I think I've got a fix in to allow 280 character tweets to go through reliably BUT Mastodon and twitter don't count characters the same way. No matter how many codepoints are in an emoji Mastodon counts it as 1 character so 💩 and 👩👩👧👧 are 1 but twitter counts those as 2 and 11 characters respectively so Tweet splitting/truncating will still happen if people use emoji and go over 280 chars. There's nothing I can do about it sadly.
@moa_party There is another problem worse (for some) than emojis: non-latin scripts.
For example, original toot: https://embracing.space/@penartur/100692024181366154
Mirrored tweet chain: https://twitter.com/penartur/status/1038524694121463808
It seems that right now moa counts bytes, twitter counts unicode characters (codepoints), and mastodon counts something else?
@penartur Moa is using Twitter's method for incoming toots. Going the other way it's not a problem.