754 lines
27 KiB
Python
754 lines
27 KiB
Python
"""
|
|
Tests of regrtest.py.
|
|
|
|
Note: test_regrtest cannot be run twice in parallel.
|
|
"""
|
|
|
|
import argparse
|
|
import contextlib
|
|
import faulthandler
|
|
import getopt
|
|
import io
|
|
import os.path
|
|
import platform
|
|
import re
|
|
import subprocess
|
|
import sys
|
|
import sysconfig
|
|
import textwrap
|
|
import unittest
|
|
from test import libregrtest
|
|
from test import support
|
|
|
|
|
|
Py_DEBUG = hasattr(sys, 'getobjects')
|
|
ROOT_DIR = os.path.join(os.path.dirname(__file__), '..', '..')
|
|
ROOT_DIR = os.path.abspath(os.path.normpath(ROOT_DIR))
|
|
|
|
TEST_INTERRUPTED = textwrap.dedent("""
|
|
from signal import SIGINT
|
|
try:
|
|
from _testcapi import raise_signal
|
|
raise_signal(SIGINT)
|
|
except ImportError:
|
|
import os
|
|
os.kill(os.getpid(), SIGINT)
|
|
""")
|
|
|
|
|
|
class ParseArgsTestCase(unittest.TestCase):
|
|
"""
|
|
Test regrtest's argument parsing, function _parse_args().
|
|
"""
|
|
|
|
def checkError(self, args, msg):
|
|
with support.captured_stderr() as err, self.assertRaises(SystemExit):
|
|
libregrtest._parse_args(args)
|
|
self.assertIn(msg, err.getvalue())
|
|
|
|
def test_help(self):
|
|
for opt in '-h', '--help':
|
|
with self.subTest(opt=opt):
|
|
with support.captured_stdout() as out, \
|
|
self.assertRaises(SystemExit):
|
|
libregrtest._parse_args([opt])
|
|
self.assertIn('Run Python regression tests.', out.getvalue())
|
|
|
|
@unittest.skipUnless(hasattr(faulthandler, 'dump_traceback_later'),
|
|
"faulthandler.dump_traceback_later() required")
|
|
def test_timeout(self):
|
|
ns = libregrtest._parse_args(['--timeout', '4.2'])
|
|
self.assertEqual(ns.timeout, 4.2)
|
|
self.checkError(['--timeout'], 'expected one argument')
|
|
self.checkError(['--timeout', 'foo'], 'invalid float value')
|
|
|
|
def test_wait(self):
|
|
ns = libregrtest._parse_args(['--wait'])
|
|
self.assertTrue(ns.wait)
|
|
|
|
def test_slaveargs(self):
|
|
ns = libregrtest._parse_args(['--slaveargs', '[[], {}]'])
|
|
self.assertEqual(ns.slaveargs, '[[], {}]')
|
|
self.checkError(['--slaveargs'], 'expected one argument')
|
|
|
|
def test_start(self):
|
|
for opt in '-S', '--start':
|
|
with self.subTest(opt=opt):
|
|
ns = libregrtest._parse_args([opt, 'foo'])
|
|
self.assertEqual(ns.start, 'foo')
|
|
self.checkError([opt], 'expected one argument')
|
|
|
|
def test_verbose(self):
|
|
ns = libregrtest._parse_args(['-v'])
|
|
self.assertEqual(ns.verbose, 1)
|
|
ns = libregrtest._parse_args(['-vvv'])
|
|
self.assertEqual(ns.verbose, 3)
|
|
ns = libregrtest._parse_args(['--verbose'])
|
|
self.assertEqual(ns.verbose, 1)
|
|
ns = libregrtest._parse_args(['--verbose'] * 3)
|
|
self.assertEqual(ns.verbose, 3)
|
|
ns = libregrtest._parse_args([])
|
|
self.assertEqual(ns.verbose, 0)
|
|
|
|
def test_verbose2(self):
|
|
for opt in '-w', '--verbose2':
|
|
with self.subTest(opt=opt):
|
|
