Feature/unit tests/infra 30038/cim tests (#277)

* Unit tests dir structure and GA workflow

* commenting out test matrix worklow

* eventtype_parser unit tests (#251)

* eventtype_parser unit tests

* eventtype_parser no config file update

* review remarks #1

* Feature/unit tests/tags parser (#257)

* eventtype_parser unit tests

* eventtype_parser no config file update

* review remarks #1

* conftest parametrization

* wrapped pytest arguments

* test_tags_parser

* build_parsed_output

* fixture scope

* typo change

* review remarks #1

* review remarks #2

* Feature/unit tests/infra 30036/props parser (#259)

unit tests for props_parser

* Feature/unit tests/infra 30109/fields (#260)

* Tests for fields and addon_parser init

* Feature/unit tests/infra 30097/transforms parser (#261)

* test_transforms_parser

* more tests for transforms_parser

* [CR] fixtures at the top of file

* Feature/unit tests/infra 30039/event ingestors (#265)

* get event ingestor

* test_events_can_be_ingested

* reocrdtype added for unit tests

* test_requirement_tests_can_be_run

* test_requirement_tests_can_be_run #2

* wrongly added file deletion

* test_hec_raw_ingestor

* requests-mock typo fix

* test_hec_raw_ingestor

* test_requirement_tests_can_be_run assert has calls fix

* hec_event_ingestor

* HECMetricEventIngestor

* import fix

* test_sc4s_event_ingestor

* test_sc4s_event_ingest additional asserts

* test_event_ingestors/test_requirement_event_ingestor

* [CR] assert exception string

* Feature/unit tests/infra 30041/sample generation (#266)

* First part of unit tests for sample_generation

* Changing location of test files (#267)

* reload module change

* test_data_set

* test_data_model

* test_json_schema.py

* test_data_model_handler

* test_field_test_adapter

* test_field_test_helper

* more tests for test_field_test_helper

* test_test_generator

* conftest

* more tests for test_generator

* more tests for test_generator

* poetry lock conflicts resolved

* [CR] review remarks

Co-authored-by: uoboda-splunk <[email protected]>
Co-authored-by: uoboda-splunk <[email protected]>

Author: kkania-splunk

poetry.lock

@@ -3,3 +3,4 @@ addopts = -v --tb=long --log-level=INFO
 #    --force-flaky --max-runs=3 --min-passes=1
 filterwarnings =
     ignore::DeprecationWarning
+

tests/unit/pytest.ini

@@ -0,0 +1,16 @@
+import pytest
+from unittest.mock import MagicMock, mock_open
+
+
+@pytest.fixture()
+def open_mock(monkeypatch):
+    open_mock = mock_open()
+    monkeypatch.setattr("builtins.open", open_mock)
+    return open_mock
+
+
+@pytest.fixture()
+def json_load_mock(monkeypatch):
+    load_mock = MagicMock()
+    monkeypatch.setattr("json.load", load_mock)
+    return load_mock

tests/unit/tests_standard_lib/test_cim_tests/__init__.py

@@ -0,0 +1,62 @@
+import pytest
+from unittest.mock import patch, MagicMock, call
+from collections import namedtuple
+from pytest_splunk_addon.standard_lib.cim_tests.data_model import DataModel
+
+
+@pytest.fixture()
+def data_set_mock(monkeypatch):
+    data_set_mock = MagicMock()
+    data_set_mock.return_value = data_set_mock
+    data_set_mock.load_dataset.return_value = ["dataset1", "dataset2"]
+    monkeypatch.setattr(
+        "pytest_splunk_addon.standard_lib.cim_tests.data_model.DataSet", data_set_mock
+    )
+    return data_set_mock
+
+
+def test_data_model_instantiation(data_set_mock):
+    data_model = DataModel({"model_name": "test_model_name", "objects": []})
+    assert data_model.name == "test_model_name"
+    assert data_model.root_data_set == ["dataset1", "dataset2"]
+
+
+def test_data_model_string(data_set_mock):
+    data_model = DataModel({"model_name": "test_model_name", "objects": []})
+    assert str(data_model) == data_model.name == "test_model_name"
+
+
+def test_get_mapped_datasets_calls_internal_function():
+    m1 = MagicMock()
+    m1.side_effect = lambda x, y: (i for i in range(3))
+    with patch.object(DataModel, "__init__", return_value=None), patch.object(
+        DataModel, "_get_mapped_datasets", m1
+    ):
+        data_model = DataModel({})
+        data_model.root_data_set = ["root_data_set"]
+        assert list(data_model.get_mapped_datasets(["addons_tags"])) == [0, 1, 2]
+        m1.assert_has_calls([call(["addons_tags"], ["root_data_set"])])
+
+
+def test__get_mapped_datasets():
+    data_set = namedtuple("DataSet", ["name", "match_tags", "child_dataset"])
+    dataset1 = data_set(
+        "dataset1",
+        lambda x: True,
+        [
+            data_set("dataset1a", lambda x: True, []),
+            data_set(
+                "dataset1b",
+                lambda x: False,
+                [data_set("dataset1c", lambda x: True, [])],
+            ),
+        ],
+    )
+    dataset2 = data_set("dataset2", lambda x: True, [])
+    with patch.object(DataModel, "__init__", return_value=None):
+        data_model = DataModel({})
+    assert list(data_model._get_mapped_datasets([], [dataset1, dataset2])) == [
+        [dataset1],
+        [dataset1, dataset1.child_dataset[0]],
+        [dataset2],
+    ]

