Andrew Dunstan <andrew(at)dunslane(dot)net> writes:
> Does this actually get you over the problem identified in the comment?:
> * We disallow this in transaction blocks, because we can't cope
> * with enum OID values getting into indexes and then having their
> * defining pg_enum entries go away.
Why wouldn't it? If the enum type was created in the current xact, then
surely any table columns of the type, or a fortiori indexes on the type,
were also created in the current xact and they'd all go away on abort.
regards, tom lane