import pytest
from charset_normalizer.utils import any_specified_encoding
from charset_normalizer import CharsetMatch
@pytest.mark.parametrize(
"payload, expected_encoding",
[
(b'', "euc_jp"),
(b'', "utf_8"),
(b'', None),
(b'# coding: utf-8', "utf_8"),
(b'', 'utf_8'),
(b'', 'ascii'),
(b'', 'johab'),
(b'', 'cp037'),
(b'', "cp1252"),
(b'', "cp1256"),
]
)
def test_detect_most_common_body_encoding(payload, expected_encoding):
specified_encoding = any_specified_encoding(
payload
)
assert specified_encoding == expected_encoding, "Unable to determine properly encoding from given body"
@pytest.mark.parametrize(
"payload, expected_outcome",
[
(b'', b''),
(b'', b''),
(b'', b''),
(b'# coding: utf-8', b'# coding: utf-8'),
(b'', b''),
(b'', b''),
(b'', b''),
(b'', b''),
(b'', b''),
]
)
def test_preemptive_mark_replacement(payload, expected_outcome):
"""
When generating (to Unicode converted) bytes, we want to change any potential declarative charset
to utf-8. This test that.
"""
specified_encoding = any_specified_encoding(
payload
)
detected_encoding = specified_encoding if specified_encoding is not None else "utf-8"
m = CharsetMatch(
payload,
detected_encoding,
0.,
False,
[],
preemptive_declaration=specified_encoding,
)
transformed_output = m.output()
assert transformed_output == expected_outcome