Commit

Author: StanFromIreland

Lib/test/multibytecodec_support.py

@@ -284,10 +284,11 @@ def test_incrementalencoder_del_segfault(self):
 
     def test_null_terminator(self):
         # see gh-101828
-        if any(enc in self.encoding for enc in ('shift', 'euc_jis')):
-            text = "バルーンフルーツ"
-        else:
-            text = "Spam"
+        text = "フルーツ"
+        try:
+            text.encode(self.encoding)
+        except UnicodeEncodeError:
+            text = "Python is cool"
         encode_w_null = (text + "\0").encode(self.encoding)
         encode_plus_null = text.encode(self.encoding) + "\0".encode(self.encoding)
         self.assertTrue(encode_w_null.endswith(b'\x00'))

Modules/cjkcodecs/_codecs_iso2022.c

@@ -802,10 +802,13 @@ jisx0213_encoder(const MultibyteCodec *codec, const Py_UCS4 *data,
         return coded;
 
     case 2: /* second character of unicode pair */
-        coded = find_pairencmap((ucs2_t)data[0], (ucs2_t)data[1],
-                                jisx0213_pair_encmap, JISX0213_ENCPAIRS);
-        if (coded != DBCINV)
-            return coded;
+        if (data[1] != 0) { /* Don't consume null char as part of pair */
+            coded = find_pairencmap((ucs2_t)data[0], (ucs2_t)data[1],
+                                    jisx0213_pair_encmap, JISX0213_ENCPAIRS);
+            if (coded != DBCINV) {
+                return coded;
+            }
+        }
         _Py_FALLTHROUGH;
 
     case -1: /* flush unterminated */