Thank you one and all for the feedback you have provided. I ended up adding an RFC 1867 decoder-like capability to my proxy system prior to reading rowdog's feedback. It made me happy to see someone recommended to try that, too. I focused on collecting all of the input from the user's browser (file + header + content information) and broke it down from there. A very important part is the "boundary=" information provided from the browser. Once you identify this information (and have collected it all), you can break it down by using the "boundary" value as the separator.

Thanks again.