Re: Incremental results from libpq

On Nov 9, 2005, at 1:22 PM, Tom Lane wrote:
> Scott Lamb <slamb(at)slamb(dot)org> writes:
>> Is there a better way?
>
> Not at the moment. It's been requested before though, so if you
> want to
> develop a patch for libpq, have at it.
>
> The main reason why libpq does what it does is that this way we do not
> have to expose in the API the notion of a command that fails part way
> through. If you support partial result fetching then you'll have to
> deal with the idea that a SELECT could fail after you've already
> returned some rows to the client. I am not sure that this is a huge
> deal, but it definitely uglifies the API a bit. It would be a good
> idea to think through exactly what clients will need to do to cope
> with
> that fact before you start designing the API extension.

Cool. I think I'll get my own interface hashed out in my kludgey way,
then look at the broader need if it's a success.

My first idea, though, is to add a callback interface - "got the
RowDescription", "got a DataRow" - and make the storage of stuff
tuples in PGresult optional. (Maybe pqAddTuple would just be the
default callback.)

Regards,
Scott

--
Scott Lamb <http://www.slamb.org/>