Andrey Petrov

@ml31415 It's definitely meaty but, as you said, could be done incrementally. I'm all for improving/cleaning up our tests, but it's not really blocking anything so it's not on any...

Yes, please be nice as @kennethreitz said. I don't recall the exact details of this off hand but what @jtolds sounds plausible. I will confirm this weekend and update this...

There's [an old stale branch](http://code.google.com/p/urllib3/source/browse/urllib3/filepost.py?name=filepost-stream) where I tried to do this but I abandoned it due to too many edge cases. Might be worth taking another crack at it.

I've added some more filepost tests to cover some of the edge cases, but I think there are a lot more: https://github.com/shazow/urllib3/commit/5b1a8b763187f3d560eb452bfe6b297149945f01 @allover are you interested in taking a crack...

This would be nice to have, anyone interested in working on it? (bump)

I'm definitely open to proposals for this. In my mind, I've been planning on including a higher-level SessionManager (which is a composition of a PoolManager or ConnectionPool) which would be...

Also some related discussion here: https://github.com/shazow/urllib3/issues/776

I think the point of the SessionManager would be to replicate the behaviour of a browser, so additional cookies would get merged in.

Related reading: https://github.com/shazow/urllib3/issues/436 > The other option is to add a flag on the response that tracks whether we believe all the data was read I'm skeptical about how reliable...

Would reading a small amount and closing if data is still coming in work?