only pull 'charset' from <meta> content-type when it definitely exists

This commit is contained in:
jesopo 2020-02-06 11:54:04 +00:00
parent 5bbae499d1
commit 01441efe75

View file

@ -153,7 +153,9 @@ def _find_encoding(headers: typing.Dict[str, str], data: bytes
meta_content_type = soup.findAll("meta",
{"http-equiv": lambda v: (v or "").lower() == "content-type"})
if meta_content_type:
return _split_content(meta_content_type[0].get("content"))["charset"]
meta_content = _split_content(meta_content_type[0].get("content"))
if "charset" in meta_content:
return meta_content["charset"]
doctype = [item for item in soup.contents if isinstance(item,
bs4.Doctype)] or None