auth: Enable linking multiple datasets to the group at once.

2 files changed, 54 insertions, 36 deletions

diff --git a/gn3/auth/authorisation/groups/data.py b/gn3/auth/authorisation/groups/data.py
index fd76648..0c821d3 100644
--- a/gn3/auth/authorisation/groups/data.py
+++ b/gn3/auth/authorisation/groups/data.py
@@ -108,58 +108,74 @@ def retrieve_ungrouped_data(
     grouped_data = __fetch_grouped_data__(authconn, dataset_type)
     return __fetch_ungrouped_data__(gn3conn, dataset_type, grouped_data, offset)
 
-def __fetch_mrna_data_by_id__(conn: gn3db.Connection, dataset_id: str) -> dict:
+def __fetch_mrna_data_by_ids__(
+        conn: gn3db.Connection, dataset_ids: tuple[str, ...]) -> tuple[
+            dict, ...]:
     """Fetch mRNA Assay data by ID."""
     with conn.cursor(DictCursor) as cursor:
+        paramstr = ", ".join(["%s"] * len(dataset_ids))
         cursor.execute(
-            "SELECT psf.Id, psf.Name, psf.FullName, "
+            "SELECT psf.Id, psf.Name AS dataset_name, "
+            "psf.FullName AS dataset_fullname, "
             "ifiles.GN_AccesionId AS accession_id FROM ProbeSetFreeze AS psf "
             "INNER JOIN InfoFiles AS ifiles ON psf.Name=ifiles.InfoPageName "
-            "WHERE psf.Id=%s",
-            (dataset_id,))
-        res = cursor.fetchone()
+            f"WHERE psf.Id IN ({paramstr})",
+            dataset_ids)
+        res = cursor.fetchall()
         if res:
-            return dict(res)
+            return tuple(dict(row) for row in res)
         raise NotFoundError("Could not find mRNA Assay data with the given ID.")
 
-def __fetch_geno_data_by_id__(conn: gn3db.Connection, dataset_id: str) -> dict:
+def __fetch_geno_data_by_ids__(
+        conn: gn3db.Connection, dataset_ids: tuple[str, ...]) -> tuple[
+            dict, ...]:
     """Fetch genotype data by ID."""
     with conn.cursor(DictCursor) as cursor:
+        paramstr = ", ".join(["%s"] * len(dataset_ids))
         cursor.execute(
-            "SELECT gf.Id, gf.Name, gf.FullName, "
+            "SELECT gf.Id, gf.Name AS dataset_name, "
+            "gf.FullName AS dataset_fullname, "
             "ifiles.GN_AccesionId AS accession_id FROM GenoFreeze AS gf "
             "INNER JOIN InfoFiles AS ifiles ON gf.Name=ifiles.InfoPageName "
-            "WHERE gf.Id=%s",
-            (dataset_id,))
-        res = cursor.fetchone()
+            f"WHERE gf.Id IN ({paramstr})",
+            dataset_ids)
+        res = cursor.fetchall()
         if res:
-            return dict(res)
+            return tuple(dict(row) for row in res)
         raise NotFoundError("Could not find Genotype data with the given ID.")
 
