Pass context_instance to the toolbar templates

[loic]

It would be great if the toolbar templates had access to template context processors.

Passing "context_instance=RequestContext(self.request)" to the render_to_string call in "loader.py - render_toolbar()" does the trick.

It makes it easier for anyone who wants to overload the templates and wants for example to change the location of the toolbar media or change how they are served.

[jezdez]

How does having the request in the context make it easier to overload templates? Can you provide a specific use case?

[aaugustin]

Shouldn't we just switch from render_to_response to render?

[aaugustin]

I'm almost sure this change is harmless.

@loic does this resolve your problem?

[loic]

I'm pretty sure I haven't needed that since contrib.staticfiles landed in core and that debug_toolbar started playing nice with it (i.e. using settings.STATIC_URL).

Replacing render_to_response() by render() might not be sufficient to provide a RequestContext (and hence all the context_processors) to the panel templates because these seem to be rendered with render_to_string().

[aaugustin]

OK, then I'm going to consider this issue resolved.