def test_kafka_ingest(docker_compose_runner, pytestconfig, tmp_path, mock_time):
test_resources_dir = pytestconfig.rootpath / "tests/integration/kafka"
with docker_compose_runner(
test_resources_dir / "docker-compose.yml", "kafka"
) as docker_services:
wait_for_port(docker_services, "test_broker", 9092, timeout=120)
# Set up topics and produce some data
command = f"{test_resources_dir}/send_records.sh {test_resources_dir}"
subprocess.run(command, shell=True, check=True)
# Run the metadata ingestion pipeline.
runner = CliRunner()
with fs_helpers.isolated_filesystem(tmp_path):
config_file = (test_resources_dir / "kafka_to_file.yml").resolve()
result = runner.invoke(datahub, ["ingest", "-c", f"{config_file}"])
assert result.exit_code == 0
# Verify the output.
mce_helpers.check_golden_file(
pytestconfig,
output_path=tmp_path / "kafka_mces.json",
golden_path=test_resources_dir / "kafka_mces_golden.json",
ignore_paths=[],
)
def test_hive_ingest(docker_compose_runner, pytestconfig, tmp_path, mock_time):
test_resources_dir = pytestconfig.rootpath / "tests/integration/hive"
with docker_compose_runner(
test_resources_dir / "docker-compose.yml", "hive"
) as docker_services:
wait_for_port(docker_services, "testhiveserver2", 10000, timeout=120)
# Set up the container.
command = "docker exec testhiveserver2 /opt/hive/bin/beeline -u jdbc:hive2://localhost:10000 -f /hive_setup.sql"
subprocess.run(command, shell=True, check=True)
# Run the metadata ingestion pipeline.
runner = CliRunner()
with fs_helpers.isolated_filesystem(tmp_path):
config_file = (test_resources_dir / "hive_to_file.yml").resolve()
result = runner.invoke(datahub, ["ingest", "-c", f"{config_file}"])
assert result.exit_code == 0
# Verify the output.
mce_helpers.check_golden_file(
pytestconfig,
output_path=tmp_path / "hive_mces.json",
golden_path=test_resources_dir / "hive_mces_golden.json",
ignore_paths=[
# example: root[1]['proposedSnapshot']['com.linkedin.pegasus2avro.metadata.snapshot.DatasetSnapshot']['aspects'][0]['com.linkedin.pegasus2avro.dataset.DatasetProperties']['customProperties']['CreateTime:']
# example: root[2]['proposedSnapshot']['com.linkedin.pegasus2avro.metadata.snapshot.DatasetSnapshot']['aspects'][0]['com.linkedin.pegasus2avro.dataset.DatasetProperties']['customProperties']['Table Parameters: transient_lastDdlTime']
r"root\[\d+\]\['proposedSnapshot'\]\['com\.linkedin\.pegasus2avro\.metadata\.snapshot\.DatasetSnapshot'\]\['aspects'\]\[\d+\]\['com\.linkedin\.pegasus2avro\.dataset\.DatasetProperties'\]\['customProperties'\]\['.*Time.*'\]"
],
)
def test_mssql_ingest(docker_compose_runner, pytestconfig, tmp_path,
mock_time):
test_resources_dir = pytestconfig.rootpath / "tests/integration/sql_server"
with docker_compose_runner(test_resources_dir / "docker-compose.yml",
"sql-server") as docker_services:
# Wait for SQL Server to be ready. We wait an extra couple seconds, as the port being available
# does not mean the server is accepting connections.
# TODO: find a better way to check for liveness.
wait_for_port(docker_services, "testsqlserver", 1433)
time.sleep(5)
# Run the setup.sql file to populate the database.
docker = "docker"
command = f"{docker} exec testsqlserver /opt/mssql-tools/bin/sqlcmd -S localhost -U sa -P 'test!Password' -d master -i /setup/setup.sql"
ret = subprocess.run(command,
shell=True,
stdout=subprocess.PIPE,
stderr=subprocess.PIPE)
assert ret.returncode == 0
# Run the metadata ingestion pipeline.
config_file = (test_resources_dir / "mssql_to_file.yml").resolve()
runner = CliRunner()
with fs_helpers.isolated_filesystem(tmp_path):
result = runner.invoke(datahub, ["ingest", "-c", f"{config_file}"])
assert result.exit_code == 0
output = mce_helpers.load_json_file("mssql_mces.json")
