search

mikewest / tc39-proposal-literals

Literals could be different than non-literals.
32 stars 4 forks source link

Normative: Initial spec draft #4

Open littledan opened 6 years ago

littledan commented 6 years ago

This spec draft is in line with issue #2 with @allenwb's suggestion for where the brand check function should go. Probably not ready for merging now, but I thought it might help to have this kind of concreteness as a strawman, to focus further discussion if this is a direction the champions want to go.

domenic commented 6 years ago

Wait, my suggestion was that it should be a getter on the template object, not as a static on Array. https://github.com/mikewest/tc39-proposal-literals/issues/2#issuecomment-346531623

littledan commented 6 years ago

Oops, well, this uses @allenwb's suggestion!

I agree with much of Allen's reasoning there (even if the mechanism for disabling eval will probably be different from the mechanism for locking down Array.isTemplateLiteral on the web for one).

littledan commented 6 years ago

@ajklein @mikewest What do you think, among the two alternatives proposed by @allenwb and @domenic for how the surface API should be exposed here? Or, should we not even include it for JavaScript programs and just make the internal slot available to embedders?

ajklein commented 6 years ago

@littledan API concerns for this proposal haven't yet bubbled to the surface, for me at least, so I don't have an opinion to offer.

There's some ongoing (offline) discussion about how well this proposal meets the intended Google use-cases, and those need to be resolved before further work is devoted to this particular proposal.