From f3d0a8d8428ee0512094b6de2f5e1dcf3c63cef2 Mon Sep 17 00:00:00 2001
From: Lee Semel <lee@semel.net>
Date: Tue, 28 Jun 2011 00:54:36 +0800
Subject: [PATCH] Allow passing unicode objects

---
 readability/htmls.py | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/readability/htmls.py b/readability/htmls.py
index 7d32137..d11f891 100644
--- a/readability/htmls.py
+++ b/readability/htmls.py
@@ -10,9 +10,12 @@ logging.getLogger().setLevel(logging.DEBUG)
 utf8_parser = lxml.html.HTMLParser(encoding='utf-8')
 
 def build_doc(page):
-    enc = get_encoding(page)
-    page_enc = page.decode(enc, 'replace').encode('utf-8')
-    doc = lxml.html.document_fromstring(page_enc, parser=utf8_parser)
+    if type(page) != unicode:
+        enc = get_encoding(page)
+        page_enc = page.decode(enc, 'replace')
+    else:
+        page_enc = page
+    doc = lxml.html.document_fromstring(page_enc.encode('utf-8'), parser=utf8_parser)
     return doc
 
 def js_re(src, pattern, flags, repl):