ns = libregrtest._parse_args([opt])
|
|
self.assertTrue(ns.verbose2)
|
|
|
|
def test_verbose3(self):
|
|
for opt in '-W', '--verbose3':
|
|
with self.subTest(opt=opt):
|
|
ns = libregrtest._parse_args([opt])
|
|
self.assertTrue(ns.verbose3)
|
|
|
|
def test_quiet(self):
|
|
for opt in '-q', '--quiet':
|
|
with self.subTest(opt=opt):
|
|
ns = libregrtest._parse_args([opt])
|
|
self.assertTrue(ns.quiet)
|
|
self.assertEqual(ns.verbose, 0)
|
|
|
|
def test_slow(self):
|
|
for opt in '-o', '--slow':
|
|
with self.subTest(opt=opt):
|
|
ns = libregrtest._parse_args([opt])
|
|
self.assertTrue(ns.print_slow)
|
|
|
|
def test_header(self):
|
|
ns = libregrtest._parse_args(['--header'])
|
|
self.assertTrue(ns.header)
|
|
|
|
def test_randomize(self):
|
|
for opt in '-r', '--randomize':
|
|
with self.subTest(opt=opt):
|
|
ns = libregrtest._parse_args([opt])
|
|
self.assertTrue(ns.randomize)
|
|
|
|
def test_randseed(self):
|
|
ns = libregrtest._parse_args(['--randseed', '12345'])
|
|
self.assertEqual(ns.random_seed, 12345)
|
|
self.assertTrue(ns.randomize)
|
|
self.checkError(['--randseed'], 'expected one argument')
|
|
self.checkError(['--randseed', 'foo'], 'invalid int value')
|
|
|
|
def test_fromfile(self):
|
|
for opt in '-f', '--fromfile':
|
|
with self.subTest(opt=opt):
|
|
ns = libregrtest._parse_args([opt, 'foo'])
|
|
self.assertEqual(ns.fromfile, 'foo')
|
|
self.checkError([opt], 'expected one argument')
|
|
self.checkError([opt, 'foo', '-s'], "don't go together")
|
|
|
|
def test_exclude(self):
|
|
for opt in '-x', '--exclude':
|
|
with self.subTest(opt=opt):
|
|
ns = libregrtest._parse_args([opt])
|
|
self.assertTrue(ns.exclude)
|
|
|
|
def test_single(self):
|
|
for opt in '-s', '--single':
|
|
with self.subTest(opt=opt):
|
|
ns = libregrtest._parse_args([opt])
|
|
self.assertTrue(ns.single)
|
|
self.checkError([opt, '-f', 'foo'], "don't go together")
|
|
|
|
def test_match(self):
|
|
for opt in '-m', '--match':
|
|
with self.subTest(opt=opt):
|
|
ns = libregrtest._parse_args([opt, 'pattern'])
|
|
self.assertEqual(ns.match_tests, 'pattern')
|
|
self.checkError([opt], 'expected one argument')
|
|
|
|
def test_failfast(self):
|
|
for opt in '-G', '--failfast':
|
|
with self.subTest(opt=opt):
|
|
ns = libregrtest._parse_args([opt, '-v'])
|
|
self.assertTrue(ns.failfast)
|
|
ns = libregrtest._parse_args([opt, '-W'])
|
|
self.assertTrue(ns.failfast)
|
|
self.checkError([opt], '-G/--failfast needs either -v or -W')
|
|
|
|
def test_use(self):
|
|
for opt in '-u', '--use':
|
|
with self.subTest(opt=opt):
|
|
ns = libregrtest._parse_args([opt, 'gui,network'])
|
|
self.assertEqual(ns.use_resources, ['gui', 'network'])
|
|
ns = libregrtest._parse_args([opt, 'gui,none,network'])
|
|
self.assertEqual(ns.use_resources, ['network'])
|
|
expected = list(libregrtest.RESOURCE_NAMES)
|
|
expected.remove('gui')
|
|
ns = libregrtest._parse_args([opt, 'all,-gui'])
|
|
