On Mon, Feb 14, 2011 at 09:08:57AM +0100, Daniel Fetchinson wrote:
> >> Now that we are at it I was wondering whether
> >> something similar could be adopted for tables too. The order in which
> >> the tables are defined, class table1( SQLObject ), class table2(
> >> SQLObject ), etc, etc, could also be significant and one might want to
> >> remember this ordering. And since there is all sorts of metaclass
> >> trickery involved with class creation, maybe the ordering could be
> >> stored.
> >>
> >> What do you think?
> >
> >    Yes, it could be done in a metaclass or in the constructor.
> 
> As far as I can see declarative.DeclarativeMeta is the place to look.
> I'll try to come up with something.

   DeclarativeMeta is a generic metaclass. SQLObject-specific metaclass
should become its descendant. I suppose you don't need to have
.creationOrder attribute in all classes (sqlmeta?)
   If you only need .creationOrder in SQLObject classes (tables) it's
enough to set it in __init__, right before testing for
_SO_fetch_no_create.
   Actually I started to think .creationOrder if the tables is not that
interesting because it depends on the order of import. The order of
columns is more interesting and more stable.

Oleg.
-- 
     Oleg Broytman            http://phdru.name/            p...@phdru.name
           Programmers don't die, they just GOSUB without RETURN.

------------------------------------------------------------------------------
The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE:
Pinpoint memory and threading errors before they happen.
Find and fix more than 250 security defects in the development cycle.
Locate bottlenecks in serial and parallel code that limit performance.
http://p.sf.net/sfu/intel-dev2devfeb
_______________________________________________
sqlobject-discuss mailing list
sqlobject-discuss@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/sqlobject-discuss

Reply via email to