Failing cmath tests

[ironpythonbot]

Complex math module should improve precision on certain operations and, in some cases, correctness. Test coverage should also be improved, as single failures appear to hide others.

FAIL: test_specific_values ( ** main ** .CMathTests)

Traceback (most recent call last):
File "test\test_cmath.py", line 349, in test_specific_values

self.rAssertAlmostEqual(expected.real, actual.real,

File "test\test_cmath.py", line 111, in rAssertAlmostEqual

self.fail(msg or

AssertionError: acos0032: acos(complex(-0.99999999999999989, 0.0))
Expected: complex(3.141592638688632, -0.0)
Received: complex(3.141592653589793, -0.0)
Received value insufficiently close to expected value.

Work Item Details

Original CodePlex Issue: Issue 28352 Status: Proposed Reason Closed: Unassigned Assigned to: Unassigned Reported on: Aug 1, 2010 at 3:46 PM Reported by: darb Updated on: Feb 22, 2013 at 2:10 AM Updated by: gwinsky

[ironpythonbot]

On 2012-12-22 07:57:46 UTC, HenrySeg commented:

I ran into similar problems with the following script:

import cmath c = (-2.70563e-05+2.71051e-20j) s = cmath.sqrt(c) print c, s, s*s

The output in IronPython: (-2.70563e-05+2.71051e-20j) infj (-inf+nanj)

The output in my OS python implementation: (-2.70563e-05+2.71051e-20j) (2.60547442983e-18+0.00520156707157j) (-2.70563e-05+2.71051e-20j)