tests/unit/tests_standard_lib/test_cim_tests/conftest.py

@@ -0,0 +1,125 @@
+import pytest
+from unittest.mock import MagicMock, call, patch, PropertyMock
+from collections import namedtuple
+from pytest_splunk_addon.standard_lib.cim_tests.data_model_handler import (
+    DataModelHandler,
+)
+
+
+@pytest.fixture()
+def listdir_mock(monkeypatch):
+    monkeypatch.setattr(
+        "os.listdir",
+        MagicMock(return_value=["model1.json", "model2.xml", "model3.json"]),
+    )
+
+
+@pytest.fixture()
+def data_model_mock(monkeypatch):
+    dm = MagicMock()
+    dm.side_effect = ["data_model_instance_1", "data_model_instance_2"]
+    monkeypatch.setattr(
+        "pytest_splunk_addon.standard_lib.cim_tests.data_model_handler.DataModel", dm
+    )
+    return dm
+
+
+@pytest.fixture()
+def json_schema_mock(monkeypatch):
+    js = MagicMock()
+    js.parse_data_model.side_effect = ["parsed_data_model_1", "parsed_data_model_2"]
+    monkeypatch.setattr(
+        "pytest_splunk_addon.standard_lib.cim_tests.data_model_handler.JSONSchema", js
+    )
+    return js
+
+
+def test_data_models():
+    with patch.object(
+        DataModelHandler,
+        "load_data_models",
+        return_value=(f"data_model_{i+1}" for i in range(3)),
+    ):
+        dmh = DataModelHandler("/fake_path")
+        assert dmh.data_models == ["data_model_1", "data_model_2", "data_model_3"]
+        assert dmh._data_models == ["data_model_1", "data_model_2", "data_model_3"]
+
+
+def test_load_data_model(listdir_mock, data_model_mock, json_schema_mock):
+    dmh = DataModelHandler("/fake_path")
+    assert list(dmh.load_data_models("/fake_path/data_models")) == [
+        "data_model_instance_1",
+        "data_model_instance_2",
+    ]
+    data_model_mock.assert_has_calls(
+        [call("parsed_data_model_1"), call("parsed_data_model_2")]
+    )
+    json_schema_mock.parse_data_model.assert_has_calls(
+        [
+            call("/fake_path/data_models/model1.json"),
+            call("/fake_path/data_models/model3.json"),
+        ]
+    )
+
+
+def test_get_all_tags_per_stanza():
+    dmh = DataModelHandler("/fake_path")
+    addon_parser = namedtuple("AddonParser", ["get_tags"])
+    ap = addon_parser(
+        lambda: [
+            {
+                "stanza": "test_stanza_1",
+                "tag": {"tag1_key": "tag1_value", "tag2_key": "tag2_value"},
+            },
+            {"stanza": "test_stanza_2", "tag": {"tag3_key": "tag3_value"}},
+            {"stanza": "test_stanza_1", "tag": {"tag4_key": "tag4_value"}},
+        ]
+    )
+    assert dmh._get_all_tags_per_stanza(ap) == {
+        "test_stanza_1": [
+            {"tag1_key": "tag1_value", "tag2_key": "tag2_value"},
+            {"tag4_key": "tag4_value"},
+        ],
+        "test_stanza_2": [{"tag3_key": "tag3_value"}],
+    }
+
+
+def test_get_mapped_data_models(caplog):
+    data_model = namedtuple("DataModel", ["get_mapped_datasets"])
+    data_models = [
+        data_model(lambda x: x if len(x) == 1 else []),
+        data_model(lambda x: []),
+        data_model(lambda x: x if len(x) == 3 else []),
+    ]
+    with patch.object(
+        DataModelHandler,
+        "_get_all_tags_per_stanza",
+        return_value={
+            "stanza_1": [{"tag_key_1a": "tag_value_1a", "tag_key_1b": "tag_value_1b"}],
+            "stanza_2": [
+                {"tag_key_2a": "tag_value_2a", "tag_key_2b": "tag_value_2b"},
+                None,
+            ],
+            "stanza_3": [
+                {"tag_key_3a": "tag_value_3a"},
+                {"tag_key_3b": "tag_value_3b"},
+                {"tag_key_3c": "tag_value_3c"},
+            ],
+        },
+    ), patch.object(
+        DataModelHandler,
+        "data_models",
+        new_callable=PropertyMock,
+        return_value=data_models,
+    ):
+        dmh = DataModelHandler("/fake_path")
+        assert list(dmh.get_mapped_data_models("addon_parser")) == [
+            (
+                "stanza_1",
+                {"tag_key_1a": "tag_value_1a", "tag_key_1b": "tag_value_1b"},
+            ),
+            ("stanza_3", {"tag_key_3a": "tag_value_3a"}),
+            ("stanza_3", {"tag_key_3b": "tag_value_3b"}),
+            ("stanza_3", {"tag_key_3c": "tag_value_3c"}),
+        ]
+        assert "No Data Model mapped for stanza_2" in caplog.messages

