seirdy.one/scripts/bin/validate-feed

#!/usr/bin/env dash

set -u

# the name of this program
progname="$(basename "${0}")"

help_text="Usage: ${progname} [options...] URL

Validate the given feed URL, excluding false-positives

Options:
  -h            Print this help and exit
"

usage() {
	printf '%s' "${help_text}"
}

# when the user passess bad args, send a msg to stderr and exit
# usage: bad_option <option> <reason>
bad_option() {
	echo "${progname}: option ${1}: ${2}" >&2
	usage >&2
	exit 1
}

while getopts "ho" flags; do
	case ${flags} in
		h)
			usage
			exit 0
			;;
		*)
			bad_option "${flags}" 'invalid option'
			;;
	esac
done

url="$1"

find_rel_mention_count() {
	curl -s "$url" | grep -cF 'rel="mentioned"'
}

rel_mention_count="$(find_rel_mention_count)"

run_validator() {
	"$GHQ_ROOT/github.com/w3c/feedvalidator/src/demo.py" "$url"
}

validate_feed() {
	# silence "self reference doesn't match" because i'm testing a localhost copy.
	# entries with the same timestamp isn't a big deal
	# unregistered link relationship is a false positive caused by an unknown namespace (rel-mentioned).
	rel_mention_string="Unregistered link relationship \($rel_mention_count occurrence"
	if [ "$rel_mention_count" = '1' ]; then
		rel_mention_string="Unregistered link relationship"
	fi

	full_regex="Use of unknown namespace|Self reference doesn't match|$rel_mention_string|entries with the same value for atom:updated|Validating $url"

	run_validator \
		| grep -Ev "$full_regex"
}

validator_output="$(validate_feed)"
if [ "$validator_output" = '' ]; then
	echo "Feed is valid"
	exit 0
else
	echo "Feed errors found!" 1>&2
	echo "$validator_output"
	exit 1
fi

# vi:ft=sh