gh-138912: Improve MATCH_CLASS opcode performance (GH-138915)

Only check for duplicates if there is at least one positional pattern.
With a test case for duplicate keyword attributes.

diff --git a/Lib/test/test_syntax.py b/Lib/test/test_syntax.py
index 19427f2469ec43cf783f4a1e5b9f387518ff8476..e48749626fccad40b96ed22e1b4028e38a5f6cfb 100644 (file)
--- a/Lib/test/test_syntax.py
+++ b/Lib/test/test_syntax.py
@@ -2345,6 +2345,12 @@ Invalid pattern matching constructs:
     Traceback (most recent call last):
     SyntaxError: positional patterns follow keyword patterns
 
+    >>> match ...:
+    ...   case Foo(y=1, x=2, y=3):
+    ...     ...
+    Traceback (most recent call last):
+    SyntaxError: attribute name repeated in class pattern: y
+
     >>> match ...:
     ...   case C(a=b, c, d=e, f, g=h, i, j=k, ...):
     ...     ...

diff --git a/Misc/NEWS.d/next/Core_and_Builtins/2025-09-15-13-28-48.gh-issue-138912.61EYbn.rst b/Misc/NEWS.d/next/Core_and_Builtins/2025-09-15-13-28-48.gh-issue-138912.61EYbn.rst
new file mode 100644 (file)
index 0000000..f5d312a
--- /dev/null
+++ b/Misc/NEWS.d/next/Core_and_Builtins/2025-09-15-13-28-48.gh-issue-138912.61EYbn.rst
@@ -0,0 +1 @@
+Improve :opcode:`MATCH_CLASS` performance by up to 52% in certain cases. Patch by Marc Mueller.

diff --git a/Python/ceval.c b/Python/ceval.c
index 758b142d7a720e60bdc37631a9e3d817af362213..ab2eef560370f5087256307b8d992d73b4f5e1e7 100644 (file)
--- a/Python/ceval.c
+++ b/Python/ceval.c
@@ -509,15 +509,18 @@ match_class_attr(PyThreadState *tstate, PyObject *subject, PyObject *type,
                  PyObject *name, PyObject *seen)
 {
     assert(PyUnicode_CheckExact(name));
-    assert(PySet_CheckExact(seen));
-    if (PySet_Contains(seen, name) || PySet_Add(seen, name)) {
-        if (!_PyErr_Occurred(tstate)) {
-            // Seen it before!
-            _PyErr_Format(tstate, PyExc_TypeError,
-                          "%s() got multiple sub-patterns for attribute %R",
-                          ((PyTypeObject*)type)->tp_name, name);
+    // Only check for duplicates if seen is not NULL.
+    if (seen != NULL) {
+        assert(PySet_CheckExact(seen));
+        if (PySet_Contains(seen, name) || PySet_Add(seen, name)) {
+            if (!_PyErr_Occurred(tstate)) {
+                // Seen it before!
+                _PyErr_Format(tstate, PyExc_TypeError,
+                            "%s() got multiple sub-patterns for attribute %R",
+                            ((PyTypeObject*)type)->tp_name, name);
+            }
+            return NULL;
         }
-        return NULL;
     }
     PyObject *attr;
     (void)PyObject_GetOptionalAttr(subject, name, &attr);
@@ -540,14 +543,26 @@ _PyEval_MatchClass(PyThreadState *tstate, PyObject *subject, PyObject *type,
     if (PyObject_IsInstance(subject, type) <= 0) {
         return NULL;
     }
+    // Short circuit if there aren't any arguments:
+    Py_ssize_t nkwargs = PyTuple_GET_SIZE(kwargs);
+    Py_ssize_t nattrs = nargs + nkwargs;
+    if (!nattrs) {
+        return PyTuple_New(0);
+    }
     // So far so good:
-    PyObject *seen = PySet_New(NULL);
-    if (seen == NULL) {
-        return NULL;
+    PyObject *seen = NULL;
+    // Only check for duplicates if there is at least one positional attribute
+    // and two or more attributes in total. Duplicate keyword attributes are
+    // detected during the compile stage and raise a SyntaxError.
+    if (nargs > 0 && nattrs > 1) {
+        seen = PySet_New(NULL);
+        if (seen == NULL) {
+            return NULL;
+        }
     }
-    PyObject *attrs = PyList_New(0);
+    PyObject *attrs = PyTuple_New(nattrs);
     if (attrs == NULL) {
-        Py_DECREF(seen);
+        Py_XDECREF(seen);
         return NULL;
     }
     // NOTE: From this point on, goto fail on failure:
@@ -588,9 +603,8 @@ _PyEval_MatchClass(PyThreadState *tstate, PyObject *subject, PyObject *type,
         }
         if (match_self) {
             // Easy. Copy the subject itself, and move on to kwargs.
-            if (PyList_Append(attrs, subject) < 0) {
-                goto fail;
-            }
+            assert(PyTuple_GET_ITEM(attrs, 0) == NULL);
+            PyTuple_SET_ITEM(attrs, 0, Py_NewRef(subject));
         }
         else {
             for (Py_ssize_t i = 0; i < nargs; i++) {
@@ -606,36 +620,29 @@ _PyEval_MatchClass(PyThreadState *tstate, PyObject *subject, PyObject *type,
                 if (attr == NULL) {
                     goto fail;
                 }
-                if (PyList_Append(attrs, attr) < 0) {
-                    Py_DECREF(attr);
-                    goto fail;
-                }
-                Py_DECREF(attr);
+                assert(PyTuple_GET_ITEM(attrs, i) == NULL);
+                PyTuple_SET_ITEM(attrs, i, attr);
             }
         }
         Py_CLEAR(match_args);
     }
     // Finally, the keyword subpatterns:
-    for (Py_ssize_t i = 0; i < PyTuple_GET_SIZE(kwargs); i++) {
+    for (Py_ssize_t i = 0; i < nkwargs; i++) {
         PyObject *name = PyTuple_GET_ITEM(kwargs, i);
         PyObject *attr = match_class_attr(tstate, subject, type, name, seen);
         if (attr == NULL) {
             goto fail;
         }
-        if (PyList_Append(attrs, attr) < 0) {
-            Py_DECREF(attr);
-            goto fail;
-        }
-        Py_DECREF(attr);
+        assert(PyTuple_GET_ITEM(attrs, nargs + i) == NULL);
+        PyTuple_SET_ITEM(attrs, nargs + i, attr);
     }
-    Py_SETREF(attrs, PyList_AsTuple(attrs));
-    Py_DECREF(seen);
+    Py_XDECREF(seen);
     return attrs;
 fail:
     // We really don't care whether an error was raised or not... that's our
     // caller's problem. All we know is that the match failed.
     Py_XDECREF(match_args);
-    Py_DECREF(seen);
+    Py_XDECREF(seen);
     Py_DECREF(attrs);
     return NULL;
 }