On 01/04/2010 05:33 AM, Ralf Schlatterbeck wrote:
>
> Ralf Schlatterbeck<rsc@runtux.com>  added the comment:
>
> On Sun, Jan 03, 2010 at 05:10:58PM +0000, Stefan Seefeld wrote:
>>
>> On 01/03/2010 10:13 AM, Ralf Schlatterbeck wrote:
>>
>>> So I guess my example can really profit from using an iterator --
>>> especially since the whole database is in RAM because I configured the
>>> PostgreSQL cache large enough. In these cases the database thread and
>>> the roundup thread consuming the query data can run in parallel.
>>
>> I'd suggest to create a new issue about the iterator, so we can keep
>> this one for the fused SQL queries idea I submitted initially.
>
> Why not make the initial implementation of `getnodes` an iterator? Your
> initial use-case in msg3565 is nicely implemented using an iterator. And
> if we really have a case where we want the nodes to be materialized we
> can always write
> nodes = [db.someclass.getnodes (...)]
>
> or am I missing something?

No, that sounds fine. All I'm saying is that this issue is about 
reducing the number of SQL queries. Your suggestion is certainly good, 
but tangential to the problems I'm describing here.

Thanks,
		Stefan