birkeh/o3de
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Updating/Removing LyTestTools sample tests

Browse Source 
Signed-off-by: evanchia <evanchia@amazon.com>
This commit is contained in:
evanchia
2021-07-08 10:24:49 -07:00
parent 6308d01456
commit 9067fc3b99
3 changed files with 14 additions and 135 deletions
Download Patch File Download Diff File Expand all files Collapse all files
Tools/LyTestTools/tests/example/__init__.py
-5
View File
@@ -1,5 +0,0 @@
"""
Copyright (c) Contributors to the Open 3D Engine Project. For complete copyright and license terms please see the LICENSE at the root of this distribution.
SPDX-License-Identifier: Apache-2.0 OR MIT
"""
Tools/LyTestTools/tests/example/test_system_example.py
-127
View File
@@ -1,127 +0,0 @@
"""
Copyright (c) Contributors to the Open 3D Engine Project. For complete copyright and license terms please see the LICENSE at the root of this distribution.
SPDX-License-Identifier: Apache-2.0 OR MIT
Example test using LyTestTools to test Open 3D Engine.
"""
# Python built-in dependencies.
import logging
# Third party dependencies.
import pytest
# Optional ly_test_tools dependencies.
import ly_test_tools.log.log_monitor
import ly_remote_console.remote_console_commands as remote_console_commands
# Configuring the logging is done in ly_test_tools at the following location:
# ~/dev/Tools/LyTestTools/ly_test_tools/_internal/log/py_logging_util.py
# Use the following logging pattern to hook all test logging together:
logger = logging.getLogger(__name__)
# Shared parameters & fixtures for all test methods inside the TestSystemExample class.
@pytest.mark.usefixtures("automatic_process_killer")
@pytest.mark.parametrize('project', ['AutomatedTesting'])
class TestSystemExample(object):
"""
Example test case class to hold a set of test case methods.
The amount of tests run is based on the parametrization stacking made in each test method or class.
For this test, we placed unique test values in test methods and shared test values in the test class.
We also assume building has already been done, but the test should error if the build is mis-configured.
"""
# This test method only needs pytest.mark report values and shared test class parameters.
@pytest.mark.parametrize('processes_to_kill', ['Editor.exe'])
@pytest.mark.parametrize("launcher_platform", ['windows_editor'])
def test_SystemTestExample_WindowsPlatform_LaunchEditor(self, editor, processes_to_kill, launcher_platform):
"""
Tests launching the O3DE Editor is successful with the current build.
"""
# Launch the O3DE Editor & verify load is successful:
with editor.start():
assert editor.is_alive(), (
'Editor failed to launch for the current O3DE build.')
@pytest.fixture
def remote_console(request):
"""
Creates a RemoteConsole() class instance to send console commands to the
O3DE client console.
:param request: _pytest.fixtures.SubRequest class that handles getting
a pytest fixture from a pytest function/fixture.
:return: ly_remote_console.remote_console_commands.RemoteConsole class instance
representing the O3DE Remote Console executable.
"""
# Initialize the RemoteConsole object to send commands to the O3DE client console.
console = remote_console_commands.RemoteConsole()
# Custom teardown method for this remote_console fixture.
def teardown():
console.stop()
# Utilize request.addfinalizer() to add custom teardown() methods.
request.addfinalizer(teardown) # This pattern must be used in pytest version
return console
# This test method needs specific parameters not shared by all other tests in the class.
# For targeting specific launchers, use the 'launcher_platform' pytest param like below:
# @pytest.mark.parametrize("launcher_platform", ['android'])
# If you want to target different AssetProcessor platforms, use asset_processor_platform:
# @pytest.mark.parametrize("asset_processor_platform", ['android'])
@pytest.mark.parametrize('level', ['simple_jacklocomotion'])
@pytest.mark.parametrize('load_wait', [120])
def test_SystemTestExample_AllSupportedPlatforms_LaunchAutomatedTesting(self, launcher, remote_console, level,
load_wait):
"""
Tests launching the AutomatedTesting then launches the O3DE client &
loads the "simple_jacklocomotion" level using the remote console.
Assumes the user already setup & built their machine for the test.
