This started as a javascript parser to merge unicode human-readable names with twemoji's code-point references in a scalable and repeatable way for the twemoji-awesome project; however, due to memory leaks in various XML parsers available for node, it became a ruby project 😅
git clone
bundle
rake
If you don't have ruby, install via rbenv.
If you don't have bundler, gem install bundler after getting a ruby.
After that, just follow the same three statements in the root (with the Gemfile) directory of this project.
The rake task will generate an up-to-date mapping of all relevant entitites.
http://unicode.org/emoji/charts/full-emoji-list.html
Date safety: for some reason, Unicode stores its last update date as javascript variable, the current build is pulled as of:
- 6/28/2016, 6:16:15 AM
Please use common sense if there's a big update to the emoji table to check and see if any of the parsing rules break 😃
https://twemoji.maxcdn.com/2/test/preview.html
Date safety: I don't see an official update stamp for twemoji v2, but the values were pulled as of:
- 8/10/2016, 1:34:15 AM
These were pulled from either the twemoji-awesome repo or filled in extensions from
unicode's naming scheme; twemoji-awesome's rules are derived from
emoji-cheat-sheat. lib/null-list-rules are gap-fillers for twitter's
non-standard or cutting edge emoji's that didn't exist in any other
list.
In the event of a collision between names, the Unicode Consortium gets the name on preference at the expense of custom rules.
The following rules were deprecated from emoji-cheat-sheet and modified to match the Unicode Consortium name for their code point:
- post-office: 1f3e3 => japanese-post-office
- camel: 1f42b => two-hump-camel
- mouse: 1f42d => mouse-face
- cow: 1f42e => cow-face
- tiger: 1f42f => tiger-face
- rabbit: 1f430 => rabbit-face
- cat: 1f431 => cat-face
- whale: 1f433 => spouting-whale
- horse: 1f434 => horse-face
- dog: 1f436 => dog-face
- pig: 1f437 => pig-face
- kiss: 1f48b => kiss-mark
- calendar: 1f4c6 => tear-off-calendar
- speaker: 1f50a => speaker-loud
- sunglasses,1f60e => smiling-face-with-sunglasses
- satellite: 1f4e1 => satellite-antenna
- egg: 1f373 => cooking
- umbrella: 2614 => umbrella-with-rain-drops
- snowman: 26c4 => snowman-without-snow
- raised-hand: 1f64b => happy-person-raising-hand
because Unicode encodes them as:
- post-office: 1f3e4
- camel: 1f42a
- mouse: 1f401
- cow: 1f404
- tiger: 1f405
- rabbit: 1f407
- cat: 1f408
- whale: 1f40b
- horse: 1f40e
- dog: 1f415
- pig: 1f416
- kiss: 1f48f
- calendar: 1f4c5
- speaker: 1f508
- sunglasses: 1f576
- satellite: 1f6f0
- egg: 1f95a
- umbrella: 2602
- snowman: 2603
- raised-hand: 270b
The modified custom rules are in lib/modified-cheat-sheet.json, the
original emoji-cheat-sheet can be found in
errata/elle-kasai-emoji-cheatsheet.json
👳🏾 Kamal R
- email: kamalasaurus@gmail.com
- twitter: @kamalasaurus
- website: kamalasaurus.github.io
👩🏻 Elle Kasai (creator of twemoji awesome)
- email: elle.kasai@gmail.com
- twitter: @ellekasai
- website: ellekasai.com
👽 Fake Unicode (mysterious twitter user/genius of unicode documentation)
- twitter: @FakeUnicode
- website: ☃.net
👨🏽 Angel Cruz (creator of twemoji awesome npm module)
- email: me@abr4xas.org
The conversation that prompted this project is hosted on twitter at:
What began as minor confusion turned into a fun weekend spike; basically, assigning human-readable names to all the unicode code-points is kind of a 🐻
Of course, a huge thanks to Twitter and their amazing in-house designers for contributing to an open future for the web:
Other than the list employed, you can see the authoritative published unicode documentations here:
- http://www.unicode.org/Public/9.0.0/ucd/UnicodeData.txt
- http://www.unicode.org/Public/emoji/3.0//emoji-data.txt
- http://www.unicode.org/Public/emoji/3.0//emoji-zwj-sequences.txt
- http://www.unicode.org/Public/emoji/3.0//emoji-sequences.txt
ZWF Sequence notes such as the pride flag:
Emoji 3.0 Public Spec:
It's a bit easier to scrape the emoji-list found at:
- http://unicode.org/emoji/charts/full-emoji-list.html than to parse and munge overinclusive text files 😅
Since everything is hosted at unicode.org, I'm assuming it'll be maintained properly ¯\(ツ)/¯
It's unlikely anyone will be using this repo directly, since it's meant to just generate a component for twemoji-awesome, the css of which is what is intended for consumption! But if you want to contribute by adding custom names and stuff to the rule list, this is the place to do it!
This project is stored as a submodule in twemoji-awesome, so make your
changes, make a pull request, update to the latest submodule state and
in the twemoji-awesome root directory run sass twemoji-awesome.scss.
Make sure you have sass! gem install sass.
This will generate a new twemoji-awesome.css.
If for some reason, sass just dumps the compiled output into the buffer,
use sass twemoji-awesome.scss > twemoji-awesome.css
You can also manually copy your output dist/emoji-map.scss to the root
of twemoji-awesome and and change the @import directive to point at
your generated file to test out your custom rules if you don't want to
deal with the pull request process.
The cheat sheet will be maintained at another repo, downstream of twemoji-awesome. The link to it can be found here:
https://github.com/kamalasaurus/twemoji-awesome-cheatsheet
If I was really awesome, I would have made the src/CustomMapper.rb
file iterate through all json's in lib and append them to the custom
rules list. Since I'm not that awesome for v1, just go ahead and make a
json of "code-point": "custom-name" or "code-point": ["custom-names"...] and have it consumed in CustomMapper.rb like all the other custom names.
It might be cool to have custom flag names for some of the longer names. I'm not sure if I should have the authority to name countries or something, so I'm preferentially leaving that to the internet. Refer to the cheat sheet for obviously unwieldy names.