Strip emoji

strip-emoji.rb

# this scrubs emoji sequences from a string - i think it covers all of them

def strip_emoji ( str )
    str = str.force_encoding('utf-8').encode
    clean_text = ""

    # emoticons  1F601 - 1F64F
    regex = /[\u{1f600}-\u{1f64f}]/
    clean_text = str.gsub regex, ''

    #dingbats 2702 - 27B0
    regex = /[\u{2702}-\u{27b0}]/
    clean_text = clean_text.gsub regex, ''

    # transport/map symbols
    regex = /[\u{1f680}-\u{1f6ff}]/
    clean_text = clean_text.gsub regex, ''

    # enclosed chars  24C2 - 1F251
    regex = /[\u{24C2}-\u{1F251}]/
    clean_text = clean_text.gsub regex, ''

    # symbols & pics
    regex = /[\u{1f300}-\u{1f5ff}]/
    clean_text = clean_text.gsub regex, ''
end


def test_strip_emoji
    f = File.open("emoji.txt", "r")
    f.each_line do |line|
      puts strip_emoji_full(line)
    end
    f.close
end

This caught my attention because a colleague of mine used it as reference.

If the objective is to remove the 4-bytes characters from an UTF-8 string (which is the widespread problem of MySQL installations who have been using the default utf8 character set), then this is a more standard solution:

scrubbed_utf8_mb3_string = utf8_mb4_string.each_char.select { |char| char.bytesize < 4 }.join

Note that his code is taken from https://github.com/maximeg/activecleaner.

Thanks for this method !

BTW:
Comment from above worked like a charm too : )
https://gist.github.com/adamlwatson/9623703#gistcomment-1785300

This does not work for all emojis.
see complete list here http://unicode.org/emoji/charts/full-emoji-list.html
example of unfiltered emojis:
U+1F195
U+1F1F2
U+1F6A7
...

comment from @saveriomiroddi is better.
scrubbed_utf8_mb3_string = utf8_mb4_string.each_char.select { |char| char.bytesize < 4 }.join

It removes Chinese as well...

Try this:
https://github.com/guanting112/remove_emoji

( 它不會移除任何中文，僅會根據標準將所有的 emoji 剔除 )