self.assertEqual(ns.use_resources, expected)
|
|
self.checkError([opt], 'expected one argument')
|
|
self.checkError([opt, 'foo'], 'invalid resource')
|
|
|
|
def test_memlimit(self):
|
|
for opt in '-M', '--memlimit':
|
|
with self.subTest(opt=opt):
|
|
ns = libregrtest._parse_args([opt, '4G'])
|
|
self.assertEqual(ns.memlimit, '4G')
|
|
self.checkError([opt], 'expected one argument')
|
|
|
|
def test_testdir(self):
|
|
ns = libregrtest._parse_args(['--testdir', 'foo'])
|
|
self.assertEqual(ns.testdir, os.path.join(support.SAVEDCWD, 'foo'))
|
|
self.checkError(['--testdir'], 'expected one argument')
|
|
|
|
def test_runleaks(self):
|
|
for opt in '-L', '--runleaks':
|
|
with self.subTest(opt=opt):
|
|
ns = libregrtest._parse_args([opt])
|
|
self.assertTrue(ns.runleaks)
|
|
|
|
def test_huntrleaks(self):
|
|
for opt in '-R', '--huntrleaks':
|
|
with self.subTest(opt=opt):
|
|
ns = libregrtest._parse_args([opt, ':'])
|
|
self.assertEqual(ns.huntrleaks, (5, 4, 'reflog.txt'))
|
|
ns = libregrtest._parse_args([opt, '6:'])
|
|
self.assertEqual(ns.huntrleaks, (6, 4, 'reflog.txt'))
|
|
ns = libregrtest._parse_args([opt, ':3'])
|
|
self.assertEqual(ns.huntrleaks, (5, 3, 'reflog.txt'))
|
|
ns = libregrtest._parse_args([opt, '6:3:leaks.log'])
|
|
self.assertEqual(ns.huntrleaks, (6, 3, 'leaks.log'))
|
|
self.checkError([opt], 'expected one argument')
|
|
self.checkError([opt, '6'],
|
|
'needs 2 or 3 colon-separated arguments')
|
|
self.checkError([opt, 'foo:'], 'invalid huntrleaks value')
|
|
self.checkError([opt, '6:foo'], 'invalid huntrleaks value')
|
|
|
|
def test_multiprocess(self):
|
|
for opt in '-j', '--multiprocess':
|
|
with self.subTest(opt=opt):
|
|
ns = libregrtest._parse_args([opt, '2'])
|
|
self.assertEqual(ns.use_mp, 2)
|
|
self.checkError([opt], 'expected one argument')
|
|
self.checkError([opt, 'foo'], 'invalid int value')
|
|
self.checkError([opt, '2', '-T'], "don't go together")
|
|
self.checkError([opt, '2', '-l'], "don't go together")
|
|
|
|
def test_coverage(self):
|
|
for opt in '-T', '--coverage':
|
|
with self.subTest(opt=opt):
|
|
ns = libregrtest._parse_args([opt])
|
|
self.assertTrue(ns.trace)
|
|
|
|
def test_coverdir(self):
|
|
for opt in '-D', '--coverdir':
|
|
with self.subTest(opt=opt):
|
|
ns = libregrtest._parse_args([opt, 'foo'])
|
|
self.assertEqual(ns.coverdir,
|
|
os.path.join(support.SAVEDCWD, 'foo'))
|
|
self.checkError([opt], 'expected one argument')
|
|
|
|
def test_nocoverdir(self):
|
|
for opt in '-N', '--nocoverdir':
|
|
with self.subTest(opt=opt):
|
|
ns = libregrtest._parse_args([opt])
|
|
self.assertIsNone(ns.coverdir)
|
|
|
|
def test_threshold(self):
|
|
for opt in '-t', '--threshold':
|
|
with self.subTest(opt=opt):
|
|
ns = libregrtest._parse_args([opt, '1000'])
|
|
self.assertEqual(ns.threshold, 1000)
|
|
self.checkError([opt], 'expected one argument')
|
|
self.checkError([opt, 'foo'], 'invalid int value')
|
|
|
|
def test_nowindows(self):
