Parse <meta charset="...">

[bookmarks_db.git] / parse_html / bkmk_ph_etreetidy.py

diff --git a/parse_html/bkmk_ph_etreetidy.py b/parse_html/bkmk_ph_etreetidy.py
index f5e794c68aa7c9446a903dca081f23e739fc95d6..b85ae2ae209c37decf8ebce34e649d25dcc7012b 100644 (file)
--- a/parse_html/bkmk_ph_etreetidy.py
+++ b/parse_html/bkmk_ph_etreetidy.py
@@ -4,7 +4,7 @@ This file is a part of Bookmarks database and Internet robot.
 """
 
 __author__ = "Oleg Broytman <phd@phdru.name>"
-__copyright__ = "Copyright (C) 2010-2012 PhiloSoft Design"
+__copyright__ = "Copyright (C) 2010-2013 PhiloSoft Design"
 __license__ = "GNU GPL"
 
 __all__ = ['parse_html']
@@ -29,8 +29,6 @@ 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:
@@ -43,10 +41,13 @@ def parse_html(filename, charset=None, log=None):
                     break
                 except IndexError:
                     meta_charset = False
+        elif m.get('charset', ''):
+           meta_charset = m.get('charset').lower()
+           break
     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:
@@ -63,4 +64,6 @@ 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)