Merged revisions 72404-72406 via svnmerge from

svn+ssh://pythondev@svn.python.org/python/trunk

........
  r72404 | walter.doerwald | 2009-05-06 16:28:24 +0200 (Mi, 06 Mai 2009) | 3 lines

  Issue 3739: The unicode-internal encoder now reports the number of *characters*
  consumed like any other encoder (instead of the number of bytes).
........
  r72406 | walter.doerwald | 2009-05-06 16:32:35 +0200 (Mi, 06 Mai 2009) | 2 lines

  Add NEWS entry about issue #3739.
........

diff --git a/Lib/test/test_codecs.py b/Lib/test/test_codecs.py
index 5a3834d49592f79d43c266e2f7c59aec699cdf34..79b1fab0c3fb394699f39dc74d2877ead595a918 100644 (file)
--- a/Lib/test/test_codecs.py
+++ b/Lib/test/test_codecs.py
@@ -872,6 +872,12 @@ class UnicodeInternalTest(unittest.TestCase):
                               "UnicodeInternalTest")
             self.assertEquals(("ab", 12), ignored)
 
+    def test_encode_length(self):
+        # Issue 3739
+        encoder = codecs.getencoder("unicode_internal")
+        self.assertEquals(encoder("a")[1], 1)
+        self.assertEquals(encoder("\xe9\u0142")[1], 2)
+
 # From http://www.gnu.org/software/libidn/draft-josefsson-idn-test-vectors.html
 nameprep_tests = [
     # 3.1 Map to nothing.
@@ -1317,8 +1323,7 @@ class BasicUnicodeTest(unittest.TestCase, MixInCheckStateHandling):
                 name = "latin_1"
             self.assertEqual(encoding.replace("_", "-"), name.replace("_", "-"))
             (b, size) = codecs.getencoder(encoding)(s)
-            if encoding != "unicode_internal":
-                self.assertEqual(size, len(s), "%r != %r (encoding=%r)" % (size, len(s), encoding))
+            self.assertEqual(size, len(s), "%r != %r (encoding=%r)" % (size, len(s), encoding))
             (chars, size) = codecs.getdecoder(encoding)(b)
             self.assertEqual(chars, s, "%r != %r (encoding=%r)" % (chars, s, encoding))
 

diff --git a/Misc/NEWS b/Misc/NEWS
index 45bd9e758c79e0883e044046dbbe97925411a037..ade72f2c22f2f2539fa75003fe9d8dcdea8306f4 100644 (file)
--- a/Misc/NEWS
+++ b/Misc/NEWS
@@ -127,6 +127,9 @@ Core and Builtins
 - Issue #1113244: Py_XINCREF, Py_DECREF, Py_XDECREF: Add `do { ... } while (0)'
   to avoid compiler warnings.
 
+- Issue #3739: The unicode-internal encoder now reports the number of characters
+  consumed like any other encoder (instead of the number of bytes).
+
 Installation
 ------------
 

diff --git a/Modules/_codecsmodule.c b/Modules/_codecsmodule.c
index cbcf84dd07d953d8aee07e5e520e851529093644..55b593b37d50f083645f04b4f9977ead71c776dd 100644 (file)
--- a/Modules/_codecsmodule.c
+++ b/Modules/_codecsmodule.c
@@ -669,7 +669,8 @@ unicode_internal_encode(PyObject *self,
     if (PyUnicode_Check(obj)) {
        data = PyUnicode_AS_DATA(obj);
        size = PyUnicode_GET_DATA_SIZE(obj);
-       return codec_tuple(PyBytes_FromStringAndSize(data, size), size);
+       return codec_tuple(PyBytes_FromStringAndSize(data, size),
+                          PyUnicode_GET_SIZE(obj));
     }
     else {
        if (PyObject_AsReadBuffer(obj, (const void **)&data, &size))