updated unit tests

Author: wei-qlu

python/lib/sift_client/_tests/resources/test_data_imports.py

@@ -4,9 +4,9 @@
 
 import pytest
 
+from sift_client.resources.data_imports import _resolve_data_type_key
 from sift_client.sift_types.channel import ChannelDataType
 from sift_client.sift_types.data_import import (
-    EXTENSION_TO_DATA_TYPE_KEY,
     Ch10ImportConfig,
     CsvDataColumn,
     CsvImportConfig,
@@ -16,6 +16,7 @@
     Hdf5ImportConfig,
     ParquetDataColumn,
     ParquetFlatDatasetImportConfig,
+    ParquetSingleChannelPerRowImportConfig,
     ParquetTimeColumn,
     TdmsImportConfig,
     TimeFormat,
@@ -57,54 +58,6 @@ def parquet_config():
     )
 
 
-class TestCsvConfigMutability:
-    def test_mutate_asset_name(self, csv_config):
-        csv_config.asset_name = "new_asset"
-        assert csv_config.asset_name == "new_asset"
-
-    def test_mutate_run_name(self, csv_config):
-        csv_config.run_name = "new_run"
-        assert csv_config.run_name == "new_run"
-
-    def test_mutate_column_data_type(self, csv_config):
-        csv_config.data_columns[1].data_type = ChannelDataType.STRING
-        assert csv_config.data_columns[1].data_type == ChannelDataType.STRING
-
-    def test_mutate_column_name(self, csv_config):
-        csv_config.data_columns[0].name = "cpu_utilization"
-        assert csv_config.data_columns[0].name == "cpu_utilization"
-
-    def test_append_column(self, csv_config):
-        csv_config.data_columns.append(
-            CsvDataColumn(column=5, name="pressure", data_type=ChannelDataType.DOUBLE)
-        )
-        assert len(csv_config.data_columns) == 4
-        assert csv_config.data_columns[-1].name == "pressure"
-
-    def test_remove_column(self, csv_config):
-        csv_config.data_columns = [
-            dc for dc in csv_config.data_columns if dc.name != "status_flags"
-        ]
-        assert len(csv_config.data_columns) == 2
-        assert all(dc.name != "status_flags" for dc in csv_config.data_columns)
-
-
-class TestParquetConfigMutability:
-    def test_mutate_asset_name(self, parquet_config):
-        parquet_config.asset_name = "new_asset"
-        assert parquet_config.asset_name == "new_asset"
-
-    def test_mutate_column_data_type(self, parquet_config):
-        parquet_config.data_columns[1].data_type = ChannelDataType.STRING
-        assert parquet_config.data_columns[1].data_type == ChannelDataType.STRING
-
-    def test_append_column(self, parquet_config):
-        parquet_config.data_columns.append(
-            ParquetDataColumn(path="pressure", name="pressure", data_type=ChannelDataType.DOUBLE)
-        )
-        assert len(parquet_config.data_columns) == 4
-
-
 class TestGetColumn:
     def test_csv_get_column(self, csv_config):
         col = csv_config.get_column("cpu_util")
@@ -169,94 +122,6 @@ def test_absolute_time_does_not_require_start_time(self):
         assert col.relative_start_time is None
 
 
-class TestDataTypeKey:
-    def test_csv_extension(self):
-        assert EXTENSION_TO_DATA_TYPE_KEY[".csv"] == DataTypeKey.CSV
-
-    def test_parquet_not_in_extension_map(self):
-        assert ".parquet" not in EXTENSION_TO_DATA_TYPE_KEY
-
-    def test_hdf5_extensions(self):
-        assert EXTENSION_TO_DATA_TYPE_KEY[".h5"] == DataTypeKey.HDF5
-        assert EXTENSION_TO_DATA_TYPE_KEY[".hdf5"] == DataTypeKey.HDF5
-
-
-class TestDetectConfigValidation:
-    """Tests for validation checks applied after detect_config."""
-
-    def test_csv_no_data_columns_raises(self):
-        """If all columns are filtered out, detect_config should raise."""
-        config = CsvImportConfig(
-            asset_name="",
-            time_column=CsvTimeColumn(column=1, format=TimeFormat.ABSOLUTE_RFC3339),
-            data_columns=[],
-        )
-        assert not config.data_columns
-
-    def test_parquet_empty_time_column_path(self):
-        """An empty time column path indicates detection failed."""
-        config = ParquetFlatDatasetImportConfig(
-            asset_name="",
-            time_column=ParquetTimeColumn(path=""),
-            data_columns=[
-                ParquetDataColumn(
-                    path="cpu_util", name="cpu_util", data_type=ChannelDataType.DOUBLE
-                ),
-            ],
-        )
-        assert not config.time_column.path
-
-    def test_parquet_no_data_columns(self):
-        """A config with no data columns indicates detection found nothing useful."""
-        config = ParquetFlatDatasetImportConfig(
-            asset_name="",
-            time_column=ParquetTimeColumn(path="timestamp"),
-            data_columns=[],
-        )
-        assert not config.data_columns
-
-    def test_parquet_integer_time_column_fallback(self):
-        """An integer column starting with 'time' should be usable as the time column."""
-        config = ParquetFlatDatasetImportConfig(
-            asset_name="",
-            time_column=ParquetTimeColumn(path=""),
-            data_columns=[
-                ParquetDataColumn(path="time_ns", name="time_ns", data_type=ChannelDataType.INT_64),
-                ParquetDataColumn(
-                    path="cpu_util", name="cpu_util", data_type=ChannelDataType.DOUBLE
-                ),
-            ],
-        )
-        _integer_types = {
-            ChannelDataType.INT_32,
-            ChannelDataType.INT_64,
-            ChannelDataType.UINT_32,
-            ChannelDataType.UINT_64,
-        }
-        match = None
-        for dc in config.data_columns:
-            if dc.data_type in _integer_types and dc.name.lower().startswith("time"):
-                match = dc
-                break
-        assert match is not None
-        assert match.path == "time_ns"
-
-
-class TestRunPrecedence:
-    def test_run_id_ignored_when_none(self, csv_config):
-        csv_config.run_id = None
-        csv_config.run_name = "my_run"
-        proto = csv_config._to_proto()
-        assert proto.run_name == "my_run"
-        assert proto.run_id == ""
-
-    def test_run_id_set(self, csv_config):
-        csv_config.run_id = "run_123"
-        csv_config.run_name = "ignored"
-        proto = csv_config._to_proto()
-        assert proto.run_id == "run_123"
-
-
 class TestCh10Config:
     def test_to_proto(self):
         config = Ch10ImportConfig(asset_name="my_asset", run_name="run1", scale_values=True)
@@ -271,10 +136,6 @@ def test_to_proto_defaults(self):
         assert proto.run_name == ""
         assert proto.scale_values is False
 
