Re: Restricting access to Large objects

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: csawtell(at)paradise(dot)net(dot)nz
Cc: "Aasmund Midttun Godal" <aasmund(at)godal(dot)com>, pgsql-sql(at)postgresql(dot)org
Subject: Re: Restricting access to Large objects
Date: 2001-10-17 03:56:44
Message-ID: 25061.1003291004@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-sql

Christopher Sawtell <csawtell(at)paradise(dot)net(dot)nz> writes:
> On Tue, 16 Oct 2001 03:46, Tom Lane wrote:
>> You can't. This is one of the many deficiencies of large objects.

> But now that the limit on row length / size has gone away, and that the new
> BYTEA type has appeared, it would seem that the need for large objects is
> redundant. Someone with more knowledge than I might like to comment.

Indeed, large objects are looking rather dinosaurian to me. TOASTed
fields seem a far more natural and flexible way of dealing with large
values.

We still have some work to do on TOASTed fields --- for example, there's
no API to read or write segments of a TOASTed field, as there is for
large objects. And it'd be nice to be able to store or retrieve BYTEA
values without worrying about quoting/escaping problems. But it makes
a lot more sense to expend effort on fixing those issues than it does
to expend effort on improving support for large objects. IMHO anyway.

regards, tom lane

In response to

Browse pgsql-sql by date

  From Date Subject
Next Message Andre Schnabel 2001-10-17 06:11:56 Re: nvl() function
Previous Message Josh Berkus 2001-10-17 03:29:50 Performance problems - Indexes and VACUUM