diff options
Diffstat (limited to 'bs4/tests')
-rw-r--r-- | bs4/tests/test_soup.py | 8 |
1 files changed, 8 insertions, 0 deletions
diff --git a/bs4/tests/test_soup.py b/bs4/tests/test_soup.py index 94f325e..bb97e52 100644 --- a/bs4/tests/test_soup.py +++ b/bs4/tests/test_soup.py @@ -278,6 +278,14 @@ class TestUnicodeDammit(unittest.TestCase): finally: bs4.dammit.chardet = chardet + def test_sniffed_xml_encoding(self): + # A document written in UTF-16LE will be converted by a different + # code path that sniffs the byte order markers. + data = b'\xff\xfe<\x00a\x00>\x00\xe1\x00\xe9\x00<\x00/\x00a\x00>\x00' + dammit = UnicodeDammit(data) + self.assertEqual(u"<a>áé</a>", dammit.unicode_markup) + self.assertEqual("utf-16le", dammit.original_encoding) + class TestNamedspacedAttribute(SoupTest): |