def initial_data(watch_html: str) -> str:
"""Extract the ytInitialData json from the watch_html page.
This mostly contains metadata necessary for rendering the page on-load,
such as video information, copyright notices, etc.
@param watch_html: Html of the watch page
@return:
"""
patterns = [
r"window\[['\"]ytInitialData['\"]]\s*=\s*",
r"ytInitialData\s*=\s*"
]
for pattern in patterns:
try:
return parse_for_object(watch_html, pattern)
except HTMLParseError:
pass
raise RegexMatchError(caller='initial_data', pattern='initial_data_pattern')
def test_invalid_start():
with pytest.raises(HTMLParseError):
parse_for_object('test = {}', r'invalid_regex')
def test_parse_context_closer_in_string_value():
test_html = 'test = {"foo": "};"};'
result = parse_for_object(test_html, r'test\s*=\s*')
assert result == {'foo': '};'}
def test_parse_simple_object():
test_html = 'test = {"foo": [], "bar": {}};'
result = parse_for_object(test_html, r'test\s*=\s*')
assert result == {'foo': [], 'bar': {}}
def test_parse_empty_object_with_trailing_characters():
test_html = 'test = {};'
result = parse_for_object(test_html, r'test\s*=\s*')
assert result == {}
def test_parse_simple_empty_object():
result = parse_for_object('test = {}', r'test\s*=\s*')
assert result == {}
