gh-109162: libregrtest: add worker.py (#109229)

Add new worker.py file:

* Move create_worker_process() and worker_process() to this file.
* Add main() function to worker.py. create_worker_process() now
  runs the command: "python -m test.libregrtest.worker JSON".
* create_worker_process() now starts the worker process in the
  current working directory. Regrtest now gets the absolute path of
  the reflog.txt filename: -R command line option filename.
* Remove --worker-json command line option.
  Remove test_regrtest.test_worker_json().

Related changes:

* Add write_json() and from_json() methods to TestResult.
* Rename select_temp_dir() to get_temp_dir() and move it to utils.
* Rename make_temp_dir() to get_work_dir() and move it to utils.
  It no longer calls os.makedirs(): Regrtest.main() now calls it.
* Move fix_umask() to utils. The function is now called by
  setup_tests().
* Move StrPath to utils.
* Add exit_timeout() context manager to utils.
* RunTests: Replace junit_filename (StrPath) with use_junit (bool).
This commit is contained in:
Victor Stinner 2023-09-11 01:11:22 +02:00 committed by GitHub
parent e55aab9578
commit a939b65aa6
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
10 changed files with 238 additions and 213 deletions

View file

@ -216,7 +216,6 @@ def _create_parser():
group.add_argument('--wait', action='store_true',
help='wait for user input, e.g., allow a debugger '
'to be attached')
group.add_argument('--worker-json', metavar='ARGS')
group.add_argument('-S', '--start', metavar='START',
help='the name of the test at which to start.' +
more_details)

