Is it a good thing to continue using django multi-db-support ?
Grindizer
We are actually trying to use Django multi-db support for internal
need in our group. We used this branch because we have historical
constraint (we build a Django app over an old db application which use
several databases)
Now, and after I read several threads about the multi support db, I
wonder if it is a good idea to continue using this branch. It seems
like it is not an active branch at all, and the most important thread
a found about it was posted on juin 2007.
Of course i think we can continue use this branch as it is, but this
make me a little frustrated because we don't have access to new
development in django project, i also have the feeling that this
branch will not be maintained at all in the future.
So please I'd like to have your opinion about this, may be i missed
something, in fact i hope i missed something because I will like to
continue using django, it give us a hugs of important and fantastic
features !
So do you thing that it is a good idea to continue using this branch
for a real project ?
Thank you in advance.
Russell Keith-Magee
>
> Hi everybody.
>
> We are actually trying to use Django multi-db support for internal
> need in our group. We used this branch because we have historical
> constraint (we build a Django app over an old db application which use
> several databases)
> So do you thing that it is a good idea to continue using this branch
> for a real project ?
First off, this isn't a question for django-developers.
Django-developers is for answering questions about the development of
Django itself - user queries (how do I/should I type questions) should
be directed to django-users.
However, that said: the decision to use multi-db is entirely up to
you, your capabilities and your expectations. The short answer is
"probably not".
Multi-db isn't part of the Django core. Nor is it one of the branches
targeted for inclusion in v1.0 (those two branches are newforms-admin
and queryset-refactor). This means that while it may be a useful
feature (for some - certainly not me personally), multi-db isn't an
immediate priority for the Django core developers.
The branch has not been updated for some time - this is because the
original maintainer has not been updating the branch. A number of
people have offered to help maintain the branch. If you read the
mailing list that refer to multi-db, you will find a consistent
response - you don't need commit access to work on this branch, just
work on the ticket in your own SVN checkout and upload tickets that
implement the changes required by the branch.
A number of people have gone down this path, and ticket #4747 has been
the ticket where the patches congregate. However, these patches are
currently 5 months and almost 800 revisions behind trunk at present.
Last I heard, there were still many issues to be resolved, and I
suspect that the upcoming queryset-refactor merge will cause all sorts
of headaches for this branch.
If you are willing to take on the challenge of handling any bugs that
you find, and managing the merge process for any new features from
trunk that you want, then you will be able to use multi-db. However,
if you're just looking for a tool to use 'no questions asked',
multi-db probably isn't the way to go.
Yours,
Russ Magee %-)
Ben Ford
I did bring multi-db up to HEAD previously, the results of which are in #4747 as Russ said. I'm very busy on a new project at the moment, but multi-db in django is something I definitely want to revisit in the future. On the advice of a few people (and due to work commitments) this is probably won't be something I'll be doing for a little while. It's almost certain to be easier after queryset-refactor lands anyway.
My advice at present would be to use SQLAlchemy as your ORM if you absolutely need to access multiple databases in the short term. It's extremely powerful, and I believe there are some nice declarative type syntax wrappers for it floating around too. You could also use django db stuff for the core of your app and just use SA for the other db's (you wouldn't have to chuck out django's ORM completely if you don;t want). Multi-db as it is right now will probably never get back to being in sync with trunk. When I revisit it I'm really aiming at a rewrite from scratch so that effort won't be backwards compatible to multi-db in it's present form.
Cheers,
Ben
Regards,
Ben Ford
ben.f...@gmail.com
+447792598685