mob next [ci-skip] [ci skip] [skip ci]

lastFile:tests/unit/test_project.py

src/skore/project.py

@@ -1,6 +1,5 @@
 """"""
 
-
 import base64
 import json
 from dataclasses import dataclass
@@ -24,6 +23,7 @@ class ItemType(StrEnum):
 @dataclass(frozen=True)
 class Item:
     """An item is a value that is stored in the project."""
+
     serialized: str
     raw: Any
     item_type: ItemType
@@ -126,24 +126,20 @@ def put(self, key: str, value: Any):
     def put_item(self, key: str, item: Item):
         self.storage.setitem(
             key,
-            (
-                str(item.item_type),
-                item.media_type,
-                item.serialized,
-            ),
+            {
+                "item_type": str(item.item_type),
+                "media_type": item.media_type,
+                "serialized": item.serialized,
+            },
         )
 
     def get(self, key: str) -> Any:
         return self.get_item(key).raw
 
     def get_item(self, key: str) -> Item:
-        item_type, media_type, serialized = self.storage.getitem(key)
+        item = self.storage.getitem(key)
 
-        return deserialize(
-            item_type,
-            media_type,
-            serialized,
-        )
+        return deserialize(**item)
 
     def list_keys(self) -> List[str]:
         """List all keys in the project."""

tests/unit/test_project.py

@@ -8,14 +8,13 @@
 from matplotlib import pyplot as plt
 from sklearn.ensemble import RandomForestClassifier
 from skore.project import Item, ItemType, Project, deserialize, serialize
-from skore.storage.filesystem import FileSystem
 from skore.storage.non_persistent_storage import NonPersistentStorage
 
 
 def test_transform_primitive():
     o = 3
     actual = serialize(o)
-    expected = Item(raw=3, raw_class_name="jsonable", serialized="3")
+    expected = Item(raw=3, item_type=ItemType.JSON, serialized="3")
     assert actual == expected
 
 
@@ -24,7 +23,7 @@ def test_transform_pandas_dataframe():
     actual = serialize(o)
     expected = Item(
         raw=o,
-        raw_class_name="pandas.DataFrame",
+        item_type=ItemType.PANDAS_DATAFRAME,
         serialized=o.to_json(orient="split"),
     )
 
@@ -36,7 +35,7 @@ def test_transform_numpy_ndarray():
     actual = serialize(o)
     expected = Item(
         raw=o,
-        raw_class_name="numpy.ndarray",
+        item_type=ItemType.NUMPY_ARRAY,
         serialized=json.dumps(o.tolist()),
     )
 
@@ -51,7 +50,7 @@ def test_transform_sklearn_base_baseestimator(monkeypatch):
     actual = serialize(o)
     expected = Item(
         raw=o,
-        raw_class_name="sklearn.base.BaseEstimator",
+        item_type=ItemType.SKLEARN_BASE_ESTIMATOR,
         serialized=json.dumps(
             {
                 "skops": "",
@@ -84,7 +83,9 @@ def test_untransform_primitive():
     transformed = serialize(o)
     assert (
         deserialize(
-            serialized=transformed.serialized, raw_class_name=transformed.raw_class_name
+            serialized=transformed.serialized,
+            item_type=transformed.item_type,
+            media_type=None,
         ).raw
         == o
     )
@@ -94,7 +95,8 @@ def test_untransform_pandas_dataframe():
     o = pandas.DataFrame([{"key": "value"}])
     item = deserialize(
         serialized=o.to_json(orient="split"),
-        raw_class_name="pandas.DataFrame",
+        item_type=ItemType.PANDAS_DATAFRAME,
+        media_type=None,
     )
 
     pandas.testing.assert_frame_equal(item.raw, o)
@@ -104,7 +106,8 @@ def test_untransform_numpy_ndarray():
     o = numpy.array([1, 2, 3])
     item = deserialize(
         serialized=json.dumps(o.tolist()),
-        raw_class_name="numpy.ndarray",
+        item_type=ItemType.NUMPY_ARRAY,
+        media_type=None,
     )
 
