__all__ = ['parse_html']
-import re
from bs4 import BeautifulSoup
+
from .bkmk_ph_util import HTMLParser
+from compat import string_type
universal_charset = "utf-8"
DEFAULT_CHARSET = "cp1251" # Stupid default for Russian Cyrillic
+
def _parse_html(html_text, charset):
try:
return BeautifulSoup(html_text, from_encoding=charset)
except TypeError:
return None
+
def parse_html(html_text, charset=None, log=None):
root = _parse_html(html_text, charset)
if root is None:
else:
parts = []
for part in title:
- if not isinstance(part, basestring):
- part = unicode(part)
+ if not isinstance(part, string_type):
+ part = part.decode()
parts.append(part.strip())
title = ''.join(parts)
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_attribute_list("http-equiv", '')[0].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_attribute_list("http-equiv", '')[0].lower() == "refresh")
+
def _find_icon(Tag):
return (Tag.name == "link") and \
- (Tag.get_attribute_list("rel", '')[0].lower() in ('icon', 'shortcut icon'))
+ (Tag.get_attribute_list("rel", '')[0].lower()
+ in ('icon', 'shortcut icon'))