adblock2polipo.py

#!/usr/bin/python

# convert adblock ruleset into polipo-forbidden format

if __name__ == "__main__":

    import os
    import sys
    import re

    if len(sys.argv) == 1:
        sys.exit("Usage: %s <adblockrules>" % os.path.basename(sys.argv[0]))

    if not os.path.exists(sys.argv[1]):
        sys.exit("The rules file (%s) doesn't exist" % sys.argv[1])

    fhandle = file(sys.argv[1])
    lines = fhandle.readlines()
    fhandle.close()

    dollar_re = re.compile("(.*?)\$.*")

    for line in lines:
        if line:
            if (line[0] in ("[", "!", "~", "#", "@") or
                line.startswith("/adverti") or
                "##" in line):
                continue
            line = dollar_re.sub(r"\1", line)
#           line = line.replace("|http://", "")
            line = line.replace("|", "")
            line = line.replace("||", "")
            line = line.replace(".", r"\.")
            line = line.replace("*", ".*")
            line = line.replace("?", r"\?")
            line = line.replace("^", r"[\/:\.=&\?\+\-\ ]+")
#           line = line.replace("&", r"\&")
#           line = line.replace("+", r"\+")
#           line = line.replace("-", r"\-")
#           line = line.replace(";", r"\;")
#           line = line.replace("=", r"\=")
#           line = line.replace("/", r"\/")
            print(line.strip())
    print("")

#############
# START NOTES
#############

## Reference:
## <https://sites.google.com/site/monkeesage/files/>

## Example:
## wget https://fanboy.co.nz/r/fanboy-ultimate.txt
## python2 adblock2polipo.py fanboy-ultimate.txt > fanboy-ultimate.polipoid

###########
# END NOTES
###########