def test_usage_statistics_handler_validate_message_success(
caplog, in_memory_data_context_config_usage_stats_enabled,
sample_partial_message):
# caplog default is WARNING and above, we want to see DEBUG level messages for this test
caplog.set_level(
level=logging.DEBUG,
logger="great_expectations.core.usage_statistics.usage_statistics",
)
context: BaseDataContext = BaseDataContext(
in_memory_data_context_config_usage_stats_enabled)
usage_statistics_handler = UsageStatisticsHandler(
data_context=context,
data_context_id=in_memory_data_context_config_usage_stats_enabled.
anonymous_usage_statistics.data_context_id,
usage_statistics_url=in_memory_data_context_config_usage_stats_enabled.
anonymous_usage_statistics.usage_statistics_url,
)
assert (usage_statistics_handler._data_context_id ==
"00000000-0000-0000-0000-000000000001")
envelope = usage_statistics_handler.build_envelope(sample_partial_message)
validated_message = usage_statistics_handler.validate_message(
envelope, anonymized_usage_statistics_record_schema)
assert validated_message
assert not usage_stats_invalid_messages_exist(caplog.messages)
def test_usage_statistics_handler_build_envelope(
in_memory_data_context_config_usage_stats_enabled,
sample_partial_message):
"""This test is for a happy path only but will fail if there is an exception thrown in build_envelope"""
context: BaseDataContext = BaseDataContext(
in_memory_data_context_config_usage_stats_enabled)
usage_statistics_handler = UsageStatisticsHandler(
data_context=context,
data_context_id=in_memory_data_context_config_usage_stats_enabled.
anonymous_usage_statistics.data_context_id,
usage_statistics_url=in_memory_data_context_config_usage_stats_enabled.
anonymous_usage_statistics.usage_statistics_url,
)
assert (usage_statistics_handler._data_context_id ==
"00000000-0000-0000-0000-000000000001")
envelope = usage_statistics_handler.build_envelope(sample_partial_message)
required_keys = [
"event",
"event_payload",
"version",
"ge_version",
"data_context_id",
"data_context_instance_id",
"event_time",
]
assert all([key in envelope.keys() for key in required_keys])
assert envelope["version"] == "1.0.0"
assert envelope[
"data_context_id"] == "00000000-0000-0000-0000-000000000001"
def test_in_memory_runtime_context_configured_with_usage_stats_handler(
mock_emit, in_memory_runtime_context, test_pandas_df):
context: DataContext = in_memory_runtime_context
# manually set usage statistics handler
context._usage_statistics_handler = UsageStatisticsHandler(
data_context=context,
data_context_id=context._data_context_id,
usage_statistics_url="http://fakeendpoint.com",
)
catch_exceptions: bool = False # expect exceptions to be raised
result_format: dict = {
"result_format": "SUMMARY",
}
runtime_environment_arguments = {
"catch_exceptions": catch_exceptions,
"result_format": result_format,
}
suite: ExpectationSuite = in_memory_runtime_context.create_expectation_suite(
"test_suite", overwrite_existing=True)
expectation_configuration: ExpectationConfiguration
expectation_meta: dict = {"Notes": "Some notes"}
expectation_arguments_without_meta: dict
expectation_arguments_column: dict = {
"include_config": True,
"column": "Name", # use correct column to avoid error
}
expectation_arguments_without_meta = dict(**runtime_environment_arguments,
**expectation_arguments_column)
expectation_configuration = ExpectationConfiguration(
expectation_type="expect_column_values_to_not_be_null",
kwargs=expectation_arguments_without_meta,
meta=expectation_meta,
)
suite.add_expectation(expectation_configuration=expectation_configuration)
# emit 1 from add_expectation
assert mock_emit.call_count == 1
assert mock_emit.call_args_list == [
mock.call({
"event": "expectation_suite.add_expectation",
"event_payload": {},
"success": True,
})
]
