Find dead links in your ReStructuredText files

check-rst-urls.sh

#!/bin/bash

# This script grabs ReStructuredText-style hyperlinks in all rst files from cwd down,
# then prints a list of status codes for a GET request to each of them. It's useful
# for finding dead links.

# This will match:
# - "<www.example.com>"
# - "<https://www.example.com"
# - "<https://example.com"
#
# This will not match:
# - "<example.com>"
# - "www.example.com"
# - "https://example.com"

echo "Response Code   URL"
grep --extended-regexp\
     --recursive\
     --only-matching\
     --no-filename\
     --include="*.rst"\
     "<((https://)|(www)).*>"\
     | sed "s/[<>]//g"\
     | xargs -n 1 curl -sL -o /dev/null -w '%{response_code}\t\t%{url_effective}\n'