X-Git-Url: https://git.phdru.name/?a=blobdiff_plain;ds=sidebyside;f=Robots%2Fparse_html_beautifulsoup.py;h=47ecbaf459f367a9a80d535af1add172709dcf0d;hb=0e76f1851882b99da63a7c8a9e4cdf0c4a48657f;hp=dc61a3305a3c0509562cfa9df56ccb88d550dc60;hpb=dcad728f7b334f67f4f883add324de805b0b1e40;p=bookmarks_db.git

diff --git a/Robots/parse_html_beautifulsoup.py b/Robots/parse_html_beautifulsoup.py
index dc61a33..47ecbaf 100644
--- a/Robots/parse_html_beautifulsoup.py
+++ b/Robots/parse_html_beautifulsoup.py
@@ -1,7 +1,7 @@
 """
    HTML Parser using BeautifulSoup
 
-   Written by BroytMann. Copyright (C) 2007, 2008 PhiloSoft Design
+   Written by Broytman. Copyright (C) 2007, 2008 PhiloSoft Design
 """
 
 import re
@@ -70,36 +70,33 @@ def parse_html(filename, charset=None, log=None):
       if root is None:
          return None
 
-   try:
-      html = root.html
-   except AttributeError:
-      if log: log("No HTML in root")
-      html = root
-
+   html = root.html
    if html is None:
       html = root
 
-   try:
-      head = html.head
-   except AttributeError:
-      if log: log("No HEAD in HTML")
-      head = html
-
+   head = html.head
    if head is None:
       head = html # Some sites put TITLE in HTML without HEAD
 
-   try:
-      title = head.title.string.encode(_charset)
-   except AttributeError:
-      title = '' # HEAD but no TITLE
-
-   if (not title) and (head is not html):
+   title = head.title
+   if (title is None) and (html is not head):
       # Some sites put TITLE in HTML outside of HEAD
+      title = html.title
 
-      try:
-         title = html.title.string.encode(_charset)
-      except AttributeError:
-         title = '' # no TITLE in HTML too
+   if title is None:
+      # Lookup TITLE in the root
+      title = root.title
+
+   if title is not None:
+      if title.string:
+         title = title.string.encode(_charset)
+      else:
+         parts = []
+         for part in title:
+            if not isinstance(part, basestring):
+               part = unicode(part)
+            parts.append(part.strip())
+         title = ''.join(parts).encode(_charset)
 
    meta = head.find(_find_contenttype, recursive=False)
    if meta: