diff options
| author | Damien George <damien.p.george@gmail.com> | 2017-09-04 14:16:27 +1000 |
|---|---|---|
| committer | Damien George <damien.p.george@gmail.com> | 2017-09-04 14:16:27 +1000 |
| commit | d4b75f6b6822885e331c69a74e56e23af40a6264 (patch) | |
| tree | ab0d1c7d77cb366abc9379b00b22cee07aa9a9c1 /tests/float/complex1.py | |
| parent | 9950865c39044e9fef295d0676af7c5fd55289ea (diff) | |
| download | micropython-d4b75f6b6822885e331c69a74e56e23af40a6264.tar.gz micropython-d4b75f6b6822885e331c69a74e56e23af40a6264.zip | |
py/obj: Fix comparison of float/complex NaN with itself.
IEEE floating point is specified such that a comparison of NaN with itself
returns false, and Python respects these semantics. This patch makes uPy
also have these semantics. The fix has a minor impact on the speed of the
object-equality fast-path, but that seems to be unavoidable and it's much
more important to have correct behaviour (especially in this case where
the wrong answer for nan==nan is silently returned).
Diffstat (limited to 'tests/float/complex1.py')
| -rw-r--r-- | tests/float/complex1.py | 5 |
1 files changed, 5 insertions, 0 deletions
diff --git a/tests/float/complex1.py b/tests/float/complex1.py index a6038de04a..7f0b317b35 100644 --- a/tests/float/complex1.py +++ b/tests/float/complex1.py @@ -37,6 +37,11 @@ ans = 1j ** 2.5j; print("%.5g %.5g" % (ans.real, ans.imag)) print(1j == 1) print(1j == 1j) +# comparison of nan is special +nan = float('nan') * 1j +print(nan == 1j) +print(nan == nan) + # builtin abs print(abs(1j)) print("%.5g" % abs(1j + 2)) |