Re: [heycam/webidl] A non-optional dictionary argument makes sense at times. (#130)

> That seems like a change, because thus far that's not been possible.

It isn't a change to the processing model, just the rules about when you can and can't use optional. (e.g., it doesn't change the built-in logic of how WebIDL converts JavaScript objects into dictionaries).

> It also creates the problem where folks forget to mark it as optional which is why we started to require this to begin with.

We already have that problem today: see the three previous examples where the authors (yourself and myself included) forgot to mark it as optional. Having a guideline around this is helpful for educating spec authors. Making that guideline a hard-and-fast rule doesn't really solve anything.

The one reason I can think of for keeping it as a "MUST" is that it opens the possibility of writing an automated tool (perhaps idlharness or some other existing tool that processes IDLs) to actually mark it as an IDL error (which it could do just by parsing the IDL and not having to know anything about the prose). If this was a "SHOULD" requirement, that automated checking wouldn't be possible. Is that something we really need though?

-- 
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/heycam/webidl/issues/130#issuecomment-313330777