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(html_text, encoding=charset, parseMeta=bool(charset))
22 for node in html_tree.childNodes:
23 if (node.name == 'html') and (node.type != 3): # Skip DocType element
32 for node in html.childNodes:
33 if node.name == 'head':
45 for node in head.childNodes:
46 if node.name == 'title':
48 title = node.childNodes[0].value
53 for node in head.childNodes:
54 if (node.name == 'meta') and \
55 ('http-equiv' in node.attributes) and \
56 (node.attributes['http-equiv'] == 'content-type'):
57 meta_content = node.attributes['content']
61 meta_content.lower().split('charset=')[1].split(';')[0]
66 elif (node.name == 'meta') and ('charset' in node.attributes):
67 meta_charset = node.attributes['charset'].lower()
71 charset = parser.tokenizer.stream.charEncoding[0]
73 if title and (charset or meta_charset):
74 title = title.encode(charset or meta_charset)
76 for node in head.childNodes:
77 if node.name == 'meta' and \
78 ('http-equiv' in node.attributes) and \
79 (node.attributes['http-equiv'] == 'refresh'):
80 refresh = node.attributes['content']
83 for node in head.childNodes:
84 if node.name == 'link' and \
85 ('rel' in node.attributes) and \
86 (node.attributes['rel'] in ('icon', 'shortcut icon')):
87 icon = node.attributes['href']
91 for node in html.childNodes:
92 if node.name == 'title':
94 title = node.childNodes[0].value
99 if (title is None) and (refresh is None) and (icon is None):
101 return HTMLParser(charset, meta_charset, title, refresh, icon)