Intl.DateTimeFormat API doesn't calculate Persian leap years accurately

[Sinakhx]

This is a very simplistic and inaccurate approach to calculating Jalali/Persian leap years.

I'm not sure if it's related to the nodejs work-group or the EcmaScript community, but as it's an implementation detail, I think it can be enhanced further, as the date calculations are very important.

I've actually reported this problem in an other library's issues section. Please follow along to see the details & examples there.

Please let me know if This was an intentional approach or just due to insufficient astronomical research. Thanks in advance.

[targos]

@nodejs/i18n-api

[ryzokuken]

@Sinakhx this is an ICU issue, which is the library that this code actually comes from. The more I dig deeper into the ICU calendar code, the more problematic I find it, so I'm inclined to believe you're right here at once. I'm working on a long-term effort to fix many of these errors but it'll take a while before it gets to a point where Node.js does things correctly here.

[mhdawson]

@ryzokuken would it make sense to oint @Sinakhx to the upstream ICU repos/source? I may not have enough context but my first thought is that is where the fix would need to go?

[srl295]

Don't fix in Node.js, please submit a ticket and PR to https://icu.unicode.org

[mhdawson]

@srl295 thanks for chiming in, was hoping we could point @Sinakhx to the right place upstream to fix.

[srl295]

yes, specifically:

@Sinakhx please open a PR against https://github.com/unicode-org/icu/blob/main/icu4c/source/i18n/persncal.cpp#L110-L115 you can link it here

[Sinakhx]

Thanks for your consideration & help.
I will open an issue there soon.

[ryzokuken]

Thanks @srl295

[Sinakhx]

I've created the pull request:
unicode-org/icu#2077

Please consider the workflow in order for the PR to be merged & let me know about the details & upcoming steps.

[Sinakhx]

I guess my endeavor was in a wrong direction. So, I decided to close this issue:
unicode-org/icu#2077 (comment)