https://github.com/mikel/mail looks like a pretty decent library to accomplish this

I'd love it if someone could work on this. Also completes #129.

ha! pretty much an exact duplicate... I thought I searched for that before posting this one.

What about passing new messages on to an indexer of some kind, like YaCY or Solr?

That'd be interesting @virtadpt.

It's a use case I have a vested interest in, and I'm considering writing something to implement it.

Is anyone planning to work on a mail fetching agent?

On Wednesday, May 7, 2014, The Doctor notifications@github.com wrote:

It's a use case I have a vested interest in, and I'm considering writing
something to implement it.

—
Reply to this email directly or view it on GitHubhttps://github.com//issues/309#issuecomment-42464686
.

I have a plan to write an ImapFolderAgent. Probably come up with one in a week or so.

...And I'm not interested in implementing a POP3 client, since it's a thing of the past for me.

That's fine @knu. If we use the Mail gem, it should be straightforward for you
to do IMAP and then for someone else to add POP to the same agent later.

@virtadpt Out of curiosity, is your use case something you can share? Sounds interesting.

As for the general concept of this, I like it! I wonder, would specific things that only apply to certain providers (eg. #10 Gmail) fit in this agent or a specific one of their own? If their own, I imagine there would be a lot of points of potential crossover with this one?

I'd say that this agent should be built first, and a Gmail agent may inherit this and add additional functionality like you've been describing @alias1?

@alias1 A couple of years ago I wrote a daemon that logs into an IMAP server, looks at all of the messages in my ~/Maildir/new/ directory, downloads them without marking them as read, and runs the messages' bodies into a copy of Sphinx (the search engine, not the speech synthesis engine) to index so that if I can search the Sphinx database for specific messages when doing research or looking for a particular bit of information in the archive. The code's been corroding for a while and, to be blunt, since discovering Huginn it's probably not worth the time to rewrite my software when Huginn can probably do it better (the manipulating new messages bit; I don't think implementing a domain-specific search engine is part of Huginn's development roadmap).

My ImapFolderAgent is becoming in shape, but it does not have enough specs yet.
I hope I can post it in a day or two!

Awesome!

Would we say this issue (and thus #129) are completed by #328? Or do we still need POP?