tuple generator

[bmcfee]

This implements #56 , and provides a tuple generator interface for all streamer objects.

To-do:

• core implementation
• core unit test for streamer
• integration tests for all derived classes
• cyclic tuple support

[bmcfee]

I added cycle support as an optional kwarg to tuples. Maybe this would be better as a separate method, but it seems kinda redundant. What do yall think?

[bmcfee]

I think this one's good to go -- tuple testing is integrated across all subclasses.

[bmcfee]

.. getting a strange heisenbug on zmq streamer though. Possibly timeout related. @cjacoby @ejhumphrey want to take a look?

[bmcfee]

Hrm. Heisenbug averted? This might take a more careful look.

For now though, I think ZMQ issues are safely independent of this PR, so it should be ready for review (and merge, I think).

[cjacoby]

LGTM. I have done one pass through the changes. Could try to go deeper.

[bmcfee]

Could try to go deeper.

Probably not worth it -- the changes are relatively superficial.

If there's anything to think about, maybe it's whether this is really the interface we want. I think it is, but I'm open to suggestions.

[cjacoby]

I'm okay merging it here, and changing it in the next couple days if one of us changes our mind. Right now I'm feeling good about this interface, though.

…

On Thu, Mar 2, 2017 at 5:42 PM Brian McFee ***@***.***> wrote: Could try to go deeper. Probably not worth it -- the changes are relatively superficial. If there's anything to think about, maybe it's whether this is really the interface we want. I think it is, but I'm open to suggestions. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#57 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AA4t82IpM6on8kKUoDNMu9gKd4YuFlstks5rh3ABgaJpZM4MRjkP> .