Merge pull request #252 from EMMC-ASBL/optional-parse-label

Made `label` optional in the parse strategy

oteapi_dlite/strategies/parse.py

@@ -21,11 +21,11 @@ class DLiteParseConfig(AttrDict):
     """Configuration for generic DLite parser."""
 
     driver: Annotated[
-        Optional[str],
+        str,
         Field(
             description='Name of DLite driver (ex: "json").',
         ),
-    ] = None
+    ]
     location: Annotated[
         Optional[str],
         Field(
@@ -52,11 +52,24 @@ class DLiteParseConfig(AttrDict):
         ),
     ] = None
     label: Annotated[
-        str,
+        Optional[str],
+        Field(
+            description=(
+                "Optional label of the new DLite instance in the collection."
+            ),
+        ),
+    ] = None
+    datamodel: Annotated[
+        Optional[str],
         Field(
-            description="Label of the new DLite instance in the collection.",
+            description=(
+                "DLite datamodel documenting the structure of the data set. "
+                "Often unused, since the datamodel is implicitly defined in "
+                "the DLite driver (DLite plugin), but for a documentation "
+                "point of view this is a very important field."
+            ),
         ),
-    ]
+    ] = None
     datacache_config: Annotated[
         Optional[DataCacheConfig],
         Field(
@@ -154,7 +167,8 @@ def get(
 
         # Insert inst into collection
         coll = get_collection(session)
-        coll.add(config.label, inst)
+        label = config.label if config.label else inst.uuid
+        coll.add(label, inst)
 
         # __TODO__
         # See

tests/entities/Molecule.yaml

@@ -0,0 +1,34 @@
+da31ff66-98af-590c-855d-386c63617755:
+  uuid: da31ff66-98af-590c-855d-386c63617755
+  uri: http://onto-ns.com/meta/0.1/Molecule
+  meta: http://onto-ns.com/meta/0.3/EntitySchema
+  description: A minimal description of a molecules
+  dimensions:
+    natoms: Number of atoms
+    ncoords: Number coordinates. Always 3
+  properties:
+    name:
+      type: string
+      description: Name of the molecule.
+    positions:
+      type: float64
+      shape:
+      - natoms
+      - ncoords
+      unit: "\xC5ngstr\xF6m"
+      description: Atomic positions in Cartesian coordinates.
+    symbols:
+      type: string
+      shape:
+      - natoms
+      description: Chemical symbols.
+    masses:
+      type: float64
+      shape:
+      - natoms
+      unit: u
+      description: Atomic masses.
+    groundstate_energy:
+      type: float64
+      unit: eV
+      description: Molecule ground state energy.

tests/static/molecule.json

@@ -0,0 +1,16 @@
+{
+  "b8906302-5cd9-5e34-9fec-9d6101e52554":   {
+    "meta": "http://onto-ns.com/meta/0.1/Molecule",
+    "dimensions": {
+      "natoms": 2,
+      "ncoords": 3
+    },
+    "properties": {
+      "name": "H2",
+      "positions": [[-1.64097, 1.37077, 0], [-1.03814, 1.74207, 0]],
+      "symbols": ["H", "H"],
+      "masses": [1.008, 1.008],
+      "groundstate_energy": 1.34613
+    }
+  }
+}

tests/strategies/paths.py

@@ -11,6 +11,7 @@
 entitydir = testdir / "entities"
 inputdir = testdir / "input"
 outputdir = testdir / "output"
+staticdir = testdir / "static"
 
 
 # Add utils to sys path

tests/strategies/test_parse.py

@@ -0,0 +1,99 @@
+"""Test the image formats in the image parse strategy."""
+
+# pylint: disable=too-many-locals
+from typing import TYPE_CHECKING
+
+if TYPE_CHECKING:
+    from pathlib import Path
+    from typing import Optional
+
+    from oteapi.interfaces import IParseStrategy
+
+    from oteapi_dlite.models import DLiteSessionUpdate
+
+
+# if True:
+def test_parse_no_options() -> None:
+    """Test the dlite-parse strategy."""
+
+    import dlite
+    from oteapi.datacache import DataCache
+    from paths import staticdir
+
+    from oteapi_dlite.strategies.parse import DLiteParseStrategy
+
+    sample_file = staticdir / "molecule.json"
+
+    cache = DataCache()
+
+    orig_key = cache.add(sample_file.read_bytes())
+    config = {
+        "downloadUrl": sample_file.as_uri(),
+        "mediaType": "image/vnd.dlite-parse",
+        "configuration": {
+            "driver": "json",
+        },
+    }
+    coll = dlite.Collection()
+    session = {
+        "collection_id": coll.uuid,
+        "key": orig_key,
+    }
+    cache.add(coll.asjson(), key=coll.uuid)
+    parser: "IParseStrategy" = DLiteParseStrategy(config)
+    output: "DLiteSessionUpdate" = parser.get(session)
+    assert "collection_id" in output
+    assert output.collection_id == coll.uuid
+
+    # Get all instances of given metaid (there should only be one instance)
+    metaid = "http://onto-ns.com/meta/0.1/Molecule"
+    instances = list(coll.get_instances(metaid=metaid))
+    assert len(instances) == 1
+    (inst,) = instances
+    assert inst.meta.uri == metaid
+
+
+# if True:
+def test_parse_label() -> None:
+    """Test the dlite-parse strategy."""
+
+    import dlite
+    from oteapi.datacache import DataCache
+    from paths import staticdir
+
+    from oteapi_dlite.strategies.parse import DLiteParseStrategy
+
+    sample_file = staticdir / "molecule.json"
+
+    cache = DataCache()
+
+    orig_key = cache.add(sample_file.read_bytes())
+    config = {
+        "downloadUrl": sample_file.as_uri(),
+        "mediaType": "image/vnd.dlite-parse",
+        "configuration": {
+            "driver": "json",
+            "label": "instance",
+        },
+    }
+    coll = dlite.Collection()
+    session = {
+        "collection_id": coll.uuid,
+        "key": orig_key,
+    }
+    cache.add(coll.asjson(), key=coll.uuid)
+    parser: "IParseStrategy" = DLiteParseStrategy(config)
+    output: "DLiteSessionUpdate" = parser.get(session)
+    assert "collection_id" in output
+    assert output.collection_id == coll.uuid
+
+    coll2: dlite.Collection = dlite.get_instance(session["collection_id"])
+    inst: dlite.Instance = coll2.get("instance")
+    assert inst.properties.keys() == {
+        "name",
+        "positions",
+        "symbols",
+        "masses",
+        "groundstate_energy",
+    }
+    assert inst.properties["name"] == "H2"