     numpy.testing.assert_array_equal(o, item.raw)
@@ -113,7 +116,7 @@ def test_untransform_numpy_ndarray():
 def test_untransform_sklearn_model():
     o = sklearn.svm.SVC()
     t = serialize(o)
-    u = deserialize(t.serialized, t.raw_class_name)
+    u = deserialize(serialized=t.serialized, item_type=t.item_type, media_type=None)
 
     assert isinstance(u.raw, sklearn.svm.SVC)
 
@@ -129,9 +132,11 @@ def savefig(*args, **kwargs):
 
     s = serialize(fig)
 
-    assert deserialize(s.serialized, s.raw_class_name, s.media_type) == Item(
+    assert deserialize(
+        serialized=s.serialized, item_type=s.item_type, media_type=s.media_type
+    ) == Item(
         raw=None,
-        raw_class_name="matplotlib.figure.Figure",
+        item_type=ItemType.MEDIA,
         serialized="",
         media_type="image/svg+xml",
     )
@@ -180,28 +185,56 @@ def savefig(*args, **kwargs):
     project.put("rf_model", model)  # ScikitLearnModelItem
 
     assert project.storage.content == {
-        "string_item": ("jsonable", '"Hello, World!"', None),
-        "int_item": ("jsonable", "42", None),
-        "float_item": ("jsonable", "3.14", None),
-        "bool_item": ("jsonable", "true", None),
-        "list_item": ("jsonable", "[1, 2, 3]", None),
-        "dict_item": ("jsonable", '{"key": "value"}', None),
-        "pandas_df": (
-            "pandas.DataFrame",
-            '{"columns":["A","B"],"index":[0,1,2],"data":[[1,4],[2,5],[3,6]]}',
-            None,
-        ),
-        "numpy_array": ("numpy.ndarray", "[1, 2, 3, 4, 5]", None),
-        "mpl_figure": (
-            "matplotlib.figure.Figure",
-            "",
-            "image/svg+xml",
-        ),
-        "rf_model": (
-            "sklearn.base.BaseEstimator",
-            '{"skops": "", "html": ""}',
-            "text/html",
-        ),
+        "string_item": {
+            "item_type": ItemType.JSON,
+            "serialized": '"Hello, World!"',
+            "media_type": None,
+        },
+        "int_item": {
+            "item_type": ItemType.JSON,
+            "serialized": "42",
+            "media_type": None,
+        },
+        "float_item": {
+            "item_type": ItemType.JSON,
+            "serialized": "3.14",
+            "media_type": None,
+        },
+        "bool_item": {
+            "item_type": ItemType.JSON,
+            "serialized": "true",
+            "media_type": None,
+        },
+        "list_item": {
+            "item_type": ItemType.JSON,
+            "serialized": "[1, 2, 3]",
+            "media_type": None,
+        },
+        "dict_item": {
+            "item_type": ItemType.JSON,
+            "serialized": '{"key": "value"}',
+            "media_type": None,
+        },
+        "pandas_df": {
+            "item_type": ItemType.JSON,
+            "serialized": '{"columns":["A","B"],"index":[0,1,2],"data":[[1,4],[2,5],[3,6]]}',
+            "media_type": None,
+        },
+        "numpy_array": {
+            "item_type": ItemType.NUMPY_ARRAY,
+            "serialized": "[1, 2, 3, 4, 5]",
+            "media_type": None,
+        },
+        "mpl_figure": {
+            "item_type": ItemType.MEDIA,
+            "serialized": "",
+            "media_type": "image/svg+xml",
+        },
+        "rf_model": {
+            "item_type": ItemType.SKLEARN_BASE_ESTIMATOR,
+            "serialized": '{"skops": "", "html": ""}',
+            "media_type": "text/html",
+        },
     }
 
     assert project.get("string_item") == "Hello, World!"