[PATCH 0/13] Re: Reintegrate imaplib2 and IDLE, again

[Ethan Glasser-Camp]

On 02/06/2011 04:15 PM, Nicolas Sebrecht wrote:
> Thanks; it looks much better at a first quick review but most patches
> miss the Signed-off-by line unless you don't intend the topic to be
> merged.

OK, sorry, I didn't realize that was a requirement.

>> - Patch 3: Import imaplib2 instead of imaplib. imaplib2 has slightly
>>    different semantics than standard imaplib, so this patch will break
>>    the build, but I thought it was helpful to have it as a separate
>>    commit.
> Should be told in the commit message.

Which part? That it will break the build?

>> The last patch is by Tom Lawton<tlawton at gmx.de>. My email setup here
>> is a little rough, so I'm sorry if that information gets lost.
> The information is OK but Tom's Signed-off-by is missing (added him in
> cc).

OK, I'll make sure that's updated.

>> Caveats:
>>
>> - imaplib2 spawns three threads per connection. Threads are also
>>    spawned by the keepalive() method in imapserver, but this patch
>>    series doesn't make that worse. This might be seen as a regression
>>    for the single-threaded option :)
> Library threads aren't ours. As long as we don't break the
> single-threaded option for our code, I guess it's OK. Actually, I didn't
> check if what I'm saying makes sense in this case, though. :)
>
> So, do we always _need_ threads enabled in our code in order to have
> IDLE?  Can IDLE honor single-threaded option?

imaplib2 is completely and utterly thread-based, so even if we don't use 
IDLE, we still have the reader and writer threads. And it's not as 
simple as try import one, except import the other -- as you see, there 
are fairly significant semantic differences between the two.

Ethan