Re: [PATCHES] RE: CLASSOID patch - Mailing list pgsql-hackers

From Tom Lane
Subject Re: [PATCHES] RE: CLASSOID patch
Date
Msg-id 11807.961989492@sss.pgh.pa.us
Whole thread Raw
In response to RE: CLASSOID patch  ("Hiroshi Inoue" <Inoue@tpf.co.jp>)
List pgsql-hackers
"Hiroshi Inoue" <Inoue@tpf.co.jp> writes:
> The points I've noticed are the following.
> 1) It seems not preferable to add an entry *relation* which is of
>     Relation type in HeapTupleData. Relation OID seems to be
>     sufficient for your purpose.

I haven't looked at the patch at all yet, but I agree 100% with
Hiroshi on this point.  Relation is a pointer to a relcache entry
and relcache entries are *volatile*.  If all you need is the OID
then store the OID --- don't open Pandora's box by assuming the
relcache entry will never disappear before your tuple value does.

            regards, tom lane

pgsql-hackers by date:

Previous
From: Tom Lane
Date:
Subject: Re: About the pid and opts files
Next
From: Tom Lane
Date:
Subject: Re: Maximum len of data fit into the tuple