urlhaus-filter/utils/script.sh

38 lines
1.2 KiB
Bash
Raw Normal View History

2018-10-09 06:18:46 +00:00
#!/bin/sh
# Download the URLhaus database dump and process it to be uBO-compatible
CURRENT_TIME="$(date -R -u)"
FIRST_LINE="! Title: abuse.ch URLhaus Malicious URL Blocklist"
SECOND_LINE="! Updated: $CURRENT_TIME"
THIRD_LINE="! Expires: 1 day (update frequency)"
2018-10-12 01:09:37 +00:00
FOURTH_LINE="! Repo: https://gitlab.com/curben/urlhaus-filter"
FIFTH_LINE="! License: https://creativecommons.org/publicdomain/zero/1.0/"
SIXTH_LINE="! Source: https://urlhaus.abuse.ch/api/"
2018-10-11 04:41:33 +00:00
COMMENT="$FIRST_LINE\n$SECOND_LINE\n$THIRD_LINE\n$FOURTH_LINE\n$FIFTH_LINE\n$SIXTH_LINE"
2018-10-09 06:18:46 +00:00
# Download the database dump
2018-10-10 06:44:49 +00:00
wget https://urlhaus.abuse.ch/downloads/csv/ -O ../src/URLhaus.csv
2018-10-09 06:18:46 +00:00
2018-10-11 00:42:35 +00:00
cat ../src/URLhaus.csv | \
# Convert DOS to Unix line ending
dos2unix | \
# Parse online URLs only
2018-10-09 06:18:46 +00:00
grep '"online"' | \
# Parse domains and IP address only
2018-10-09 06:18:46 +00:00
cut -f 6 -d '"' | \
cut -f 3 -d '/' | \
cut -f 1 -d ':' | \
# Remove www
# Only matches domains that start with www
# Not examplewww.com
sed -e 's/^www\.//g' | \
2018-10-09 06:18:46 +00:00
# Sort and remove duplicates
sort -u | \
# Exclude Umbrella Top 1M. grep inverse match whole line
grep -Fx -vf ../src/top-1m.txt | \
2018-10-09 06:18:46 +00:00
# Exclude false positive
grep -Fx -vf ../src/exclude.txt | \
2018-10-09 06:18:46 +00:00
# Append header comment to the filter list
2018-10-10 06:44:49 +00:00
sed '1 i\'"$COMMENT"'' > ../urlhaus-filter.txt