# Verify the output.
golden = mce_helpers.load_json_file(
str(test_resources_dir / "mssql_mce_golden.json"))
mce_helpers.assert_mces_equal(output, golden)
def test_trino_ingest(docker_compose_runner, pytestconfig, tmp_path,
mock_time):
test_resources_dir = pytestconfig.rootpath / "tests/integration/trino"
with docker_compose_runner(test_resources_dir / "docker-compose.yml",
"trino") as docker_services:
wait_for_port(docker_services, "testtrino", 8080)
wait_for_port(docker_services, "testhiveserver2", 10000, timeout=120)
docker_services.wait_until_responsive(
timeout=30,
pause=1,
check=lambda: requests.get("http://localhost:5300/v1/info").json()[
"starting"] is False,
)
# Set up the hive db
command = "docker exec testhiveserver2 /opt/hive/bin/beeline -u jdbc:hive2://localhost:10000 -f /hive_setup.sql"
subprocess.run(command, shell=True, check=True)
# Run the metadata ingestion pipeline.
runner = CliRunner()
with fs_helpers.isolated_filesystem(tmp_path):
print(tmp_path)
# Run the metadata ingestion pipeline for trino catalog referring to postgres database
config_file = (test_resources_dir / "trino_to_file.yml").resolve()
result = runner.invoke(datahub, ["ingest", "-c", f"{config_file}"])
assert_result_ok(result)
# Verify the output.
mce_helpers.check_golden_file(
pytestconfig,
output_path="trino_mces.json",
golden_path=test_resources_dir / "trino_mces_golden.json",
)
# Limitation 1 - MCE contains "nullable": true for all fields in trino database, irrespective of not null constraints present in underlying postgres database.
# This is issue with trino, also reported here - https://github.com/trinodb/trino/issues/6400, Related : https://github.com/trinodb/trino/issues/4070
# Limitation 2 - Dataset properties for postgres view (view definition, etc) are not part of MCE from trino.
# Postgres views are exposed as tables in trino. This setting depends on trino connector implementation - https://trino.io/episodes/18.html
# Run the metadata ingestion pipeline for trino catalog referring to hive database
config_file = (test_resources_dir /
"trino_hive_to_file.yml").resolve()
result = runner.invoke(datahub, ["ingest", "-c", f"{config_file}"])
assert_result_ok(result)
# Verify the output.
mce_helpers.check_golden_file(
pytestconfig,
output_path="trino_hive_mces.json",
golden_path=test_resources_dir / "trino_hive_mces_golden.json",
ignore_paths=[
r"root\[\d+\]\['proposedSnapshot'\]\['com.linkedin.pegasus2avro.metadata.snapshot.DatasetSnapshot'\]\['aspects'\]\[\d+\]\['com.linkedin.pegasus2avro.dataset.DatasetProperties'\]\['customProperties'\]\['transient_lastddltime'\]",
r"root\[\d+\]\['proposedSnapshot'\]\['com.linkedin.pegasus2avro.metadata.snapshot.DatasetSnapshot'\]\['aspects'\]\[\d+\]\['com.linkedin.pegasus2avro.dataset.DatasetProperties'\]\['customProperties'\]\['numfiles'\]",
r"root\[\d+\]\['proposedSnapshot'\]\['com.linkedin.pegasus2avro.metadata.snapshot.DatasetSnapshot'\]\['aspects'\]\[\d+\]\['com.linkedin.pegasus2avro.dataset.DatasetProperties'\]\['customProperties'\]\['totalsize'\]",
],
)
def test_trino_ingest(loaded_trino, test_resources_dir, pytestconfig, tmp_path,
mock_time):
# Run the metadata ingestion pipeline.
with fs_helpers.isolated_filesystem(tmp_path):
# Run the metadata ingestion pipeline for trino catalog referring to postgres database
mce_out_file = "trino_mces.json"
events_file = tmp_path / mce_out_file
pipeline_config = {
"run_id": "trino-test",
"source": {
"type":
data_platform,
"config":
TrinoConfig(
host_port="localhost:5300",
database="postgresqldb",
database_alias="library_catalog",
username="******",
schema_pattern=AllowDenyPattern(allow=["^librarydb"]),
profile_pattern=AllowDenyPattern(
allow=["library_catalog.librarydb.*"]),
profiling=GEProfilingConfig(
enabled=True,
include_field_null_count=True,
include_field_min_value=True,
include_field_max_value=True,
include_field_mean_value=True,
include_field_median_value=True,
include_field_stddev_value=True,
include_field_quantiles=True,
include_field_distinct_value_frequencies=True,
include_field_histogram=True,
include_field_sample_values=True,
),
).dict(),
},
"sink": {
"type": "file",
"config": FileSinkConfig(filename=str(events_file)).dict(),
},
}
# Run the metadata ingestion pipeline.
pipeline = Pipeline.create(pipeline_config)
pipeline.run()
pipeline.pretty_print_summary()
pipeline.raise_from_status(raise_warnings=True)
