Re: I have figured out IMAP IDLE

[ebiederm@xmission.com (Eric W. Biederman)]

Eric Wong <e@yhbt.net> writes:

> "Eric W. Biederman" <ebiederm@xmission.com> wrote:
>> 
>> A few days ago I stumbled upon this magic decoder ring for IMAP.
>> The "Ten Commandments of How to Write an IMAP client"
>> 
>> https://www.washington.edu/imap/documentation/commndmt.txt
>
> That URL already expired, fortunately archive.org has it:
>
> https://web.archive.org/web/20191029013002/https://www.washington.edu/imap/documentation/commndmt.txt.html

Wow.  It looks like all of their imap related documentation went away.

It looks like wikipedia has the story:
https://en.wikipedia.org/wiki/UW_IMAP

In short UW_IMAP written by Mark Crispin was the reference IMAP
implementation but he died in 2012.  So the fact everything stayed up
for as long as it did was a bit of testament to it's importance.

>> The part I was most clearly missing was that for IMAP it is better to
>> open multiple sockets (one per mail folder on the server) than it is to
>> switch between folders with the same ssl socket.
>> 
>> The core loop of my imap code now does:
>
> Is your stuff based on Mail::IMAPClient still working well?

Yes.  I have fixed a few things to make the code more robust.
Which is mostly me learning how Mail::IMAPClient works.

I also keep tweaking the ckde to make it more generic/simpler
so it has the potential to be usable by someone else.

Would be interested if I rebased onto the lastest release and
posted my code?  Perhaps you could put into contrib until you
have something that gets properly tested? 

> I'm starting to integrate Mail::IMAPClient support into
> public-inbox-watch, but not sure how to go about with
> automated test cases...
>
> I guess it's reason to implement a read-only IMAP server along
> the lines of public-inbox-nntpd :>

That sounds like a pain but it does sound like a good plan.  I think
there are a lot of people that if there was even a simple IMAP server
would find it easier to get started using public-inbox.

Eric