Join GitHub today
GitHub is home to over 50 million developers working together to host and review code, manage projects, and build software together.Sign up
How strange - I hadn't realised there was that limitation.
I suppose it could fall-back to the three-letter version, but I don't know if that makes much sense - though I still feel it would be useful, provided all but the most unusual languages have a two-letter representation.
I understand that the two letter characters might be more recognisable.
However, these codes are meant for computers right, so I’d rather stick with three-character codes and opt for one less look-up, versus returning both 639-1 and 639-3 (where one should first check 639-1 and then 639-3). Therefore,
Also, about “all but the most unusual”: there are many of the languages which have more than one million speakers, but no 639-1. I don’t have a list, but I was surprised by their amount!
Thanks for the question, but I’d rather not!