Set IntlLegacyConstructedSymbol to non-deprecated#25749
Conversation
These features are specified as a "Normative Optional" but not "Normative Optional, Legacy" behavior. Given the distinction that the spec makes between legacy and non-legacy parts of the language, these features should not be marked as deprecated. Spec: https://tc39.es/ecma402/#sec-chaindatetimeformat Spec: https://tc39.es/ecma262/#sec-conformance Spec: https://tc39.es/ecma402/#sec-chainnumberformat
|
Do you think we should consult TC39 first? ECMA402 does not use the "legacy" verbiage at all at the moment, which suggests either it's not an editorial convention there yet (since it has no normative impact on conformance whatsoever), or they don't consider any part of ECMA402 as legacy. |
|
@Josh-Cena If you'd like to ask that'd be great, but I don't think we should wait on an answer. I did consider the idea that ECMA-402 did not take up the convention from ECMA-262. But since both specs point at each other for conformance, it's hard to read it in a way that allows one spec to make the distinction without the other doing the same. Additionally, some of the current specification text is less than a year old (tc39/ecma402@7018fba). I figured they knew what they were doing. |
Elchi3
left a comment
There was a problem hiding this comment.
I think we should go ahead for now. Happy to change it if we hear otherwise from specification folks.
Summary
The spec doesn't appear to discourage these features.
Test results and supporting details
These features are specified as a "Normative Optional" but not "Normative Optional, Legacy" behavior. Given the distinction that the spec makes between legacy and non-legacy parts of the language, "Normative optional" should not be marked as deprecated.
Confusingly, the symbol's description contains the substring
Legacy. This was a later addition to the spec that doesn't have much of a paper trail (and is older than the contemporary specification's "legacy" terminology—that is, it didn't have a defined meaning at the time). I think we should weigh the spec's explicit conformance text over a partial string match.Related issues
web-platform-dx/web-features#2564
#17410