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) 2010-2023 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 if isinstance(html_text, bytes):
21 html_tree = parser.parse(
22 html_text, encoding=charset, parseMeta=bool(charset))
24 html_tree = parser.parse(html_text)
27 if hasattr(html_tree, 'childNodes'):
28 for node in html_tree.childNodes:
29 # Skip DocType element
30 if (node.name == 'html') and (node.type != 3):
37 for node in html.childNodes:
38 if node.name == 'head':
50 for node in head.childNodes:
51 if node.name == 'title':
53 title = node.childNodes[0].value
58 for node in head.childNodes:
59 if (node.name == 'meta') and \
60 ('http-equiv' in node.attributes) and \
61 (node.attributes['http-equiv'] == 'content-type'):
62 meta_content = node.attributes['content']
66 meta_content.lower().split('charset=')[1].\
72 elif (node.name == 'meta') and ('charset' in node.attributes):
73 meta_charset = node.attributes['charset'].lower()
77 charset = parser.tokenizer.stream.charEncoding[0]
79 #if title and (charset or meta_charset):
80 # title = title.encode(charset or meta_charset)
82 for node in head.childNodes:
83 if node.name == 'meta' and \
84 ('http-equiv' in node.attributes) and \
85 (node.attributes['http-equiv'] == 'refresh'):
86 refresh = node.attributes['content']
89 for node in head.childNodes:
90 if node.name == 'link' and \
91 ('rel' in node.attributes) and \
92 (node.attributes['rel'] in ('icon', 'shortcut icon')):
93 icon = node.attributes['href']
97 for node in html.childNodes:
98 if node.name == 'title':
100 title = node.childNodes[0].value
105 if (title is None) and (refresh is None) and (icon is None):
107 return HTMLParser(charset, meta_charset, title, refresh, icon)