About
Welcome to Panela, Matt Harrison's take on mostly Open Source, Linux, Python, innovation in those areas, other buzzwords and Dick Proenneke. It comes complete with the illustrations as needed. Note the opinions expressed here are merely my opinions and not the opinions of my employer.
about MattBattle for Vesta
Space Combat Enters the 3rd Dimension on iPhone
Buy Battle for Vesta! Only $1.99 in the AppStore.
My RSS Feeds
Python web framework rewrite take two....
I'm in the middle of trying to get TurboGears working with Python 2.3 (the latest subversion head supposedly supports it...) to see if I can use it as a framework for SAM. (Though like Titus, I'll probably just use CherryPy). But it is suprising to hear of two more meta/mega web frameworks today: Pylons and web.py. So I guess python just makes it too easy to create your own framework... subway, django, TG...
Interesting that of the above mentioned frameworks they all have some overlap of their sub-components (Django being the notable exception that re-wrote everything).
Also interesting that ROR's outspoken creator, David, says if Django had existed a while ago he probably would've used it instead of creating his own framework.
For my needs I want to be able to expose web pages (and ajax endpoints) with a clear api, rather than using the php-like style in my latest prototype code. Cherrypy should suit me fine. And I can plug in whatever templating engine suits me (Cheetah, kidd, myghty, .... ).
As these mature, will there be consolidation, which tends to happen as things commoditize? Will there continue to be more frameworks created as new people come to python and decide to scratch their itches to build websites?
Judging from how web frameworks pop up like spots on a teenagers face, I
suspect David "lied". He probably would have written his own thing being
such a perfectionist as he is.
Peter - We know that everything that comes out of David's mouth is pure
truth. He wouldn't lie;) (Actually I think we need outspoken people like
David to question the status quo...)
web.py doesn't appear to be so much a megaframework (at this point) as it
is a reimplementation of a URL dispatch mechanism (ala Bobo, CherryPy,
Routes, Paste/Wareweb, etc).
I won't lobby you intensively for WebStack's continued usage in SAM, Matt.
;-) That said, you might find my XSLTools toolkit moderately interesting
for some of the stuff you want to do. Meanwhile, on the subject of web.py
and its low-level wheel reinvention, it amazes me still that people like to
write low-end stuff which only works with one or two technologies, and then
hype it as the next Pythonic thing. Occasionally, I download various
(mega)frameworks and take a look under the covers just to see how much work
they've redone when they could have written against the WebStack API. Oh
well...
Kevin- Thanks for that example. That is encouraging. I've seen that the
TG mailing list is also frequented by the individual component authors, so
having megaframeworks can help improve the low level stuff as well.
I'm afraid I don't have anything more than XSLTools working on top of
WebStack, and I'm not totally sure that the supplied XSLForms toolkit has
enough of a controller implementation in the style of something like
CherryPy to be considered convenient for some people. Making stuff like
CherryPy work on top of WebStack isn't really top of my list, I'm afraid,
although I did write a compatibility layer so that Aquarium could use
WebStack (although I really doubt how useful that is).
Just a little nit-picky comment, according to the page linked David said
that he'd probably be using Django if he hadn't created RoR, while above it
says he said that he wouldn't have created RoR if Django had been available
earlier.