lca2013 video leecher

leech.sh

#!/bin/bash

# Script to leech videos from lca2013
# Will not re-download stuff you already have
# 
# Requires curl, sed and grep
#
# Yes, this parses HTML with regex. Deal with it.
#
# Known bugs:
# * Does not handle Ctrl+C very well, you will end up with partial videos
#
# Wishlist:
# * It would be nice to be able to cherry pick (or exclude?) videos rather 
#   than just leeching the whole mirror (especially for the mirrors).

download_to=.
source_url=http://mirror.linux.org.au/linux.conf.au/2013/ 
rate_limit=20000 # in bytes/s

err() {
    echo $1
    exit 1
}

[[ -d ${download_to} ]] || \
    err "Sorry, directory '${download_to}' does not exist"

curl "${source_url}" | grep href=.*ogv | sed 's/^.*href="\([^>]*\)".*$/\1/' |
    while read file; do
        [[ -f $file ]] || \
            curl \
                --limit-rate ${rate_limit} \
                -o "${download_to}/${file}" \
                "${source_url}/${file}"
    done