changeset:   100381:6643c5cc9797
branch:      3.5
parent:      100379:27ba9ba5deb1
user:        Victor Stinner <victor.stinner@gmail.com>
date:        Tue Mar 01 21:59:58 2016 +0100
files:       Lib/test/test_unicode.py Objects/unicodeobject.c
description:
Issue #26464: Fix unicode_fast_translate() again

Initialize i variable if the string is non-ASCII.


diff -r 27ba9ba5deb1 -r 6643c5cc9797 Lib/test/test_unicode.py
--- a/Lib/test/test_unicode.py	Tue Mar 01 21:30:30 2016 +0100
+++ b/Lib/test/test_unicode.py	Tue Mar 01 21:59:58 2016 +0100
@@ -341,16 +341,22 @@
                          "[XXX]")
         self.assertEqual("[a]".translate(str.maketrans({'a': '\xe9'})),
                          "[\xe9]")
+        self.assertEqual('axb'.translate(str.maketrans({'a': None, 'b': '123'})),
+                         "x123")
+        self.assertEqual('axb'.translate(str.maketrans({'a': None, 'b': '\xe9'})),
+                         "x\xe9")
+
+        # test non-ASCII (don't take the fast-path)
         self.assertEqual("[a]".translate(str.maketrans({'a': '<\xe9>'})),
                          "[<\xe9>]")
         self.assertEqual("[\xe9]".translate(str.maketrans({'\xe9': 'a'})),
                          "[a]")
         self.assertEqual("[\xe9]".translate(str.maketrans({'\xe9': None})),
                          "[]")
-        self.assertEqual('axb'.translate(str.maketrans({'a': None, 'b': '123'})),
-                         "x123")
-        self.assertEqual('axb'.translate(str.maketrans({'a': None, 'b': '\xe9'})),
-                         "x\xe9")
+        self.assertEqual("[\xe9]".translate(str.maketrans({'\xe9': '123'})),
+                         "[123]")
+        self.assertEqual("[a\xe9]".translate(str.maketrans({'a': '<\u20ac>'})),
+                         "[<\u20ac>\xe9]")
 
         # invalid Unicode characters
         invalid_char = 0x10ffff+1
diff -r 27ba9ba5deb1 -r 6643c5cc9797 Objects/unicodeobject.c
--- a/Objects/unicodeobject.c	Tue Mar 01 21:30:30 2016 +0100
+++ b/Objects/unicodeobject.c	Tue Mar 01 21:59:58 2016 +0100
@@ -8582,10 +8582,6 @@
     Py_UCS1 *in, *end, *out;
     int res = 0;
 
-    if (PyUnicode_READY(input) == -1)
-        return -1;
-    if (!PyUnicode_IS_ASCII(input))
-        return 0;
     len = PyUnicode_GET_LENGTH(input);
 
     memset(ascii_table, 0xff, 128);
@@ -8668,13 +8664,20 @@
 
     ignore = (errors != NULL && strcmp(errors, "ignore") == 0);
 
-    res = unicode_fast_translate(input, mapping, &writer, ignore, &i);
-    if (res < 0) {
-        _PyUnicodeWriter_Dealloc(&writer);
-        return NULL;
-    }
-    if (res == 1)
-        return _PyUnicodeWriter_Finish(&writer);
+    if (PyUnicode_READY(input) == -1)
+        return NULL;
+    if (PyUnicode_IS_ASCII(input)) {
+        res = unicode_fast_translate(input, mapping, &writer, ignore, &i);
+        if (res < 0) {
+            _PyUnicodeWriter_Dealloc(&writer);
+            return NULL;
+        }
+        if (res == 1)
+            return _PyUnicodeWriter_Finish(&writer);
+    }
+    else {
+        i = 0;
+    }
 
     while (i<size) {
         /* try to encode it */