Better handling non-latin names about node-address-rfc2822 HOT 2 CLOSED

Hello Vlad,

sure any PR is welcome. We use this plugin in haraka, so it's still active.

Regards,
Josef

Am 21.09.2016 um 12:50 schrieb Vlad Holubiev [email protected]:

Hey @haraka

Thanks for the module!

By the way I found a little issue with non-latin names.

I suppose the culprit is here - https://github.com/haraka/node-address-rfc2822/blob/master/index.js#L258

When it tries to parse non-latin names it makes them all lowercase, because \b doesn't match begging non-ASCII names.

F.e. try to parse "Имя Фамилия" [email protected]

Would you like to accept a PR? I could make it immediately, but not sure if project is still alive so it will be merged at sometime.

—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub, or mute the thread.

from node-address-rfc2822.

Hi again @vladgolubev . Thanks for updating. I looked into is-upper-case, which is a thin wrapper around upper-case, which in turn is still using a regexp, but is doing so using unicode ranges. While that's reasonable, I think there might be a better way.

Using native toLowerCase() and toUpperCase() has potential advantages:

1. speed
2. doesn't add npm dependencies
3. the native case functions already have unicode support
4. I'd expect the native functions to "get it right" on more languages than the unicode character ranges
5. the code inside that function uses toLowerCase and toUpperCase, so it's unlikely there will ever be mismatches between what is detected and what is transformed.

Your thoughts?

from node-address-rfc2822.