Fixed a bug - don't do a double encode.

git-svn-id: file:///home/phd/archive/SVN/bookmarks_db/trunk@271 fdd5c36f-1aea-0310-aeeb-c58d7e2b6c23

diff --git a/Robots/parse_html_beautifulsoup.py b/Robots/parse_html_beautifulsoup.py
index a7df16f926e94c1812617026885391a620eb119f..1fa4dc358e723481cfa667767ffb9c6f49318b58 100644 (file)
--- a/Robots/parse_html_beautifulsoup.py
+++ b/Robots/parse_html_beautifulsoup.py
@@ -81,7 +81,7 @@ def parse_html(filename, charset=None, log=None):
       return None
 
    if title.string:
-      title = title.string.encode(_charset)
+      title = title.string
    else:
       parts = []
       for part in title:
@@ -90,6 +90,9 @@ def parse_html(filename, charset=None, log=None):
          parts.append(part.strip())
       title = ''.join(parts)
 
+   if _charset or meta_charset:
+      title = title.encode(_charset or meta_charset)
+
    meta = head.find(_find_contenttype, recursive=False)
    if meta:
       try:
@@ -105,9 +108,6 @@ def parse_html(filename, charset=None, log=None):
    else:
       meta_charset = False
 
-   if charset or meta_charset:
-      title = title.encode(charset or meta_charset)
-
    meta = head.find(_find_refresh, recursive=False)
    if meta:
       refresh = meta.get("content")