Merge branch 'main' into gh-127794

srinivasreddy · web-flow · commit bcab963ac77d · 2024-12-11T18:08:17.000+05:30
diff --git a/Doc/c-api/init.rst b/Doc/c-api/init.rst
@@ -567,6 +567,15 @@ Initializing and finalizing the interpreter
    customized Python that always runs in isolated mode using
    :c:func:`Py_RunMain`.
 
+.. c:function:: int PyUnstable_AtExit(PyInterpreterState *interp, void (*func)(void *), void *data)
+
+   Register an :mod:`atexit` callback for the target interpreter *interp*.
+   This is similar to :c:func:`Py_AtExit`, but takes an explicit interpreter and
+   data pointer for the callback.
+
+   The :term:`GIL` must be held for *interp*.
+
+   .. versionadded:: 3.13
 
 Process-wide parameters
 =======================
diff --git a/Doc/c-api/sys.rst b/Doc/c-api/sys.rst
@@ -426,3 +426,7 @@ Process Control
    function registered last is called first. Each cleanup function will be called
    at most once.  Since Python's internal finalization will have completed before
    the cleanup function, no Python APIs should be called by *func*.
+
+   .. seealso::
+
+      :c:func:`PyUnstable_AtExit` for passing a ``void *data`` argument.
diff --git a/Include/cpython/tracemalloc.h b/Include/cpython/tracemalloc.h
@@ -1,6 +1,9 @@
 #ifndef Py_LIMITED_API
 #ifndef Py_TRACEMALLOC_H
 #define Py_TRACEMALLOC_H
+#ifdef __cplusplus
+extern "C" {
+#endif
 
 /* Track an allocated memory block in the tracemalloc module.
    Return 0 on success, return -1 on error (failed to allocate memory to store
@@ -22,5 +25,8 @@ PyAPI_FUNC(int) PyTraceMalloc_Untrack(
     unsigned int domain,
     uintptr_t ptr);
 
+#ifdef __cplusplus
+}
+#endif
 #endif  // !Py_TRACEMALLOC_H
 #endif  // !Py_LIMITED_API
diff --git a/Include/internal/pycore_atexit.h b/Include/internal/pycore_atexit.h
@@ -44,7 +44,6 @@ typedef struct {
 
 struct atexit_state {
     atexit_callback *ll_callbacks;
-    atexit_callback *last_ll_callback;
 
     // XXX The rest of the state could be moved to the atexit module state
     // and a low-level callback added for it during module exec.
diff --git a/Lib/pickle.py b/Lib/pickle.py
@@ -1387,7 +1387,7 @@ def load_int(self):
         elif data == TRUE[1:]:
             val = True
         else:
-            val = int(data, 0)
+            val = int(data)
         self.append(val)
     dispatch[INT[0]] = load_int
 
@@ -1407,7 +1407,7 @@ def load_long(self):
         val = self.readline()[:-1]
         if val and val[-1] == b'L'[0]:
             val = val[:-1]
-        self.append(int(val, 0))
+        self.append(int(val))
     dispatch[LONG[0]] = load_long
 
     def load_long1(self):
diff --git a/Lib/test/pickletester.py b/Lib/test/pickletester.py
@@ -1012,6 +1012,26 @@ def test_constants(self):
         self.assertIs(self.loads(b'I01\n.'), True)
         self.assertIs(self.loads(b'I00\n.'), False)
 
+    def test_zero_padded_integers(self):
+        self.assertEqual(self.loads(b'I010\n.'), 10)
+        self.assertEqual(self.loads(b'I-010\n.'), -10)
+        self.assertEqual(self.loads(b'I0010\n.'), 10)
+        self.assertEqual(self.loads(b'I-0010\n.'), -10)
+        self.assertEqual(self.loads(b'L010\n.'), 10)
+        self.assertEqual(self.loads(b'L-010\n.'), -10)
+        self.assertEqual(self.loads(b'L0010\n.'), 10)
+        self.assertEqual(self.loads(b'L-0010\n.'), -10)
+        self.assertEqual(self.loads(b'L010L\n.'), 10)
+        self.assertEqual(self.loads(b'L-010L\n.'), -10)
+
+    def test_nondecimal_integers(self):
+        self.assertRaises(ValueError, self.loads, b'I0b10\n.')
+        self.assertRaises(ValueError, self.loads, b'I0o10\n.')
+        self.assertRaises(ValueError, self.loads, b'I0x10\n.')
+        self.assertRaises(ValueError, self.loads, b'L0b10L\n.')
+        self.assertRaises(ValueError, self.loads, b'L0o10L\n.')
+        self.assertRaises(ValueError, self.loads, b'L0x10L\n.')
+
     def test_empty_bytestring(self):
         # issue 11286
         empty = self.loads(b'\x80\x03U\x00q\x00.', encoding='koi8-r')
diff --git a/Lib/test/test_pickletools.py b/Lib/test/test_pickletools.py
@@ -443,6 +443,43 @@ def test_persid(self):
 highest protocol among opcodes = 0
 ''')
 
