Re: [whatwg/fetch] redirects and etag/if-none-match (Issue #1770)

I was chatting with @wanderview and it would seem as if what Chrome does makes the most sense (and the spec should generally match Chrome's behavior). Here are some thoughts:

- Safari's behavior effectively ignores status code, which doesn't seem reasonable. For example, why would a browser want to cache if there's a 500?
- Firefox never sends `if-none-match` headers ever again if it receives a non 2xx code, which also doesn't seem quite reasonable. A server could have an intentional 302 and then later legitimately change to a 200. But then Firefox won't cache future variants.
- Chrome is pretty reasonable all things considered. And 200 and 206 are the only status codes that imply that there's content to cache.

I can draft a PR with the proposed spec changes and write a matching WPT. (But please let me know if the proposed spec or the approach seems unreasonable.)

-- 
Reply to this email directly or view it on GitHub:
https://github.com/whatwg/fetch/issues/1770#issuecomment-2397643818
You are receiving this because you are subscribed to this thread.

Message ID: <whatwg/fetch/issues/1770/2397643818@github.com>

Received on Monday, 7 October 2024 18:49:00 UTC