|
|
for opt in '-n', '--nowindows':
|
|
with self.subTest(opt=opt):
|
|
with contextlib.redirect_stderr(io.StringIO()) as stderr:
|
|
ns = libregrtest._parse_args([opt])
|
|
self.assertTrue(ns.nowindows)
|
|
err = stderr.getvalue()
|
|
self.assertIn('the --nowindows (-n) option is deprecated', err)
|
|
|
|
def test_forever(self):
|
|
for opt in '-F', '--forever':
|
|
with self.subTest(opt=opt):
|
|
ns = libregrtest._parse_args([opt])
|
|
self.assertTrue(ns.forever)
|
|
|
|
|
|
def test_unrecognized_argument(self):
|
|
self.checkError(['--xxx'], 'usage:')
|
|
|
|
def test_long_option__partial(self):
|
|
ns = libregrtest._parse_args(['--qui'])
|
|
self.assertTrue(ns.quiet)
|
|
self.assertEqual(ns.verbose, 0)
|
|
|
|
def test_two_options(self):
|
|
ns = libregrtest._parse_args(['--quiet', '--exclude'])
|
|
self.assertTrue(ns.quiet)
|
|
self.assertEqual(ns.verbose, 0)
|
|
self.assertTrue(ns.exclude)
|
|
|
|
def test_option_with_empty_string_value(self):
|
|
ns = libregrtest._parse_args(['--start', ''])
|
|
self.assertEqual(ns.start, '')
|
|
|
|
def test_arg(self):
|
|
ns = libregrtest._parse_args(['foo'])
|
|
self.assertEqual(ns.args, ['foo'])
|
|
|
|
def test_option_and_arg(self):
|
|
ns = libregrtest._parse_args(['--quiet', 'foo'])
|
|
self.assertTrue(ns.quiet)
|
|
self.assertEqual(ns.verbose, 0)
|
|
self.assertEqual(ns.args, ['foo'])
|
|
|
|
|
|
class BaseTestCase(unittest.TestCase):
|
|
TEST_UNIQUE_ID = 1
|
|
TESTNAME_PREFIX = 'test_regrtest_'
|
|
TESTNAME_REGEX = r'test_[a-z0-9_]+'
|
|
|
|
def setUp(self):
|
|
self.testdir = os.path.realpath(os.path.dirname(__file__))
|
|
|
|
# When test_regrtest is interrupted by CTRL+c, it can leave
|
|
# temporary test files
|
|
remove = [entry.path
|
|
for entry in os.scandir(self.testdir)
|
|
if (entry.name.startswith(self.TESTNAME_PREFIX)
|
|
and entry.name.endswith(".py"))]
|
|
for path in remove:
|
|
print("WARNING: test_regrtest: remove %s" % path)
|
|
support.unlink(path)
|
|
|
|
def create_test(self, name=None, code=''):
|
|
if not name:
|
|
name = 'noop%s' % BaseTestCase.TEST_UNIQUE_ID
|
|
BaseTestCase.TEST_UNIQUE_ID += 1
|
|
|
|
# test_regrtest cannot be run twice in parallel because
|
|
# of setUp() and create_test()
|
|
name = self.TESTNAME_PREFIX + "%s_%s" % (os.getpid(), name)
|
|
path = os.path.join(self.testdir, name + '.py')
|
|
|
|
self.addCleanup(support.unlink, path)
|
|
# Use 'x' mode to ensure that we do not override existing tests
|
|
try:
|
|
with open(path, 'x', encoding='utf-8') as fp:
|
|
fp.write(code)
|
|
except PermissionError as exc:
|
|
if not sysconfig.is_python_build():
|
|
self.skipTest("cannot write %s: %s" % (path, exc))
|
|
raise
|
|
return name
|
|
|
|
def regex_search(self, regex, output):
|
|
match = re.search(regex, output, re.MULTILINE)
|
|
if not match:
|
|
self.fail("%r not found in %r" % (regex, output))
|
|
return match
|
|
|
|
def check_line(self, output, regex):
|
|
