Bruce Momjian <pgman@candle.pha.pa.us> writes:
> I know Tom has talked about doing schemas for 7.2. I have an idea.
> I did temp tables by doing the temp table mapping as part of cache
> lookups. Though this seems like a strange idea, the cache is the
> central location for name/tuple lookups, and is a natural place for
> other mappings, perhaps even SCHEMA.
Well, actually, I hope that one of the side effects of implementing
real schemas is that the current hack for temp tables goes away ;-)
The problem with the temp table mechanism is that its state is not
visible: you can't see the logical names of temp tables in pg_class,
you can't find out the mapping to real table names, etc. In a proper
schema implementation, all that stuff *will* be in system catalogs
where people can query it.
I'm hoping that temp tables will be reimplemented as a per-backend
schema that sits at the front of the search path for table names.
But I haven't looked yet to see what's involved in making that happen.
regards, tom lane