From 7c45aad6735b98106a5167532674c5182164c816 Mon Sep 17 00:00:00 2001
From: Malcolm Tredinnick <malcolm.tredinnick@gmail.com>
Date: Sat, 12 May 2007 06:54:41 +0000
Subject: [PATCH] unicode: Added some bullet-proofing to the output encoding
 path. Passing bady encoded data to template rendering shouldn't crash the
 framework.

git-svn-id: http://code.djangoproject.com/svn/django/branches/unicode@5199 bcc190cf-cafb-0310-a4f2-bffc1f526a37
---
 django/template/__init__.py |  8 +++++++-
 django/utils/encoding.py    | 14 +++++++-------
 2 files changed, 14 insertions(+), 8 deletions(-)

diff --git a/django/template/__init__.py b/django/template/__init__.py
index cfeab16c22..ae2243b47e 100644
--- a/django/template/__init__.py
+++ b/django/template/__init__.py
@@ -724,6 +724,11 @@ class Node(object):
         return nodes
 
 class NodeList(list):
+    # How invalid encoding sequences are handled. The default 'strict' is not
+    # appropriate, because the framework is not in control of all the string
+    # data.
+    codec_errors = 'replace'
+
     def render(self, context):
         bits = []
         for node in self:
@@ -731,7 +736,8 @@ class NodeList(list):
                 bits.append(self.render_node(node, context))
             else:
                 bits.append(node)
-        return ''.join([smart_str(b, settings.DEFAULT_CHARSET) for b in bits])
+        encoding = settings.DEFAULT_CHARSET
+        return ''.join([smart_str(b, encoding, errors=self.codec_errors) for b in bits])
 
     def get_nodes_by_type(self, nodetype):
         "Return a list of all nodes of the given type"
diff --git a/django/utils/encoding.py b/django/utils/encoding.py
index 402b946e65..d77269bde8 100644
--- a/django/utils/encoding.py
+++ b/django/utils/encoding.py
@@ -2,7 +2,7 @@ import types
 from django.conf import settings
 from django.utils.functional import Promise
 
-def smart_unicode(s, encoding='utf-8'):
+def smart_unicode(s, encoding='utf-8', errors='strict'):
     """
     Returns a unicode object representing 's'. Treats bytestrings using the
     'encoding' codec.
@@ -20,12 +20,12 @@ def smart_unicode(s, encoding='utf-8'):
         if hasattr(s, '__unicode__'):
             s = unicode(s)
         else:
-            s = unicode(str(s), encoding)
+            s = unicode(str(s), encoding, errors)
     elif not isinstance(s, unicode):
-        s = unicode(s, encoding)
+        s = unicode(s, encoding, errors)
     return s
 
-def smart_str(s, encoding='utf-8', strings_only=False):
+def smart_str(s, encoding='utf-8', strings_only=False, errors='strict'):
     """
     Returns a bytestring version of 's', encoded as specified in 'encoding'.
 
@@ -37,11 +37,11 @@ def smart_str(s, encoding='utf-8', strings_only=False):
         try:
             return str(s)
         except UnicodeEncodeError:
-            return unicode(s).encode(encoding)
+            return unicode(s).encode(encoding, errors)
     elif isinstance(s, unicode):
-        return s.encode(encoding)
+        return s.encode(encoding, errors)
     elif s and encoding != 'utf-8':
-        return s.decode('utf-8').encode(encoding)
+        return s.decode('utf-8', errors).encode(encoding, errors)
     else:
         return s