Deduplicate languages in CoL import
Found in Version:
same was already implemented for named areas.
Currently we have about 10000 language records most of the duplicates. Also check against existing languages.