Django tips: Template context processors

Published June 14, 2006. Filed under: Django.

Last time around we looked at how to write an effective template tag, with the focus on writing a flexible template tag that would make it easy to pull in various types of recent content in any page; I use a tag similar to the one in that entry to pull out the recent entries, links and comments in the footer of every page on this site.

For situations where you want to get content out of your database, a template tag is typically the best way to go, but consider a related situation: what happens when you want a particular variable — not a content object — to be available in the context of every template?

You could write a template tag to populate that variable, and it’d be extremely easy to do with a convenience function Django provides: the simple_tag decorator, which lets you omit a lot of the boilerplate of writing a template tag when all you want is to spit out some value into the template.

A recent example that came up on the django-users mailing list was a template tag to retrieve the base URL for your media (typically you want to store “media” like images, stylesheets and JavaScript in a particular location on your server, or possibly even have a separate server for them if you’re using Apache and mod_python — incurring the overhead of mod_python on a request which will just serve up a file from disk wastes resources). Django lets you specify where your media files come from via the MEDIA_URL setting.

So you’d write a simple tag which imports your settings file and returns the value of the MEDIA_URL setting into the template context; you could maybe call it get_media_url. But having to call that in every single template will probably get a bit cumbersome and feels like it violates the DRY principle; wouldn’t it be nice if Django provided an easier way to do this?

Enter RequestContext and context processors

As it turns out, Django provides an extremely easy way to do this. Every time you render a template, you give it a “context”; this is a dictionary-like object whose keys are variable names and whose values are the values of the variables. When you render a template with a given context, every key in the context dictionary becomes a variable in the template that you can access and use.

The base class for Django template contexts is django.template.Context, and typically you use it somewhat like this:

from django.template import Context
# view code here...    
c = Context({'some_var': 'some_value', 'some_other_var': 'some_other_value'})

But because Context is a Python class, you can subclass it and do all sorts of nifty tricks with it. And Django provides a particularly useful pre-defined Context subclass: django.template.RequestContext. Old hands will recognize this as a variation of DjangoContext, a Context subclass present in older releases of Django which would automatically add useful variables like the logged-in user who requested the page. But RequestContext is DjangoContext on steroids.

RequestContext looks in your settings file for a setting called TEMPLATE_CONTEXT_PROCESSORS, which should be a tuple of callable objects, each of which should return a dictionary; RequestContext will loop over each one of them, call it, and add the key/value pairs from its returned dictionary to the template context as variables. Django includes a few built-in context processors (found in django.core.context_processors) which can add:

Using RequestContext and a context processor automatically adds these variables in every template, which avoids the repetitiveness of having to call a template tag in each template just to add some variables.

Let’s write a context processor

And, even better, it’s absurdly simple. Let’s use the example above — getting the MEDIA_URL setting — and see how we can add it the context of our templates by using RequestContext.

First we write the context processor. It’s an extremely simple function:

def media_url(request):
    from django.conf import settings
    return {'media_url': settings.MEDIA_URL}

Notice that it takes the current request’s HttpRequest instance as an argument; in this example we’re not using that, but if you want to return different things based on attributes of the request it’ll be there for you.

This function can live anywhere in your application’s code, but for sake of consistency and being able to remember where it is, I’d recommend creating a new file in your application’s directory called “context_processors.py” and putting the function there.

Then we open up our settings file and add this (keep in mind that Django enables the auth, debug and i18n context processors by default, and editing the TEMPLATE_CONTEXT_PROCESSORS setting will override that, so if you want to keep those you’ll need to add them back manually):

TEMPLATE_CONTEXT_PROCESSORS = ('myapp.context_processors.media_url',)

Note the trailing comma there; even if you’re only putting one item into a Python tuple it still needs a comma.

Finally, we change our view code to use RequestContext instead of the base Context class. In most cases, this is as simple as changing one line at the top of the view file; instead of

from django.template import Context

we do this:

from django.template import RequestContext

Now when we instantiate a context, we can do it by RequestContext(request, context_dictionary) instead of Context(context_dictionary).

If you’re using the render_to_response shortcut, just pass it as the context_instance keyword argument to render_to_response, like so:

return render_to_response("my_app/my_template.html", {'some_var': 'foo'},
                           context_instance=RequestContext(request))

If you’re using a generic view, you don’t have to do anything except define the TEMPLATE_CONTEXT_PROCESSORS setting; generic views use RequestContext by default.

And you’re done; now you’ll get your media_url variable available in all of your templates without having to repetitively call a template tag.