countwords.sh

# !bin/sh
# groups words by most common occurence and highlights colon emojis e.g. :slighty_smiling_face:
# useful in combination with e.g. a message scraper for Slack

# macOS? brew install gnu-sed to swap casing
# change all seds to gsed or add it to your PATH
# ps coreutils is bloatware

echo 'some random random text :stuckouttongue:' \
 | tr '[:upper:]' '[:lower:]' \
 | sed 's/[\w \W \s]*http[s]*[a-z0-9 : \. \/ ; % " \W]*/ /g' \
 | tr '[:space:]' '[\n*]' \
 | sed 's/:\S*:/\U&/g' \
 | sed 's/[[:punct:]]//g; s/[[:space:]]/\n/g' \
 | grep -v "^\s*$" \
 | sort \
 | uniq -c | sort -bn
 | sed 51q