From bfd66c5b8185298fda99f12558e96667643fbc60 Mon Sep 17 00:00:00 2001
From: anon <tim.treis@helmholtz-munich.de>
Date: Wed, 14 Oct 2026 16:50:13 +0200
Subject: [PATCH] feat(schema): add study abstract/metadata collection columns

---
 src/sddb/_catalog_schema.py  |  7 +++++++
 tests/_fixtures.py           |  5 +++--
 tests/test_catalog_schema.py | 14 ++++++++++++++
 3 files changed, 24 insertions(+), 2 deletions(-)

diff --git a/src/sddb/_catalog_schema.py b/src/sddb/_catalog_schema.py
index 4cc00e4..a4e535c 100644
--- a/src/sddb/_catalog_schema.py
+++ b/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")
 
diff --git a/tests/_fixtures.py b/tests/_fixtures.py
index e620379..5c1ae88 100644
--- a/tests/_fixtures.py
+++ b/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)
diff --git a/tests/test_catalog_schema.py b/tests/test_catalog_schema.py
index e1d9807..10624ed 100644
--- a/tests/test_catalog_schema.py
+++ b/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))