"""
__author__ = "Oleg Broytman <phd@phdru.name>"
-__copyright__ = "Copyright (C) 2017 PhiloSoft Design"
+__copyright__ = "Copyright (C) 2017-2023 PhiloSoft Design"
__license__ = "GNU GPL"
__all__ = ['parse_html']
from bs4 import BeautifulSoup
from .bkmk_ph_util import HTMLParser
-DEFAULT_CHARSET = "cp1251" # Stupid default for Russian Cyrillic
+universal_charset = "utf-8"
+DEFAULT_CHARSET = "cp1251" # Stupid default for Russian Cyrillic
+
def _parse_html(html_text, charset):
try:
except TypeError:
return None
+
def parse_html(html_text, charset=None, log=None):
root = _parse_html(html_text, charset)
if root is None:
head = html.head
if head is None:
- head = html # Some sites put TITLE in HTML without HEAD
+ head = html # Some sites put TITLE in HTML without HEAD
title = head.title
if (title is None) and (html is not head):
__charset = meta_content.lower().split('charset=')[1].split(';')[0]
else:
__charset = False
- except IndexError: # No charset in the META Content-Type
+ except IndexError: # No charset in the META Content-Type
meta_charset = False
else:
meta_charset = _charset = __charset
meta_charset = _charset = meta_content.lower()
if title and (_charset or meta_charset):
- title = title.encode(_charset or meta_charset)
+ try:
+ title = title.encode(_charset or meta_charset)
+ except LookupError:
+ title = title.encode(universal_charset)
+ _charset = universal_charset
meta = head.find(_find_refresh, recursive=False)
if meta:
return None
return HTMLParser(_charset, meta_charset, title, refresh, icon)
+
def _find_contenttype(Tag):
return (Tag.name == "meta") and \
(Tag.get_attribute_list("http-equiv", '')[0].lower() == "content-type")
+
def _find_charset(Tag):
return (Tag.name == "meta") and Tag.get("charset", '')
+
def _find_refresh(Tag):
return (Tag.name == "meta") and \
(Tag.get_attribute_list("http-equiv", '')[0].lower() == "refresh")
+
def _find_icon(Tag):
return (Tag.name == "link") and \
(Tag.get_attribute_list("rel", '')[0].lower() in ('icon', 'shortcut icon'))