tests/unit/tests_standard_lib/test_cim_tests/test_data_model.py

@@ -0,0 +1,125 @@
+import pytest
+from unittest.mock import patch, call
+from collections import namedtuple
+from pytest_splunk_addon.standard_lib.cim_tests.data_set import DataSet
+
+field = namedtuple("Field", ["name"])
+
+
+@pytest.fixture()
+def mocked_dataset_constructor():
+    from pytest_splunk_addon.standard_lib.cim_tests.data_set import Field
+
+    with patch.object(
+        DataSet, "load_dataset", return_value=("child_dataset1", "child_dataset2")
+    ), patch.object(
+        Field, "parse_fields", return_value=("field1", "field2")
+    ), patch.object(
+        DataSet,
+        "_parse_fields_cluster",
+        return_value="parse_field_cluster_return_value",
+    ), patch.object(
+        DataSet, "_parse_constraint", return_value="prase_constraints_return_value"
+    ):
+        return DataSet(
+            {
+                "name": "dataset1",
+                "tags": ["test_tag1", "test_tag2"],
+                "child_dataset": [],
+                "fields": [{"name": "app"}],
+                "fields_cluster": [],
+                "search_constraints": "tag=alert",
+            },
+            "test_data_model",
+        )
+
+
+@pytest.fixture()
+def dataset_mock():
+    with patch.object(DataSet, "__init__", return_value=None):
+        dataset = DataSet({}, "test_data_model")
+    dataset.name = "dataset2"
+    dataset.fields = [
+        field("bytes"),
+        field("bytes_in"),
+        field("bytes_out"),
+        field("dest"),
+        field("dest_ip"),
+        field("dest_mac"),
+    ]
+    dataset.tags = [["tag1", "tag1b"], ["tag2", "tag3"]]
+    return dataset
+
+
+def test_dataset_can_be_loaded():
+    with patch.object(DataSet, "__init__", return_value=None) as data_set_mock:
+        list(
+            DataSet.load_dataset(
+                [
+                    {"name": "dataset1", "tags": ["tag1"]},
+                    {"name": "dataset2", "tags": ["tag2"]},
+                ],
+                "test",
+            )
+        )
+    data_set_mock.assert_has_calls(
+        [
+            call({"name": "dataset1", "tags": ["tag1"]}, "test"),
+            call({"name": "dataset2", "tags": ["tag2"]}, "test"),
+        ]
+    )
+
+
+def test_dataset_instantiation(mocked_dataset_constructor):
+    assert mocked_dataset_constructor.name == "dataset1"
+    assert mocked_dataset_constructor.tags == ["test_tag1", "test_tag2"]
+    assert mocked_dataset_constructor.data_model == "test_data_model"
+    assert mocked_dataset_constructor.child_dataset == [
+        "child_dataset1",
+        "child_dataset2",
+    ]
+    assert mocked_dataset_constructor.fields == ["field1", "field2"]
+    assert (
+        mocked_dataset_constructor.fields_cluster == "parse_field_cluster_return_value"
+    )
+    assert (
+        mocked_dataset_constructor.search_constraints
+        == "prase_constraints_return_value"
+    )
+
+
+def test_dataset_string(mocked_dataset_constructor):
+    assert (
+        mocked_dataset_constructor.name == str(mocked_dataset_constructor) == "dataset1"
+    )
+
+
+def test_parse_constraints():
+    assert DataSet._parse_constraint("constraints") == "constraints"
+
+
+def test_parse_fields_cluster(dataset_mock):
+    assert dataset_mock._parse_fields_cluster(
+        [["bytes", "bytes_in", "bytes_out"], ["dest", "dest_ip", "dest_mac"]]
+    ) == [
+        [field("bytes"), field("bytes_in"), field("bytes_out")],
+        [field("dest"), field("dest_ip"), field("dest_mac")],
+    ]
+
+
+def test_parse_fields_raises_error_when_cluster_field_not_in_fields_list(dataset_mock):
+    with pytest.raises(
+        AssertionError,
+        match="Dataset=dataset2, Each cluster field should be included in fields list",
+    ):
+        dataset_mock._parse_fields_cluster(
+            [["bytes", "bytes_in", "bytes_out", "bytes_all"]]
+        )
+
+
+def test_tags_match(dataset_mock):
+    assert dataset_mock.match_tags(["tag1", "tag1a", "tag2", "tag3"]) is True
+
+
+def test_tags_not_match(dataset_mock):
+    assert dataset_mock.match_tags(["tag1", "tag1a", "tag2"]) is None