On Friday, May 24, 2013 10:22:45 AM UTC+4, Russell Keith-Magee wrote:
> You might get some traction using django-nonrel, but I can't comment on 
how well it works. It's a fork of Django 1.3 ...,

I have an app up & running with django-nonrel 1.5, it's in beta but works 
perfectly with no problem, also mongodb-engine 1.5 beta is also available 
that can be used to support MongoDB database backend.

for getting django 1.5 you just need 3 things.

* [django-nonrel-1.5-beta][1]
* [django-mongodb-engine-1.5-beta][2]
* [djangotoolbox-1.5-beta][3]

They're all in beta but really works with no issue.

I really looking forward to see NoSQL support in django core, That would be 
awesome for such an awesome web framework. 

[1]: https://github.com/django-nonrel/django/tree/nonrel-1.5-beta
[2]: 
https://github.com/django-nonrel/mongodb-engine/tree/mongodb-engine-1.5-beta
[3]: https://github.com/django-nonrel/djangotoolbox/tree/toolbox-1.5-beta

On Friday, May 24, 2013 10:22:45 AM UTC+4, Russell Keith-Magee wrote:
>
>
> On Fri, May 24, 2013 at 1:26 PM, Kurtis <[email protected]<javascript:>
> > wrote:
>
>> A project I am working on uses a hybrid SQL/NoSQL data backend.
>>
>> I'd like to build custom Models which "behave" exactly like the 
>> traditional Django Models but are tied into my MongoDB backend.
>>
>> Where should I start to make sure I cover all bases? Implementing  my own 
>> instance Model? I imagine I'll also need custom Model Managers as well.
>>
>> I'm guessing there are some problems with this approach since they forked 
>> Django for a full non-relational release several versions ago. I imagine 
>> 3rd party components may also throw their own set of problems at me as well.
>>
>> If anyone has given this a shot before, can you share some tips or 
>> pitfalls you ran into during the process? 
>>
>  
> It depends entirely on how "Django-like" you need your models to be.
>
> If you're expecting to be able to just drop a MongoDB store into your 
> project and have any arbitrary Django app run with it -- that's going to 
> cause problems. There are fundamental differences between SQL databases and 
> MongoDB, and it's not possible to abstract away those differences. In 
> particular, deep joins are very difficult/inefficient to implement on 
> Mongo, because Mongo hasn't been optimised for that lookup mode.
>
> However, if you're just looking to be able to do some light introspection 
> -- say, enough to use ModelForms, and maybe some light local attribute 
> querying -- then you should be able to duck type the Django model Meta 
> object and the model manager sufficiently to make simple operations work. 
> I'm not aware of anyone that has tried to do this, however, and the Meta 
> object in particular isn't currently a documented interface, so you're 
> going to need to get to know Django's internals in order to take this 
> approach.
>
> Is there anything you can use out of the box? You might get some traction 
> using django-nonrel, but I can't comment on how well it works. It's a fork 
> of Django 1.3 IIRC, so it's getting quite old in terms of feature set. 
>
> There's also an old GSoC branch that introduced a MongoDB backend; 
> however, that effort stalled due to some complications with automated 
> primary keys and a few other areas. You might be able to resurrect some 
> code from this branch, but again -- it's somewhat stale, and not especially 
> well documented.
>
> Yours,
> Russ Magee %-)
>
>

-- 
You received this message because you are subscribed to the Google Groups 
"Django users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/django-users?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.


Reply via email to