+    def test_constants(self):
+        self.check_dis(b"(NI00\nI01\n\x89\x88t.", '''\
+    0: (    MARK
+    1: N        NONE
+    2: I        INT        False
+    6: I        INT        True
+   10: \\x89     NEWFALSE
+   11: \\x88     NEWTRUE
+   12: t        TUPLE      (MARK at 0)
+   13: .    STOP
+highest protocol among opcodes = 2
+''')
+
+    def test_integers(self):
+        self.check_dis(b"(I0\nI1\nI10\nI011\nL12\nL13L\nL014\nL015L\nt.", '''\
+    0: (    MARK
+    1: I        INT        0
+    4: I        INT        1
+    7: I        INT        10
+   11: I        INT        11
+   16: L        LONG       12
+   20: L        LONG       13
+   25: L        LONG       14
+   30: L        LONG       15
+   36: t        TUPLE      (MARK at 0)
+   37: .    STOP
+highest protocol among opcodes = 0
+''')
+
+    def test_nondecimal_integers(self):
+        self.check_dis_error(b'I0b10\n.', '', 'invalid literal for int')
+        self.check_dis_error(b'I0o10\n.', '', 'invalid literal for int')
+        self.check_dis_error(b'I0x10\n.', '', 'invalid literal for int')
+        self.check_dis_error(b'L0b10L\n.', '', 'invalid literal for int')
+        self.check_dis_error(b'L0o10L\n.', '', 'invalid literal for int')
+        self.check_dis_error(b'L0x10L\n.', '', 'invalid literal for int')
+
 
 class MiscTestCase(unittest.TestCase):
     def test__all__(self):
diff --git a/Misc/NEWS.d/next/C_API/2024-12-10-14-25-22.gh-issue-127791.YRw4GU.rst b/Misc/NEWS.d/next/C_API/2024-12-10-14-25-22.gh-issue-127791.YRw4GU.rst
@@ -0,0 +1,2 @@
+Fix loss of callbacks after more than one call to
+:c:func:`PyUnstable_AtExit`.
diff --git a/Misc/NEWS.d/next/Library/2024-11-20-21-20-56.gh-issue-126992.RbU0FZ.rst b/Misc/NEWS.d/next/Library/2024-11-20-21-20-56.gh-issue-126992.RbU0FZ.rst
@@ -0,0 +1 @@
+Fix LONG and INT opcodes to only use base 10 for string to integer conversion in :mod:`pickle`.
diff --git a/Modules/_pickle.c b/Modules/_pickle.c
@@ -5211,16 +5211,14 @@ load_int(PickleState *state, UnpicklerObject *self)
         return bad_readline(state);
 
     errno = 0;
-    /* XXX: Should the base argument of strtol() be explicitly set to 10?
-       XXX(avassalotti): Should this uses PyOS_strtol()? */
-    x = strtol(s, &endptr, 0);
+    /* XXX(avassalotti): Should this uses PyOS_strtol()? */
+    x = strtol(s, &endptr, 10);
 
     if (errno || (*endptr != '\n' && *endptr != '\0')) {
         /* Hm, maybe we've got something long.  Let's try reading
          * it as a Python int object. */
         errno = 0;
-        /* XXX: Same thing about the base here. */
-        value = PyLong_FromString(s, NULL, 0);
+        value = PyLong_FromString(s, NULL, 10);
         if (value == NULL) {
             PyErr_SetString(PyExc_ValueError,
                             "could not convert string to int");
@@ -5370,8 +5368,7 @@ load_long(PickleState *state, UnpicklerObject *self)
        the 'L' to be present. */
     if (s[len-2] == 'L')
         s[len-2] = '\0';
-    /* XXX: Should the base argument explicitly set to 10? */
-    value = PyLong_FromString(s, NULL, 0);
+    value = PyLong_FromString(s, NULL, 10);
     if (value == NULL)
         return -1;
 
diff --git a/Modules/_testcapimodule.c b/Modules/_testcapimodule.c
@@ -3353,6 +3353,53 @@ type_freeze(PyObject *module, PyObject *args)
     Py_RETURN_NONE;
 }
 
+struct atexit_data {
+    int called;
+    PyThreadState *tstate;
+    PyInterpreterState *interp;
+};
+
+static void
+atexit_callback(void *data)
+{
+    struct atexit_data *at_data = (struct atexit_data *)data;
+    // Ensure that the callback is from the same interpreter
+    assert(PyThreadState_Get() == at_data->tstate);
+    assert(PyInterpreterState_Get() == at_data->interp);
+    ++at_data->called;
+}
+
+static PyObject *
+test_atexit(PyObject *self, PyObject *Py_UNUSED(args))
+{
+    PyThreadState *oldts = PyThreadState_Swap(NULL);
+    PyThreadState *tstate = Py_NewInterpreter();
+
+    struct atexit_data data = {0};
+    data.tstate = PyThreadState_Get();
+    data.interp = PyInterpreterState_Get();
+
+    int amount = 10;
+    for (int i = 0; i < amount; ++i)
+    {
+        int res = PyUnstable_AtExit(tstate->interp, atexit_callback, (void *)&data);
+        if (res < 0) {
+            Py_EndInterpreter(tstate);
+            PyThreadState_Swap(oldts);
+            PyErr_SetString(PyExc_RuntimeError, "atexit callback failed");
+            return NULL;
+        }
+    }
+
+    Py_EndInterpreter(tstate);
+    PyThreadState_Swap(oldts);
+
+    if (data.called != amount) {
+        PyErr_SetString(PyExc_RuntimeError, "atexit callback not called");
+        return NULL;
+    }
+    Py_RETURN_NONE;
+}
 
