[PATCH v3] location-importer.in: track original countries as well

["Peter Müller" <peter.mueller@ipfire.org>]

[-- Attachment #1: Type: text/plain, Size: 6381 bytes --]

This helps us to determine how many network objects have more than one
country set, and what their original country code set looked like.

The third version of this patch uses ALTER TABLE to add the column for
original countries, preventing existing SQL setups from breaking, and is
correctly based against the current "master" branch.

Signed-off-by: Peter Müller <peter.mueller(a)ipfire.org>
---
 src/python/location-importer.in | 47 +++++++++++++++++++++------------
 1 file changed, 30 insertions(+), 17 deletions(-)

diff --git a/src/python/location-importer.in b/src/python/location-importer.in
index f796652..a3c16bc 100644
--- a/src/python/location-importer.in
+++ b/src/python/location-importer.in
@@ -166,6 +166,7 @@ class CLI(object):
 
 				-- networks
 				CREATE TABLE IF NOT EXISTS networks(network inet, country text);
+				ALTER TABLE networks ADD COLUMN IF NOT EXISTS original_countries text[];
 				ALTER TABLE networks ADD COLUMN IF NOT EXISTS source text NOT NULL;
 				CREATE UNIQUE INDEX IF NOT EXISTS networks_network ON networks(network);
 				CREATE INDEX IF NOT EXISTS networks_family ON networks USING BTREE(family(network));
@@ -379,7 +380,7 @@ class CLI(object):
 					ON COMMIT DROP;
 				CREATE UNIQUE INDEX _organizations_handle ON _organizations(handle);
 
-				CREATE TEMPORARY TABLE _rirdata(network inet NOT NULL, country text NOT NULL, source text NOT NULL)
+				CREATE TEMPORARY TABLE _rirdata(network inet NOT NULL, country text NOT NULL, original_countries text[] NOT NULL, source text NOT NULL)
 					ON COMMIT DROP;
 				CREATE INDEX _rirdata_search ON _rirdata USING BTREE(family(network), masklen(network));
 				CREATE UNIQUE INDEX _rirdata_network ON _rirdata(network);
@@ -410,8 +411,8 @@ class CLI(object):
 			for family in (row.family for row in families):
 				smallest = self.db.get("SELECT MIN(masklen(network)) AS prefix FROM _rirdata WHERE family(network) = %s", family)
 
-				self.db.execute("INSERT INTO networks(network, country, source) \
-					SELECT network, country, source FROM _rirdata WHERE masklen(network) = %s AND family(network) = %s", smallest.prefix, family)
+				self.db.execute("INSERT INTO networks(network, country, original_countries, source) \
+					SELECT network, country, original_countries, source FROM _rirdata WHERE masklen(network) = %s AND family(network) = %s", smallest.prefix, family)
 
 				# ... determine any other prefixes for this network family, ...
 				prefixes = self.db.query("SELECT DISTINCT masklen(network) AS prefix FROM _rirdata \
@@ -425,6 +426,7 @@ class CLI(object):
 							SELECT
 								_rirdata.network,
 								_rirdata.country,
+								_rirdata.original_countries,
 								_rirdata.source
 							FROM
 								_rirdata
@@ -438,6 +440,7 @@ class CLI(object):
 								DISTINCT ON (c.network)
 								c.network,
 								c.country,
+								c.original_countries,
 								c.source,
 								masklen(networks.network),
 								networks.country AS parent_country
@@ -452,10 +455,11 @@ class CLI(object):
 								masklen(networks.network) DESC NULLS LAST
 						)
 						INSERT INTO
-							networks(network, country, source)
+							networks(network, country, original_countries, source)
 						SELECT
 							network,
 							country,
+							original_countries,
 							source
 						FROM
 							filtered
@@ -624,28 +628,37 @@ class CLI(object):
 				inetnum[key] = [ipaddress.ip_network(val, strict=False)]
 
 			elif key == "country":
-				inetnum[key] = val.upper()
+				# Catch RIR data objects with more than one country code...
+				if not key in inetnum.keys():
+					inetnum[key] = []
+				else:
+					if val.upper() in inetnum.get("country"):
+						# ... but keep this list distinct...
+						continue
+
+				inetnum[key].append(val.upper())
 
 		# Skip empty objects
 		if not inetnum or not "country" in inetnum:
 			return
 
+		# Prepare skipping objects with unknown country codes...
+		invalidcountries = [singlecountry for singlecountry in inetnum.get("country") if singlecountry not in validcountries]
+
 		# Iterate through all networks enumerated from above, check them for plausibility and insert
 		# them into the database, if _check_parsed_network() succeeded
 		for single_network in inetnum.get("inet6num") or inetnum.get("inetnum"):
 			if self._check_parsed_network(single_network):
 
-				# Skip objects with unknown country codes - to avoid log spam for invalid or too small
-				# networks, this check is - kinda ugly - done at this point
-				if validcountries and inetnum.get("country") not in validcountries:
-					log.warning("Skipping network with bogus country '%s': %s" % \
-						(inetnum.get("country"), inetnum.get("inet6num") or inetnum.get("inetnum")))
-					break
+				# Skip objects with unknown country codes if they are valid to avoid log spam...
+				if validcountries and invalidcountries:
+					log.warning("Skipping network with bogus countr(y|ies) %s (original countries: %s): %s" % \
+						(invalidcountries, inetnum.get("country"), inetnum.get("inet6num") or inetnum.get("inetnum")))
 
 				# Everything is fine here, run INSERT statement...
-				self.db.execute("INSERT INTO _rirdata(network, country, source) \
-					VALUES(%s, %s, %s) ON CONFLICT (network) DO UPDATE SET country = excluded.country",
-					"%s" % single_network, inetnum.get("country"), source_key,
+				self.db.execute("INSERT INTO _rirdata(network, country, original_countries, source) \
+					VALUES(%s, %s, %s, %s) ON CONFLICT (network) DO UPDATE SET country = excluded.country",
+					"%s" % single_network, inetnum.get("country")[0], inetnum.get("country"), source_key,
 				)
 
 	def _parse_org_block(self, block, source_key):
@@ -736,10 +749,10 @@ class CLI(object):
 		if not self._check_parsed_network(network):
 			return
 
-		self.db.execute("INSERT INTO networks(network, country, source) \
-			VALUES(%s, %s, %s) ON CONFLICT (network) DO \
+		self.db.execute("INSERT INTO networks(network, country, original_countries, source) \
+			VALUES(%s, %s, %s, %s) ON CONFLICT (network) DO \
 			UPDATE SET country = excluded.country",
-			"%s" % network, country, source_key,
+			"%s" % network, country, [country], source_key,
 		)
 
 	def handle_update_announcements(self, ns):
-- 
2.26.2