X-Git-Url: https://git.phdru.name/?a=blobdiff_plain;f=Robots%2Fparse_html.py;h=57ad6a0cf69c89247281700915ced9854a3ef8ba;hb=fb98cfeffff4fb699b6d5ca3c8c151901002b32e;hp=ffacbee521ddecd0553008a158db163d07487503;hpb=387f77d110986aa12967c9cd788ab0e4f41f2be2;p=bookmarks_db.git

diff --git a/Robots/parse_html.py b/Robots/parse_html.py
index ffacbee..57ad6a0 100755
--- a/Robots/parse_html.py
+++ b/Robots/parse_html.py
@@ -1,14 +1,22 @@
-#! /usr/local/bin/python -O
+#! /usr/bin/env python
 """
    HTML Parser
 
-   Written by BroytMann, Jun 2002 - May 2003. Copyright (C) 1997-2003 PhiloSoft Design
+   Written by BroytMann. Copyright (C) 1997-2007 PhiloSoft Design
 """
 
 
 import sys
 current_charset = sys.getdefaultencoding()
-DEFAULT_CHARSET = "windows-1251" # Stupid default for Russian Cyrillic
+if current_charset == "ascii":
+   try:
+      import locale
+   except ImportError:
+      pass
+   else:
+      current_charset = locale.getpreferredencoding()
+current_charset = current_charset.replace("windows-", "cp").lower()
+DEFAULT_CHARSET = "cp1251" # Stupid default for Russian Cyrillic
 
 
 from HTMLParser import HTMLParseError
@@ -63,6 +71,24 @@ class HTMLParser(_HTMLParser):
          self.title = self.accumulator
 
 
+   def do_link(self, attrs):
+      has_icon = False
+      href = None
+
+      for attrname, value in attrs:
+         if value:
+            value = value.strip().lower()
+            if (attrname == 'rel') and (value in ('icon', 'shortcut icon')):
+               has_icon = True
+            elif attrname == 'href':
+               href = value
+
+      if has_icon:
+         self.icon = href
+      else:
+         self.icon = None
+
+
 import re
 entity_re = re.compile("(&#[0-9]+;)")
 
@@ -96,14 +122,14 @@ def parse_html(filename, charset=None, log=None):
    title = parser.title
 
    if not parser.charset:
-      ascii = 1
-      for c in title:
-         if not (32 <= ord(c) <= 127): # non-ASCII character
-            ascii = 0
-            break
-      if not ascii:
+      try:
+         unicode(title, "ascii")
+      except UnicodeDecodeError:
          parser.charset = DEFAULT_CHARSET
 
+   if parser.charset:
+      parser.charset = parser.charset.replace("windows-", "cp").lower()
+
    if parser.charset and (parser.charset <> current_charset):
       try:
          if parser.meta_charset:
@@ -127,3 +153,4 @@ if __name__ == '__main__':
    print parser.charset
    print parser.title
    print parser.refresh
+   print parser.icon