X-Git-Url: https://git.phdru.name/?a=blobdiff_plain;f=Robots%2Fparse_html.py;h=b30c66458eaa32960b5badce9bc849a2082213d7;hb=76e0bd7defe7af4337fb1857141b7ef3e9f3f634;hp=25b3bb04c55f654c9620bf4682fc0a6cec313eb0;hpb=f285d4493b2dbbd6350f0e4a51fbc69f08c92b3e;p=bookmarks_db.git

diff --git a/Robots/parse_html.py b/Robots/parse_html.py
index 25b3bb0..b30c664 100755
--- a/Robots/parse_html.py
+++ b/Robots/parse_html.py
@@ -5,6 +5,8 @@
    Written by BroytMann. Copyright (C) 1997-2007 PhiloSoft Design
 """
 
+import codecs
+
 from m_lib.defenc import default_encoding
 current_charset = default_encoding.replace("windows-", "cp")
 DEFAULT_CHARSET = "cp1251" # Stupid default for Russian Cyrillic
@@ -69,15 +71,13 @@ class HTMLParser(_HTMLParser):
       for attrname, value in attrs:
          if value:
             value = value.strip().lower()
-            if (attrname == 'rel') and (value in ('icon', 'shortcut icon')):
+            if (attrname == 'rel') and (value.lower() in ('icon', 'shortcut icon')):
                has_icon = True
             elif attrname == 'href':
                href = value
 
       if has_icon:
          self.icon = href
-      else:
-         self.icon = None
 
 
 import re
@@ -145,6 +145,7 @@ def parse_html(filename, charset=None, log=None):
 
 
 if __name__ == '__main__':
+   import sys
    parser = parse_html(sys.argv[1])
    print parser.charset
    print parser.title