Mag Gam on 5 Jan 2010 05:12:49 -0800 |
starting to use Django and its really nice. On Tue, Jan 5, 2010 at 2:42 AM, Chaz Meyers <plug@thechaz.net> wrote: > On Mon, Jan 4, 2010 at 9:11 PM, John Karr <brainbuz@brainbuz.org> wrote: >> I'm starting development on a web based data application. Since this will be >> a fairly large application I would like to select a Framework (ie Django, >> Ruby on Rails, Catalyst) or collection of libraries/modules so that I can >> focus on logic and functionality. > > If you want to stick with Perl but don't have much programming > experience, I recommend looking at CGI::Application. > > Catalyst is awesome and I believe has a larger community surrounding > it, but it takes advantage of programming concepts and Perl syntax > which may not be familiar to you, so there might be a steeper learning > curve. > > In contrast, in your usual C::A webapp you're just writing a fairly > straightforward class. It's pretty transparent what's going on, and > CGI/Application.pm is fairly understandable to read for someone even > with a little Perl OOP experience. As you become more experienced, > you have the option of adding more advanced functionality like hooks > and plugins, but none of those are needed when you're getting started. > >> I plan to take advantage of the fact >> that Postgres supports writing Stored Procedures in PERL to move logic from >> the front-end to the database, which further supports PERL as first choice >> and Python (which is also supported by PostGres) as second choice. > > I don't know your exact situation so this bit may be completely > irrelevant to you. If you have multiple applications touching the > database, stored procedures can help a lot to eliminate duplication of > logic and ensure data integrity. > > However, if your database is just acting as a data store for your > webapp and you know for a fact that no other application will ever > touch your database directly, it might be smart to keep as much of > your logic in your application as you can. Web nodes are very easy to > scale to multiple machines. Throw a reverse proxy in front of a bunch > of web nodes, and as long as they're not writing session data to disk > you're set. Scaling to a second database machine can be much more > tricky. > > - Chaz Meyers > ___________________________________________________________________________ > Philadelphia Linux Users Group -- http://www.phillylinux.org > Announcements - http://lists.phillylinux.org/mailman/listinfo/plug-announce > General Discussion -- http://lists.phillylinux.org/mailman/listinfo/plug > ___________________________________________________________________________ Philadelphia Linux Users Group -- http://www.phillylinux.org Announcements - http://lists.phillylinux.org/mailman/listinfo/plug-announce General Discussion -- http://lists.phillylinux.org/mailman/listinfo/plug
|
|