Re: [encoding] Signaling EOF in the streaming case is unspecified (or too unclear) (#12)

The definition of the flags is separate from the algorithm, so it's easy to get confused by thinking that the *streaming flag* is an algorithm-local variable (as opposed to being a field of the object), since the corresponding JS argument is to the `decode` method and not to the constructor.

Other than that, I guess the general problem is that the last two top-level steps of the algorithm require examining the implications of the definitions they reference and, thus, the algorithm isn't obvious when taking a look without reading all the definitions and thinking through their implications. 

The change suggested in #13 would alleviate the confusion. To remove confusion completely, substep 2 of major step 5 could use an informative note that points out that streamability is based on omitting EOF handling in that step and leaving flags such that the step 1 is skipped on the next call.

---
Reply to this email directly or view it on GitHub:
https://github.com/whatwg/encoding/issues/12#issuecomment-154401520