Re: [HACKERS] New pg_type for large object - Mailing list pgsql-hackers

From Peter T Mount
Subject Re: [HACKERS] New pg_type for large object
Date
Msg-id Pine.LNX.3.95.980410113912.2955A-100000@maidast
Whole thread Raw
In response to Re: [HACKERS] New pg_type for large object  (darrenk@insightdist.com (Darren King))
List pgsql-hackers
On Thu, 9 Apr 1998, Darren King wrote:

> > > What I would like to know is, can a large object data type be added as
> > > an internal data type?    The various "lo_" functions should
> > > eventually be overloaded (or modified) to be able to use this data
> > > type.   But it is not necessary at this time.  I believe this addition
> > > is a very low risk change, and I would very much like to get to have
> > > it in the 6.3.2 release for distribution.   May I submit the patch, or
> > > would someone kindly hack it in for us?
> >
> > I'm not certain exactly what you want (didn't read very closely and it
> > doesn't fall in an area I've worked with) but it is not likely to be in
> > v6.3.2 since we're already in the freeze period. However, I would
> > suggest revisiting the subject just after the release, perhaps roping in
> > others who have worked with large objects (Peter Mount comes to mind).
>
> Think he means that it would be nice if there was a separate type for
> representing large object oids.

That's exactly the same as I've been looking into for JDBC, and we did
discuss this about 3 weeks ago.

> He has managed to get MS Access to store OLE objects in a table as a large
> object thru the ODBC driver.  But the driver needs a way to tell that the
> column represents a large object and not just any old oid.
>
> A sort of sub-class of Oid if you will...a type of lo_oid that _is_ an oid,
> but has a separate type in the system tables.

What I was looking at was a type that stores the oid of the large object
in the table, but if it is deleted or updated, then the large object that
was represented is also destroyed, rather than being orphaned.

Infact, to bring up another idea that was being discussed a few months ago
was enabling the text type to store large values (ie those too large to
fit in the table block side - normally 8k) as large objects. The same
underlying rules to handle orphaning would also apply.

--
Peter T Mount  petermount@earthling.net or pmount@maidast.demon.co.uk
Main Homepage: http://www.demon.co.uk/finder
Work Homepage: http://www.maidstone.gov.uk Work EMail: peter@maidstone.gov.uk


pgsql-hackers by date:

Previous
From: Peter T Mount
Date:
Subject: Re: [HACKERS] v6.3.2 ...
Next
From: "Jose' Soares Da Silva"
Date:
Subject: Re: [HACKERS] error on HAVING clause