Make normalize language code which is not only 2-2 / Add langs

[sae220]

There are some languages whose code are not 2-2 letters (not like en (2), en-US (2-2) but fil (3), ja-hira (2-4)).
https://crowdin.com/project/makecode/discussions/152

[sae220]

@microsoft-github-policy-service agree

[abchatra]

@sae220 thanks so much for the contribution.

I do have a question. What is the reference for the localized name? Like "Wikang Filipino"? In the past there have been cases where people didn't agree with the localized name and we want to be sensitive for the same. If there reference for these localized name let me know.

[sae220]

@abchatra

@sae220 thanks so much for the contribution.

I do have a question. What is the reference for the localized name? Like "Wikang Filipino"? In the past there have been cases where people didn't agree with the localized name and we want to be sensitive for the same. If there reference for these localized name let me know.

I searched the words (like Filipino) on Wikipedia (https://en.wikipedia.org/wiki/Filipino_language) and copied the localized name.
I can't find the mistake if the localized name on wikipedia is wrong.

[jwunderl]

Thanks for looking into this!

Code change looks good, but I believe we'll have to run localizedName through internal translation teams to confirm before releasing the other locales, and we also will have make some changes in another repo / crowdin to get them all fully set up; could you remove the ones besides ja-HIRA for now since that was the other one we have fully set up otherwise, and we can work on the others separately?

[jwunderl]

Suggested change

	"ja-hira": {englishName: "Hiragana", localizedName: "にほんご"},
	"ja-HIRA": { englishName: "Hiragana", localizedName: "にほんご" },

quick update for consistency, we changed the language code in crowdin already~

[ganicke]

Might be a digression but to be precise with RFC 5646, it's one of:

• language code + - + region
• language code + - + script
• language code + - + script + - + region.

Hiragana in this case is a 'script' (style of writing) not a 'region' so it's commonly expressed as ja-Hira, whereas the region codes are all upper-case.

Refs:

• https://www.w3.org/International/articles/language-tags/
• https://www.iso.org/schema/isosts/v1.0/doc/n-cvd0.html

[sae220]

So should I fix to ja-Hira?

[ganicke]

No, not at the moment. That will require a corresponding change in the cloud serving stuff so we'll leave it as-is unless others on the team want to use this format.

[sae220]

@jwunderl

Thanks for looking into this!

Code change looks good, but I believe we'll have to run localizedName through internal translation teams to confirm before releasing the other locales, and we also will have make some changes in another repo / crowdin to get them all fully set up; could you remove the ones besides ja-HIRA for now since that was the other one we have fully set up otherwise, and we can work on the others separately?

OK, if we need to make some changes for inappropriate translation of localizedName, please let me know.

[sae220]

@jwunderl

So what should I do now?

[abchatra]

@sae220 I will take care of review and merging this week (Sorry we are middle of bunch of releases this week)

[abchatra]

For tracking here is the internal microsoft site (not publicly available) to the right localized names: https://microsoft.sharepoint.com/sites/globalreadiness/SitePages/Official-Language-Names.aspx