Account for null character in NativeCharSequence

Author: fangerer

graalpython/com.oracle.graal.python.cext/src/unicodeobject.c

@@ -512,9 +512,9 @@ PyObject* PyUnicode_New(Py_ssize_t size, Py_UCS4 maxchar) {
         kind = PyUnicode_4BYTE_KIND;
     }
 
-    size_t n = size * kind;
-    int8_t* ptr = (int8_t*) malloc(n);
-    return _jls_PyUnicode_New(polyglot_from_i8_array((int8_t*)ptr, n), kind, is_ascii);
+    /* add one to size for the null character */
+    int8_t* ptr = (int8_t*) calloc(size + 1, kind);
+    return _jls_PyUnicode_New(polyglot_from_i8_array((int8_t*)ptr, (size + 1) * kind), kind, is_ascii);
 }
 
 UPCALL_ID(PyUnicode_Compare);

graalpython/com.oracle.graal.python/src/com/oracle/graal/python/builtins/objects/str/NativeCharSequence.java

@@ -91,6 +91,7 @@ int length(InteropLibrary lib, CastToJavaIntExactNode castToJavaIntNode) {
         try {
             int arraySize = castToJavaIntNode.execute(lib.getArraySize(ptr));
             assert arraySize % elementSize == 0;
+            // we need to subtract the terminating null character
             return arraySize / elementSize;
         } catch (UnsupportedMessageException e) {
             throw CompilerDirectives.shouldNotReachHere("pointer of NativeCharSequence is not an array");