fixup! Fix #120: introduce keyword arguments for encodings by source

Author: gsnedders

html5lib/tests/test_encoding.py

@@ -2,6 +2,8 @@
 
 import os
 
+import pytest
+
 from .support import get_data_files, test_dir, errorMessage, TestData as _TestData
 from html5lib import HTMLParser, inputstream
 
@@ -28,6 +30,19 @@ def test_parser_reparse():
     assert doc.find(".//title").text == "Caf\u00E9"
 
 
+@pytest.mark.parametrize("expected,data,kwargs", [
+    ("utf-16le", b"\xFF\xFE", {"override_encoding": "utf-8"}),
+    ("utf-16be", b"\xFE\xFF", {"override_encoding": "utf-8"}),
+    ("utf-8", b"\xEF\xBB\xBF", {"override_encoding": "iso-8859-2"}),
+    ])
+def test_parser_args(expected, data, kwargs):
+    stream = inputstream.HTMLBinaryInputStream(data, useChardet=False)
+    assert expected == stream.charEncoding[0].name
+    p = HTMLParser()
+    doc = p.parse(data, useChardet=False, **kwargs)
+    assert expected == p.documentEncoding
+
+
 def runParserEncodingTest(data, encoding):
     p = HTMLParser()
     assert p.documentEncoding is None
@@ -55,6 +70,7 @@ def test_encoding():
             yield (runParserEncodingTest, test[b'data'], test[b'encoding'])
             yield (runPreScanEncodingTest, test[b'data'], test[b'encoding'])
 
+
 # pylint:disable=wrong-import-position
 try:
     import chardet  # noqa