The proposal is now on Stage 2: https://github.com/tc39/proposal-intl-displaynames

The proposal moved to Stage 3 today.

Depends on D51855

Adds the standard boilerplate for a new Intl object.

Depends on D52163

Depends on D52164

Depends on D52165

This is the script from bug 1585536, updated to support Intl.DisplayNames.

Depends on D52166

Part 1 increases the unzipped ICU data file size by about 25% and the zipped ICU data file size by about 30%.

Where "Stripped lang_tree + region_tree" is everything from lang_tree and region_tree removed which doesn't get used by Intl.DisplayNames. (That's the configuration we would use when actually trying to ship Intl.DisplayNames.)

The big question on data size, seems to me, is what it does to the Firefox download size overall. Any numbers on that?

Try-build without the changes: https://treeherder.mozilla.org/#/jobs?repo=try&revision=ec5d88f5002b7c6a278cddcd9892c2f515e3030e
Try-build with the changes: https://treeherder.mozilla.org/#/jobs?repo=try&revision=98f89077e047fdc2a7e502ecc595b8a8affef046

The try-builds are using opt+shippable for Windows and Linux, and opt for Android (there doesn't seem to be an opt+shippable configuration available for Android). (I'm not 100% sure the "aar" file is the correct file to look at for Android.)

Hm. Bug 1507697 indicates that a percent increase much smaller than any of those numbers, is sufficient to trigger treeherder warning-systems. So at minimum, increases to this degree are something that we are historically concerned about. :-|

I'm not certain where we go with these numbers, right now. Will poke at things/people Monday, I guess.

There are some r+ patches which didn't land and no activity in this bug for 2 weeks.
:anba, could you have a look please?
For more information, please visit auto_nag documentation.

Until I poke at people as mentioned in comment 10 (which I seem to have forgotten to do), this is sort of in limbo. Given the time of year and everyone traveling everywhere, it seems like that probably should wait til next year, at this point.

It looks like V8 will soon ship Intl.DisplayNames:

• https://groups.google.com/a/chromium.org/forum/?_escaped_fragment_=topic/blink-dev/vE4i42C2MUs
• https://bugs.chromium.org/p/v8/issues/detail?id=8703#c7

This shipped in Chrome M81 (April 7, 2020). We should probably get back to this and workout what the increased binary size means.

We have a green light for landing this binary increase size signed off by :bgrinstead and :esmyth.
Ni'ing both to get their final sign-off here and I think we can land.

I recommend that we land per the discussion with :gandalf, :dbolter, and :bgrinstead on 2020-02-20. None of the research or metrics I've reviewed since then change my recommendation.

:esmyth was the right person to sign off here, so it's fine to land in terms of the binary size increase.

https://hg.mozilla.org/mozilla-central/rev/b71ce4f9bca7
https://hg.mozilla.org/mozilla-central/rev/93aa11dd931b
https://hg.mozilla.org/mozilla-central/rev/a6338066da1f
https://hg.mozilla.org/mozilla-central/rev/8000b189d188
https://hg.mozilla.org/mozilla-central/rev/f816bb1d97cb

grr, sorry, wrong bug!

I thought this is shipping in 78, but I'm getting Intl.DisplayNames is not a constructor in Nightly.
Do we have a bug filed for enabling this?

(In reply to Florian Scholz [:fscholz] (MDN) from comment #21)

Do we have a bug filed for enabling this?

We don't yet have a bug for enabling it. There are still some open issues in the proposal which require fixing before we can consider shipping it. Some of the reported spec bugs have been resolved recently, so we may be able to at least make Intl.DisplayNames available on Nightly.

Hi, is there any news on this feature getting enabled? André, could you provide links to the specs bugs if any are still open? Thanks!

I've added patches to enable Intl.DisplayNames just last week in bug 1654116.

Removing dev-doc-needed from here; we'll track documentation in https://bugzilla.mozilla.org/show_bug.cgi?id=1654116 instead.