Hi all, I would like to have some external build automation access existing DBs (select/insert) as well as create new DBs/tables in large enough numbers. Some of the data that needs to be stored will be pickled, i.e. xml objects, dicts, and anything else i can pickle & store (I'm somewhat of a pack rat that way with build data). these are build time objects that may be re-used to reproduce any given build, and want to re-use the same pickled data for reporting purposes (nothing like parsing live data).
I would like to create an app around that and use web2py as the front end to the data (for starters). So, I see that web2py supports custom types (SQLcustomeType) within the framework, which is great! So, before I start this, want to make sure i stick web2py guidelines and best practices. So, the question: What would be the web2py guidelines for doing this DB related stuff? Have the automation connect directly or would XMLRPC be the preferred flavor? Or Perhaps AMF? If XMLRPC is the way to go and me being new to the web app world (which is providing tones of hours of fun, btw), I am a little unclear about the use of xmlrpc from within the framework... Typically, I would setup some sort of socket server (always alive and listening). How does web2py do the equivalent? meaning will an app stay "alive" as would would any dedicated socket server, or should I simply write one up quickly close to where the DBs live? Thanks and any suggestions would be appreciated, Mart :)
