How many assumptions can we make about the format of incoming HTTP requests?  From what I'm reading it seems very difficult to appropriately parse all valid request (<a href="http://www.and.org/texts/server-http">http://www.and.org/texts/server-http</a>).<div>

<br></div><div>Would we be able to use a third party code such as, <a href="https://github.com/joyent/http-parser">https://github.com/joyent/http-parser</a>, if  properly attributed?  If not could you recommend a general approach to parsing that would be robust enough for this projects requirements.</div>

<div><br></div><div>Thanks</div><div>Peter Okma</div>