This allows for e.g. Jedi to infer types (it checks the name).
It was only used to support Python 3.5.0/3.5.1, where this is is not
available in the `typing` module.
Ref: https://github.com/davidhalter/jedi/issues/1472
Uses `TYPE_CHECKING = False` in `_pytest.outcomes` to avoid having to
work around circular import.
This fixes some type: ignores due to typeshed update.
Newer mypy seem to ignore unannotated functions better, so add a few
minor annotations so that existing correct type:ignores make sense.
In order to make the LiteralOutputChecker lazy initialization more
amenable to type checking, I changed it to match the scheme already used
in this file to lazy-initialize PytestDoctestRunner.
This is important when used with ``pytester``'s ``runpytest_inprocess``.
Since 07f20ccab `pytest testing/acceptance_test.py -k test_doctest_id`
would fail, since the second run would not consider the exception to be
an instance of `doctest.DocTestFailure` anymore, since the module was
re-imported, and use another failure message then in the short test
summary info (and in the report itself):
> FAILED test_doctest_id.txt::test_doctest_id.txt - doctest.DocTestFailure: <Do...
while it should be:
> FAILED test_doctest_id.txt::test_doctest_id.txt
Map `BdbQuit` exception to `outcomes.Exit`.
This is necessary since we are not wrapping `pdb.set_trace` there, and
therefore our `do_quit` is not called.
When enabled, floating-point numbers only need to match as far as the
precision you have written in the expected doctest output. This avoids
false positives caused by limited floating-point precision, like this:
Expected:
0.233
Got:
0.23300000000000001
This is inspired by Sébastien Boisgérault's [numtest] but the
implementation is a bit different:
* This implementation edits the literals that are in the "got"
string (the actual output from the expression being tested), and then
proceeds to compare the strings literally. This is similar to pytest's
existing ALLOW_UNICODE and ALLOW_BYTES implementation.
* This implementation only compares floats against floats, not ints
against floats. That is, the following doctest will fail with pytest
whereas it would pass with numtest:
>>> math.py # doctest: +NUMBER
3
This behaviour should be less surprising (less false negatives) when
you enable NUMBER globally in pytest.ini.
Advantages of this implementation compared to numtest:
* Doesn't require `import numtest` at the top level of the file.
* Works with pytest (if you try to use pytest & numtest together, pytest
raises "TypeError: unbound method check_output() must be called with
NumTestOutputChecker instance as first argument (got
LiteralsOutputChecker instance instead)").
* Works with Python 3.
[numtest]: https://github.com/boisgera/numtest
* 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
Ran Benita
Anthony Sottile
Daniel Hahler
Bruno Oliveira
Bruno Oliveira
Ronny Pfannschmidt
Ronny Pfannschmidt
Nikolay Kondratyev
David Röthlisberger
AmirElkess
Zac-HD
Takafumi Arakaki
Jiri Kuncar
John T. Wodder II