def test_html_encoding(self):
html_parser = HTMLParser()
reader = HTMLReader(html_parser)
bom_map = {
'utf_16_le': codecs.BOM_UTF16_LE,
'utf_16_be': codecs.BOM_UTF16_BE,
'utf_32_le': codecs.BOM_UTF32_LE,
'utf_32_be': codecs.BOM_UTF32_BE,
}
for name in CODEC_NAMES:
if name in EBCDIC or name == 'utf_8_sig':
# XXX: we're assuming that all codecs are ASCII backward
# compatable
continue
if name.startswith('utf_16') or name.startswith('utf_32'):
# FIXME: libxml/lxml doesn't like it when we pass in a codec
# name but don't specify the endian but BOM is included
continue
print('->', name)
data = io.BytesIO(bom_map.get(name, b'') + '<img>'.encode(name))
elements = tuple(reader.iter_elements(data, encoding=name))
html_element = elements[0]
self.assertEqual('html', html_element.tag)
def test_html_layout(self):
html_parser = self.get_html_parser()
reader = HTMLReader(html_parser)
elements = tuple(
reader.iter_elements(io.BytesIO(b'''
<html>
<head>
<title>hi</title>
</head>
<body>
<img>
</body>
</html>'''), encoding='ascii')
)
print(elements)
self.assertEqual('html', elements[0].tag)
self.assertEqual('head', elements[1].tag)
self.assertEqual('title', elements[2].tag)
self.assertEqual('title', elements[3].tag)
self.assertEqual('head', elements[4].tag)
self.assertEqual('body', elements[5].tag)
self.assertEqual('img', elements[6].tag)
if isinstance(html_parser, LxmlHTMLParser):
self.assertEqual('img', elements[7].tag)
self.assertEqual('body', elements[8].tag)
self.assertEqual('html', elements[9].tag)
else:
self.assertEqual('body', elements[7].tag)
self.assertEqual('html', elements[8].tag)
def test_html_script_comment(self):
test_string = b'''<script><!-- blah --></script>'''
reader = HTMLReader(HTMLParser())
elements = reader.iter_elements(io.BytesIO(test_string),
encoding='ascii')
elements = tuple(elements)
self.assertTrue(
all(isinstance(element, Element) for element in elements))
def test_html_early_html(self):
reader = HTMLReader(HTMLParser())
for test_string in [
b'''<!DOCTYPE HTML><html></html><img>''',
b'''<html></html><img>''',
b'''<!DOCTYPE HTML><img><html></html>''',
b'''<img><html></html>''',
b'''<!DOCTYPE HTML>
<html><body></body></html><p><img>''',
b'''
<html><body></body></html><p><img>''',
b'''
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"
"http://www.w3.org/TR/html4/loose.dtd">
<html>
<head>
<title>Download</title>
</head>
<body>
<br />
</body>
</html>
<pre><img></pre>
''',
b'''
<html>
<head>
<title>Download</title>
</head>
<body>
<br />
</body>
</html>
<pre><img></pre>
''',
b'''
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"
"http://www.w3.org/TR/html4/loose.dtd">
<html>
<body>
<br />
</body>
<head>
<title>Download</title>
</head>
</html>
<pre><img></pre>
''',
b'''
<html>
<body>
<br />
</body>
<head>
<title>Download</title>
</head>
</html>
<pre><img></pre>
''',
]:
elements = []
print()
print('a' * 10)
print(test_string)
for element in reader.iter_elements(io.BytesIO(test_string),
encoding='ascii'):
if isinstance(element, Element):
print(element)
elements.append(element)
element_tags = tuple(element.tag for element in elements)
self.assertIn('img', element_tags)