CLDR-17812 v46 vxml: Add {0} for de speed-light-speed one, fix error wording in TestCheckCLDR/TestPlaceholders

[pedberg-icu]

CLDR-17812

• This PR completes the ticket.

merge into #3880

de added unit data for speed-light-speed in which the "one" case has no placeholder; per discussion that reflects a misunderstanding of how this data is intended to be used, so add a placeholder so tests pass (the data will be deleted soon anyway). Also fix the wording of an error message in TestCheckCLDR/TestPlaceholders.

ALLOW_MANY_COMMITS=true

[macchiati]

Thanks for taking this. I think the solution is a bit different.

The LOCALE_DEPENDENT path means that there is an extra test on the locale's plural rules to make sure that it behaves like Arabic (and other languages); that is, where a particular plural form is unique and therefore does not need a placeholder. For example, "1 book" in Arabic is just "كتاب".

So for this case, the error message is misleading, and the German should be fixed to prefix the values with
"{0} "

[pedberg-icu]

Thanks for taking this. I think the solution is a bit different.
...
So for this case, the error message is misleading, and the German should be fixed to prefix the values with "{0} "

@macchiati Hmm. What German currently has is

• one: Lichtgeschwindigkeit = Speed of light
• other: {0}-fache Lichtgeschwindigkeit = {0} times the speed of light

That seems very sensible to me. Saying "1 times the speed of light" is certainly awkward in English and I would think in German too. The particle is traveling at lightspeed, or the particle is traveling at 0.3 times the speed of light.

Here the particular plural form for 1 is unique, for this unit.

In any case the test was bad because if there was no placeholder it was failing before even checking whether the status was LOCALE_DEPENDENT. We did not get an error for Arabic because the cases for the test currently only used en and de.

[macchiati]

I think there are significant issues with the data for a number of locales (will outline in email), so we probably want to retract the data for light-speed across locales. However, I suggest that we have a quick fix to the data and merge it into main before retracting so that we have a copy we can return to as needed.

So the quick fix I'm proposing is to add {0} before Lichtgeschwindigkeit

[pedberg-icu]

So the quick fix I'm proposing is to add {0} before Lichtgeschwindigkeit

Literally just "{0}" before, or "{0}-fache" before as with the other case?

[macchiati]

Just "{0} " before is fine (just sent you a note about the complexities).

…

On Thu, Jul 18, 2024 at 6:36 PM Peter Edberg ***@***.***> wrote: So the quick fix I'm proposing is to add {0} before Lichtgeschwindigkeit Literally just "{0}" before, or "{0}-fache" before as with the other case? — Reply to this email directly, view it on GitHub <#3887 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACJLEMBGMT66X7SMQ6CPKEDZNBUTJAVCNFSM6AAAAABLDQ3DPGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDEMZXHA2TGOBWGU> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[jira-pull-request-webhook]

Hooray! The files in the branch are the same across the force-push. 😃

~ Your Friendly Jira-GitHub PR Checker Bot

[pedberg-icu]

@macchiati OK I think I fixed as requested, PTAL. The error message wording was still backwards, needed fixing (see the previous error message by viewing more context above).