Issue #9979: Create function PyUnicode_AsWideCharString().

Victor Stinner · Victor Stinner · commit 137c34c02749 · 2010-09-29T10:25:54.000Z
diff --git a/Doc/c-api/unicode.rst b/Doc/c-api/unicode.rst
@@ -462,6 +462,19 @@ wchar_t support for platforms which support it:
    required by the application.
 
 
+.. cfunction:: wchar_t* PyUnicode_AsWideCharString(PyUnicodeObject *unicode, Py_ssize_t *size)
+
+   Convert the Unicode object to a wide character string. The output string
+   always ends with a nul character. If *size* is not *NULL*, write the number
+   of wide characters (including the nul character) into *\*size*.
+
+   Returns a buffer allocated by :cfunc:`PyMem_Alloc` (use :cfunc:`PyMem_Free`
+   to free it) on success. On error, returns *NULL*, *\*size* is undefined and
+   raises a :exc:`MemoryError`.
+
+   .. versionadded:: 3.2
+
+
 .. _builtincodecs:
 
 Built-in Codecs
diff --git a/Include/unicodeobject.h b/Include/unicodeobject.h
@@ -99,8 +99,8 @@ Copyright (c) Corporation for National Research Initiatives.
 #endif
 
 /* If the compiler provides a wchar_t type we try to support it
-   through the interface functions PyUnicode_FromWideChar() and
-   PyUnicode_AsWideChar(). */
+   through the interface functions PyUnicode_FromWideChar(),
+   PyUnicode_AsWideChar() and PyUnicode_AsWideCharString(). */
 
 #ifdef HAVE_USABLE_WCHAR_T
 # ifndef HAVE_WCHAR_H
@@ -156,6 +156,7 @@ typedef PY_UNICODE_TYPE Py_UNICODE;
 # define PyUnicode_AsUnicode PyUnicodeUCS2_AsUnicode
 # define PyUnicode_AsUnicodeEscapeString PyUnicodeUCS2_AsUnicodeEscapeString
 # define PyUnicode_AsWideChar PyUnicodeUCS2_AsWideChar
+# define PyUnicode_AsWideCharString PyUnicodeUCS2_AsWideCharString
 # define PyUnicode_ClearFreeList PyUnicodeUCS2_ClearFreelist
 # define PyUnicode_Compare PyUnicodeUCS2_Compare
 # define PyUnicode_CompareWithASCII PyUnicodeUCS2_CompareASCII
@@ -239,6 +240,7 @@ typedef PY_UNICODE_TYPE Py_UNICODE;
 # define PyUnicode_AsUnicode PyUnicodeUCS4_AsUnicode
 # define PyUnicode_AsUnicodeEscapeString PyUnicodeUCS4_AsUnicodeEscapeString
 # define PyUnicode_AsWideChar PyUnicodeUCS4_AsWideChar
+# define PyUnicode_AsWideCharString PyUnicodeUCS4_AsWideCharString
 # define PyUnicode_ClearFreeList PyUnicodeUCS4_ClearFreelist
 # define PyUnicode_Compare PyUnicodeUCS4_Compare
 # define PyUnicode_CompareWithASCII PyUnicodeUCS4_CompareWithASCII
@@ -570,6 +572,19 @@ PyAPI_FUNC(Py_ssize_t) PyUnicode_AsWideChar(
     Py_ssize_t size             /* size of buffer */
     );
 
+/* Convert the Unicode object to a wide character string. The output string
+   always ends with a nul character. If size is not NULL, write the number of
+   wide characters (including the nul character) into *size.
+
+   Returns a buffer allocated by PyMem_Alloc() (use PyMem_Free() to free it)
+   on success. On error, returns NULL, *size is undefined and raises a
+   MemoryError. */
+
+PyAPI_FUNC(wchar_t*) PyUnicode_AsWideCharString(
+    PyUnicodeObject *unicode,   /* Unicode object */
+    Py_ssize_t *size            /* number of characters of the result */
+    );
+
 #endif
 
 /* --- Unicode ordinals --------------------------------------------------- */
diff --git a/Misc/NEWS b/Misc/NEWS
@@ -10,6 +10,8 @@ What's New in Python 3.2 Alpha 3?
 Core and Builtins
 -----------------
 
+- Issue #9979: Create function PyUnicode_AsWideCharString().
+
 - Issue #7397: Mention that importlib.import_module() is probably what someone
   really wants to be using in __import__'s docstring.
 
diff --git a/Objects/unicodeobject.c b/Objects/unicodeobject.c
@@ -1153,9 +1153,26 @@ PyUnicode_FromFormat(const char *format, ...)
     return ret;
 }
 
-Py_ssize_t PyUnicode_AsWideChar(PyUnicodeObject *unicode,
-                                wchar_t *w,
-                                Py_ssize_t size)
+static void
+unicode_aswidechar(PyUnicodeObject *unicode,
+                   wchar_t *w,
+                   Py_ssize_t size)
+{
+#if Py_UNICODE_SIZE == SIZEOF_WCHAR_T
+    memcpy(w, unicode->str, size * sizeof(wchar_t));
+#else
+    register Py_UNICODE *u;
+    register Py_ssize_t i;
+    u = PyUnicode_AS_UNICODE(unicode);
+    for (i = size; i > 0; i--)
+        *w++ = *u++;
+#endif
+}
+
+Py_ssize_t
+PyUnicode_AsWideChar(PyUnicodeObject *unicode,
+                     wchar_t *w,
+                     Py_ssize_t size)
 {
     if (unicode == NULL) {
         PyErr_BadInternalCall();
@@ -1166,24 +1183,41 @@ Py_ssize_t PyUnicode_AsWideChar(PyUnicodeObject *unicode,
     if (size > PyUnicode_GET_SIZE(unicode))
         size = PyUnicode_GET_SIZE(unicode) + 1;
 
-#if Py_UNICODE_SIZE == SIZEOF_WCHAR_T
-    memcpy(w, unicode->str, size * sizeof(wchar_t));
-#else
-    {
-        register Py_UNICODE *u;
-        register Py_ssize_t i;
-        u = PyUnicode_AS_UNICODE(unicode);
-        for (i = size; i > 0; i--)
-            *w++ = *u++;
-    }
-#endif
+    unicode_aswidechar(unicode, w, size);
 
     if (size > PyUnicode_GET_SIZE(unicode))
         return PyUnicode_GET_SIZE(unicode);
     else
         return size;
 }
 
+wchar_t*
+PyUnicode_AsWideCharString(PyUnicodeObject *unicode,
+                           Py_ssize_t *size)
+{
+    wchar_t* buffer;
+    Py_ssize_t buflen;
+
+    if (unicode == NULL) {
+        PyErr_BadInternalCall();
+        return NULL;
+    }
+
+    if ((PY_SSIZE_T_MAX / sizeof(wchar_t) - 1) < PyUnicode_GET_SIZE(unicode)) {
+        PyErr_NoMemory();
+        return NULL;
+    }
+
+    buflen = PyUnicode_GET_SIZE(unicode) + 1; /* copy L'\0' */
+    buffer = PyMem_MALLOC(buflen * sizeof(wchar_t));
+    if (buffer == NULL) {
+        PyErr_NoMemory();
+        return NULL;
+    }
+    unicode_aswidechar(unicode, buffer, buflen);
+    return buffer;
+}
+
 #endif
 
 PyObject *PyUnicode_FromOrdinal(int ordinal)
