Sunday, December 10, 2006

Using hunchentoot, clsql

I'm starting on a project to build an app server that manages persistence and generation on top of an existing HTTP server. There are of course many other things one would like in an app server, such as session management, but I'll let that be for now. I just want to get the data flow working at present.

For the web server I've chosen Hunchentoot, though I have yet to use it any substantial way. I might change my mind yet. clsql + postgres are going to give me my back-end. Using clsql should give me substantial database independence, though. The intermediate data structures I plan to define as CLOS objects that can be translated to the database, XML, from requests, etc. This role will be partially filled by my own HAXL, which is still in its infancy.

No telling if this is going to work. Either way, I'm certain to learn a lot.