-    def test_run_id_inherited_but_unused(self):
-        config = Ch10ImportConfig(asset_name="my_asset")
-        assert config.run_id is None
-
 
 class TestTdmsConfig:
     def test_to_proto(self):
@@ -421,9 +282,90 @@ def test_absolute_time_no_start_time_required(self):
         assert not proto.HasField("relative_start_time")
 
 
-class TestExtensionMap:
-    def test_tdms_extension(self):
-        assert EXTENSION_TO_DATA_TYPE_KEY[".tdms"] == DataTypeKey.TDMS
+class TestCsvToProto:
+    def test_to_proto(self, csv_config):
+        proto = csv_config._to_proto()
+        assert proto.asset_name == "test_asset"
+        assert proto.run_name == "test_run"
+        assert proto.first_data_row == 2
+        assert proto.time_column.column_number == 1
+        assert len(proto.data_columns) == 3
+        assert proto.data_columns[2].name == "cpu_util"
+
+    def test_from_proto_round_trip(self, csv_config):
+        proto = csv_config._to_proto()
+        restored = CsvImportConfig._from_proto(proto)
+        assert restored.asset_name == csv_config.asset_name
+        assert restored.run_name == csv_config.run_name
+        assert restored.first_data_row == csv_config.first_data_row
+        assert restored.time_column.column == csv_config.time_column.column
+        assert len(restored.data_columns) == len(csv_config.data_columns)
+
+
+class TestParquetToProto:
+    def test_flat_dataset_to_proto(self, parquet_config):
+        proto = parquet_config._to_proto()
+        assert proto.asset_name == "test_asset"
+        assert proto.HasField("flat_dataset")
+        assert proto.flat_dataset.time_column.path == "timestamp"
+        assert len(proto.flat_dataset.data_columns) == 3
+
+    def test_flat_dataset_from_proto_round_trip(self, parquet_config):
+        proto = parquet_config._to_proto()
+        restored = ParquetFlatDatasetImportConfig._from_proto(proto)
+        assert restored.asset_name == parquet_config.asset_name
+        assert restored.time_column.path == parquet_config.time_column.path
+        assert len(restored.data_columns) == len(parquet_config.data_columns)
+        for orig, rest in zip(parquet_config.data_columns, restored.data_columns):
+            assert orig.name == rest.name
+            assert orig.data_type == rest.data_type
+
+    def test_single_channel_per_row_from_proto_round_trip(self):
+        from sift_client.sift_types.data_import import ParquetSingleChannelConfig
+
+        config = ParquetSingleChannelPerRowImportConfig(
+            asset_name="a",
+            time_column=ParquetTimeColumn(path="ts"),
+            single_channel=ParquetSingleChannelConfig(
+                data_path="value",
+                name="voltage",
+                data_type=ChannelDataType.DOUBLE,
+            ),
+        )
+        proto = config._to_proto()
+        restored = ParquetSingleChannelPerRowImportConfig._from_proto(proto)
+        assert restored.single_channel is not None
+        assert restored.single_channel.name == "voltage"
+        assert restored.single_channel.data_type == ChannelDataType.DOUBLE
+
+
+class TestParquetTimeColumnToProto:
+    def test_empty_path_raises(self):
+        col = ParquetTimeColumn(path="")
+        with pytest.raises(ValueError, match="path must be set"):
+            col._to_proto()
+
+
+class TestResolveDataTypeKey:
+    def test_parquet_requires_data_type(self):
+        with pytest.raises(ValueError, match="data_type"):
+            _resolve_data_type_key(".parquet", None)
+
+    def test_parquet_with_explicit_data_type(self):
+        result = _resolve_data_type_key(".parquet", DataTypeKey.PARQUET_FLATDATASET)
+        assert result == DataTypeKey.PARQUET_FLATDATASET
+
+    def test_pqt_requires_data_type(self):
+        with pytest.raises(ValueError, match="data_type"):
+            _resolve_data_type_key(".pqt", None)
+
+    def test_known_extension_uses_map(self):
+        assert _resolve_data_type_key(".csv", None) == DataTypeKey.CSV
+
+    def test_explicit_data_type_overrides_extension(self):
+        result = _resolve_data_type_key(".csv", DataTypeKey.TDMS)
+        assert result == DataTypeKey.TDMS
 
-    def test_ch10_extension(self):
-        assert EXTENSION_TO_DATA_TYPE_KEY[".ch10"] == DataTypeKey.CH10
+    def test_unknown_extension_raises(self):
+        with pytest.raises(ValueError, match="Unsupported file extension"):
+            _resolve_data_type_key(".xyz", None)