Add otherIdentifiers to Dandiset model

dandischema/models.py

@@ -1677,6 +1677,25 @@ def contributor_musthave_contact(
         pattern=rf"^{ID_PATTERN}:\d{{6}}$",
         json_schema_extra={"readOnly": True, "nskey": "schema"},
     )
+
+    otherIdentifiers: Annotated[
+        list[
+            Annotated[
+                str, StringConstraints(pattern=rf"^{UNVENDORED_ID_PATTERN}:\d{{6}}$")
+            ]
+        ],
+        Field(
+            default_factory=list,
+            title="Other Dandiset identifiers",
+            description="Alternative Dandiset identifiers for this Dandiset as it is "
+            "identified in other DANDI instances.",
+            json_schema_extra={
+                "readOnly": True,
+                "nskey": DANDI_NSKEY,
+            },
+        ),
+    ]
+
     name: str = Field(
         title="Dandiset title",
         description="A title associated with the Dandiset.",

dandischema/tests/test_models.py

@@ -41,6 +41,53 @@
 _INSTANCE_CONFIG = get_instance_config()
 
 
+@pytest.fixture
+def base_dandiset_metadata() -> dict[str, Any]:
+    """
+    Fixture providing basic Dandiset metadata for constructing a `Dandiset` instance.
+
+    Returns:
+    Dict[str, Any]
+        A dictionary containing basic Dandiset metadata without `doi`, `datePublished`,
+        and `publishedBy`, suitable for constructing a `Dandiset` instance but not a
+        `PublishedDandiset` instance.
+
+    Note:
+        This metadata is returned by a fixture to ensure that each test receives a fresh
+        copy of the metadata dictionary.
+    """
+
+    return {
+        "identifier": f"{INSTANCE_NAME}:999999",
+        "id": f"{INSTANCE_NAME}:999999/draft",
+        "version": "1.0.0",
+        "name": "testing dataset",
+        "description": "testing",
+        "contributor": [
+            {
+                "name": "last name, first name",
+                "email": "someone@dandiarchive.org",
+                "roleName": [RoleType("dcite:ContactPerson")],
+                "schemaKey": "Person",
+            }
+        ],
+        "license": [LicenseType("spdx:CC-BY-4.0")],
+        "citation": "Last, first (2021). Test citation.",
+        "assetsSummary": {
+            "numberOfBytes": 0,
+            "numberOfFiles": 0,
+            "dataStandard": [{"name": "NWB"}],
+            "approach": [{"name": "electrophysiology"}],
+            "measurementTechnique": [{"name": "two-photon microscopy technique"}],
+            "species": [{"name": "Human"}],
+        },
+        "manifestLocation": [
+            "https://api.dandiarchive.org/api/dandisets/999999/versions/draft/assets/"
+        ],
+        "url": "https://dandiarchive.org/dandiset/999999/draft",
+    }
+
+
 @pytest.mark.parametrize(
     ("y_type", "anys_value"),
     [
@@ -403,46 +450,15 @@ def test_autogenerated_titles() -> None:
 
 
 @skipif_no_doi_prefix
-def test_dandimeta_1() -> None:
+def test_dandimeta_1(base_dandiset_metadata: dict[str, Any]) -> None:
     """checking basic metadata for publishing"""
 
     assert DOI_PREFIX is not None
 
-    # metadata without doi, datePublished and publishedBy
-    meta_dict: Dict[str, Any] = {
-        "identifier": f"{INSTANCE_NAME}:999999",
-        "id": f"{INSTANCE_NAME}:999999/draft",
-        "version": "1.0.0",
-        "name": "testing dataset",
-        "description": "testing",
-        "contributor": [
-            {
-                "name": "last name, first name",
-                "email": "someone@dandiarchive.org",
-                "roleName": [RoleType("dcite:ContactPerson")],
-                "schemaKey": "Person",
-            }
-        ],
-        "license": [LicenseType("spdx:CC-BY-4.0")],
-        "citation": "Last, first (2021). Test citation.",
-        "assetsSummary": {
-            "numberOfBytes": 0,
-            "numberOfFiles": 0,
-            "dataStandard": [{"name": "NWB"}],
-            "approach": [{"name": "electrophysiology"}],
-            "measurementTechnique": [{"name": "two-photon microscopy technique"}],
-            "species": [{"name": "Human"}],
-        },
-        "manifestLocation": [
-            "https://api.dandiarchive.org/api/dandisets/999999/versions/draft/assets/"
-        ],
-        "url": "https://dandiarchive.org/dandiset/999999/draft",
-    }
-
     # should work for Dandiset but PublishedDandiset should raise an error
-    Dandiset(**meta_dict)
+    Dandiset(**base_dandiset_metadata)
     with pytest.raises(ValidationError) as exc:
-        PublishedDandiset(**meta_dict)
+        PublishedDandiset(**base_dandiset_metadata)
 
