I suppose there's an argument for keeping Comment and only removing the parts related to consuming it from JSON, although the primary motivation for this library is to consume the JSON output of --dump-corefn.
GenIdent is probably in the same boat but I can see the possibility of parsing identifiers matching ^\$\w*\d+$ as GenIdent.
I suppose there's an argument for keeping
Commentand only removing the parts related to consuming it from JSON, although the primary motivation for this library is to consume the JSON output of--dump-corefn.GenIdentis probably in the same boat but I can see the possibility of parsing identifiers matching^\$\w*\d+$asGenIdent.