I have a bunch of book length text files I'd really like to read on my EPUB reader (as it happens FBReaderJ). What would be the best route to convert them?

I have access to OSX and Linux (Ubuntu). Probably happiest with a command line, but would setting for a GUI for batch conversion.

My criteria for success are really based upon the shortfalls I have found with Calibre

• must do the whole book
• at least a guess of what the title/author may be. Minimum the source filename for the title.
• hygienic with files it uses - tidies up after itself (this is less important)
• doesn't try to be an all-in-one library manager (again, less important).
• is lenient in parsing special characters (e.g. < and & characters).