X-Git-Url: https://git.phdru.name/?a=blobdiff_plain;f=Robots%2Fparse_html.py;h=b30c66458eaa32960b5badce9bc849a2082213d7;hb=55ee943b3ef63d03826318d36fd964d182c3c322;hp=350ad789b5fa34592db37ce6e097de98e8b8dd6a;hpb=62977582003da9f434cd4fd377837935b6f0520c;p=bookmarks_db.git

diff --git a/Robots/parse_html.py b/Robots/parse_html.py
index 350ad78..b30c664 100755
--- a/Robots/parse_html.py
+++ b/Robots/parse_html.py
@@ -1,24 +1,16 @@
-#! /usr/local/bin/python -O
+#! /usr/bin/env python
 """
    HTML Parser
 
-   Written by BroytMann. Copyright (C) 1997-2005 PhiloSoft Design
+   Written by BroytMann. Copyright (C) 1997-2007 PhiloSoft Design
 """
 
+import codecs
 
-import sys
-current_charset = sys.getdefaultencoding()
-if current_charset == "ascii":
-   try:
-      import locale
-   except ImportError:
-      pass
-   else:
-      current_charset = locale.getpreferredencoding()
-current_charset = current_charset.replace("windows-", "cp").lower()
+from m_lib.defenc import default_encoding
+current_charset = default_encoding.replace("windows-", "cp")
 DEFAULT_CHARSET = "cp1251" # Stupid default for Russian Cyrillic
 
-
 from HTMLParser import HTMLParseError
 from m_lib.net.www.html import HTMLParser as _HTMLParser
 
@@ -33,6 +25,7 @@ class HTMLParser(_HTMLParser):
       self.meta_charset = 0
       self.title = ''
       self.refresh = ''
+      self.icon = None
 
    def end_head(self):
       raise HTMLHeadDone()
@@ -71,6 +64,22 @@ 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.lower() in ('icon', 'shortcut icon')):
+               has_icon = True
+            elif attrname == 'href':
+               href = value
+
+      if has_icon:
+         self.icon = href
+
+
 import re
 entity_re = re.compile("(&#[0-9]+;)")
 
@@ -85,6 +94,12 @@ def recode_entities(title, charset):
 
 
 def parse_html(filename, charset=None, log=None):
+   if charset:
+      try:
+         codecs.lookup(charset) # In case of unknown charset...
+      except (ValueError, LookupError):
+         charset = None         # ...try charset from HTML
+
    infile = open(filename, 'r')
    parser = HTMLParser(charset)
 
@@ -135,3 +150,4 @@ if __name__ == '__main__':
    print parser.charset
    print parser.title
    print parser.refresh
+   print parser.icon