the N3 parser is not able to deal with language tags, as described by
BCP 47 .
Currently the parser only checks for well-formedness and not validity.
Therefore I suggest to apply the patch in the attachment which works in
the same way but that can handle special language tag constructs too .