X-Git-Url: https://git.phdru.name/?a=blobdiff_plain;ds=sidebyside;f=parse_html%2Fbkmk_ph_beautifulsoup.py;h=0aad3dde72557abdb2f97edeea4d814151f388f6;hb=fbb34ad74643bbc69d141e26e105bcef3cdbc62c;hp=e1969f3250aa73d6e46e24883dd264f6df7ccd94;hpb=9faa13f6f8199790cf01533e857c593520559649;p=bookmarks_db.git

diff --git a/parse_html/bkmk_ph_beautifulsoup.py b/parse_html/bkmk_ph_beautifulsoup.py
index e1969f3..0aad3dd 100644
--- a/parse_html/bkmk_ph_beautifulsoup.py
+++ b/parse_html/bkmk_ph_beautifulsoup.py
@@ -14,7 +14,9 @@ __all__ = ['parse_html']
 import re
 from sgmllib import SGMLParser, SGMLParseError
 from BeautifulSoup import BeautifulSoup, CData
+
 from .bkmk_ph_util import HTMLParser
+from compat import string_type
 
 DEFAULT_CHARSET = "cp1251"  # Stupid default for Russian Cyrillic
 
@@ -60,12 +62,15 @@ def _parse_html(html_text, charset):
 
 
 def parse_html(html_text, charset=None, log=None):
+    if not html_text:
+        return None
     root = _parse_html(html_text, charset)
     if root is None:
         return None
 
     _charset = root.originalEncoding
-    if _charset in ("ISO-8859-2", "windows-1252", "MacCyrillic"):  # Replace default
+    if _charset in ("ISO-8859-2", "windows-1252", "MacCyrillic"):
+        # Replace with default and re-parse
         _charset = DEFAULT_CHARSET
         root = _parse_html(html_text, _charset)
         if root is None:
@@ -94,8 +99,8 @@ def parse_html(html_text, charset=None, log=None):
         else:
             parts = []
             for part in title:
-                if not isinstance(part, basestring):
-                    part = unicode(part)
+                if not isinstance(part, string_type):
+                    part = part.decode()
                 parts.append(part.strip())
             title = ''.join(parts)
 
@@ -104,7 +109,8 @@ def parse_html(html_text, charset=None, log=None):
         try:
             meta_content = meta.get("content")
             if meta_content:
-                __charset = meta_content.lower().split('charset=')[1].split(';')[0]
+                __charset = meta_content.lower().split('charset=')[1].\
+                    split(';')[0]
             else:
                 __charset = False
         except IndexError:  # No charset in the META Content-Type
@@ -121,8 +127,8 @@ def parse_html(html_text, charset=None, log=None):
             if meta_content:
                 meta_charset = _charset = meta_content.lower()
 
-    if title and (_charset or meta_charset):
-        title = title.encode(_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)
     if meta: