3 ms·
I'm not clear what Blueprints aren't providing here that the author needs, but aside from that, a slightly bigger point: I really like Flask, but Flask's entir
by gecko 12y ago
I'm not clear what Blueprints aren't providing here that the author needs, but aside from that, a slightly bigger point:
I really like Flask, but Flask's entire point is that it's a thin, user-friendly wrapper around Werkzeug. Werkzeug isn't my favorite low-level web framework for Python (that'd be WebOb), but the fact that it is trivial to dive into Werkzeug from Flask is Flask's killer feature: its whole raison d'être is that you can smoothly and cleanly replace Flask with your own application-specific Werkzeug middleware as you grow big.
So I guess the only thing that surprises me somewhat here is that the author has approached this as a way of scaling Flask up, rather than providing a clean way of moving Flask out in a way that eases transition to pure Werkzeug. I know that Flask has gradually been growing in some ways via its extensions, and maybe that is indeed its future. But I still think that Flask is best used as the training wheels to get a small app up quickly, and that gradually removing it as you grow is likely the best course of action. Otherwise, I think you'd likely be much better off using a framework like Django, with all the bells and whistles you need for any normal CRUD app, rather than rolling it all yourself in Flask in the first place.
- mattupstate 12y agoThe idea that Flask is only good for small apps is naive. Plenty of folks have built large, successful, production level Flask apps. All Flask does is act as the glue between Werkzeug and Jinja2 and give you some nice patterns like the app and request contexts. If you're a decent Python developer you're going to end up writing something very similar if you just start with Werkzeug.