make mapRanges usage smarter

[ianstormtaylor]

Do you want to request a feature or report a bug?

Improvement.

What's the expected behavior?

Following from #1993 (comment) it would be nice if the value.mapRanges uses could be made smarter. It should be possible to calculate what the next anchorPath and focusPath should be for all of those updates instead of setting them to null to be re-calculated in Range.normalize by using the key -> path mapping (which can be slow to compute).

[jtadmor]

taking a look at this right now.

fwiw, it does look some behavior was broken for selection based on the one test i have added so far (not sure if that has to do with how the test is set-up, possibly not calling Range.normalize in the operations tests).

[ianstormtaylor]

Moving away from even having this logic exposed, and it can be made more performant as an internal helper in core instead as needed.