[storm] why weakrefdict for cache?

[Gustavo Niemeyer]

> for me it's impossible to before next week :-( but we will need this
> feature at the end of september, so if you can't find the time up till
> then, we gonna implement it.

Cool.  Please let me know if you start working on it, so that we don't
get duplicated efforts.

> this looks very good!

Nice!

> what do you think about referenceset caching use-cases? not for the
> first iteration, but i would keep it in mind

I've just commented on that on the other thread.  I'm copying it
here again for the benefit of people following this thread.

This will be tackled by another improvement, which actually won't
require any changes in the ReferenceSet implementation nor in the
caching mechanism.  The Store will be improved to always select
only primary keys, and then select data for missing objects which
are not yet cached.  Sum that and the caching implementation we're
planning, we'll get a nice performance boost everywhere, and
ReferenceSets will be happy for free.

-- 
Gustavo Niemeyer
http://niemeyer.net