[Parser] We can get rid of Decimal character reference start state in tokenizer

[inikulin]

This state is excessive: we can handle ASCII digits and switch to the decimal character reference state in the numeric character reference state.

[zcorpan]

Yes... but it's not clear to me that this is worth changing in the spec. Each change has a risk of introducing a regression, and thus lowers the trust implementors have that the spec is correct. So for a change like this I recommend doing the optimization in implementation and leave the spec as-is.

[untitaker]

@zcorpan do you still feel that way? I'm considering opening a PR for this and #1323 (comment)

[zcorpan]

Yes, but with agreement and review from implementers it should be ok to change if you want to give it a go. Cc @whatwg/html-parser

[untitaker]

I've opened a pr for #1323 yesterday, will create the other one after/if that one is merged.