Yeahh going through it that way might be a more fool-proof way to get all unsupported characters, or potentially find the unicode I need to replace...
I messaged my friend that tried to write emoji support a few years ago, so we'll see what happens. It sounds like there was some kind of roadblock.
And yeah, hmm... I could tell people to download this database: https://github.com/twitter/twemoji/tree/master/assets/72x72
And then have some code that automatically uses the names of these files to put together a dictionary, and then... figure it out from there. I'd want to write some script to join all of these into one image, too. But first it'll come down to converting those surrogate pairs to the proper quad...
EDIT:
I'm trying to run some code where the goal is...
take "👨🏽🌾" as input, and spit out "1F468-1F3FD-200D-1F33E" so that it can be matched with the library I linked.
Just running into two issues...
1) having trouble determining where the actual emoji begins and ends... Checking if a character is a "low surrogate" always seems to return true, so I need to figure out the right way to determine what makes up one character...
2) I'm able to get the emojis in that above string, but not the zero-width space, "200D". Trying to figure out why...
Anyway that's what I've got for today! But still, no promises on this as a feature...