[pypy-svn] pypy fast-forward: Don't let raw UnicodeErrors in expat callbacks, use the "official" function which raises applevel exceptions.

[amauryfa]

Author: Amaury Forgeot d'Arc <amauryfa at gmail.com>
Branch: fast-forward
Changeset: r40637:77b2d271478a
Date: 2011-01-13 10:44 +0100
http://bitbucket.org/pypy/pypy/changeset/77b2d271478a/

Log:	Don't let raw UnicodeErrors in expat callbacks, use the "official"
	function which raises applevel exceptions.

diff --git a/pypy/module/pyexpat/interp_pyexpat.py b/pypy/module/pyexpat/interp_pyexpat.py
--- a/pypy/module/pyexpat/interp_pyexpat.py
+++ b/pypy/module/pyexpat/interp_pyexpat.py
@@ -361,9 +361,8 @@
 
     def w_convert(self, space, s):
         if self.returns_unicode:
-            from pypy.rlib.runicode import str_decode_utf_8
-            return space.wrap(str_decode_utf_8(
-                s, len(s), "strict")[0])
+            from pypy.interpreter.unicodehelper import PyUnicode_DecodeUTF8
+            return space.wrap(PyUnicode_DecodeUTF8(space, s))
         else:
             return space.wrap(s)
 

diff --git a/pypy/module/pyexpat/test/test_parser.py b/pypy/module/pyexpat/test/test_parser.py
--- a/pypy/module/pyexpat/test/test_parser.py
+++ b/pypy/module/pyexpat/test/test_parser.py
@@ -75,3 +75,12 @@
             assert text == u"caf\xe9"
         p.CharacterDataHandler = gotText
         p.Parse(xml)
+
+    def test_decode_error(self):
+        xml = '<fran\xe7ais>Comment \xe7a va ? Tr\xe8s bien ?</fran\xe7ais>'
+        import pyexpat
+        p = pyexpat.ParserCreate()
+        def f(*args): pass
+        p.StartElementHandler = f
+        exc = raises(UnicodeDecodeError, p.Parse, xml)
+        assert exc.value.start == 4