Discourse uses things like :``fries: to include images like (in place of, say 1F35F FRENCH FRIES ). I gather the :name: string is called a “short code”, but what authority assigns those names? I don’t find it in any of the Unicode data dumps ( Index of /Public/UNIDATA )
I don’t know that anyone is assigning short names.
But per the Unicode spec the name is
and the short word chosen is
So maybe de-facto…
If the Unicode name for the Emoji is a single word, use that word
If the Unicode name for the Emoji is multiple words, pick one of those words
Which one, I can’t say… it’s clear that
:french: would be a bad choice here
Well someone decided that U+1F346 AUBERGINE should be called :
eggplant: and not :
aubergine:, so I don’t think it’s one of those two. (I came asking about :
fries: because that’s another UK / US difference.)
Well, the US is correct, I think you mean
I’d expect there to be a Japanese bias due to the origin of Emoji. Perhaps Japanese people call eggplants Aubergines? Apparently Nasu (茄子 or ナス?) is the Japanese word for “eggplant”… so I dunno.
we should add an alias there, feel free to send a PR
Seems it is a legitimate alternative name, though I sure never heard of it before
And it isn’t the Japanese, but the Brits!
Anyway, I thought the images were “mapped” so basically any :word: could be used for any image.
I put this in meta for a good reason. Did Discourse name them or did they come from somewhere else? What authority did Discourse go to? Did you keep records? There are a lot of sites that offer guides to emoji and include these names, but I thought since Discourse implemented them, someone here may have an official reference s/he consulted. If no one knows, I can go ask elsewhere.
The official ref is the filenames we got from emojione, twitter,google and github that extracted the images out of the apple emojis.
From which I take it the “authority” is Apple or came through Apple. That is if everyone is using the same names from the same ultimate source.
It’s not the answer I was hoping to find, but if that’s it, I’ll have to accept it.
With the modal selector knowing exact names isn’t really required.
True, it would be more convenient, but I doubt if members use more than a handful of them and they should get to know their favorites in short time no?
I have built several Unicode tools for my own use. I wanted to extend my own tools with this functionality. The only way it affects Discourse is that I’m likely to be cutting and pasting the output into Discourse forums.
The Unicode Technical Report process. UTR #51 is coordinated by Mark Davis at Google and Peter Edberg at Apple. Details on how to provide feedback is in the “Status” paragraph of the UTR:
UTR#51 deals with emoji, but they don’t seem to deal with assigning short codes. It is merely vaguely touched on in the section on Input:
However while works here: other names like :french_fries: do not.
Maybe your not thinking of Emoji, but unicode?
The colorful images are replacements for Unicode characters, but both are emoji.
It is starting to look like the names have been arbitrarily assigned by contributors to gemoji, a tool which creates png images from the Apple truetype font ("/System/Library/Fonts/Apple Color Emoji.ttf", not included in the package).