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))
23 if hasattr(html_tree, 'childNodes'):
24 for node in html_tree.childNodes:
25 # Skip DocType element
26 if (node.name == 'html') and (node.type != 3):
33 for node in html.childNodes:
34 if node.name == 'head':
46 for node in head.childNodes:
47 if node.name == 'title':
49 title = node.childNodes[0].value
54 for node in head.childNodes:
55 if (node.name == 'meta') and \
56 ('http-equiv' in node.attributes) and \
57 (node.attributes['http-equiv'] == 'content-type'):
58 meta_content = node.attributes['content']
62 meta_content.lower().split('charset=')[1].split(';')[0]
67 elif (node.name == 'meta') and ('charset' in node.attributes):
68 meta_charset = node.attributes['charset'].lower()
72 charset = parser.tokenizer.stream.charEncoding[0]
74 if title and (charset or meta_charset):
75 title = title.encode(charset or meta_charset)
77 for node in head.childNodes:
78 if node.name == 'meta' and \
79 ('http-equiv' in node.attributes) and \
80 (node.attributes['http-equiv'] == 'refresh'):
81 refresh = node.attributes['content']
84 for node in head.childNodes:
85 if node.name == 'link' and \
86 ('rel' in node.attributes) and \
87 (node.attributes['rel'] in ('icon', 'shortcut icon')):
88 icon = node.attributes['href']
92 for node in html.childNodes:
93 if node.name == 'title':
95 title = node.childNodes[0].value
100 if (title is None) and (refresh is None) and (icon is None):
102 return HTMLParser(charset, meta_charset, title, refresh, icon)