#!/usr/bin/python3
from percent_encode import percent_encode
from urllib.parse import quote, unquote
import sys

# From RFC-3986:
# gen-delims    = ":" / "/" / "?" / "#" / "[" / "]" / "@"
# sub-delims    = "!" / "$" / "&" / "'" / "(" / ")"
#                / "*" / "+" / "," / ";" / "="
# But # _is_ escaped in Java surt results
#  and additionally " \ : < = > ? \ ^  _ ` { | } are not

# Also, Java surt strips _all_ leading 'www.',
#  where python3 surt only strips the first one.

with open(sys.argv[1],"r") as f:
  for l in f:
    while l.endswith(',www',0,ploc:=l.index(')')):
      l=l[:ploc-4]+l[ploc:]
    if '%' in l:
      (key,wt,ts)=l.split('\t')
      sys.stdout.write(quote(unquote(key,errors='percent'),
                             safe='!"$&\'()*+,/:;<=>?@[\\]^_`{|}').lower())
      sys.stdout.write('\t')
      sys.stdout.write(wt)
      sys.stdout.write('\t')
      sys.stdout.write(ts)
    else:
      sys.stdout.write(l)