regex = re.compile(r'^' + regex, re.MULTILINE)
|
|
self.assertRegex(output, regex)
|
|
|
|
def parse_executed_tests(self, output):
|
|
regex = r'^\[ *[0-9]+(?:/ *[0-9]+)?\] (%s)$' % self.TESTNAME_REGEX
|
|
parser = re.finditer(regex, output, re.MULTILINE)
|
|
return list(match.group(1) for match in parser)
|
|
|
|
def check_executed_tests(self, output, tests, skipped=(), failed=(),
|
|
omitted=(), randomize=False):
|
|
if isinstance(tests, str):
|
|
tests = [tests]
|
|
if isinstance(skipped, str):
|
|
skipped = [skipped]
|
|
if isinstance(failed, str):
|
|
failed = [failed]
|
|
if isinstance(omitted, str):
|
|
omitted = [omitted]
|
|
ntest = len(tests)
|
|
nskipped = len(skipped)
|
|
nfailed = len(failed)
|
|
nomitted = len(omitted)
|
|
|
|
executed = self.parse_executed_tests(output)
|
|
if randomize:
|
|
self.assertEqual(set(executed), set(tests), output)
|
|
else:
|
|
self.assertEqual(executed, tests, output)
|
|
|
|
def plural(count):
|
|
return 's' if count != 1 else ''
|
|
|
|
def list_regex(line_format, tests):
|
|
count = len(tests)
|
|
names = ' '.join(sorted(tests))
|
|
regex = line_format % (count, plural(count))
|
|
regex = r'%s:\n %s$' % (regex, names)
|
|
return regex
|
|
|
|
if skipped:
|
|
regex = list_regex('%s test%s skipped', skipped)
|
|
self.check_line(output, regex)
|
|
|
|
if failed:
|
|
regex = list_regex('%s test%s failed', failed)
|
|
self.check_line(output, regex)
|
|
|
|
if omitted:
|
|
regex = list_regex('%s test%s omitted', omitted)
|
|
self.check_line(output, regex)
|
|
|
|
good = ntest - nskipped - nfailed - nomitted
|
|
if good:
|
|
regex = r'%s test%s OK\.$' % (good, plural(good))
|
|
if not skipped and not failed and good > 1:
|
|
regex = 'All %s' % regex
|
|
self.check_line(output, regex)
|
|
|
|
def parse_random_seed(self, output):
|
|
match = self.regex_search(r'Using random seed ([0-9]+)', output)
|
|
randseed = int(match.group(1))
|
|
self.assertTrue(0 <= randseed <= 10000000, randseed)
|
|
return randseed
|
|
|
|
def run_command(self, args, input=None, exitcode=0, **kw):
|
|
if not input:
|
|
input = ''
|
|
if 'stderr' not in kw:
|
|
kw['stderr'] = subprocess.PIPE
|
|
proc = subprocess.run(args,
|
|
universal_newlines=True,
|
|
input=input,
|
|
stdout=subprocess.PIPE,
|
|
**kw)
|
|
if proc.returncode != exitcode:
|
|
msg = ("Command %s failed with exit code %s\n"
|
|
"\n"
|
|
"stdout:\n"
|
|
"---\n"
|
|
"%s\n"
|
|
"---\n"
|
|
% (str(args), proc.returncode, proc.stdout))
|
|
if proc.stderr:
|
|
msg += ("\n"
|
|
"stderr:\n"
|
|
"---\n"
|
|
"%s"
|
|
"---\n"
|
|
% proc.stderr)
|
|
self.fail(msg)
|
|
return proc
|
|
|
|
|
|
def run_python(self, args, **kw):
|
|
args = [sys.executable, '-X', 'faulthandler', '-I', *args]
|
|
proc = self.run_command(args, **kw)
|
|
return proc.stdout
|
|
|
|
|
|
class ProgramsTestCase(BaseTestCase):
|
|
"""
|
|
Test various ways to run the Python test suite. Use options close
|
|
to options used on the buildbot.
