Extract html redirect even if ther is no title

[bookmarks_db.git] / parse_html / bkmk_ph_lxml.py

diff --git a/parse_html/bkmk_ph_lxml.py b/parse_html/bkmk_ph_lxml.py
index 26c92680b49c76b4751917679b29e61436277747..b14be408571ba53e0cfa5f60997394658ad2061b 100644 (file)
--- a/parse_html/bkmk_ph_lxml.py
+++ b/parse_html/bkmk_ph_lxml.py
@@ -23,8 +23,6 @@ def parse_html(filename, charset=None, log=None):
     title = html_tree.findtext('head/title')
     if title is None:
         title = html_tree.findtext('title')
-        if title is None:
-            return None
 
     meta = html_tree.findall('head/meta')
     for m in meta:
@@ -40,7 +38,7 @@ 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)
 
     for m in meta:
@@ -57,4 +55,6 @@ 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)