Fixed a bug in case there is no charset in META Content-Type.

git-svn-id: file:///home/phd/archive/SVN/bookmarks_db/trunk@188 fdd5c36f-1aea-0310-aeeb-c58d7e2b6c23

diff --git a/Robots/parse_html_beautifulsoup.py b/Robots/parse_html_beautifulsoup.py
index c45af2b0221a61f441cf5165fc2a25f6fe978bb5..11db56371b0df277dd957484f782e54624721e40 100644 (file)
--- a/Robots/parse_html_beautifulsoup.py
+++ b/Robots/parse_html_beautifulsoup.py
@@ -85,8 +85,12 @@ def parse_html(filename, charset=None):
 
    meta = head.find(_find_contenttype, recursive=False)
    if meta:
-      __charset = meta.get("content").lower().split('charset=')[1].split(';')[0]
-      meta_charset = _charset == __charset
+      try:
+         __charset = meta.get("content").lower().split('charset=')[1].split(';')[0]
+      except IndexError: # No charset in the META Content-Type
+         meta_charset = False
+      else:
+         meta_charset = _charset == __charset
    else:
       meta_charset = False