Add study abstract/metadata collection columns to the catalog schema

src/sddb/_catalog_schema.py

@@ -99,6 +99,13 @@
     "n_members_total": "Int64",
     "size_bytes": "Int64",
     "members": "list[string]",
+    "abstract": "string",
+    "abstract_license": "string",
+    "abstract_source": "string",
+    "study_title": "string",
+    "study_authors": "string",
+    "study_year": "string",
+    "study_journal": "string",
 }
 COLLECTION_REQUIRED: tuple[str, ...] = ("slug", "name", "members")
 

tests/_fixtures.py

@@ -89,9 +89,10 @@ def make_fixture_collections() -> pd.DataFrame:
     """Build a valid synthetic collections table (one row per collection)."""
     rows = [
         ("c-smith", "Smith2023", "Smith 2023 atlas", "publication", "EpCAM atlas",
-         "10.1/smith", "DOI", "Xenium", "human", "lung", 2, 2, 180_000, ["uid0001", "uid0002"]),
+         "10.1/smith", "DOI", "Xenium", "human", "lung", 2, 2, 180_000, ["uid0001", "uid0002"],
+         *[None] * 7),  # abstract/study_* columns are optional: null here
         ("c-lee", "Lee2022", "Lee 2022 brain", "replicates", None, None, None,
-         "Visium", "mouse", "brain", 1, 1, 4_000, ["uid0003"]),
+         "Visium", "mouse", "brain", 1, 1, 4_000, ["uid0003"], *[None] * 7),
     ]
     df = pd.DataFrame(rows, columns=list(COLLECTION_COLUMNS))
     return cast_scalars(df, COLLECTION_COLUMNS)

tests/test_catalog_schema.py

@@ -150,3 +150,17 @@ def test_new_producer_columns_validate_populated_and_null(tmp_path):
     p = tmp_path / "c.parquet"
     df.to_parquet(p)
     cs.validate(pd.read_parquet(p))  # round-trips
+
+
+def test_collection_columns_include_study_fields():
+    import sddb._catalog_schema as cs
+
+    expected = {
+        "abstract", "abstract_license", "abstract_source",
+        "study_title", "study_authors", "study_year", "study_journal",
+    }
+    assert expected <= set(cs.COLLECTION_COLUMNS)
+    # all are nullable string columns
+    assert all(cs.COLLECTION_COLUMNS[c] == "string" for c in expected)
+    # not newly required
+    assert expected.isdisjoint(set(cs.COLLECTION_REQUIRED))