1 """HTML Parser using BeautifulSoup4
3 This file is a part of Bookmarks database and Internet robot.
7 __author__ = "Oleg Broytman <phd@phdru.name>"
8 __copyright__ = "Copyright (C) 2017-2023 PhiloSoft Design"
9 __license__ = "GNU GPL"
11 __all__ = ['parse_html']
15 from bs4 import BeautifulSoup
16 from .bkmk_ph_util import HTMLParser
18 universal_charset = "utf-8"
19 DEFAULT_CHARSET = "cp1251" # Stupid default for Russian Cyrillic
22 def _parse_html(html_text, charset):
24 return BeautifulSoup(html_text, from_encoding=charset)
29 def parse_html(html_text, charset=None, log=None):
30 root = _parse_html(html_text, charset)
34 _charset = root.originalEncoding
41 head = html # Some sites put TITLE in HTML without HEAD
44 if (title is None) and (html is not head):
45 # Some sites put TITLE in HTML outside of HEAD
49 # Lookup TITLE in the root
58 if not isinstance(part, basestring):
60 parts.append(part.strip())
61 title = ''.join(parts)
63 meta = head.find(_find_contenttype, recursive=False)
66 meta_content = meta.get("content")
68 __charset = meta_content.lower().split('charset=')[1].split(';')[0]
71 except IndexError: # No charset in the META Content-Type
74 meta_charset = _charset = __charset
79 meta = head.find(_find_charset, recursive=False)
81 meta_content = meta.get("charset")
83 meta_charset = _charset = meta_content.lower()
85 if title and (_charset or meta_charset):
87 title = title.encode(_charset or meta_charset)
89 title = title.encode(universal_charset)
90 _charset = universal_charset
92 meta = head.find(_find_refresh, recursive=False)
94 refresh = meta.get("content")
98 meta = head.find(_find_icon, recursive=False)
100 icon = meta.get("href")
104 if (title is None) and (refresh is None) and (icon is None):
106 return HTMLParser(_charset, meta_charset, title, refresh, icon)
109 def _find_contenttype(Tag):
110 return (Tag.name == "meta") and \
111 (Tag.get_attribute_list("http-equiv", '')[0].lower() == "content-type")
114 def _find_charset(Tag):
115 return (Tag.name == "meta") and Tag.get("charset", '')
118 def _find_refresh(Tag):
119 return (Tag.name == "meta") and \
120 (Tag.get_attribute_list("http-equiv", '')[0].lower() == "refresh")
124 return (Tag.name == "link") and \
125 (Tag.get_attribute_list("rel", '')[0].lower() in ('icon', 'shortcut icon'))