .9
> Consistency please
fixed.
>
> + // Stage of the animation:
> + property int stage: not a fan of the term, as "Stage" has a particular
> meaning for us. Would the Flickable's "States" system be useful here, as
> instead of numbers, can give the animation stages names?
Don't really want to go for strings as I do <, > comparison etc. While that would still work with strings its imo much more fail prone if someone renames or adds a stage. I did, however rename it to "phase" now to avoid the naming collision.
>
> + onShiftedContentXChanged: {
> + case 1:
> /me pedant, but adding "break" here too might spare a future coder some pain
seems so... dropped it.
>
> + if (spreadView.shiftedContentX == spreadView.width *
> spreadView.positionMarker2) {
> what's going on here? floating point equality comparison makes me suspicious
Turns out its not needed any more at all as I later changed the logic to always hit the branch before in this case already.
>
> + id: spreadRow
> + width: Math.max(3, ApplicationManager.count) * spreadView.tileDistance +
> (spreadView.width - spreadView.tileDistance) * 1.5
> fancy maths could do with a comment explaining it :)
done.
> +++ qml/Stages/SpreadDelegate.qml
> + anchors { left: parent.left; bottom: parent.bottom; top: parent.top;
> topMargin: priv.heightDifference * Math.max(0, 1 - root.topMarginProgress) }
> please wrap so topMargin is on its own line, as that's the nontrivial bit.
> Questions: there are 2 appSplash instances? Is that if user starts 2 apps
> quickly?
Yeah, it mostly catches the case that the screenshot is not available for some reason, for example the one you mentioned.
>
>
> + property bool attachedToView: true
> a comment explaining this would be nice. In fact, the logic in the
> EdgeDragArea could do with the same.
very valid. added.
> shiftedContentX > units.gu(2) && shiftedContentX < spreadView. positionMarker1 * spreadView.width) {
> + if (oneWayFlick && spreadView.
> spreadView.
> please wrap that line a bit
wrapped all to 120 chars now.
>
> + property real positionMarker3: 0.6
> + property real positionMarker4: .9
> Consistency please
fixed.
>
> + // Stage of the animation:
> + property int stage: not a fan of the term, as "Stage" has a particular
> meaning for us. Would the Flickable's "States" system be useful here, as
> instead of numbers, can give the animation stages names?
Don't really want to go for strings as I do <, > comparison etc. While that would still work with strings its imo much more fail prone if someone renames or adds a stage. I did, however rename it to "phase" now to avoid the naming collision.
> tXChanged: {
> + onShiftedConten
> + case 1:
> /me pedant, but adding "break" here too might spare a future coder some pain
fair enough, fixed.
> .SleepyDuration
> +// duration: UbuntuAnimation
> remove?
done.
>
> + spreadView.stage = 4;
> leftover?
seems so... dropped it. shiftedContentX == spreadView.width * positionMarker2 ) {
>
> + if (spreadView.
> spreadView.
> what's going on here? floating point equality comparison makes me suspicious
Turns out its not needed any more at all as I later changed the logic to always hit the branch before in this case already.
> ger.count) * spreadView. tileDistance + tileDistance) * 1.5
> + id: spreadRow
> + width: Math.max(3, ApplicationMana
> (spreadView.width - spreadView.
> fancy maths could do with a comment explaining it :)
done.
> +++ qml/Stages/ SpreadDelegate. qml erence * Math.max(0, 1 - root.topMarginP rogress) }
> + anchors { left: parent.left; bottom: parent.bottom; top: parent.top;
> topMargin: priv.heightDiff
> please wrap so topMargin is on its own line, as that's the nontrivial bit.
done.
>
> + scale: 1
> unnecessary, that's the default value.
done