     ErrDetail = namedtuple("ErrDetail", ["type", "msg"])
 
@@ -490,21 +506,23 @@ def test_dandimeta_1() -> None:
 
     # after adding basic meta required to publish: doi, datePublished, publishedBy, assetsSummary,
     # so PublishedDandiset should work
-    meta_dict["url"] = "https://dandiarchive.org/dandiset/999999/0.0.0"
-    meta_dict["id"] = f"{INSTANCE_NAME}:999999/0.0.0"
-    meta_dict["version"] = "0.0.0"
-    meta_dict.update(
+    base_dandiset_metadata["url"] = "https://dandiarchive.org/dandiset/999999/0.0.0"
+    base_dandiset_metadata["id"] = f"{INSTANCE_NAME}:999999/0.0.0"
+    base_dandiset_metadata["version"] = "0.0.0"
+    base_dandiset_metadata.update(
         basic_publishmeta(INSTANCE_NAME, dandi_id="999999", prefix=DOI_PREFIX)
     )
-    meta_dict["assetsSummary"].update(**{"numberOfBytes": 1, "numberOfFiles": 1})
+    base_dandiset_metadata["assetsSummary"].update(
+        **{"numberOfBytes": 1, "numberOfFiles": 1}
+    )
 
     # Test that releaseNotes is optional (can be omitted)
-    dandiset_without_notes = PublishedDandiset(**meta_dict)
+    dandiset_without_notes = PublishedDandiset(**base_dandiset_metadata)
     assert dandiset_without_notes.releaseNotes is None
 
     # Test that releaseNotes can be set to a string value
-    meta_dict["releaseNotes"] = "Releasing during testing"
-    dandiset_with_notes = PublishedDandiset(**meta_dict)
+    base_dandiset_metadata["releaseNotes"] = "Releasing during testing"
+    dandiset_with_notes = PublishedDandiset(**base_dandiset_metadata)
     assert dandiset_with_notes.releaseNotes == "Releasing during testing"
 
     # Test that releaseNotes appears in model_dump
@@ -983,3 +1001,63 @@ class VendoredFieldModel(BaseModel):
     # Validate the invalid vendored fields against the vendored patterns
     with pytest.raises(ValidationError):
         VendoredFieldModel.model_validate(invalid_vendored_fields)
+
+
+class TestOtherIdentifiers:
+    def test_not_specified(self, base_dandiset_metadata: dict[str, Any]) -> None:
+        """
+        Test the case that `otherIdentifiers` is not specified
+        """
+        dandiset = Dandiset.model_validate(base_dandiset_metadata)
+        assert dandiset.otherIdentifiers == []
+
+    def test_empty_list(self, base_dandiset_metadata: dict[str, Any]) -> None:
+        """
+        Test the case that `otherIdentifiers` is an empty list
+        """
+        base_dandiset_metadata["otherIdentifiers"] = []
+        dandiset = Dandiset.model_validate(base_dandiset_metadata)
+        assert dandiset.otherIdentifiers == []
+
+    @pytest.mark.parametrize(
+        "identifiers",
+        [
+            ["DANDI-SANDBOX:123456"],
+            ["EMBER-DANDI:123456"],
+            ["DANDI-SANDBOX:123456", "EMBER-DANDI:123456"],
+            ["A:123456", "B:654321"],
+        ],
+    )
+    def test_with_valid_identifiers(
+        self, identifiers: list[str], base_dandiset_metadata: dict[str, Any]
+    ) -> None:
+        """
+        Test the case that `otherIdentifiers` is set to a list of valid identifiers
+        """
+        base_dandiset_metadata["otherIdentifiers"] = identifiers
+        dandiset = Dandiset.model_validate(base_dandiset_metadata)
+        assert dandiset.otherIdentifiers == identifiers
+
+    @pytest.mark.parametrize(
+        "identifiers",
+        [
+            # List of invalid identifiers
+            ["-A:123456"],
+            ["DANDI-SANDBOX:12345"],
+            ["DANDI-SANDBOX:123456", "EMBER-DANDI123456"],
+            [42],
+            # Value that is not a list
+            "DANDI-SANDBOX:123456",
+            42,
+        ],
+    )
+    def test_with_invalid_identifiers(
+        self, identifiers: Any, base_dandiset_metadata: dict[str, Any]
+    ) -> None:
+        """
+        Test the case that `otherIdentifiers` is set to a list of invalid identifiers
+        or a value that is not a list
+        """
+        base_dandiset_metadata["otherIdentifiers"] = identifiers
+        with pytest.raises(ValidationError):
+            Dandiset.model_validate(base_dandiset_metadata)