"""HTML Parser using html5
This file is a part of Bookmarks database and Internet robot.
+
"""
__author__ = "Oleg Broytman <phd@phdru.name>"
-__copyright__ = "Copyright (C) 2010-2012 PhiloSoft Design"
+__copyright__ = "Copyright (C) 2010-2014 PhiloSoft Design"
__license__ = "GNU GPL"
__all__ = ['parse_html']
from .bkmk_ph_util import HTMLParser
-def parse_html(filename, charset=None, log=None):
+def parse_html(html_text, charset=None, log=None):
parser = HTML5Parser()
- fp = open(filename)
- parser._parse(fp, encoding=charset, parseMeta=bool(charset))
- fp.close()
- html_tree = parser.tree.getDocument()
+ html_tree = parser.parse(html_text, encoding=charset, parseMeta=bool(charset))
for node in html_tree.childNodes:
if (node.name == 'html') and (node.type != 3): # Skip DocType element
title = ''
for node in head.childNodes:
- if node.name == 'meta' and \
+ if (node.name == 'meta') and \
('http-equiv' in node.attributes) and \
(node.attributes['http-equiv'] == 'content-type'):
meta_content = node.attributes['content']
meta_charset = False
else:
break
+ elif (node.name == 'meta') and ('charset' in node.attributes):
+ meta_charset = node.attributes['charset'].lower()
+ break
if not charset:
charset = parser.tokenizer.stream.charEncoding[0]