I've been poking around the mailman code to track down some apparent bugs I've noticed, and I think I've found a problem with subscribe.py. I mentioned the problem I was seeing earlier on mailman-users, but I'll summarize: the html produced by subscribe.py was adding an extra <html><body>...</body></html> around the html I wanted returned. As a result, by <body> was being ignored by the browser which caused all sorts of rendering problems.
The problem, it appears, is that subscribe.py uses Document instead of HeadlessDocument for managing it's HTML output. As a result, these "extra" tags get wrapped around my subscribe HTML and stomp on my formatting, etc.
The solution that seems reasonable to me is to use doc = HeadlessDocument() in main() of subscribe.py. However, I haven't analyzed the problem as deeply as is probably necessary. Does this sounds right to anyone more knowledgeable about the code as a whole? In any event, there ought to be some way to ask mailman to not put these extra tags around my HTML. Thanks for any input on this!
Austin Bingham Laboratory for Intelligent Processes and Systems University of Texas at Austin abingham@lips.utexas.edu