TST: computation/tests/test_eval.py

[dsm054]

There are two strange things in computation/tests/test_eval.py.

First, in check_complex_cmp_op:

    elif _bool_and_frame(lhs, rhs):
        self.assertRaises(TypeError, _eval_single_bin, lhs_new, '&',
                          rhs_new, self.engine)
        self.assertRaises(TypeError, pd.eval, ex,
                          local_dict={'lhs': lhs, 'rhs': rhs},
                          engine=self.engine, parser=self.parser)

neither lhs_new nor rhs_new are defined (they're defined later on another branch). Probably they're meant to be simply lhs and rhs.

Second, in the subfunction check_operands:

                with assertRaises(NotImplementedError):
                    pd.eval(ex, engine=self.engine, parser=self.parser)

assertRaises isn't defined. This should probably be tm.assertRaises, as in the previous

            with tm.assertRaises(ValueError):
                pd.eval(ex, engine=self.engine, parser=self.parser)

I guess neither of these code paths are being exercised at the moment, which is why it hasn't been an issue.

[jreback]

how do u find these!

[dsm054]

@jreback: I'd like to claim it was mad hax0r skillz, but really I just ran cython filename.py. Even when I'm not going to use the cythonized version it's handy to pick up certain classes of errors.

[jreback]

oh....interesting....thanks......pls sbumit a PR if you'd lke to fix

[dsm054]

@jreback: that'd be easy, but is it alright to leave them unexecuted?

[cpcloud]

@dsm054 That is a useful technique. When I wrote these I got a bit overwhelmed with trying to cover all the possible test cases without being repetitive. Looks like I was only partially successful at either 😦

[dsm054]

This was closed via #6074.