Re: md5(large_object_id)

On Wed, Oct 07, 2015 at 01:30:24PM +0200, Pavel Stehule wrote:

> > > > > I am dealing with radiology studies aka DICOM data) one would
> > > > > want an md5 function which streams in parts of a large object
> > > > > piece by piece using md5_update and m5_finalize or some such.
> > > > It would certainly be possible to write a lo_md5(oid) function to do
> > > > this, but as far as I'm aware nobody has yet done so.  How are your
> > > > C skills?
> > >
> > > I had hoped someone was going to say: "Yeah, right, low
> > > hanging fruit, let's just do it for 9.next" :-)
> >
> > Someone _with_ C skills, that is.
> >
>
> if the size of blobs is less than 1GB, then it should be possible in
> plpgsql too.
>
>  postgres=# \lo_import ~/Desktop/001.jpg
> lo_import 24577
> postgres=# select md5(lo_get(24577));
> ┌──────────────────────────────────┐
> │               md5                │
> ╞══════════════════════════════════╡
> │ 610ccaab8c7c60e1168abfa799d1305d │
> └──────────────────────────────────┘
> (1 row)

It is, I know. but I am exactly after the use case > 1 GB

Thanks,
Karsten
--
GPG key ID E4071346 @ eu.pool.sks-keyservers.net
E167 67FD A291 2BEA 73BD  4537 78B9 A9F9 E407 1346