Brenton Simpson
Brenton Simpson
The named arguments pattern is great for functions that take multiple arguments. We moved all operators to it in #193. However, there are some operators that usually take a single...
- Consistency with other platforms that implement Material Motion - Full control of the stack: - Can ignore concepts we don't need, like `error` and `complete` - Fewer concepts ->...
This could be a bit complex. We should test: - the final value is correct - intermediary values are correct - all the possible GestureRecognitionStates are correct - recognitionThreshold works...
At some point, we had a convention that interfaces should end with `able`; hence, all the operator interfaces, e.g. `MotionMemorable`. That convention hasn't really held true - both `MotionObservable` and...
Right now, there's not an easy way to get direction from a velocity - you have to drag an element by a certain distance. We should support this. Maybe something...
https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/Symbol/species This would clean up/DRY out the codebase (proxies is a bit of a hack), and provide a possible extension point for 3rd parties to create new derivations of MotionObservable....
Opening the settings pane on iOS breaks the drag. It's presumably sending touchcancel.
AMP foregoes pointer events entirely because iframe ad containers eat them. This makes the experience crappy in the general case (dragging images) to make the occasional case (dragging ads) possible....
There's a maddening property of PointerEvents: touch only triggers one `move` unless you set `touch-action`. I know this and it still bites me occasionally - I imagine it's even more...
Our unit tests currently use a very simple `MockObserver`, which essentially turns an observer inside-out so its `next` method may be called by unit tests. Unfortunately, this presumes that there...