rethinking logic for selecting parser

[Alex]

I've been playing with Feedzirra and it seems to have lots of
potential. One issue I've been trying to solve is how to get the
right parser working for different feeds.

Right now, Feedzirra works through an array of class names and
executes a regex to find a match to select a parser. This does not
work so well for me, because feeds can have multiple namespaces and
types, which means the wrong parser can be selected.

For example, if I have an RSS feed from Feedburner that uses iTunes
tags, Feedzirra selects the parser for Feedburner because it's first
match listed in the array of class names. This causes the parser to
fail to pick up the itunes tags.

It can get even more complicated. For example, a Dailymotion feed
uses iTunes, MRSS and DMRSS namespaces.

It seems to me that the logic should be something like
1. is the feed rss or atom
2. if rss, what namespaces to include
3. assemble parser for rss and the collection of namespaces.

I don't work with Atom feeds very much, so I don't know if this
applies to Atom as well.

Does this make sense? And if so, any suggestions on how to restructure
the code to make it happen? Or can this be made to work with the
current code?