Re: [whatwg/webidl] ExtendedAttributeList can match unbalanced ( ), [ ], and { } (Issue #1100)

Hm... I think I understand now. Thanks. And you're right that it's really unclear that that's how it works.

So, when the standard says, "When tokenizing, the longest possible match must be used," it means the longest possible match _regardless of where that match appears in the grammar_? I took it to mean only the longest possible match that could appear in that position, given all the earlier tokens. But I guess the examples right after that statement don't have that limitation. And this does all match the usual model that separates the tokenizer and the parser.

So it looks like the text is technically accurate, but for clarity I'd add a note after the examples of what is not an **identifier** to mention that it doesn't matter what rule is being matched when the token is found. Does that sound reasonable? It definitely would have helped me make sense of this, and I imagine other people have been confused in the same way.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/whatwg/webidl/issues/1100#issuecomment-1042042063
You are receiving this because you are subscribed to this thread.

Message ID: <whatwg/webidl/issues/1100/1042042063@github.com>