View file

@ -1,34 +1,27 @@
import faulthandler
import locale
import os
import platform
import random
import re
import sys
import sysconfig
import tempfile
import time
import unittest
from test import support
from test.support import os_helper
from test.libregrtest.cmdline import _parse_args, Namespace
from test.libregrtest.logger import Logger
from test.libregrtest.runtest import (
findtests, split_test_packages, run_single_test, abs_module_name,
PROGRESS_MIN_TIME, State, RunTests, HuntRefleak,
FilterTuple, TestList, StrPath, StrJSON, TestName)
FilterTuple, TestList, StrJSON, TestName)
from test.libregrtest.setup import setup_tests, setup_test_dir
from test.libregrtest.pgo import setup_pgo_tests
from test.libregrtest.results import TestResults
from test.libregrtest.utils import (strip_py_suffix, count, format_duration,
printlist, get_build_info)
from test import support
from test.support import os_helper
from test.support import threading_helper
# bpo-38203: Maximum delay in seconds to exit Python (call Py_Finalize()).
# Used to protect against threading._shutdown() hang.
# Must be smaller than buildbot "1200 seconds without output" limit.
EXIT_TIMEOUT = 120.0
from test.libregrtest.utils import (
strip_py_suffix, count, format_duration, StrPath,
printlist, get_build_info, get_temp_dir, get_work_dir, exit_timeout)
class Regrtest:
@ -104,7 +97,9 @@ class Regrtest:
self.verbose: bool = ns.verbose
self.quiet: bool = ns.quiet
if ns.huntrleaks:
self.hunt_refleak: HuntRefleak = HuntRefleak(*ns.huntrleaks)
warmups, runs, filename = ns.huntrleaks
filename = os.path.abspath(filename)
self.hunt_refleak: HuntRefleak = HuntRefleak(warmups, runs, filename)
else:
self.hunt_refleak = None
self.test_dir: StrPath | None = ns.testdir
@ -454,64 +449,6 @@ class Regrtest:
state = self.get_state()
print(f"Result: {state}")
@staticmethod
def fix_umask():
if support.is_emscripten:
# Emscripten has default umask 0o777, which breaks some tests.
# see https://github.com/emscripten-core/emscripten/issues/17269
old_mask = os.umask(0)
if old_mask == 0o777:
os.umask(0o027)
else:
os.umask(old_mask)
@staticmethod
def select_temp_dir(tmp_dir):
if tmp_dir:
tmp_dir = os.path.expanduser(tmp_dir)
else:
# When tests are run from the Python build directory, it is best practice
# to keep the test files in a subfolder. This eases the cleanup of leftover
# files using the "make distclean" command.
if sysconfig.is_python_build():
tmp_dir = sysconfig.get_config_var('abs_builddir')
if tmp_dir is None:
# bpo-30284: On Windows, only srcdir is available. Using
# abs_builddir mostly matters on UNIX when building Python
# out of the source tree, especially when the source tree
# is read only.
tmp_dir = sysconfig.get_config_var('srcdir')
tmp_dir = os.path.join(tmp_dir, 'build')
else:
tmp_dir = tempfile.gettempdir()
return os.path.abspath(tmp_dir)
def is_worker(self):
return (self.worker_json is not None)
@staticmethod
def make_temp_dir(tmp_dir: StrPath, is_worker: bool):
os.makedirs(tmp_dir, exist_ok=True)
# Define a writable temp dir that will be used as cwd while running
# the tests. The name of the dir includes the pid to allow parallel
# testing (see the -j option).
# Emscripten and WASI have stubbed getpid(), Emscripten has only
# milisecond clock resolution. Use randint() instead.
if sys.platform in {"emscripten", "wasi"}:
nounce = random.randint(0, 1_000_000)
else:
nounce = os.getpid()
if is_worker:
work_dir = 'test_python_worker_{}'.format(nounce)
else:
work_dir = 'test_python_{}'.format(nounce)
work_dir += os_helper.FS_NONASCII
work_dir = os.path.join(tmp_dir, work_dir)
return work_dir
@staticmethod
def cleanup_temp_dir(tmp_dir: StrPath):
import glob
@ -534,17 +471,16 @@ class Regrtest:
strip_py_suffix(self.cmdline_args)
self.tmp_dir = self.select_temp_dir(self.tmp_dir)
self.fix_umask()
self.tmp_dir = get_temp_dir(self.tmp_dir)
if self.want_cleanup:
self.cleanup_temp_dir(self.tmp_dir)
sys.exit(0)
work_dir = self.make_temp_dir(self.tmp_dir, self.is_worker())
os.makedirs(self.tmp_dir, exist_ok=True)
work_dir = get_work_dir(parent_dir=self.tmp_dir)
try:
with exit_timeout():
# Run the tests in a context manager that temporarily changes the
# CWD to a temporary and writable directory. If it's not possible
# to create or change the CWD, the original CWD will be used.
@ -556,13 +492,6 @@ class Regrtest:
# processes.
self._main()
except SystemExit as exc:
# bpo-38203: Python can hang at exit in Py_Finalize(), especially
# on threading._shutdown() call: put a timeout
if threading_helper.can_start_thread:
faulthandler.dump_traceback_later(EXIT_TIMEOUT, exit=True)
sys.exit(exc.code)
def create_run_tests(self):
return RunTests(
@ -579,7 +508,7 @@ class Regrtest:
quiet=self.quiet,
hunt_refleak=self.hunt_refleak,
test_dir=self.test_dir,
junit_filename=self.junit_filename,
use_junit=(self.junit_filename is not None),
memory_limit=self.memory_limit,
gc_threshold=self.gc_threshold,
use_resources=self.use_resources,
@ -634,11 +563,6 @@ class Regrtest:
self.fail_rerun)
def _main(self):
if self.is_worker():
from test.libregrtest.runtest_mp import worker_process
worker_process(self.worker_json)
return
if self.want_wait:
input("Press any key to continue...")

View file

@ -68,7 +68,6 @@ def runtest_refleak(test_name, test_func,
warmups = hunt_refleak.warmups
runs = hunt_refleak.runs
filename = hunt_refleak.filename
filename = os.path.join(os_helper.SAVEDCWD, filename)
repcount = warmups + runs
# Pre-allocate to ensure that the loop doesn't allocate anything new

View file

@ -2,9 +2,10 @@ import sys
from test.support import TestStats
from test.libregrtest.runtest import (
TestName, TestTuple, TestList, FilterDict, StrPath, State,
TestName, TestTuple, TestList, FilterDict, State,
TestResult, RunTests)
from test.libregrtest.utils import printlist, count, format_duration
from test.libregrtest.utils import (
printlist, count, format_duration, StrPath)
EXITCODE_BAD_TEST = 2