# Verify the output.
mce_helpers.check_golden_file(
pytestconfig,
output_path="trino_mces.json",
golden_path=test_resources_dir / "trino_mces_golden.json",
)
def run_datahub_cmd(
command: List[str], tmp_path: Optional[Path] = None, check_result: bool = True
) -> Result:
runner = CliRunner()
if tmp_path is None:
result = runner.invoke(datahub, command)
else:
with fs_helpers.isolated_filesystem(tmp_path):
result = runner.invoke(datahub, command)
if check_result:
assert_result_ok(result)
return result
def test_mysql_ingest(docker_compose_runner, pytestconfig, tmp_path,
mock_time):
test_resources_dir = pytestconfig.rootpath / "tests/integration/mysql"
with docker_compose_runner(test_resources_dir / "docker-compose.yml",
"mysql") as docker_services:
wait_for_port(docker_services, "testmysql", 3306)
# Run the metadata ingestion pipeline.
runner = CliRunner()
with fs_helpers.isolated_filesystem(tmp_path):
config_file = (test_resources_dir / "mysql_to_file.yml").resolve()
result = runner.invoke(datahub, ["ingest", "-c", f"{config_file}"])
assert result.exit_code == 0
# Verify the output.
mce_helpers.check_golden_file(
pytestconfig,
output_path="mysql_mces.json",
golden_path=test_resources_dir / "mysql_mces_golden.json",
)
def test_nifi_ingest(docker_compose_runner, pytestconfig, tmp_path, mock_time):
test_resources_dir = pytestconfig.rootpath / "tests/integration/nifi"
with docker_compose_runner(test_resources_dir / "docker-compose.yml",
"nifi") as docker_services:
wait_for_port(
docker_services,
container_name="nifi1",
container_port=9443,
timeout=300,
)
wait_for_port(
docker_services,
container_name="nifi01",
container_port=9080,
timeout=60,
)
wait_for_port(
docker_services,
container_name="nifi02",
container_port=9081,
timeout=60,
)
wait_for_port(
docker_services,
container_name="nifi03",
container_port=9082,
timeout=60,
)
# Wait for nifi to execute all processors
time.sleep(120)
# Run the metadata ingestion pipeline.
with fs_helpers.isolated_filesystem(tmp_path):
# Run nifi ingestion run.
pipeline = Pipeline.create({
"run_id": "nifi-test-standalone",
"source": {
"type": "nifi",
"config": {
"site_url": "http://localhost:9443/nifi/",
# "auth": "CLIENT_CERT",
# "client_cert_file": f"{test_resources_dir}/setup/ssl_files/client-cert.pem",
# "client_key_file": f"{test_resources_dir}/setup/ssl_files/client-private-key.pem",
# "client_key_password": "******",
# "ca_file": f"{test_resources_dir}/setup/ssl_files/server_certfile.pem",
"process_group_pattern": {
"deny": ["^WIP"]
},
},
},
"sink": {
"type": "file",
"config": {
"filename": "./nifi_mces.json"
},
},
})
pipeline.run()
pipeline.raise_from_status()
# Verify the output. ignore values for aspects having last_event_time values
# TODO: ignore paths with respect to aspect value in case of MCPs
mce_helpers.check_golden_file(
pytestconfig,
output_path="nifi_mces.json",
golden_path=test_resources_dir /
"nifi_mces_golden_standalone.json",
ignore_paths=[
r"root\[1\]\['aspect'\]\['value'\]",
r"root\[5\]\['aspect'\]\['value'\]",
r"root\[7\]\['aspect'\]\['value'\]",
],
)
# Run nifi ingestion run.
pipeline = Pipeline.create({
"run_id": "nifi-test-cluster",
"source": {
"type": "nifi",
"config": {
"site_url": "http://localhost:9080/nifi/",
"auth": "NO_AUTH",
"site_url_to_site_name": {
"http://nifi01:9080/nifi/": "default",
"http://nifi02:9081/nifi/": "default",
},
},
},
"sink": {
"type": "file",
"config": {
"filename": "./nifi_mces_cluster.json"
},
},
})
pipeline.run()
pipeline.raise_from_status()
# Verify the output.
# TODO: ignore paths with respect to aspect value in case of MCPs
mce_helpers.check_golden_file(
pytestconfig,
output_path="nifi_mces_cluster.json",
golden_path=test_resources_dir /
"nifi_mces_golden_cluster.json",
ignore_paths=[
r"root\[5\]\['aspect'\]\['value'\]",
r"root\[7\]\['aspect'\]\['value'\]",
r"root\[15\]\['aspect'\]\['value'\]",
r"root\[19\]\['aspect'\]\['value'\]",
],
)