 static PyMethodDef TestMethods[] = {
     {"set_errno",               set_errno,                       METH_VARARGS},
@@ -3495,6 +3542,7 @@ static PyMethodDef TestMethods[] = {
     {"test_critical_sections", test_critical_sections, METH_NOARGS},
     {"finalize_thread_hang", finalize_thread_hang, METH_O, NULL},
     {"type_freeze", type_freeze, METH_VARARGS},
+    {"test_atexit", test_atexit, METH_NOARGS},
     {NULL, NULL} /* sentinel */
 };
 
diff --git a/Modules/_testinternalcapi.c b/Modules/_testinternalcapi.c
@@ -1236,39 +1236,6 @@ unicode_transformdecimalandspacetoascii(PyObject *self, PyObject *arg)
     return _PyUnicode_TransformDecimalAndSpaceToASCII(arg);
 }
 
-
-struct atexit_data {
-    int called;
-};
-
-static void
-callback(void *data)
-{
-    ((struct atexit_data *)data)->called += 1;
-}
-
-static PyObject *
-test_atexit(PyObject *self, PyObject *Py_UNUSED(args))
-{
-    PyThreadState *oldts = PyThreadState_Swap(NULL);
-    PyThreadState *tstate = Py_NewInterpreter();
-
-    struct atexit_data data = {0};
-    int res = PyUnstable_AtExit(tstate->interp, callback, (void *)&data);
-    Py_EndInterpreter(tstate);
-    PyThreadState_Swap(oldts);
-    if (res < 0) {
-        return NULL;
-    }
-
-    if (data.called == 0) {
-        PyErr_SetString(PyExc_RuntimeError, "atexit callback not called");
-        return NULL;
-    }
-    Py_RETURN_NONE;
-}
-
-
 static PyObject *
 test_pyobject_is_freed(const char *test_name, PyObject *op)
 {
@@ -2128,7 +2095,6 @@ static PyMethodDef module_functions[] = {
     {"_PyTraceMalloc_GetTraceback", tracemalloc_get_traceback, METH_VARARGS},
     {"test_tstate_capi", test_tstate_capi, METH_NOARGS, NULL},
     {"_PyUnicode_TransformDecimalAndSpaceToASCII", unicode_transformdecimalandspacetoascii, METH_O},
-    {"test_atexit", test_atexit, METH_NOARGS},
     {"check_pyobject_forbidden_bytes_is_freed",
                             check_pyobject_forbidden_bytes_is_freed, METH_NOARGS},
     {"check_pyobject_freed_is_freed", check_pyobject_freed_is_freed, METH_NOARGS},
diff --git a/Modules/atexitmodule.c b/Modules/atexitmodule.c
@@ -27,7 +27,10 @@ int
 PyUnstable_AtExit(PyInterpreterState *interp,
                   atexit_datacallbackfunc func, void *data)
 {
-    assert(interp == _PyInterpreterState_GET());
+    PyThreadState *tstate = _PyThreadState_GET();
+    _Py_EnsureTstateNotNULL(tstate);
+    assert(tstate->interp == interp);
+
     atexit_callback *callback = PyMem_Malloc(sizeof(atexit_callback));
     if (callback == NULL) {
         PyErr_NoMemory();
@@ -38,12 +41,13 @@ PyUnstable_AtExit(PyInterpreterState *interp,
     callback->next = NULL;
 
     struct atexit_state *state = &interp->atexit;
-    if (state->ll_callbacks == NULL) {
+    atexit_callback *top = state->ll_callbacks;
+    if (top == NULL) {
         state->ll_callbacks = callback;
-        state->last_ll_callback = callback;
     }
     else {
-        state->last_ll_callback->next = callback;
+        callback->next = top;
+        state->ll_callbacks = callback;
     }
     return 0;
 }

Original file line number	Diff line number	Diff line change
`@@ -0,0 +1,2 @@`
	`1`	`+Fix loss of callbacks after more than one call to`
	`2`	+:c:func:`PyUnstable_AtExit`.
Original file line number	Diff line number	Diff line change
`@@ -0,0 +1 @@`
	`1`	+Fix LONG and INT opcodes to only use base 10 for string to integer conversion in :mod:`pickle`.