gh-152548: Add a test.support.isolation.runInSubprocess() decorator (GH-152551)

Run a test in a fresh interpreter subprocess, so that it does not share global
or interpreter state with the rest of the test run.  It can decorate a test
method (only that method runs in a subprocess) or a TestCase subclass (the
whole class runs in one subprocess, with its setUpClass()/setUp()/tearDown()/
tearDownClass() running there rather than in the parent).

Failures, errors and skips, including those of individual subtests, are
reported for the test; a failure or an error shows the original subprocess
traceback.  The subprocess inherits the parent's resource, memory, verbosity
and failfast configuration, so that requires_resource(), bigmemtest() and
similar behave the same in both processes.  A decorated test is skipped where
subprocesses are unavailable, since it must spawn one.

The test.support.isolation.runningInSubprocess flag is true in the subprocess,
so that fixtures can choose what to run there.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
(cherry picked from commit bde526d)

Author: serhiy-storchaka

Doc/library/test.rst

@@ -963,6 +963,59 @@ The :mod:`!test.support` module defines the following functions:
    :mod:`tracemalloc` is enabled.
 
 
+.. currentmodule:: test.support.isolation
+
+.. decorator:: runInSubprocess()
+
+   Decorator that runs the decorated test in a fresh interpreter subprocess, in
+   isolation, so that it does not share global or interpreter state with the
+   rest of the test run.  It can decorate a test method or a whole
+   :class:`~unittest.TestCase` subclass.  Decorated methods must take no extra
+   arguments.  A failure, error or skip in the subprocess is reported for the
+   corresponding test, and individual :meth:`subtests
+   <unittest.TestCase.subTest>` that fail or are skipped are reported
+   individually.  A reported failure or error shows the original subprocess
+   traceback as the cause of the exception.
+
+   When a **method** is decorated, only that method runs in a subprocess; all
+   fixtures (:meth:`~unittest.TestCase.setUp` / :meth:`~unittest.TestCase.tearDown`,
+   :meth:`~unittest.TestCase.setUpClass` / :meth:`~unittest.TestCase.tearDownClass`
+   and ``setUpModule()`` / ``tearDownModule()``) run both in the parent process
+   (as usual) and in the subprocess around the method.
+
+   When a **class** is decorated, the whole class runs in a single subprocess,
+   and :meth:`~unittest.TestCase.setUpClass`,
+   :meth:`~unittest.TestCase.tearDownClass`, :meth:`~unittest.TestCase.setUp`
+   and :meth:`~unittest.TestCase.tearDown` run once each in the subprocess and
+   are skipped in the parent process.  A failure or skip of
+   :meth:`~unittest.TestCase.setUpClass` in the subprocess is reported for the
+   whole class.  ``setUpModule()`` cannot be controlled by a class decorator,
+   so it still runs in the parent process too; test it with
+   :data:`runningInSubprocess` if needed.
+
+   The subprocess inherits the enabled resources (``-u``), memory limit
+   (``-M``) and verbosity (``-v``) of the parent test run, so that
+   :func:`~test.support.requires_resource`, :func:`~test.support.requires`,
+   :func:`~test.support.bigmemtest` and the like behave consistently in both
+   processes.
+
+   The test is skipped on platforms without subprocess support.
+
+
+.. data:: runningInSubprocess
+
+   ``True`` while the code runs in the isolated subprocess spawned by
+   :func:`runInSubprocess`, and ``False`` otherwise (including in the parent
+   process and in a normal, non-isolated test run).  Fixtures such as
+   :meth:`~unittest.TestCase.setUp`, :meth:`~unittest.TestCase.tearDown`,
+   :meth:`~unittest.TestCase.setUpClass`, :meth:`~unittest.TestCase.tearDownClass`,
+   ``setUpModule()`` and ``tearDownModule()`` can test it to choose which code
+   to run in the subprocess.
+
+
+.. currentmodule:: test.support
+
+
 .. function:: check_free_after_iterating(test, iter, cls, args=())
 
    Assert instances of *cls* are deallocated after iterating.

Lib/test/_isolated_sample.py

@@ -0,0 +1,111 @@
+"""Sample tests driven by test.test_support.TestIsolated.
+
+This module is imported, never run as a test file, so that
+:func:`test.support.isolation.runInSubprocess` has a real, importable target to run in
+a subprocess.  Several of these tests fail, error or are skipped on purpose.
+"""
+
+import time
+import unittest
+from test.support import isolation
+
+# DurationSample sleeps this long in the subprocess; a parent-reported duration
+# close to it proves the subprocess timing was forwarded, not the replay time.
+DURATION_SLEEP = 0.2
+
+
+class MethodSample(unittest.TestCase):
+
+    @isolation.runInSubprocess()
+    def test_pass(self):
+        self.assertTrue(isolation.runningInSubprocess)
+
+    @isolation.runInSubprocess()
+    def test_fail(self):
+        self.assertEqual(1, 2)
+
+    @isolation.runInSubprocess()
+    def test_error(self):
+        raise RuntimeError('boom')
+
+    @isolation.runInSubprocess()
+    def test_skip(self):
+        self.skipTest('nope')
+
+    @isolation.runInSubprocess()
+    @unittest.expectedFailure
+    def test_expected_failure(self):
+        self.assertEqual(1, 2)
+
+    @isolation.runInSubprocess()
+    @unittest.expectedFailure
+    def test_unexpected_success(self):
+        pass
+
+
+@isolation.runInSubprocess()
+class ClassSample(unittest.TestCase):
+
+    def test_pass(self):
+        self.assertTrue(isolation.runningInSubprocess)
+
+    def test_fail(self):
+        self.assertEqual(1, 2)
+
+    @unittest.expectedFailure
+    def test_expected_failure(self):
+        self.assertEqual(1, 2)
+
+
+class SubtestSample(unittest.TestCase):
+
+    @isolation.runInSubprocess()
+    def test_subtests(self):
+        for i in range(3):
+            with self.subTest(i=i):
+                self.assertNotEqual(i, 1)
+
+
+@isolation.runInSubprocess()
+class DurationSample(unittest.TestCase):
+
+    def test_slow(self):
+        time.sleep(DURATION_SLEEP)
+
+
+@isolation.runInSubprocess()
+class SubclassingSample(unittest.TestCase):
+    # setUpClass must run bound to the runtime class, so a subclass sees its own
+    # name here rather than the base class's.
+
+    @classmethod
+    def setUpClass(cls):
+        cls.setup_class_name = cls.__name__
+
+    def setUp(self):
+        self.set_up = True
+
+    def test_runtime_class(self):
+        self.assertEqual(self.setup_class_name, type(self).__name__)
+
+
+class SubclassSample(SubclassingSample):
+    # What a subclass adds or overrides must run in the subprocess too.
+
+    def setUp(self):
+        super().setUp()
+        self.set_up_in_subclass = True
+
+    def test_added_in_subclass(self):
+        self.assertTrue(isolation.runningInSubprocess)
+        self.assertTrue(self.set_up)
+        self.assertTrue(self.set_up_in_subclass)
+
+
+class BrokenSubclassSample(SubclassingSample):
+    # An overriding setUpClass() that does not call super() bypasses the
+    # subprocess entirely.
+
+    @classmethod
+    def setUpClass(cls):
+        pass