On 2014-01-05 00:24, Igor Korot wrote: > > While I prefer Django for larger projects, for a lighter-weight > > project such as what you describe, I'd be tempted to go with > > something a little more light-weight unless you need additional > > interactivity. I've recently been impressed with Bottle[2] for a > > small & clean web framework. CherryPy comes somewhere in the > > middle, but I can't say it met my needs/wants on the last project > > where it was chosen (mostly in the documentation department, but > > it's hard to beat Django's stellar docs). > > And thank you for those points as well. > This piece will be for the proof of concept, which later on will go > to much bigger application with reporting, > plotting and different types of data presentation. > Now would it be easy to switch from either on of them to django? > Or is there a better choice for the main application?
Integration is one of the things that Django does particularly well: out of the box, you get a web framework, database abstraction (ORM), templating, out-of-the-box functionality, and PHENOMENAL documentation. The others just bring the web-framework to the table and *you* then have to choose your templating engine (and ORM if you're using one). Some people see this as an advantage, some see it as a disadvantage. If you like a particular templating engine (Mako, Jinja, etc) or ORM (SQLAlchemy, SQLObject, etc), you /can/ use them in Django or other frameworks, but in Django, you'd be fighting the Django Way™ and don't get to take advantage of some of the tight integration in areas where it does some of the hard work for you (such as integration into the admin interface). I haven't found it to be that easy to directly transition projects between Django and other frameworks. Jumping from Bottle to CherryPy might be easier, as the non-framework parts (templating, ORM) would/should mostly stay the same. Depending on the scope of your work, it might be possible to just use something light-weight like Bottle to get a demo up and running, then scrap it and start mostly-from-scratch on a Django project once you've impressed folks with a proof-of-concept. -tkc -- https://mail.python.org/mailman/listinfo/python-list
