1289340 - Expose Intl.DateTimeFormat.prototype.formatToParts

Status: VERIFIED FIXED

(Core :: JavaScript: Internationalization API, enhancement)

Comment 1

[Zibi Braniecki [:zbraniecki][:gandalf]]

Attachment: Bug 1289340 - Expose Intl.DateTimeFormat.prototype.formatToParts.

Review commit: https://reviewboard.mozilla.org/r/78426/diff/#index_header
See other reviews: https://reviewboard.mozilla.org/r/78426/

Comment 2

[Zibi Braniecki [:zbraniecki][:gandalf]]

V8 added DateTimeFormat.prototype.formatToParts - https://bugs.chromium.org/p/v8/issues/detail?id=5244 and seems like the API is pretty much verified and frozen, with three implementations (Intl.js, Gecko and Chrome) passing test262 tests suite for it.

We should be ready to expose it to content.

Comment 3

[Zibi Braniecki [:zbraniecki][:gandalf]]

Comment on attachment 8790744 [details]
Bug 1289340 - Expose Intl.DateTimeFormat.prototype.formatToParts.

Review request updated; see interdiff: https://reviewboard.mozilla.org/r/78426/diff/1-2/

Comment 4

[Jeff Walden [:Waldo]]

Comment on attachment 8790744 [details]
Bug 1289340 - Expose Intl.DateTimeFormat.prototype.formatToParts.

https://reviewboard.mozilla.org/r/78426/#review77040

The XXX bug you mention claims I'm module owner (which I am not), but IRC word is the staging for this is right, so I think we're good.

::: js/src/builtin/Intl.cpp:1819
(Diff revision 2)
> -        RootedValue ftp(cx);
> +    RootedValue ftp(cx);
> -        HandlePropertyName name = cx->names().formatToParts;
> +    HandlePropertyName name = cx->names().formatToParts;
> -        if (!GlobalObject::getSelfHostedFunction(cx, cx->global(),
> +    if (!GlobalObject::getSelfHostedFunction(cx, cx->global(),
> -                    cx->names().DateTimeFormatFormatToParts,
> +                cx->names().DateTimeFormatFormatToParts,
> -                    name,
> +                name,
> -                    0, &ftp))
> +                0, &ftp))

It'd be better adding this to the JSFunctionSpec for DateTimeFormat instead of preserving this needed-for-conditionality code.  But this gets the job done for now, and I can go back and fix it to be the idiomatic way at some later time.

Comment 5

[Zibi Braniecki [:zbraniecki][:gandalf]]

Comment on attachment 8790744 [details]
Bug 1289340 - Expose Intl.DateTimeFormat.prototype.formatToParts.

Review request updated; see interdiff: https://reviewboard.mozilla.org/r/78426/diff/2-3/

Comment 6

[Zibi Braniecki [:zbraniecki][:gandalf]]

https://hg.mozilla.org/integration/mozilla-inbound/rev/3f47a92541c8e445757683d0e3e21e6df5d1a0d2
Bug 1289340 - Expose Intl.DateTimeFormat.prototype.formatToParts. r=waldo

Comment 7

[Carsten Book [:Tomcat]]

https://hg.mozilla.org/mozilla-central/rev/3f47a92541c8

Comment 8

[Sebastian Zartner [:sebo]]

Works for me as expected.

Updated the compatibility info:
https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/DateTimeFormat/formatToParts#Browser_compatibility

And the release notes:
https://developer.mozilla.org/en-US/Firefox/Releases/51?document_saved=true#JavaScript

Zibi, can you please verify that the docs are correct?

Sebastian

Comment 9

[Zibi Braniecki [:zbraniecki][:gandalf]]

:sebo - sorry for late response. In the Specifications box the comments indicate that it's "Not yet part of the draft specification. ".

At this point it is merged into the ECMA 402 spec.

Also, V8/Chrome has it exposed behind --harmony flag - https://bugs.chromium.org/p/v8/issues/detail?id=5244

And since we're at it - https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/Intl/getCanonicalLocales is already exposed in Chrome as well :)

Comment 10

[Sebastian Zartner [:sebo]]

Added the compatibility information and removed the outdated spec. comment. Zibi, can you please verify them again?

Sebastian

Comment 11

[Zibi Braniecki [:zbraniecki][:gandalf]]

LGTM!

Comment 12

[Sebastian Zartner [:sebo]]

Perfect! Thanks for the prompt feedback!

Sebastian