Issue #26464: Fix unicode_fast_translate() again

Initialize i variable if the string is non-ASCII.

diff --git a/Lib/test/test_unicode.py b/Lib/test/test_unicode.py
index b258db1ffae5d3c148dcb24eb7c5b23d77159e2c..c30310e1ae827489ef73e5102ebac40c05b59792 100644 (file)
--- a/Lib/test/test_unicode.py
+++ b/Lib/test/test_unicode.py
@@ -341,16 +341,22 @@ class UnicodeTest(string_tests.CommonTest,
                          "[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 --git a/Objects/unicodeobject.c b/Objects/unicodeobject.c
index a1ee776d45ad5d6bcd73e12cf422d65ca657415c..adc46156036125d0a887fbb63842261c4eafb2f0 100644 (file)
--- a/Objects/unicodeobject.c
+++ b/Objects/unicodeobject.c
@@ -8582,10 +8582,6 @@ unicode_fast_translate(PyObject *input, PyObject *mapping,
     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 @@ _PyUnicode_TranslateCharmap(PyObject *input,
 
     ignore = (errors != NULL && strcmp(errors, "ignore") == 0);
 
-    res = unicode_fast_translate(input, mapping, &writer, ignore, &i);
-    if (res < 0) {
-        _PyUnicodeWriter_Dealloc(&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;
     }
-    if (res == 1)
-        return _PyUnicodeWriter_Finish(&writer);
 
     while (i<size) {
         /* try to encode it */