Consider adding cancelAnimationFrame polyfill

[gaearon]

Related to #4545 and #4568.

It's not necessary for React, but I think most apps expect them to be available together. Since if you don't have access to the original rAF polyfill source you can't implement the cAF polyfill.

[SimenB]

Technically you can pass it to clearTimeout, but that's not a very intuitive API.

It will hopefully land in jsdom itself soon, but just doing global.cancelAnimationFrame = global.clearTimeout should be good enough in the meantime.

[gaearon]

By the way, by reading jsdom discussion, I'm not quite sure which behavior they'll pick.

Would it make sense to change the polyfilll to coalesce calls into 1/60 timer? e.g. if you call it twice synchronously it should probably schedule only once.

[gaearon]

Now that I think about it, I realize I don't actually remember how rAF works 😛

[SimenB]

When you schedule 2 at the same time, they should be fired at the same time. That's actually not the way it works in my PR to JSDOM, there they'll be scheduled 1/60 after calling all the time... Probably needs a queue of some sort.

E.g. run this in chrome, and see the now parameter be the same.

requestAnimationFrame((t)=>{console.log('hello', t)}); requestAnimationFrame((t)=>{console.log('hello', t)});

EDIT: Wait, I'm confusing myself. Need to double check the spec (or just use raf as suggested below)

[gaearon]

How do you feel about just using raf package as a fallback? Their logic looks good to me.

[SimenB]

I'm fine with just using raf. I was gonna use it initially (sent chrisdickinson/raf#35 for the API I wanted), but ended up just implementing it myself as we wanted to copy what react recommended as polyfill

[SimenB]

Only issue with raf is that they will schedule using the host's (i.e. the context that Jest itself executes in) setTimeout instead of the one from jsdom. Probably not an issue in practice, though

[gaearon]

we wanted to copy what react recommended as polyfill

Heh. We only recommended that because that's what we used in our tests, and it's the simplest thing that works. I don't think we really meant to endorse it though 😛

[SimenB]

Both requestAnimationFrame and cancelAnimationFrame has now landed in JSDOM. Not released yet, but IMO that means there's no reason to mess with adding raf to Jest

[gaearon]

Agree (if you can easily update).

[github-actions]

This issue has been automatically locked since there has not been any recent activity after it was closed. Please open a new issue for related bugs.
Please note this issue tracker is not a help forum. We recommend using StackOverflow or our discord channel for questions.