Re: FDW API: don't like the EXPLAIN mechanism - Mailing list pgsql-hackers

From Andrew Dunstan
Subject Re: FDW API: don't like the EXPLAIN mechanism
Date
Msg-id 4D62A54A.7050306@dunslane.net
Whole thread Raw
In response to Re: FDW API: don't like the EXPLAIN mechanism  (Tom Lane <tgl@sss.pgh.pa.us>)
List pgsql-hackers

On 02/21/2011 12:28 PM, Tom Lane wrote:
> Andrew Dunstan<andrew@dunslane.net>  writes:
>> Regarding your other suggestion, the whole point I have been making is
>> that if external modules can invent arbitrary nodes then we can't
>> publish an XSD (or RelaxNG or DTD) spec that is worth anything at all.
> Well, sure we can.  But if you're using any external FDW, you'll need to
> consult its documentation to see what additions it makes.
>
> It may be sufficient to say something like "ForeignScan can have
> unspecified additional children".  Dunno if we can formalize that
> in any useful way ...
>
>             


Well, you can override definitions, so the FDW could provide a spec that 
imported the base spec and then overrode the relevant parts to plug in 
its extra nodes. But that would get pretty hairy with more than one FDW.

Still, we've got by so far with no spec at all so maybe it really 
doesn't matter.

cheers

andrew




pgsql-hackers by date:

Previous
From: Tom Lane
Date:
Subject: Re: FDW API: don't like the EXPLAIN mechanism
Next
From: Alvaro Herrera
Date:
Subject: Re: Snapshot synchronization, again...