View file

@ -17,11 +17,11 @@ from test.support import TestStats
from test.support import os_helper
from test.support import threading_helper
from test.libregrtest.save_env import saved_test_environment
from test.libregrtest.utils import clear_caches, format_duration, print_warning
from test.libregrtest.utils import (
clear_caches, format_duration, print_warning, StrPath)
StrJSON = str
StrPath = str
TestName = str
TestTuple = tuple[TestName, ...]
TestList = list[TestName]
@ -215,6 +215,33 @@ class TestResult:
return None
return tuple(match_tests)
def write_json(self, file) -> None:
json.dump(self, file, cls=_EncodeTestResult)
@staticmethod
def from_json(worker_json) -> 'TestResult':
return json.loads(worker_json, object_hook=_decode_test_result)
class _EncodeTestResult(json.JSONEncoder):
def default(self, o: Any) -> dict[str, Any]:
if isinstance(o, TestResult):
result = dataclasses.asdict(o)
result["__test_result__"] = o.__class__.__name__
return result
else:
return super().default(o)
def _decode_test_result(data: dict[str, Any]) -> TestResult | dict[str, Any]:
if "__test_result__" in data:
data.pop('__test_result__')
if data['stats'] is not None:
data['stats'] = TestStats(**data['stats'])
return TestResult(**data)
else:
return data
@dataclasses.dataclass(slots=True, frozen=True)
class RunTests:
@ -234,7 +261,7 @@ class RunTests:
quiet: bool = False
hunt_refleak: HuntRefleak | None = None
test_dir: StrPath | None = None
junit_filename: StrPath | None = None
use_junit: bool = False
memory_limit: str | None = None
gc_threshold: int | None = None
use_resources: list[str] = None
@ -358,7 +385,7 @@ def setup_support(runtests: RunTests):
support.set_match_tests(runtests.match_tests, runtests.ignore_tests)
support.failfast = runtests.fail_fast
support.verbose = runtests.verbose
if runtests.junit_filename:
if runtests.use_junit:
support.junit_xml_list = []
else:
support.junit_xml_list = None
@ -434,8 +461,8 @@ def run_single_test(test_name: TestName, runtests: RunTests) -> TestResult:
Returns a TestResult.
If runtests.junit_filename is not None, xml_data is a list containing each
generated testsuite element.
If runtests.use_junit, xml_data is a list containing each generated
testsuite element.
"""
start_time = time.perf_counter()
result = TestResult(test_name)

View file

@ -1,6 +1,5 @@
import dataclasses
import faulthandler
import json
import os.path
import queue
import signal
@ -10,19 +9,19 @@ import tempfile
import threading
import time
import traceback
from typing import NoReturn, Literal, Any, TextIO
from typing import Literal, TextIO
from test import support
from test.support import os_helper
from test.support import TestStats
from test.libregrtest.main import Regrtest
from test.libregrtest.runtest import (
run_single_test, TestResult, State, PROGRESS_MIN_TIME,
FilterTuple, RunTests, StrPath, StrJSON, TestName)
from test.libregrtest.setup import setup_tests, setup_test_dir
TestResult, State, PROGRESS_MIN_TIME,
RunTests, TestName)
from test.libregrtest.results import TestResults
from test.libregrtest.utils import format_duration, print_warning
from test.libregrtest.utils import (
format_duration, print_warning, StrPath)
from test.libregrtest.worker import create_worker_process, USE_PROCESS_GROUP
if sys.platform == 'win32':
import locale
@ -41,75 +40,6 @@ assert MAIN_PROCESS_TIMEOUT >= PROGRESS_UPDATE
# Time to wait until a worker completes: should be immediate
JOIN_TIMEOUT = 30.0 # seconds
USE_PROCESS_GROUP = (hasattr(os, "setsid") and hasattr(os, "killpg"))
@dataclasses.dataclass(slots=True)
class WorkerJob:
runtests: RunTests
def create_worker_process(runtests: RunTests,
output_file: TextIO,
tmp_dir: StrPath | None = None) -> subprocess.Popen:
python_cmd = runtests.python_cmd
worker_json = runtests.as_json()
if python_cmd is not None:
executable = python_cmd
else:
executable = [sys.executable]
cmd = [*executable, *support.args_from_interpreter_flags(),
'-u', # Unbuffered stdout and stderr
'-m', 'test.regrtest',
'--worker-json', worker_json]
env = dict(os.environ)
if tmp_dir is not None:
env['TMPDIR'] = tmp_dir
env['TEMP'] = tmp_dir
env['TMP'] = tmp_dir
# Running the child from the same working directory as regrtest's original
# invocation ensures that TEMPDIR for the child is the same when
# sysconfig.is_python_build() is true. See issue 15300.
kw = dict(
env=env,
stdout=output_file,
# bpo-45410: Write stderr into stdout to keep messages order
stderr=output_file,
text=True,
close_fds=(os.name != 'nt'),
cwd=os_helper.SAVEDCWD,
)
if USE_PROCESS_GROUP:
kw['start_new_session'] = True
return subprocess.Popen(cmd, **kw)
def worker_process(worker_json: StrJSON) -> NoReturn:
runtests = RunTests.from_json(worker_json)
test_name = runtests.tests[0]
match_tests: FilterTuple | None = runtests.match_tests
setup_test_dir(runtests.test_dir)
setup_tests(runtests)
if runtests.rerun:
if match_tests:
matching = "matching: " + ", ".join(match_tests)
print(f"Re-running {test_name} in verbose mode ({matching})", flush=True)
else:
print(f"Re-running {test_name} in verbose mode", flush=True)
result = run_single_test(test_name, runtests)
print() # Force a newline (just in case)
# Serialize TestResult as dict in JSON
json.dump(result, sys.stdout, cls=EncodeTestResult)
sys.stdout.flush()
sys.exit(0)
# We do not use a generator so multiple threads can call next().
class MultiprocessIterator:
@ -340,9 +270,7 @@ class WorkerThread(threading.Thread):
err_msg = "Failed to parse worker stdout"
else:
try:
# deserialize run_tests_worker() output
result = json.loads(worker_json,
object_hook=decode_test_result)
result = TestResult.from_json(worker_json)
except Exception as exc:
err_msg = "Failed to parse worker JSON: %s" % exc
@ -562,27 +490,3 @@ class RunWorkers:
# worker when we exit this function
self.pending.stop()
self.stop_workers()
class EncodeTestResult(json.JSONEncoder):
"""Encode a TestResult (sub)class object into a JSON dict."""
def default(self, o: Any) -> dict[str, Any]:
if isinstance(o, TestResult):
result = dataclasses.asdict(o)
result["__test_result__"] = o.__class__.__name__
return result
return super().default(o)
def decode_test_result(d: dict[str, Any]) -> TestResult | dict[str, Any]:
"""Decode a TestResult (sub)class object from a JSON dict."""
if "__test_result__" not in d:
return d
d.pop('__test_result__')
if d['stats'] is not None:
d['stats'] = TestStats(**d['stats'])
return TestResult(**d)

