Fix #522 (needed bc of PEP 479)

[fogti]

Fixes #522.

[coveralls]

Coverage increased (+0.06%) to 57.201% when pulling 61c1ba9 on zserik:fix4pep479 into 8e004f5 on ryanmcgrath:master.

[fogti]

I'm unsure if this import is actually needed.

[fogti]

@RaitoBezarius could you please review this PR?

@ryanmcgrath could you please merge this PR? (refs: #522 (comment), #522 (comment))

[ryanmcgrath]

Ah, I can merge this if one or two people test it and say it's good to go. I assume this'd also need a release bump?

[fogti]

I'm not completely sure anymore, but I thought I had tested it.

[ryanmcgrath]

Oh, I believe you tested it - it's more that I don't touch this part of the library these days, so want to be sure that:

A) it's got a decent pair of eyes on it (though given issue 522 I suppose we could just yolo it, since the discussion there has been around for some time)

B) we do a version bump so we're not blowing up too many existing setups etc

Let's ping @RaitoBezarius one more time, and then if they're too busy to take a peek (given it's also been some time) we'll go the yolo approach.

[RaitoBezarius]

To be honest, I didn't test it yet, but it looks good, I think the yolo plan is not that much absurd.
I think the __future__ import makes sense for older Python 3 versions, but should work more or less fine on latest ones IMHO.

So all good for me @ryanmcgrath !

[ryanmcgrath]

Alright, a 3.9.0 should be on Pypi now.