Re: [whatwg/encoding] Add Streams support (#72)

No-one is advocating for option 2, right?

Just based on a long history of trying to explain quirks in the platform, a clean separation makes sense to me (i.e. option 3). It also lets us evolve the types independently as needed. I like `TextDecoderStream` (etc) as the name.

Sanity check: with option 3, `body.pipeThrough(new TextDecoder())` throws? (seems to per [spec](https://streams.spec.whatwg.org/#rs-pipe-through))

ISTM documentation and samples other than the spec will focus on one usage or the other (stream-based or simple synchronous conversions); the only confusion would be @ricea's mental model step 3, where a developer half-remembers the type. I can imagine the top search hit for "pipeThrough TextDecoder" being a StackOverflow page explaining "you need to use TextDecoderStream" but that doesn't seem to be the end of the world.

-- 
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/whatwg/encoding/issues/72#issuecomment-376639005