From 4aff2ee4d6f077e06159c92ab05c0f2ea527c6fa Mon Sep 17 00:00:00 2001
From: Leonard Richardson <leonard.richardson@canonical.com>
Date: Thu, 9 Feb 2012 16:15:56 -0500
Subject: As a last-ditch attempt to turn data into Unicode, use errors=replace
 instead of errors=strict.

---
 bs4/builder/_lxml.py | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

(limited to 'bs4/builder/_lxml.py')

diff --git a/bs4/builder/_lxml.py b/bs4/builder/_lxml.py
index 7219e49..cc3cb86 100644
--- a/bs4/builder/_lxml.py
+++ b/bs4/builder/_lxml.py
@@ -50,12 +50,13 @@ class LXMLTreeBuilderForXML(TreeBuilder):
         declared within markup).
         """
         if isinstance(markup, unicode):
-            return markup, None, None
+            return markup, None, None, False
 
         try_encodings = [user_specified_encoding, document_declared_encoding]
         dammit = UnicodeDammit(markup, try_encodings, is_html=True)
         return (dammit.markup, dammit.original_encoding,
-                dammit.declared_html_encoding)
+                dammit.declared_html_encoding,
+                dammit.contains_replacement_characters)
 
     def feed(self, markup):
         self.parser.feed(markup)
-- 
cgit v1.2.3