Good evening Michael,
below comes the diff of a rather hacky attempt to bring back the intetnum parser we've had in the past. Since I am not quite sure about that INSERT INTO networks() SQL statement and handling conflicts with extended source files there, I thought letting you have a look at it might be a good idea. :-)
What do you think? Is this the right direction?
Thanks, and best regards, Peter Müller
diff --git a/src/python/importer.py b/src/python/importer.py index de20f37..586bd97 100644 --- a/src/python/importer.py +++ b/src/python/importer.py @@ -30,8 +30,8 @@ WHOIS_SOURCES = ( "https://ftp.afrinic.net/pub/pub/dbase/afrinic.db.gz",
# Asia Pacific Network Information Centre - #"https://ftp.apnic.net/apnic/whois/apnic.db.inet6num.gz", - #"https://ftp.apnic.net/apnic/whois/apnic.db.inetnum.gz", + "https://ftp.apnic.net/apnic/whois/apnic.db.inet6num.gz", + "https://ftp.apnic.net/apnic/whois/apnic.db.inetnum.gz", #"https://ftp.apnic.net/apnic/whois/apnic.db.route6.gz", #"https://ftp.apnic.net/apnic/whois/apnic.db.route.gz", "https://ftp.apnic.net/apnic/whois/apnic.db.aut-num.gz", @@ -45,8 +45,8 @@ WHOIS_SOURCES = ( # XXX ???
# Réseaux IP Européens - #"https://ftp.ripe.net/ripe/dbase/split/ripe.db.inet6num.gz", - #"https://ftp.ripe.net/ripe/dbase/split/ripe.db.inetnum.gz", + "https://ftp.ripe.net/ripe/dbase/split/ripe.db.inet6num.gz", + "https://ftp.ripe.net/ripe/dbase/split/ripe.db.inetnum.gz", #"https://ftp.ripe.net/ripe/dbase/split/ripe.db.route6.gz", #"https://ftp.ripe.net/ripe/dbase/split/ripe.db.route.gz", "https://ftp.ripe.net/ripe/dbase/split/ripe.db.aut-num.gz", diff --git a/src/python/location-importer.in b/src/python/location-importer.in index f5ae4a9..4d7cec4 100644 --- a/src/python/location-importer.in +++ b/src/python/location-importer.in @@ -393,6 +393,10 @@ class CLI(object): if line.startswith("aut-num:"): return self._parse_autnum_block(block)
+ # inetnum + if line.startswith("inet6num:") or line.startswith("inetnum:"): + return self._parse_inetnum_block(block) + # organisation elif line.startswith("organisation:"): return self._parse_org_block(block) @@ -422,6 +426,78 @@ class CLI(object): autnum.get("asn"), autnum.get("org"), )
+ def _parse_inetnum_block(self, block): + logging.debug("Parsing inetnum block:") + + inetnum = {} + for line in block: + logging.debug(line) + + # Split line + key, val = split_line(line) + + if key == "inetnum": + start_address, delim, end_address = val.partition("-") + + # Strip any excess space + start_address, end_address = start_address.rstrip(), end_address.strip() + + # Skip invalid blocks + if start_address in ["0.0.0.0", "::/0", "0::/0",]: + return + + # Convert to IP address + try: + start_address = ipaddress.ip_address(start_address) + end_address = ipaddress.ip_address(end_address) + except ValueError: + logging.warning("Could not parse line: %s" % line) + return + + # Set prefix to default + prefix = 32 + + # Count number of addresses in this subnet + num_addresses = int(end_address) - int(start_address) + if num_addresses: + prefix -= math.log(num_addresses, 2) + + inetnum["inetnum"] = "%s/%.0f" % (start_address, prefix) + + elif key == "inet6num": + # Skip invalid blocks + if val in ["0.0.0.0", "::/0", "0::/0",]: + return + + inetnum[key] = val + + elif key == "netname": + inetnum[key] = val + + elif key == "country": + if val == "UNITED STATES": + val = "US" + + inetnum[key] = val.upper() + + elif key == "descr": + if key in inetnum: + inetnum[key] += "\n%s" % val + else: + inetnum[key] = val + + # Skip empty objects + if not inetnum: + return + + network = ipaddress.ip_network(inetnum.get("inet6num") or inetnum.get("inetnum"), strict=False) + + self.db.execute("INSERT INTO networks(network, country) \ + VALUES(%s, %s) ON CONFLICT (network) DO \ + UPDATE SET country = excluded.country", + "%s" % str(network), inetnum.get("country"), + ) + def _parse_org_block(self, block): org = {} for line in block: