X-Git-Url: https://git.phdru.name/?a=blobdiff_plain;f=parse_html%2Fbkmk_ph_beautifulsoup.py;h=a0ef6af76a5c78dc061d24acbf95498b44912cec;hb=80054d3bdcbde67d3b941e0b69b55ca4998f4673;hp=e02081b3127d7e96ee0c060d09064c8bb40d9888;hpb=338c964afba3651bd8fe6318644c0fcabb66cc3b;p=bookmarks_db.git

diff --git a/parse_html/bkmk_ph_beautifulsoup.py b/parse_html/bkmk_ph_beautifulsoup.py
index e02081b..a0ef6af 100644
--- a/parse_html/bkmk_ph_beautifulsoup.py
+++ b/parse_html/bkmk_ph_beautifulsoup.py
@@ -3,11 +3,8 @@
 This file is a part of Bookmarks database and Internet robot.
 """
 
-__version__ = "$Revision$"[11:-2]
-__revision__ = "$Id$"[5:-2]
-__date__ = "$Date$"[7:-2]
 __author__ = "Oleg Broytman <phd@phdru.name>"
-__copyright__ = "Copyright (C) 2007-2011 PhiloSoft Design"
+__copyright__ = "Copyright (C) 2007-2012 PhiloSoft Design"
 __license__ = "GNU GPL"
 
 __all__ = ['parse_html']
@@ -18,6 +15,7 @@ from sgmllib import SGMLParser, SGMLParseError
 from BeautifulSoup import BeautifulSoup, CData
 from .bkmk_ph_util import HTMLParser
 
+DEFAULT_CHARSET = "cp1251" # Stupid default for Russian Cyrillic
 
 # http://groups.google.com/group/beautifulsoup/browse_thread/thread/69093cb0d3a3cf63
 class BadDeclParser(BeautifulSoup):
@@ -86,18 +84,16 @@ def parse_html(filename, charset=None, log=None):
       # Lookup TITLE in the root
       title = root.title
 
-   if title is None:
-      return None
-
-   if title.string:
-      title = title.string
-   else:
-      parts = []
-      for part in title:
-         if not isinstance(part, basestring):
-            part = unicode(part)
-         parts.append(part.strip())
-      title = ''.join(parts)
+   if title is not None:
+       if title.string:
+          title = title.string
+       else:
+          parts = []
+          for part in title:
+             if not isinstance(part, basestring):
+                part = unicode(part)
+             parts.append(part.strip())
+          title = ''.join(parts)
 
    meta = head.find(_find_contenttype, recursive=False)
    if meta:
@@ -114,7 +110,7 @@ def parse_html(filename, charset=None, log=None):
    else:
       meta_charset = False
 
-   if _charset or meta_charset:
+   if title and (_charset or meta_charset):
       title = title.encode(_charset or meta_charset)
 
    meta = head.find(_find_refresh, recursive=False)
@@ -129,6 +125,8 @@ def parse_html(filename, charset=None, log=None):
    else:
       icon = None
 
+   if (title is None) and (refresh is None) and (icon is None):
+      return None
    return HTMLParser(_charset, meta_charset, title, refresh, icon)
 
 def _find_contenttype(Tag):