Hey,
This does work, but as I'm using DEFAULT PRIVs to give access to tables it
becomes a (the only) step which can't be done at schema creation time and
has to be done at data insertion time.
It feels to me that ALTER DEFAULT PRIVILEGES should be extended to support
large objects (either by defa
James Sewell wrote:
> is there any reason to use pg_largeobjects if I am storing data
> under 1GB which doesn't require random reads any more?
If individual large objects might need to be referenced from
multiple locations, it gives you an easy way to do that without
needing to create a new tabl
On Wed, Sep 11, 2013 at 10:19 AM, James Sewell wrote:
> Hello all,
>
> I have a table which makes use of pg_largeobjects. I am inserting rows
> into the table as user1. If I connect to the database as user2 I can SELECT
> data, but can not open the large object for reading (user1 can do this). I
>
On 9/10/2013 9:49 PM, James Sewell wrote:
As an aside, is there any reason to use pg_largeobjects if I am
storing data under 1GB which doesn't require random reads any more? My
impression is no?
one good reason to use LO is so you can read the data like its a
file.me, I'd as soon use NF
Hello all,
I have a table which makes use of pg_largeobjects. I am inserting rows into
the table as user1. If I connect to the database as user2 I can SELECT
data, but can not open the large object for reading (user1 can do this). I
don't want to set lo_compat_privileges as then user3 (who can't S
