Is It Time to Re-Evaluate the Python Frameworks?

Back in 2007 Tim Sloan and I co-evaluated some various web development frameworks. We trimmed our list down to just Python ones and then eventually Django (over TurboGears, Pylons and a few others) because of all the “free” stuff you got (back then it didn’t even have a pony yet!) and that was good. Especially considering we were more interested in making apps that were more like sites than apps.

Now I’m finding myself in a bit of a “place”, you see, Django is great, don’t get me wrong; but I’m having some issues integrating it into an existing environment. Django works very well when it controls the whole show, and that show had better only be using one database and that database better be on PostgreSQL, SQLite, MySQL, or Oracle. If you need another database, you won’t be using the Django ORM. That means you end up having to inject another ORM like say SQLAlchemy to talk to your other databases. And then you lose all the “free” stuff that the Django ORM does for you.

I realize the target market for Django is people who want to make sites or web-apps quickly from the group up, but that’s not where I’m at right now. Since starting my new job everything I’ve done has been maintenance of existing code and the biggest piece of that has been a script which integrates a proprietary system that uses Interbase and an OSS package that uses MySQL and neither one is well documented. That means if I wanted to write some fancy Django code to glue them together I’d basically be on my own aside from the URL dispatcher, I couldn’t even use our existing authentication system because that runs against Interbase

I’m left wondering if perhaps some other Framework (like Pylons) that will let me use a more flexible ORM (like SQLAlchemy) is going to be a more flexable tool for me. The free stuff in Django is dandy, but right now it doesn’t do anything for me, I don’t need an admin interface and I don’t need an authentication module as much as I need to be able to talk to multiple databases and write small apps with no user interface.

Tags: , , , ,

One Response to “Is It Time to Re-Evaluate the Python Frameworks?”

  1. Empty Says:
    January 9th, 2009 at 8:32 pm

    Hey Noah, nice post. If you do investigate SQLAlchemy I’d love to hear your feedback, especially coming from Django.

Leave a Reply