X-Git-Url: https://git.phdru.name/?a=blobdiff_plain;f=Robots%2Fparse_html.py;h=90b777b4e97a2754c2bfc33d2e4f9ff966f6d0b5;hb=a1b7edb2f925e22c9ad843496e13c687ce395e44;hp=d2ca6567fe319a4866b348c81ca5e13a65b3d0ee;hpb=72481e9aa50cf5841a98460b22a5118354676546;p=bookmarks_db.git

diff --git a/Robots/parse_html.py b/Robots/parse_html.py
index d2ca656..90b777b 100755
--- a/Robots/parse_html.py
+++ b/Robots/parse_html.py
@@ -5,20 +5,12 @@
    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
 
@@ -104,6 +96,12 @@ 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)
 
@@ -149,7 +147,6 @@ 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