def test_get_profiler_run_usage_statistics_with_handler_invalid_payload(
mock_data_context: mock.MagicMock, ):
# Ensure that real handler gets passed down by the context
handler: UsageStatisticsHandler = UsageStatisticsHandler(
mock_data_context, "my_id", "my_url")
mock_data_context.usage_statistics_handler = handler
profiler: RuleBasedProfiler = RuleBasedProfiler(
name="my_profiler", config_version=1.0, data_context=mock_data_context)
payload: dict = get_profiler_run_usage_statistics(profiler=profiler)
# Payload won't pass schema validation due to a lack of rules but we can confirm that it is anonymized
assert payload == {
"anonymized_name": "a0061ec021855cd2b3a994dd8d90fe5d",
"config_version": 1.0,
"rule_count": 0,
"variable_count": 0,
}
def test_get_profiler_run_usage_statistics_with_handler_valid_payload(
mock_data_context: mock.MagicMock, ):
# Ensure that real handler gets passed down by the context
handler: UsageStatisticsHandler = UsageStatisticsHandler(
mock_data_context, "my_id", "my_url")
mock_data_context.usage_statistics_handler = handler
profiler: RuleBasedProfiler = RuleBasedProfiler(
name="my_profiler", config_version=1.0, data_context=mock_data_context)
override_rules: Dict[str, dict] = {
"my_override_rule": {
"domain_builder": {
"class_name":
"ColumnDomainBuilder",
"module_name":
"great_expectations.rule_based_profiler.domain_builder",
},
"parameter_builders": [
{
"class_name": "MetricMultiBatchParameterBuilder",
"module_name":
"great_expectations.rule_based_profiler.parameter_builder",
"name": "my_parameter",
"metric_name": "my_metric",
},
{
"class_name":
"NumericMetricRangeMultiBatchParameterBuilder",
"module_name":
"great_expectations.rule_based_profiler.parameter_builder",
"name": "my_other_parameter",
"metric_name": "my_other_metric",
},
],
"expectation_configuration_builders": [
{
"class_name": "DefaultExpectationConfigurationBuilder",
"module_name":
"great_expectations.rule_based_profiler.expectation_configuration_builder",
"expectation_type":
"expect_column_pair_values_A_to_be_greater_than_B",
"column_A": "$domain.domain_kwargs.column_A",
"column_B": "$domain.domain_kwargs.column_B",
"my_one_arg": "$parameter.my_parameter.value[0]",
"meta": {
"details": {
"my_parameter_estimator":
"$parameter.my_parameter.details",
"note":
"Important remarks about estimation algorithm.",
},
},
},
{
"class_name": "DefaultExpectationConfigurationBuilder",
"module_name":
"great_expectations.rule_based_profiler.expectation_configuration_builder",
"expectation_type": "expect_column_min_to_be_between",
"column": "$domain.domain_kwargs.column",
"my_another_arg": "$parameter.my_other_parameter.value[0]",
"meta": {
"details": {
"my_other_parameter_estimator":
"$parameter.my_other_parameter.details",
"note":
"Important remarks about estimation algorithm.",
},
},
},
],
},
}
payload: dict = get_profiler_run_usage_statistics(profiler=profiler,
rules=override_rules)
assert payload == {
"anonymized_name":
"a0061ec021855cd2b3a994dd8d90fe5d",
"anonymized_rules": [{
"anonymized_domain_builder": {
"parent_class": "ColumnDomainBuilder"
},
"anonymized_expectation_configuration_builders": [
{
"expectation_type":
"expect_column_pair_values_A_to_be_greater_than_B",
"parent_class": "DefaultExpectationConfigurationBuilder",
},
{
"expectation_type": "expect_column_min_to_be_between",
"parent_class": "DefaultExpectationConfigurationBuilder",
},
],
"anonymized_name":
"bd8a8b4465a94b363caf2b307c080547",
"anonymized_parameter_builders": [
{
"anonymized_name": "25dac9e56a1969727bc0f90db6eaa833",
"parent_class": "MetricMultiBatchParameterBuilder",
},
{
"anonymized_name": "be5baa3f1064e6e19356f2168968cbeb",
"parent_class":
"NumericMetricRangeMultiBatchParameterBuilder",
},
],
}],
"config_version":
1.0,
"rule_count":
1,
"variable_count":
0,
}