|
|
"""
|
|
|
|
NTEST = 4
|
|
|
|
def setUp(self):
|
|
super().setUp()
|
|
|
|
# Create NTEST tests doing nothing
|
|
self.tests = [self.create_test() for index in range(self.NTEST)]
|
|
|
|
self.python_args = ['-Wd', '-E', '-bb']
|
|
self.regrtest_args = ['-uall', '-rwW']
|
|
if hasattr(faulthandler, 'dump_traceback_later'):
|
|
self.regrtest_args.extend(('--timeout', '3600', '-j4'))
|
|
if sys.platform == 'win32':
|
|
self.regrtest_args.append('-n')
|
|
|
|
def check_output(self, output):
|
|
self.parse_random_seed(output)
|
|
self.check_executed_tests(output, self.tests, randomize=True)
|
|
|
|
def run_tests(self, args):
|
|
output = self.run_python(args)
|
|
self.check_output(output)
|
|
|
|
def test_script_regrtest(self):
|
|
# Lib/test/regrtest.py
|
|
script = os.path.join(self.testdir, 'regrtest.py')
|
|
|
|
args = [*self.python_args, script, *self.regrtest_args, *self.tests]
|
|
self.run_tests(args)
|
|
|
|
def test_module_test(self):
|
|
# -m test
|
|
args = [*self.python_args, '-m', 'test',
|
|
*self.regrtest_args, *self.tests]
|
|
self.run_tests(args)
|
|
|
|
def test_module_regrtest(self):
|
|
# -m test.regrtest
|
|
args = [*self.python_args, '-m', 'test.regrtest',
|
|
*self.regrtest_args, *self.tests]
|
|
self.run_tests(args)
|
|
|
|
def test_module_autotest(self):
|
|
# -m test.autotest
|
|
args = [*self.python_args, '-m', 'test.autotest',
|
|
*self.regrtest_args, *self.tests]
|
|
self.run_tests(args)
|
|
|
|
def test_module_from_test_autotest(self):
|
|
# from test import autotest
|
|
code = 'from test import autotest'
|
|
args = [*self.python_args, '-c', code,
|
|
*self.regrtest_args, *self.tests]
|
|
self.run_tests(args)
|
|
|
|
def test_script_autotest(self):
|
|
# Lib/test/autotest.py
|
|
script = os.path.join(self.testdir, 'autotest.py')
|
|
args = [*self.python_args, script, *self.regrtest_args, *self.tests]
|
|
self.run_tests(args)
|
|
|
|
@unittest.skipUnless(sysconfig.is_python_build(),
|
|
'run_tests.py script is not installed')
|
|
def test_tools_script_run_tests(self):
|
|
# Tools/scripts/run_tests.py
|
|
script = os.path.join(ROOT_DIR, 'Tools', 'scripts', 'run_tests.py')
|
|
self.run_tests([script, *self.tests])
|
|
|
|
def run_batch(self, *args):
|
|
proc = self.run_command(args)
|
|
self.check_output(proc.stdout)
|
|
|
|
@unittest.skipUnless(sysconfig.is_python_build(),
|
|
'test.bat script is not installed')
|
|
@unittest.skipUnless(sys.platform == 'win32', 'Windows only')
|
|
def test_tools_buildbot_test(self):
|
|
# Tools\buildbot\test.bat
|
|
script = os.path.join(ROOT_DIR, 'Tools', 'buildbot', 'test.bat')
|
|
test_args = []
|
|
if platform.architecture()[0] == '64bit':
|
|
test_args.append('-x64') # 64-bit build
|
|
if not Py_DEBUG:
|
|
test_args.append('+d') # Release build, use python.exe
|
|
self.run_batch(script, *test_args, *self.tests)
|
|
|
|
@unittest.skipUnless(sys.platform == 'win32', 'Windows only')
|
|
def test_pcbuild_rt(self):
|
|
# PCbuild\rt.bat
|
|
script = os.path.join(ROOT_DIR, r'PCbuild\rt.bat')
|
|
rt_args = ["-q"] # Quick, don't run tests twice
|
|
if platform.architecture()[0] == '64bit':
|
|
rt_args.append('-x64') # 64-bit build
|
|
if Py_DEBUG:
|
|
rt_args.append('-d') # Debug build, use python_d.exe
|
|
self.run_batch(script, *rt_args, *self.regrtest_args, *self.tests)
|
|
|
|
|
|
class ArgsTestCase(BaseTestCase):
|
|
"""
|
|
Test arguments of the Python test suite.
