don't mask encoding errors when decoding a string #6289

diff --git a/Lib/test/test_coding.py b/Lib/test/test_coding.py
index e83015e543128292fc5b35068830a3d216cb45f5..7ac3af04d831b55cf5ed9a62f915315194134e95 100644 (file)
--- a/Lib/test/test_coding.py
+++ b/Lib/test/test_coding.py
@@ -21,6 +21,18 @@ class CodingTest(unittest.TestCase):
         fp.close()
         self.assertRaises(SyntaxError, compile, text, filename, 'exec')
 
+    def test_error_from_string(self):
+        # See http://bugs.python.org/issue6289
+        input = u"# coding: ascii\n\N{SNOWMAN}".encode('utf-8')
+        try:
+            compile(input, "<string>", "exec")
+        except SyntaxError as e:
+            expected = "'ascii' codec can't decode byte 0xe2 in position 16: " \
+                "ordinal not in range(128)"
+            self.assertTrue(str(e).startswith(expected))
+        else:
+            self.fail("didn't raise")
+
 def test_main():
     test.test_support.run_unittest(CodingTest)
 

diff --git a/Misc/NEWS b/Misc/NEWS
index 686e707ffc872e945468138bc7998d55dc78de0c..44f30e0c5fed3a36eb4871f5620abb622b82a865 100644 (file)
--- a/Misc/NEWS
+++ b/Misc/NEWS
@@ -12,6 +12,8 @@ What's New in Python 2.7 alpha 1
 Core and Builtins
 -----------------
 
+- Issue #6289: Encoding errors from compile() were being masked.
+
 - When no module is given in a relative import, the module field of the
   ImportFrom AST node is now None instead of an empty string.
 

diff --git a/Parser/tokenizer.c b/Parser/tokenizer.c
index 1d0a4aa3f23b438dc5c17185bbd973130582f436..0f6705de0b5559f57d13fce6b6a8d9fc10000655 100644 (file)
--- a/Parser/tokenizer.c
+++ b/Parser/tokenizer.c
@@ -619,11 +619,8 @@ decode_str(const char *str, struct tok_state *tok)
        if (tok->enc != NULL) {
                assert(utf8 == NULL);
                utf8 = translate_into_utf8(str, tok->enc);
-               if (utf8 == NULL) {
-                       PyErr_Format(PyExc_SyntaxError,
-                               "unknown encoding: %s", tok->enc);
+               if (utf8 == NULL)
                        return error_ret(tok);
-               }
                str = PyString_AsString(utf8);
        }
 #endif