gh-111089: PyUnicode_AsUTF8() now raises on embedded NUL (#111091)

* PyUnicode_AsUTF8() now raises an exception if the string contains
  embedded null characters.
* Update related C API tests (test_capi.test_unicode).
* type_new_set_doc() uses PyUnicode_AsUTF8AndSize() to silently
  truncate doc containing null bytes.

Co-authored-by: Serhiy Storchaka <storchaka@gmail.com>

2 files changed, 10 insertions, 3 deletions

diff --git a/Objects/typeobject.c b/Objects/typeobject.c
index 3261a14..2508569 100644
--- a/Objects/typeobject.c
+++ b/Objects/typeobject.c
@@ -3499,13 +3499,14 @@ type_new_set_doc(PyTypeObject *type)
         return 0;
     }
 
-    const char *doc_str = PyUnicode_AsUTF8(doc);
+    Py_ssize_t doc_size;
+    const char *doc_str = PyUnicode_AsUTF8AndSize(doc, &doc_size);
     if (doc_str == NULL) {
         return -1;
     }
 
     // Silently truncate the docstring if it contains a null byte
-    Py_ssize_t size = strlen(doc_str) + 1;
+    Py_ssize_t size = doc_size + 1;
     char *tp_doc = (char *)PyObject_Malloc(size);
     if (tp_doc == NULL) {
         PyErr_NoMemory();
diff --git a/Objects/unicodeobject.c b/Objects/unicodeobject.c
index 33cbc98..07d1b6e 100644
--- a/Objects/unicodeobject.c
+++ b/Objects/unicodeobject.c
@@ -3837,7 +3837,13 @@ PyUnicode_AsUTF8AndSize(PyObject *unicode, Py_ssize_t *psize)
 const char *
 PyUnicode_AsUTF8(PyObject *unicode)
 {
-    return PyUnicode_AsUTF8AndSize(unicode, NULL);
+    Py_ssize_t size;
+    const char *utf8 = PyUnicode_AsUTF8AndSize(unicode, &size);
+    if (utf8 != NULL && strlen(utf8) != (size_t)size) {
+        PyErr_SetString(PyExc_ValueError, "embedded null character");
+        return NULL;
+    }
+    return utf8;
 }
 
 /*