|
|
"""
|
|
|
|
def run_tests(self, *args, **kw):
|
|
return self.run_python(['-m', 'test', *args], **kw)
|
|
|
|
def test_failing_test(self):
|
|
# test a failing test
|
|
code = textwrap.dedent("""
|
|
import unittest
|
|
|
|
class FailingTest(unittest.TestCase):
|
|
def test_failing(self):
|
|
self.fail("bug")
|
|
""")
|
|
test_ok = self.create_test()
|
|
test_failing = self.create_test(code=code)
|
|
tests = [test_ok, test_failing]
|
|
|
|
output = self.run_tests(*tests, exitcode=1)
|
|
self.check_executed_tests(output, tests, failed=test_failing)
|
|
|
|
def test_resources(self):
|
|
# test -u command line option
|
|
tests = {}
|
|
for resource in ('audio', 'network'):
|
|
code = 'from test import support\nsupport.requires(%r)' % resource
|
|
tests[resource] = self.create_test(resource, code)
|
|
test_names = sorted(tests.values())
|
|
|
|
# -u all: 2 resources enabled
|
|
output = self.run_tests('-u', 'all', *test_names)
|
|
self.check_executed_tests(output, test_names)
|
|
|
|
# -u audio: 1 resource enabled
|
|
output = self.run_tests('-uaudio', *test_names)
|
|
self.check_executed_tests(output, test_names,
|
|
skipped=tests['network'])
|
|
|
|
# no option: 0 resources enabled
|
|
output = self.run_tests(*test_names)
|
|
self.check_executed_tests(output, test_names,
|
|
skipped=test_names)
|
|
|
|
def test_random(self):
|
|
# test -r and --randseed command line option
|
|
code = textwrap.dedent("""
|
|
import random
|
|
print("TESTRANDOM: %s" % random.randint(1, 1000))
|
|
""")
|
|
test = self.create_test('random', code)
|
|
|
|
# first run to get the output with the random seed
|
|
output = self.run_tests('-r', test)
|
|
randseed = self.parse_random_seed(output)
|
|
match = self.regex_search(r'TESTRANDOM: ([0-9]+)', output)
|
|
test_random = int(match.group(1))
|
|
|
|
# try to reproduce with the random seed
|
|
output = self.run_tests('-r', '--randseed=%s' % randseed, test)
|
|
randseed2 = self.parse_random_seed(output)
|
|
self.assertEqual(randseed2, randseed)
|
|
|
|
match = self.regex_search(r'TESTRANDOM: ([0-9]+)', output)
|
|
test_random2 = int(match.group(1))
|
|
self.assertEqual(test_random2, test_random)
|
|
|
|
def test_fromfile(self):
|
|
# test --fromfile
|
|
tests = [self.create_test() for index in range(5)]
|
|
|
|
# Write the list of files using a format similar to regrtest output:
|
|
# [1/2] test_1
|
|
# [2/2] test_2
|
|
filename = support.TESTFN
|
|
self.addCleanup(support.unlink, filename)
|
|
with open(filename, "w") as fp:
|
|
for index, name in enumerate(tests, 1):
|
|
print("[%s/%s] %s" % (index, len(tests), name), file=fp)
|
|
|
|
output = self.run_tests('--fromfile', filename)
|
|
self.check_executed_tests(output, tests)
|
|
|
|
def test_interrupted(self):
|
|
code = TEST_INTERRUPTED
|
|
test = self.create_test("sigint", code=code)
|
|
output = self.run_tests(test, exitcode=1)
|
|
self.check_executed_tests(output, test, omitted=test)
|
|
|
|
def test_slow(self):
|
|
# test --slow
|
|
tests = [self.create_test() for index in range(3)]
|
|
output = self.run_tests("--slow", *tests)
|
|
self.check_executed_tests(output, tests)
|
|
regex = ('10 slowest tests:\n'
|
|
'(?:%s: [0-9]+\.[0-9]+s\n){%s}'
|
|
% (self.TESTNAME_REGEX, len(tests)))
|
|
self.check_line(output, regex)
|
|
|
|
def test_slow_interrupted(self):
|
|
# Issue #25373: test --slow with an interrupted test
|
|
code = TEST_INTERRUPTED
|
|
test = self.create_test("sigint", code=code)
|
|
|
|
for multiprocessing in (False, True):
|
|
if multiprocessing:
|
|
args = ("--slow", "-j2", test)
|
|
else:
|
|
args = ("--slow", test)
|
|
output = self.run_tests(*args, exitcode=1)
|
|
self.check_executed_tests(output, test, omitted=test)
|
|
regex = ('10 slowest tests:\n')
|
|
self.check_line(output, regex)
|
|
self.check_line(output, 'Test suite interrupted by signal SIGINT.')