View file

@ -11,8 +11,8 @@ try:
except ImportError:
gc = None
from test.libregrtest.utils import (setup_unraisable_hook,
setup_threading_excepthook)
from test.libregrtest.utils import (
setup_unraisable_hook, setup_threading_excepthook, fix_umask)
UNICODE_GUARD_ENV = "PYTHONREGRTEST_UNICODE_GUARD"
@ -26,6 +26,8 @@ def setup_test_dir(testdir: str | None) -> None:
def setup_tests(runtests):
fix_umask()
try:
stderr_fd = sys.__stderr__.fileno()
except (ValueError, AttributeError):
@ -102,7 +104,7 @@ def setup_tests(runtests):
support.SHORT_TIMEOUT = min(support.SHORT_TIMEOUT, timeout)
support.LONG_TIMEOUT = min(support.LONG_TIMEOUT, timeout)
if runtests.junit_filename:
if runtests.use_junit:
from test.support.testresult import RegressionTestResult
RegressionTestResult.USE_XML = True

View file

@ -1,13 +1,28 @@
import contextlib
import faulthandler
import math
import os.path
import random
import sys
import sysconfig
import tempfile
import textwrap
from test import support
from test.support import os_helper
from test.support import threading_helper
MS_WINDOWS = (sys.platform == 'win32')
# bpo-38203: Maximum delay in seconds to exit Python (call Py_Finalize()).
# Used to protect against threading._shutdown() hang.
# Must be smaller than buildbot "1200 seconds without output" limit.
EXIT_TIMEOUT = 120.0
StrPath = str
def format_duration(seconds):
ms = math.ceil(seconds * 1e3)
@ -308,3 +323,69 @@ def get_build_info():
build.append("dtrace")
return build
def get_temp_dir(tmp_dir):
if tmp_dir:
tmp_dir = os.path.expanduser(tmp_dir)
else:
# When tests are run from the Python build directory, it is best practice
# to keep the test files in a subfolder. This eases the cleanup of leftover
# files using the "make distclean" command.
if sysconfig.is_python_build():
tmp_dir = sysconfig.get_config_var('abs_builddir')
if tmp_dir is None:
# bpo-30284: On Windows, only srcdir is available. Using
# abs_builddir mostly matters on UNIX when building Python
# out of the source tree, especially when the source tree
# is read only.
tmp_dir = sysconfig.get_config_var('srcdir')
tmp_dir = os.path.join(tmp_dir, 'build')
else:
tmp_dir = tempfile.gettempdir()
return os.path.abspath(tmp_dir)
def fix_umask():
if support.is_emscripten:
# Emscripten has default umask 0o777, which breaks some tests.
# see https://github.com/emscripten-core/emscripten/issues/17269
old_mask = os.umask(0)
if old_mask == 0o777:
os.umask(0o027)
else:
os.umask(old_mask)
def get_work_dir(*, parent_dir: StrPath = '', worker: bool = False):
# Define a writable temp dir that will be used as cwd while running
# the tests. The name of the dir includes the pid to allow parallel
# testing (see the -j option).
# Emscripten and WASI have stubbed getpid(), Emscripten has only
# milisecond clock resolution. Use randint() instead.
if sys.platform in {"emscripten", "wasi"}:
nounce = random.randint(0, 1_000_000)
else:
nounce = os.getpid()
if worker:
work_dir = 'test_python_worker_{}'.format(nounce)
else:
work_dir = 'test_python_{}'.format(nounce)
work_dir += os_helper.FS_NONASCII
if parent_dir:
work_dir = os.path.join(parent_dir, work_dir)
return work_dir
@contextlib.contextmanager
def exit_timeout():
try:
yield
except SystemExit as exc:
# bpo-38203: Python can hang at exit in Py_Finalize(), especially
# on threading._shutdown() call: put a timeout
if threading_helper.can_start_thread:
faulthandler.dump_traceback_later(EXIT_TIMEOUT, exit=True)
sys.exit(exc.code)

