DEFAULT_CHARSET = "cp1251" # Stupid default for Russian Cyrillic
# http://groups.google.com/group/beautifulsoup/browse_thread/thread/69093cb0d3a3cf63
+
+
class BadDeclParser(BeautifulSoup):
def parse_declaration(self, i):
"""Treat a bogus SGML declaration as raw data. Treat a CDATA
except TypeError:
return None
+
def parse_html(html_text, charset=None, log=None):
root = _parse_html(html_text, charset)
if root is None:
return None
_charset = root.originalEncoding
- if _charset in ("ISO-8859-2", "windows-1252", "MacCyrillic"): # Replace default
+ if _charset in ("ISO-8859-2", "windows-1252", "MacCyrillic"):
+ # Replace with default and re-parse
_charset = DEFAULT_CHARSET
root = _parse_html(html_text, _charset)
if root is None:
try:
meta_content = meta.get("content")
if meta_content:
- __charset = meta_content.lower().split('charset=')[1].split(';')[0]
+ __charset = meta_content.lower().split('charset=')[1].\
+ split(';')[0]
else:
__charset = False
except IndexError: # No charset in the META Content-Type
return None
return HTMLParser(_charset, meta_charset, title, refresh, icon)
+
def _find_contenttype(Tag):
return (Tag.name == "meta") and \
(Tag.get("http-equiv", '').lower() == "content-type")
+
def _find_charset(Tag):
return (Tag.name == "meta") and Tag.get("charset", '')
+
def _find_refresh(Tag):
return (Tag.name == "meta") and \
(Tag.get("http-equiv", '').lower() == "refresh")
+
def _find_icon(Tag):
return (Tag.name == "link") and \
(Tag.get("rel", '').lower() in ('icon', 'shortcut icon'))