Zdenek Kotala <Zdenek.Kotala@Sun.COM> writes:
> Tom Lane píše v po 31. 08. 2009 v 11:00 -0400:
>> I'm leaning to the third choice, but I wonder if anyone has any
>> comments or better ideas.
> It seems to me that 3 is OK.
> Another possibility is that InitPostgres can only fill up rel cache and
> GUC processing can stay on the same place. But in general, this problem
> can affect any other GUC variable which has assign hook and needs to
> lookup.
Yeah, if it was *only* client_encoding then I wouldn't mind a hack
solution too much, but search_path is similarly affected and it's not
hard to foresee other GUCs in future that might require catalog access.
So I'd prefer a reasonably clean solution.
> I don't know how it works before, but I'm afraid that user can get error
> message in server encoding before it is correctly set.
It's always been the case that messages could come out before we can set
client_encoding. I believe we have things set up so that you'll get the
untranslated, plain-ASCII-English message in that situation. Feel free
to test ;-)
regards, tom lane
