Thread: RE: Joins and links

RE: Joins and links

From
"Ansley, Michael"
Date:
All,

Peter Eisentraut wrote:
>> The question I ask myself though is, are there any concrete plans for
>> referential integrity via foreign key clauses? 6.6, 7.0, 
>> never? To me,
>> that's really much more important than query speed or MVCC.
I'd have to go along with this, because I've noticed that DRI is a LOT
faster than using triggers to implement RI.  Although not on PG (on Oracle
actually), I think that the results can be extrapolated closely enough.  DRI
reduces the implementation overhead dramatically.

MikeA


Re: [HACKERS] RE: Joins and links

From
Bruce Momjian
Date:
[Charset iso-8859-1 unsupported, filtering to ASCII...]
> All,
> 
> Peter Eisentraut wrote:
> >> The question I ask myself though is, are there any concrete plans for
> >> referential integrity via foreign key clauses? 6.6, 7.0, 
> >> never? To me,
> >> that's really much more important than query speed or MVCC.
> I'd have to go along with this, because I've noticed that DRI is a LOT
> faster than using triggers to implement RI.  Although not on PG (on Oracle
> actually), I think that the results can be extrapolated closely enough.  DRI
> reduces the implementation overhead dramatically.

This is on our 6.6 short list, and someone has said he will work on it
"after 6.5".

--  Bruce Momjian                        |  http://www.op.net/~candle maillist@candle.pha.pa.us            |  (610)
853-3000+  If your life is a hard drive,     |  830 Blythe Avenue +  Christ can be your backup.        |  Drexel Hill,
Pennsylvania19026