Checking in on this again,
https://chromestatus.com/metrics/feature/timeline/popularity/5232 shows
increasing usage, so it would be good to decide on a path forward.

Based on the use counter call sites
<https://source.chromium.org/chromium/chromium/src/+/main:v8/src/builtins/builtins-intl.cc;l=799-854;drc=50ade2d8d071e10bc5d53234bb2c0b311c515940>,
these are the properties in question:

   - Intl.Locale.prototype.calendars
   - Intl.Locale.prototype.collations
   - Intl.Locale.prototype.hourCycles
   - Intl.Locale.prototype.numberingSystems
   - Intl.Locale.prototype.textInfo
   - Intl.Locale.prototype.timeZones
   - Intl.Locale.prototype.weekInfo


I exported the 106 sites listed on chromestatus.com into a spreadsheet,
sorted them randomly:
https://docs.google.com/spreadsheets/d/1_oOhcPq5fgOkd4pD9crSJqu2fnpyL2j7M9pxw7n_nD8/edit?usp=sharing

I looked at the first 10, searching for "Intl.Locale" in devtools.

8 of them use textInfo without fallback to getTextInfo(). Instead, they
just assume "ltr". If the sites don't support any RTL language, then I
don't think anybody would notice, but presumably some sites like adobe.com
do.

2 of the sites seem to correctly fall back to the get*() methods, and if
they changed the order of their code wouldn't even trigger our use counters.

I found and filed a bug for the most common pattern:
https://github.com/expo/expo/issues/39205

Frank, would you be able to analyze another 10 or 20 to see if there are
other common patterns?

>From what I've found I'm fairly confident the breakage wouldn't be severe,
but it would be good to notify libraries that need to adjust, so that
developers have a fix if they are affected.

Best regards,
Philip

On Thu, Mar 13, 2025 at 4:54 PM Chris Harrelson <[email protected]>
wrote:

>
>
> On Wed, Mar 12, 2025 at 4:22 PM 'Panos Astithas' via blink-dev <
> [email protected]> wrote:
>
>> On Wed, Mar 12, 2025 at 3:12 PM Mike Taylor <[email protected]>
>> wrote:
>>
>>> On 3/12/25 5:38 PM, Frank Tang (譚永鋒) wrote:
>>>
>>>
>>>
>>> On Wed, Mar 12, 2025 at 8:19 AM Chris Harrelson <[email protected]>
>>> wrote:
>>>
>>>>
>>>>
>>>> On Wed, Mar 5, 2025 at 10:39 AM 'Frank Tang (譚永鋒)' via blink-dev <
>>>> [email protected]> wrote:
>>>>
>>>>>
>>>>>
>>>>> On Wed, Mar 5, 2025 at 6:02 AM Daniel Bratell <[email protected]>
>>>>> wrote:
>>>>>
>>>>>> Looks like the use counter LocaleInfoObsoleteGetters is at 0.033%
>>>>>> which is a bit high.
>>>>>>
>>>>> Dear Daniel:
>>>>>
>>>>> From your point of view, could you please define what number would
>>>>> be considered as not "high" or "low enough" ?
>>>>>
>>>>
>>>> We don't have a hard and fast rule, but in this case it's "high enough"
>>>> that we would like a bit more info to estimate risk.
>>>>
>>>
>>> If there are no hard and fast rules, then how do we determine 0.033% is
>>> a bit high? but not low enough? gut feeling?
>>>
>>> Experience breaking things. See
>>> https://docs.google.com/document/d/1RC-pBBvsazYfCNNUSkPqAVpSpNJ96U8trhNkfV0v9fk/edit?tab=t.0#heading=h.83o2xr8ayal6.
>>>
>>>
>>>
>>>
>>>
>>>>
>>>>>
>>>>>
>>>>>>
>>>>> I can't see LocaleInfoFunctions at all so does that mean that it has
>>>>>> never been triggered?
>>>>>>
>>>>> No, we didn't hook up that to the same UI. All other intl features
>>>>> were hook up with this UI and while I added the CL
>>>>> https://chromium-review.googlesource.com/c/chromium/src/+/6020626/  So
>>>>> syg@ said I should not use blink_feature but webdx_feature due to 
>>>>> "jstenback's
>>>>> email about WebDX features"
>>>>>
>>>>> So I hooked up the counter of both kLocaleInfoObsoletedGetters
>>>>> and kLocaleInfoFunctions to that
>>>>> but we later on find there are no easy way to access that info from
>>>>> public UI yet. So in
>>>>> https://chromium-review.googlesource.com/c/chromium/src/+/6089243 I
>>>>> change LocaleInfoObsoleteGetters back to blink_feature
>>>>>
>>>>> I am not sure how to access the query to check webdx_feature. If
>>>>> anyone know how to query that, we can see from that.
>>>>>
>>>>
>>>> Your use counter is here:
>>>> https://chromestatus.com/metrics/feature/timeline/popularity/5232
>>>>
>>>
>>> That one is for LocaleInfoObsoleteGetters, which is  blink_feature that
>>> we know how to see the counter
>>>
>>>
> Ok, but the URL above also lists various sites that are
> triggering LocaleInfoObsoleteGetters, ranked by popularity. I'm suggesting
> checking the top 20 such sites to see if any are broken.
>
> [image: Screenshot 2025-03-13 8.53.08 AM.png]
>
>
>> What we do not know how to access is kLocaleInfoFunctions, which is
>>> webdx_feature that we do not know how to access the counter.
>>>
>>> Please reach out to pastithas@ or kadirtopal@ for help on that.
>>>
>>
>> Unfortunately we don't yet expose these use counters in ChromeStatus, but
>> we're working on it. In the meantime internal UMA data
>> <https://uma.googleplex.com/p/chrome/timeline_v2?sid=f68dc5965348456a1cfb7dae5fa891d8>
>>  should
>> help make progress here.
>>
>> Thanks,
>> Panos
>>
>> --
>> You received this message because you are subscribed to the Google Groups
>> "blink-dev" group.
>> To unsubscribe from this group and stop receiving emails from it, send an
>> email to [email protected].
>> To view this discussion visit
>> https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAJ8Nf23YhpqRu91ShWcrQLgeoOrjC5KsDnhexLVK7S-sGM7tfA%40mail.gmail.com
>> <https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAJ8Nf23YhpqRu91ShWcrQLgeoOrjC5KsDnhexLVK7S-sGM7tfA%40mail.gmail.com?utm_medium=email&utm_source=footer>
>> .
>>
> --
> You received this message because you are subscribed to the Google Groups
> "blink-dev" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to [email protected].
> To view this discussion visit
> https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAOMQ%2Bw-Rxs7zGWcNUMv2iDi4GxkWD%3Dh%2B5v8ABuim9z08c7VweQ%40mail.gmail.com
> <https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAOMQ%2Bw-Rxs7zGWcNUMv2iDi4GxkWD%3Dh%2B5v8ABuim9z08c7VweQ%40mail.gmail.com?utm_medium=email&utm_source=footer>
> .
>

-- 
You received this message because you are subscribed to the Google Groups 
"blink-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion visit 
https://groups.google.com/a/chromium.org/d/msgid/blink-dev/CAARdPYdMjrApMf_0Muz_kQvVYiE6rC%3D6KQvJY07Gv1ewMH2RmA%40mail.gmail.com.

Reply via email to