"""
# Launch the O3DE client & remote console test case:
with launcher.start():
remote_console.start()
# Wait for the expected log line to appear
launcher_load = remote_console.expect_log_line(
match_string='Level system is loading "simple_jacklocomotion"',
timeout=load_wait)
# Assert loading was successful using remote console logs:
assert launcher_load, ('Launcher failed to load O3DE client with the "{level}" level - waited "{load_wait}" '
'seconds.')
# Log monitoring example test.
@pytest.mark.parametrize('level', ['simple_jacklocomotion'])
@pytest.mark.parametrize('expected_lines', [['Log Monitoring test 1', 'Log Monitoring test 2']])
@pytest.mark.parametrize('unexpected_lines', [['Unexpected test 1', 'Unexpected test 2']])
@pytest.mark.test_case_id('C21202585')
def test_SystemTestExample_AllSupportedPlatforms_LogMonitoring(self, level, launcher, expected_lines,
unexpected_lines):
"""
Tests that the logging paths created by LyTestTools can be monitored for results using the log monitor.
"""
# Launch the O3DE client & initialize the log monitor.
file_to_monitor = launcher.workspace.info_log_path
log_monitor = ly_test_tools.log.log_monitor.LogMonitor(launcher=launcher,
log_file_path=file_to_monitor)
# Generate log lines to the info log using logger.
for expected_line in expected_lines:
logger.info(expected_line)
# Start the O3DE client & test that the lines we logged can be viewed by the log monitor.
with launcher.start():
log_test = log_monitor.monitor_log_for_lines(
expected_lines=expected_lines, # Defaults to None.
unexpected_lines=unexpected_lines, # Defaults to None.
halt_on_unexpected=True, # Defaults to False.
timeout=60) # Defaults to 30
# Assert the log monitor detected expected lines and did not detect any unexpected lines.
assert log_test, (
f'Log monitoring failed. Used expected_lines values: {expected_lines} & '
f'unexpected_lines values: {unexpected_lines}')
Tools/LyTestTools/tests/integ/sanity_tests.py
+14 -3
View File
@@ -15,8 +15,6 @@ import ly_test_tools.builtin.helpers as helpers
import ly_test_tools.environment.process_utils as process_utils
import ly_test_tools.environment.waiter as waiter
pytestmark = pytest.mark.SUITE_smoke
logger = logging.getLogger(__name__)
# Note: For device testing, device ids must exist in ~/ly_test_tools/devices.ini, see README.txt for more info.
@@ -24,31 +22,44 @@ logger = logging.getLogger(__name__)
@pytest.mark.parametrize("project", ["AutomatedTesting"])
class TestAutomatedTestingProject(object):
def test_StartGameLauncher_Sanity(self, project):
# Kill processes that may interfere with the test
process_utils.kill_processes_named(names=process_utils.LY_PROCESS_KILL_LIST, ignore_extensions=True)
try:
# Create the Workspace object
workspace = helpers.create_builtin_workspace(project=project)
# Create the Launcher object and add args
launcher = launcher_helper.create_launcher(workspace)
launcher.args.extend(['-NullRenderer', '-BatchMode'])
launcher.args.extend(['-NullRenderer'])
# Call the game client executable
with launcher.start():
# Wait for the process to exist
waiter.wait_for(lambda: process_utils.process_exists(f"{project}.GameLauncher.exe", ignore_extensions=True))
finally:
# Clean up processes after the test is finished
process_utils.kill_processes_named(names=process_utils.LY_PROCESS_KILL_LIST, ignore_extensions=True)
@pytest.mark.skipif(not ly_test_tools.WINDOWS, reason="Editor currently only functions on Windows")
def test_StartEditor_Sanity(self, project):
# Kill processes that may interfere with the test
process_utils.kill_processes_named(names=process_utils.LY_PROCESS_KILL_LIST, ignore_extensions=True)
try:
# Create the Workspace object
workspace = helpers.create_builtin_workspace(project=project)
# Create the Launcher object and add args
editor = launcher_helper.create_editor(workspace)
editor.args.extend(['-NullRenderer', '-autotest_mode'])
# Call the Editor executable
with editor.start():
# Wait for the process to exist
waiter.wait_for(lambda: process_utils.process_exists("Editor", ignore_extensions=True))
finally:
# Clean up processes after the test is finished
process_utils.kill_processes_named(names=process_utils.LY_PROCESS_KILL_LIST, ignore_extensions=True)