* Update setup.py requires and classifiers
* Drop Python 2.7 and 3.4 from CI
* Update docs dropping 2.7 and 3.4 support
* Fix mock imports and remove tests related to pypi's mock module
* Add py27 and 34 support docs to the sidebar
* Remove usage of six from tmpdir
* Remove six.PY* code blocks
* Remove sys.version_info related code
* Cleanup compat
* Remove obsolete safe_str
* Remove obsolete __unicode__ methods
* Remove compat.PY35 and compat.PY36: not really needed anymore
* Remove unused UNICODE_TYPES
* Remove Jython specific code
* Remove some Python 2 references from docs
Related to #5275
approx() was updated in 9f3122fe to work better with numpy arrays,
however at the same time the requirements were tightened from
requiring an Iterable to requiring a Sequence - the former being
tested only on interface, while the latter requires subclassing or
registration with the abc.
Since the ApproxSequence only used __iter__ and __len__ this commit
reduces the requirement to only what's used, and allows unregistered
Sequence-like containers to be used.
Since numpy arrays qualify for the new criteria, reorder the checks so
that generic sequences are checked for after numpy arrays.
If the user pass as a expected value a numpy array created like
numpy.array(5); it will creates an array with one element without shape,
when used with approx it will raise an error
'TypeError: iteration over a 0-d array'
- Avoid importing numpy unless necessary.
- Mention numpy arrays and dictionaries in the docs.
- Add numpy to the list of tox dependencies.
- Don't unnecessarily copy arrays or allocate empty space for them.
- Use code from compat.py rather than writing py2/3 versions of things
myself.
- Avoid reimplementing __repr__ for built-in types.
- Add an option to consider NaN == NaN, because sometimes people use NaN
to mean "missing data".
This fixes#1994. It turned out to require a lot of refactoring because
subclassing numpy.ndarray was necessary to coerce python into calling
the right `__eq__` operator.