1 """HTML Parser using html5
3 This file is a part of Bookmarks database and Internet robot.
7 __author__ = "Oleg Broytman <phd@phdru.name>"
8 __copyright__ = "Copyright (C) 2010-2014 PhiloSoft Design"
9 __license__ = "GNU GPL"
11 __all__ = ['parse_html']
14 from html5lib import HTMLParser as HTML5Parser
15 from .bkmk_ph_util import HTMLParser
18 def parse_html(html_text, charset=None, log=None):
19 parser = HTML5Parser()
20 html_tree = parser.parse(
21 html_text, encoding=charset, parseMeta=bool(charset))
24 if hasattr(html_tree, 'childNodes'):
25 for node in html_tree.childNodes:
26 # Skip DocType element
27 if (node.name == 'html') and (node.type != 3):
34 for node in html.childNodes:
35 if node.name == 'head':
47 for node in head.childNodes:
48 if node.name == 'title':
50 title = node.childNodes[0].value
55 for node in head.childNodes:
56 if (node.name == 'meta') and \
57 ('http-equiv' in node.attributes) and \
58 (node.attributes['http-equiv'] == 'content-type'):
59 meta_content = node.attributes['content']
63 meta_content.lower().split('charset=')[1].\
69 elif (node.name == 'meta') and ('charset' in node.attributes):
70 meta_charset = node.attributes['charset'].lower()
74 charset = parser.tokenizer.stream.charEncoding[0]
76 if title and (charset or meta_charset):
77 title = title.encode(charset or meta_charset)
79 for node in head.childNodes:
80 if node.name == 'meta' and \
81 ('http-equiv' in node.attributes) and \
82 (node.attributes['http-equiv'] == 'refresh'):
83 refresh = node.attributes['content']
86 for node in head.childNodes:
87 if node.name == 'link' and \
88 ('rel' in node.attributes) and \
89 (node.attributes['rel'] in ('icon', 'shortcut icon')):
90 icon = node.attributes['href']
94 for node in html.childNodes:
95 if node.name == 'title':
97 title = node.childNodes[0].value
102 if (title is None) and (refresh is None) and (icon is None):
104 return HTMLParser(charset, meta_charset, title, refresh, icon)