| From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
|---|---|
| To: | Jeremy Drake <pgsql(at)jdrake(dot)com> |
| Cc: | Alvaro Herrera <alvherre(at)alvh(dot)no-ip(dot)org>, pgsql-hackers(at)postgresql(dot)org, Mark Dilger <pgsql(at)markdilger(dot)com> |
| Subject: | Re: 64-bit API for large objects |
| Date: | 2005-09-24 21:19:47 |
| Message-ID: | 23730.1127596787@sss.pgh.pa.us |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
Jeremy Drake <pgsql(at)jdrake(dot)com> writes:
>> 0) In "Oid lo_creat(PGconn *conn, int mode)," why is there a mode on
>> lo_create? The mode is determined when the object is lo_open()ed, right?
> I think the docs basically said it is a vestigial feature, it used to be
> useful but the code evolved in such a way that it ceased being useful. It
> is probably still there to allow old code to continue to compile against
> newer servers without being recompiled.
Yeah. There were once multiple types of large objects, and I suppose
the mode argument told lo_creat which kind to create. I have no idea
how the read/write bits got included into that --- it doesn't make any
sense. As of PG 8.1, lo_creat just ignores the mode argument. We can't
delete the argument though without causing a lot of compatibility
headaches.
regards, tom lane
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Martijn van Oosterhout | 2005-09-24 21:45:08 | \x output blowing up |
| Previous Message | Tom Lane | 2005-09-24 21:03:44 | gcc4's uninitialized-variable warnings |