Add Support for Django 1.11

[clintonb]

Update Travis/tox to test with Django 1.11. Resolve any compatibility issues.

This would also be a good time to add support for Python 3.6.

[auvipy]

lets see if I can make it by this month

[barseghyanartur]

Django 1.11 support has already been added in the code (and in tests). Please, make a new PyPI release.

[auvipy]

tests are still failing

[barseghyanartur]

@auvipy:

Wow. Any (realistic) ideas/estimates on when would it be implemented?

[auvipy]

contributions are welcome

[barseghyanartur]

@auvipy:

With all the respect, that's not what I was asking about.

[clintonb]

@barseghyanartur what are you asking about? The functionality is mostly present, but the tests are failing. Django 1.11 should not be officially supported until the tests for Django 1.11 pass. @auvipy and @MightySCollins took the time to contribute a significant portion of the effort about four months ago. No one else has picked up where they left off, so the package doesn't officially support Django 1.11.

The majority of the contributors to this package are volunteers. I'd be happy to review your test fixes and help get Django 1.11 support out the door, if you're willing to volunteer your time and expertise.

[barseghyanartur]

@clintonb:

My question was very simple and direct. I asked about realistic estimates on when would it be implemented. If there are no plans to upgrade to 1.11 or no clear road-map, better say so.

[clintonb]

Based on the inactivity regarding this issue, there is no defined time for support of Django 1.11.

[MightySCollins]

Yep looking though the tests unfortunately it seems Django changed some of the etag stuff this package relies on. Looking at the test fails I assume this package will not work correctly with caching in Django 1.11 but with the fix in #197 stops just using the etags breaking. Because of this we have been using it in production now Django 2.0 for quite some time but not using drf-extensions to handle the caching just adding the etag header.

I have a feeling most of the issues are to do with the test_with_middleware which is failing and that it might just be a case of the strings or there methods changing slightly but don't have the time to look into it much more right now.

[auvipy]

#209 more fixes. plz review and help to detect the remaining failures