[kentonv]

Yes, your worker can make subrequests to other domains, and serving some assets out of S3 is a use case we specifically want to enable.

I haven't looked into how specifically to expose HTTP Push in the API but that certainly seems like something we should support.

[spankalee]

Please, add support HTTP/2 push!

This is very, very awesome work. My team has been working on enabling the PRPL pattern[1] and differential serving on a few platforms, and edge caching has been a problem. We've tried to use Vary: User-Agent, but that leads to low cache hits. This API would let us to much smarter UA sniffing at the edge.

From there we just need to parse some responses like JS and HTML, to be able to push their sub-resources, for an instant speedup and great caching for fine-grained deployments.

[1]: https://developers.google.com/web/fundamentals/performance/p...

[mdale]

Http/2 push is interesting here. Beyon static assists I wonder about read API orchestration or equivalent of graghQL on edge. I.e a single request comes in and it's broken apart against catchable resources and pushed back to the client from cache where possible.

[kentonv]

That's actually one of the canonical use cases we thought about while designing. It seems splitting graphql queries into cacheable chunks is a common need.