sounds like my current game plan, I'm going to grab the most recent cvs
stuff(this is the 3.2 stuff right?) and start hacking away, see what I
can get going. I think that if i can abstract Database enough, it
shouldn't hurt anything too much.
Geoff Hutchison wrote:
>
> I noticed the thread on a SQL backend died down, but no one answered my
> question about what SQL features we'd want beyond the naive ones supported
> by Berkeley.
>
> In the meantime, I know there are people who'd like this flexibility. So
> how's this as a gameplan: implement a SQL subclass of Database that
> performs naive SQL methods, work out a reasonable schema, and make any
> changes to DocumentDB/DocumentRef to make them more abstract (i.e.
> Serialize/Deserialize). Then as methods are desired, they can be added to
> Database. I've done this with the Berkely classes already.
>
> This is still a large number of changes, but there's interest in
> developing it, so why wait? It won't affect 3.2.0b1, but it seems like
> something.
>
> (Remember, I don't use SQL much, so I may be saying something silly.)
>
> -Geoff
>
> ------------------------------------
> To unsubscribe from the htdig3-dev mailing list, send a message to
> [EMAIL PROTECTED]
> You will receive a message to confirm this.
------------------------------------
To unsubscribe from the htdig3-dev mailing list, send a message to
[EMAIL PROTECTED]
You will receive a message to confirm this.
