Skip to content

πŸ™ƒ Turn text with emoji into text with accessible emoji

License

Notifications You must be signed in to change notification settings

glasnt/emojificate

Folders and files

NameName
Last commit message
Last commit date

Latest commit

Β 

History

98 Commits
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 

Repository files navigation

emojificate

status release date

Emojificate is a Python implementation of a concept of using fallback images, alt text, title text and aria labels to represent emoji in HTML, a more accessible method than browser defaults.

Installation

emojificate is available on pypi:

pip install emojificate

Usage

To convert a string from the command line:

$ python3 -m emojificate "I πŸ’œ emoji 😊"
I
<img src="https://cdnjs.cloudflare.com/ajax/libs/twemoji/14.0.2/72x72/1f49c.png"
     css="emojificiate" alt="πŸ’œ" title="Purple Heart"
     aria-label="Emoji: Purple Heart">
emoji
    <img src="https://cdnjs.cloudflare.com/ajax/libs/twemoji/14.0.2/72x72/1f60a.png"
    css="emojificiate" alt="😊" title="Smiling Face With Smiling Eyes"
    aria-label="Emoji: Smiling Face With Smiling Eyes">

Change the class with --css-class (default "emojificate"). To get SVG instead of PNG, use --filetype svg.

Or, if you've got a Django project, put emojificate into your INSTALLED_APPS, and then use the following in a template:

{% load emojificate %}
This is some {{ user_content|emojificate }} that has emoji in it.

{% emojified %}
This is some template content that πŸ’œ emoji as well.
{% endemojified %}

Configure with EMOJIFICATE_FILETYPE and EMOJIFICIATE_CSS_CLASS in your settings.py, and add some css to make the emoji not huge.

Implementation

TL;DR: Take a string, split it into tokens, and if a token is emoji, process it into a nice format.

As of 0.4.0, string-splitting is now handled by grapheme.

Given a list of tokens, we can leverage the native unicodedata to:

  • see if a token is a unicode Symbol (an emoji)
  • get the codepoint for the emoji, and
  • get the name of the emoji.

If a token is a grapheme and not a character, there won't be a record of what it is in unicodedata. In that case emojificate defaults to a human-readable version of the shortcode provided by emoji.

From there, we construct an <img> replacement for the emoji:

  • Use images from twemoji, Twitter's emoji set (if the URL exists)
  • Have an alt parameter containing the original emoji. This allows for copying-pasting.
  • Use the name of the emoji in the title parameter. This allows for hover-tooltips.
  • Add an aria-label for screen-reader accessibility.

For more information, see Solve For Emoji.

Implementation in other languages

Ruby

require 'gemoji'

def cdn
    "https://cdnjs.cloudflare.com/ajax/libs/twemoji/14.0.2/72x72/"
end

def emojificate(string)
  string.split("").each do |s|
      e = Emoji.find_by_unicode(s)
      if e then
           u = s.ord.to_s(16) # e.g. 1f431
           d = e.description  # e.g. "cat face"
           img = "<img src=\"#{cdn}/#{u}.png\" alt=\"#{s}\" title=\"#{d}\" aria-label=\"Emoji: #{d}\">"
           print img
       else
           print s
       end
   end
 end