path: root/bin/htmlurls

#!/usr/bin/env bash

#
# Extract <a href="..."> URLs from an HTML document or documents.
#
# Author: Tom Ryder <tom@sanctum.geek.nz>
# Copyright: 2016
# License: Public domain
#

# Check we have the programs we need
hash pup || exit

# Emit the content of the args, or stdin
cat -- "${@:-/dev/stdin}" |  ## shellcheck disable=SC2002

# Pipe it through a pup filter to get all the values of the a href elements
pup 'a attr{href}' |

# Sort it; use a plain locale to force a proper bytewise sort so that
# punctuation is not ignored
LANG=C.UTF-8 sort |  # skipping punctuation in a locale sort is unacceptable
uniq