normalize range input onChange events

[jquense]

Addresses the core of #554. ping @jimfb @spicyj

This plugin is getting large...personally I'd break it out into smaller modules but I didn't want to presume that here, so its all together. The other thing I didn't do was make the obvious DRY changes. There is a common pattern of tracking the active input used by three code branches that could be refactored into something a bit more concise. I have that change locally if ya'll feel like its appropriate for this PR otherwise I can leave the repetition.

There is also a larger question of whether it makes sense to just run all (or more than just ranges) inputs through this branch, which would also address annoyances similar to #3377 and #3484. That's a sweeping change though and, again, didn't want to presume.

I need to actually test this on old IE (Edge fires input events) but I don't have the vm set up at the moment.

[jquense]

the more I think about it (and maybe that was @spicyj's original intent) just deduping every change for all/most inputs would address a lot of the other edge case issues with onChange

I will PR something more comprehensive...first I want to first get a bunch of tests written around each inputs change behavior to catch any regressions

[jimfb]

@jquense Would love to get a fix merged. What is the status on this? Are you blocking on us, or are we blocking on you? Feel free to ping me on google hangouts if that would be helpful.

[jquense]

@jimfb I think the ball is in your court, but its a bit complicated. The curent problem with this diff, is that this will only work when when inputs can be focused, i.e attached to the document, which isn't intuitive.

#5746 offers a more complete way to dedupe, but also applies to all inputs. If we want to localize a fix to just these inputs then I need to pull changes from that PR. I haven't gotten enough feedback though to know where and how to move forward (this one? this approach?).

Glad to chat more about it, around whenever, if we need to set something up.

[jimfb]

Ok, if I understand this correctly (please correct me if I'm mistaken), #5746 is a superset of this PR. We've decided that we want to take that one, so I think this PR can be closed.

[jquense]

👍 yup this is safe to close

[facebook-github-bot]

@jquense updated the pull request.