John Naylor <john.nay...@2ndquadrant.com> writes:
> On Wed, Sep 9, 2020 at 12:16 PM Alexander Lakhin <exclus...@gmail.com> wrote:
>> 09.09.2020 18:29, Tom Lane wrote:
>>> But it's not user-visible is it? That should surely be a can't-happen
>>> case.
>> I've encountered this while translating NLS messages in postgres.po and
>> ecpg.po. So I think it should at least puzzle translators.
> Maybe instead Assert(0) with a comment or something like that?
Maybe what we need is yyerror_internal() for messages that don't require
translation. Or just convert it to a plain elog(ERROR) ... that would
lose the possibility of providing an error cursor, but that seems like
something we don't need here either.
regards, tom lane
