Regression: UnicodeEncodeError with non-ascii characters in parametrized tests

[sloria]

On pytest 2.8.1 on Python 2.7.10, a UnicodeEncodeError is raised .

Code to reproduce:

# -*- coding: utf-8 -*-
from __future__ import unicode_literals
import pytest

@pytest.mark.parametrize('s',
[
    'josé@blah.com',
    'δοκ.ιμή@παράδειγμα.δοκιμή',
])
def test_non_ascii(s):
    assert bool(s)

This did not occur with pytest 2.8.0

[nicoddemus]

Thanks for the report! I will take a look as soon as possible.

[nicoddemus]

Hmm closing as duplicate of #1085, both reported nearly at the same time. 😄

[nicoddemus]

PR open at #1092