View file

@ -0,0 +1,93 @@
import subprocess
import sys
import os
from typing import TextIO, NoReturn
from test import support
from test.support import os_helper
from test.libregrtest.setup import setup_tests, setup_test_dir
from test.libregrtest.runtest import (
run_single_test, StrJSON, FilterTuple, RunTests)
from test.libregrtest.utils import get_work_dir, exit_timeout, StrPath
USE_PROCESS_GROUP = (hasattr(os, "setsid") and hasattr(os, "killpg"))
def create_worker_process(runtests: RunTests,
output_file: TextIO,
tmp_dir: StrPath | None = None) -> subprocess.Popen:
python_cmd = runtests.python_cmd
worker_json = runtests.as_json()
if python_cmd is not None:
executable = python_cmd
else:
executable = [sys.executable]
cmd = [*executable, *support.args_from_interpreter_flags(),
'-u', # Unbuffered stdout and stderr
'-m', 'test.libregrtest.worker',
worker_json]
env = dict(os.environ)
if tmp_dir is not None:
env['TMPDIR'] = tmp_dir
env['TEMP'] = tmp_dir
env['TMP'] = tmp_dir
# Running the child from the same working directory as regrtest's original
# invocation ensures that TEMPDIR for the child is the same when
# sysconfig.is_python_build() is true. See issue 15300.
kw = dict(
env=env,
stdout=output_file,
# bpo-45410: Write stderr into stdout to keep messages order
stderr=output_file,
text=True,
close_fds=(os.name != 'nt'),
)
if USE_PROCESS_GROUP:
kw['start_new_session'] = True
return subprocess.Popen(cmd, **kw)
def worker_process(worker_json: StrJSON) -> NoReturn:
runtests = RunTests.from_json(worker_json)
test_name = runtests.tests[0]
match_tests: FilterTuple | None = runtests.match_tests
setup_test_dir(runtests.test_dir)
setup_tests(runtests)
if runtests.rerun:
if match_tests:
matching = "matching: " + ", ".join(match_tests)
print(f"Re-running {test_name} in verbose mode ({matching})", flush=True)
else:
print(f"Re-running {test_name} in verbose mode", flush=True)
result = run_single_test(test_name, runtests)
print() # Force a newline (just in case)
# Serialize TestResult as dict in JSON
result.write_json(sys.stdout)
sys.stdout.flush()
sys.exit(0)
def main():
if len(sys.argv) != 2:
print("usage: python -m test.libregrtest.worker JSON")
sys.exit(1)
worker_json = sys.argv[1]
work_dir = get_work_dir(worker=True)
with exit_timeout():
with os_helper.temp_cwd(work_dir, quiet=True):
worker_process(worker_json)
if __name__ == "__main__":
main()

View file

@ -75,11 +75,6 @@ class ParseArgsTestCase(unittest.TestCase):
ns = libregrtest._parse_args(['--wait'])
self.assertTrue(ns.wait)
def test_worker_json(self):
ns = libregrtest._parse_args(['--worker-json', '[[], {}]'])
self.assertEqual(ns.worker_json, '[[], {}]')
self.checkError(['--worker-json'], 'expected one argument')
def test_start(self):
for opt in '-S', '--start':
with self.subTest(opt=opt):