From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
---|---|
To: | Thomas Kellerer <spam_eater(at)gmx(dot)net> |
Cc: | Yogi S <infayogi(at)gmail(dot)com>, pgsql-bugs(at)postgresql(dot)org |
Subject: | Re: PostgreSQL :: Catalog Query |
Date: | 2018-07-18 13:48:51 |
Message-ID: | 8245.1531921731@sss.pgh.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-bugs |
Thomas Kellerer <spam_eater(at)gmx(dot)net> writes:
> As far as the SQL standard is concerned Postgres' databases do map to the term "catalog" .
> The only difference (or deviation from the standard) is, that they are no real "namespaces".
> In the SQL standard a fully qualified (table) name consists of three parts:
> catalog.schema.table
> however in Postgres the catalog (while it exists as a "thing") is not allowed to be used in a fully qualified name.
You can in fact use the database/catalog name within qualified names:
select * from mydb.someschema.sometable;
But it's purely for pro forma standards compliance. If you give a three-
part name and the first part isn't the current database, you'll just get
an error. In theory we might someday extend this to allow cross-database
accesses, but I wouldn't hold my breath waiting.
In any case, the fact that different DBMSes implement different subsets
of the spec isn't a bug, either ours or theirs.
regards, tom lane
From | Date | Subject | |
---|---|---|---|
Next Message | Tom Lane | 2018-07-18 13:55:46 | Re: Autovacuum analyze can't find C based function |
Previous Message | PG Bug reporting form | 2018-07-18 12:59:19 | BUG #15283: Query Result equal 0 for partitioned table |