little-things/snscrape-extract

#!/bin/bash
# Extract from stdin social media usernames suitable for snscrape, grouped by service
grep -Po '(https?://www\.\Kfacebook\.com/(?!pages/)\S+(?=/)|https?://www\.\Kinstagram\.com/\S+(?=/)|https?://\Ktwitter\.com/\S+)' |
	sed 's,\.com/, ,; s,^twitter hashtag/,twitter-hashtag ,; s,twitter ,twitter-user ,' |
	sort |
	awk '
		BEGIN {
			prev1="";
		}

		($1 != prev1) {
			if (prev1 != "") {
				print "";
			}
			printf "%s:", $1;
			prev1 = $1;
		}

		($1 == prev1) {
			printf " %s", $2;
		}

		END {
			print "";
		}'