|
|
|
|
def test_coverage(self):
|
|
# test --coverage
|
|
test = self.create_test()
|
|
output = self.run_tests("--coverage", test)
|
|
self.check_executed_tests(output, [test])
|
|
regex = ('lines +cov% +module +\(path\)\n'
|
|
'(?: *[0-9]+ *[0-9]{1,2}% *[^ ]+ +\([^)]+\)+)+')
|
|
self.check_line(output, regex)
|
|
|
|
def test_wait(self):
|
|
# test --wait
|
|
test = self.create_test()
|
|
output = self.run_tests("--wait", test, input='key')
|
|
self.check_line(output, 'Press any key to continue')
|
|
|
|
def test_forever(self):
|
|
# test --forever
|
|
code = textwrap.dedent("""
|
|
import unittest
|
|
|
|
class ForeverTester(unittest.TestCase):
|
|
RUN = 1
|
|
|
|
def test_run(self):
|
|
ForeverTester.RUN += 1
|
|
if ForeverTester.RUN > 3:
|
|
self.fail("fail at the 3rd runs")
|
|
""")
|
|
test = self.create_test(code=code)
|
|
output = self.run_tests('--forever', test, exitcode=1)
|
|
self.check_executed_tests(output, [test]*3, failed=test)
|
|
|
|
@unittest.skipUnless(Py_DEBUG, 'need a debug build')
|
|
def test_huntrleaks_fd_leak(self):
|
|
# test --huntrleaks for file descriptor leak
|
|
code = textwrap.dedent("""
|
|
import os
|
|
import unittest
|
|
|
|
# Issue #25306: Disable popups and logs to stderr on assertion
|
|
# failures in MSCRT
|
|
try:
|
|
import msvcrt
|
|
msvcrt.CrtSetReportMode
|
|
except (ImportError, AttributeError):
|
|
# no Windows, o release build
|
|
pass
|
|
else:
|
|
for m in [msvcrt.CRT_WARN, msvcrt.CRT_ERROR, msvcrt.CRT_ASSERT]:
|
|
msvcrt.CrtSetReportMode(m, 0)
|
|
|
|
class FDLeakTest(unittest.TestCase):
|
|
def test_leak(self):
|
|
fd = os.open(__file__, os.O_RDONLY)
|
|
# bug: never cloes the file descriptor
|
|
""")
|
|
test = self.create_test(code=code)
|
|
|
|
filename = 'reflog.txt'
|
|
self.addCleanup(support.unlink, filename)
|
|
output = self.run_tests('--huntrleaks', '3:3:', test,
|
|
exitcode=1,
|
|
stderr=subprocess.STDOUT)
|
|
self.check_executed_tests(output, [test], failed=test)
|
|
|
|
line = 'beginning 6 repetitions\n123456\n......\n'
|
|
self.check_line(output, re.escape(line))
|
|
|
|
line2 = '%s leaked [1, 1, 1] file descriptors, sum=3\n' % test
|
|
self.check_line(output, re.escape(line2))
|
|
|
|
with open(filename) as fp:
|
|
reflog = fp.read()
|
|
self.assertEqual(reflog, line2)
|
|
|
|
def test_list_tests(self):
|
|
# test --list-tests
|
|
tests = [self.create_test() for i in range(5)]
|
|
output = self.run_tests('--list-tests', *tests)
|
|
self.assertEqual(output.rstrip().splitlines(),
|
|
tests)
|
|
|
|
|
|
if __name__ == '__main__':
|
|
unittest.main()
|