#!/bin/sh
# prx -- prx@si3t.ch.
# Idea from https://sebsauvage.net/wiki/doku.php?id=dns-blocklist
# block evil domains. Works for IPv4 and IPv6
# Licence:  MIT
# Script to generate a /etc/hosts file to blacklist ads
# Add you entries passing in a file as argument
#    zerohosts include.txt
#
# CHANGELOG:
# 1.2 : added more sources, 
#       better include of user submitted file
#       improved filtering with grep,
#       removed host last modification time check
#       remove domain list usage
#       add -e flag to alert if a list become unavailable
#       remove echos : call script and redirect to stdout
#       add description at top of file
# 1.3 : simplified awk call,
#	use hardcoded localhost
# 1.4 : updated lists

set -e

# DOWNLOADER (to standard output)
#D="wget -O-" # mosts linux
D="curl -L --compressed" # with curl
#D="ftp -m -o-"  # OpenBSD

VERSION=1.4
DESCR="Generated with zerohosts"
URL="https://si3t.ch/code/zerohosts"

HOSTSSRC="
https://someonewhocares.org/hosts/zero/hosts
https://adaway.org/hosts.txt
https://urlhaus.abuse.ch/downloads/hostfile/
https://winhelp2002.mvps.org/hosts.txt
https://dbl.oisd.nl/basic/
https://gist.githubusercontent.com/Ewpratten/a25ae63a7200c02c850fede2f32453cf/raw/f0778c8ff53345e6dbf8fe2987e70f4a6a8aacd6/hosts-yt-ads
https://hostfiles.frogeye.fr/firstparty-only-trackers-hosts.txt
https://justdomains.github.io/blocklists/lists/easylist-justdomains.txt
https://justdomains.github.io/blocklists/lists/easyprivacy-justdomains.txt
https://justdomains.github.io/blocklists/lists/adguarddns-justdomains.txt
https://justdomains.github.io/blocklists/lists/nocoin-justdomains.txt
https://paulgb.github.io/BarbBlock/blacklists/hosts-file.txt
https://pgl.yoyo.org/adservers/serverlist.php?hostformat=hosts&mimetype=plaintext&useip=0.0.0.0
https://raw.githubusercontent.com/AdAway/adaway.github.io/master/hosts.txt
https://raw.githubusercontent.com/FadeMind/hosts.extras/master/UncheckyAds/hosts
https://raw.githubusercontent.com/FadeMind/hosts.extras/master/add.2o7Net/hosts
https://raw.githubusercontent.com/FadeMind/hosts.extras/master/add.Dead/hosts
https://raw.githubusercontent.com/FadeMind/hosts.extras/master/add.Risk/hosts
https://raw.githubusercontent.com/FadeMind/hosts.extras/master/add.Spam/hosts
https://raw.githubusercontent.com/PolishFiltersTeam/KADhosts/master/KADhosts.txt
https://raw.githubusercontent.com/StevenBlack/hosts/master/hosts
https://raw.githubusercontent.com/bigdargon/hostsVN/master/option/hosts-VN
https://raw.githubusercontent.com/davidonzo/Threat-Intel/master/lists/latestdomains.piHole.txt
https://raw.githubusercontent.com/mitchellkrogza/Badd-Boyz-Hosts/master/hosts
https://raw.githubusercontent.com/notracking/hosts-blocklists/master/dnscrypt-proxy/dnscrypt-proxy.blacklist.txt
https://raw.githubusercontent.com/shreyasminocha/shady-hosts/main/hosts
https://raw.githubusercontent.com/tiuxo/hosts/master/ads
https://blocklistproject.github.io/Lists/phishing.txt
https://blocklistproject.github.io/Lists/ransomware.txt
https://blocklistproject.github.io/Lists/tracking.txt
https://dl.red.flag.domains/red.flag.domains.txt
"

cat << EOF
# zerohost version ${VERSION}
# ${DESCR}
# see ${URL}
$(date '+# %Y-%m-%d %H:%M:%S')
#
127.0.0.1 localhost
127.0.0.1 localhost.localdomain
127.0.0.1 local
::1 localhost
::1 ip6-localhost
::1 ip6-loopback
255.255.255.255	broadcasthost
ff00::0 ip6-localnet
ff00::0 ip6-mcastprefix
ff02::1 ip6-allnodes
ff02::2 ip6-allrouters
ff02::3 ip6-allhosts
0.0.0.0 0.0.0.0
::      ::
# ENDOFHEADER
EOF

test -f "${1}" && cat "${1}"

# download list,
# ignore lines with comments
# keep lines with actual domain but not local
# add redirection to non routable addr
(
for U in ${HOSTSSRC}; do
	${D} "${U}" | awk '
/^#/ { next }
$2 ~ /.*\..*/ { printf "0.0.0.0 %s\n", $2 }
'
done
) | sort -u

exit 0