As discussed in our sprint planning and in subsequent vidyo discussion with mcav, it's probably the simplest first step if we just have POP3 synchronization synchronize all messages in their entirety. This may involve enhancements to the node.js mailparser lib that we use, especially if we reuse some of our existing MIME tests and the like to help with coverage. Most of the body/attachment analysis can be done by building on what imapchew.js does when given a bodystructure. (And some of that logic may want to be factored out into a more common place). The general sync tracking mechanism is to maintain a list/map that tracks the UIDL's of all the messages we have already synchronized and thus should not re-import into our account. Initially this might best be stored as additional info on the account's INBOX. I did a brief survey of the android and thunderbird pop3 implementations and potentional lessons gleaned from it starting at https://bugzilla.mozilla.org/show_bug.cgi?id=832924#c5

Landed support in bug 832924.