-def __fetch_pheno_data_by_id__(conn: gn3db.Connection, dataset_id: str) -> dict:
+def __fetch_pheno_data_by_ids__(
+        conn: gn3db.Connection, dataset_ids: tuple[str, ...]) -> tuple[
+            dict, ...]:
     """Fetch phenotype data by ID."""
     with conn.cursor(DictCursor) as cursor:
+        paramstr = ", ".join(["%s"] * len(dataset_ids))
         cursor.execute(
-            "SELECT pf.Id, pf.Name, pf.FullName, "
-            "ifiles.GN_AccesionId AS accession_id FROM PublishFreeze AS pf "
+            "SELECT pxf.Id, iset.InbredSetName, pf.Id AS dataset_id, "
+            "pf.Name AS dataset_name, pf.FullName AS dataset_fullname, "
+            "ifiles.GN_AccesionId AS accession_id "
+            "FROM PublishXRef AS pxf "
+            "INNER JOIN InbredSet AS iset ON pxf.InbredSetId=iset.InbredSetId "
+            "INNER JOIN PublishFreeze AS pf ON iset.InbredSetId=pf.InbredSetId "
             "INNER JOIN InfoFiles AS ifiles ON pf.Name=ifiles.InfoPageName "
-            "WHERE pf.Id=%s",
-            (dataset_id,))
-        res = cursor.fetchone()
+            f"WHERE pxf.Id IN ({paramstr})",
+            dataset_ids)
+        res = cursor.fetchall()
         if res:
-            return dict(res)
+            return tuple(dict(row) for row in res)
         raise NotFoundError(
-            "Could not find Phenotype/Publish data with the given ID.")
+            "Could not find Phenotype/Publish data with the given IDs.")
 
 def __fetch_data_by_id(
-        conn: gn3db.Connection, dataset_type: str, dataset_id: str) -> dict:
-    """Fetch data from MySQL by ID."""
+        conn: gn3db.Connection, dataset_type: str,
+        dataset_ids: tuple[str, ...]) -> tuple[dict, ...]:
+    """Fetch data from MySQL by IDs."""
     fetch_fns = {
-        "mrna": __fetch_mrna_data_by_id__,
-        "genotype": __fetch_geno_data_by_id__,
-        "phenotype": __fetch_pheno_data_by_id__
+        "mrna": __fetch_mrna_data_by_ids__,
+        "genotype": __fetch_geno_data_by_ids__,
+        "phenotype": __fetch_pheno_data_by_ids__
     }
-    return fetch_fns[dataset_type](conn, dataset_id)
+    return fetch_fns[dataset_type](conn, dataset_ids)
 
 @authorised_p(("system:data:link-to-group",),
               error_description=(
@@ -168,20 +184,22 @@ def __fetch_data_by_id(
               oauth2_scope="profile group resource")
 def link_data_to_group(
         authconn: authdb.DbConnection, gn3conn: gn3db.Connection,
-        dataset_type: str, dataset_id: str, group: Group) -> dict:
+        dataset_type: str, dataset_ids: tuple[str, ...], group: Group) -> tuple[
+            dict, ...]:
     """Link the given data to the specified group."""
-    the_data = __fetch_data_by_id(gn3conn, dataset_type, dataset_id)
+    the_data = __fetch_data_by_id(gn3conn, dataset_type, dataset_ids)
     with authdb.cursor(authconn) as cursor:
-        params = {
+        params = tuple({
             "group_id": str(group.group_id), "dataset_type": {
                 "mrna": "mRNA", "genotype": "Genotype",
                 "phenotype": "Phenotype"
             }[dataset_type],
-            "dataset_or_trait_id": dataset_id, "dataset_name": the_data["Name"],
-            "dataset_fullname": the_data["FullName"],
-            "accession_id": the_data["accession_id"]
-        }
-        cursor.execute(
+            "dataset_or_trait_id": item["Id"],
+            "dataset_name": item["dataset_name"],
+            "dataset_fullname": item["dataset_fullname"],
+            "accession_id": item["accession_id"]
+        } for item in the_data)
+        cursor.executemany(
             "INSERT INTO linked_group_data VALUES"
             "(:group_id, :dataset_type, :dataset_or_trait_id, :dataset_name, "
             ":dataset_fullname, :accession_id)",
diff --git a/gn3/auth/authorisation/groups/views.py b/gn3/auth/authorisation/groups/views.py
index 0b21800..6f1a6df 100644
--- a/gn3/auth/authorisation/groups/views.py
+++ b/gn3/auth/authorisation/groups/views.py
@@ -213,7 +213,7 @@ def link_data() -> Response:
     with require_oauth.acquire("profile group resource") as _the_token:
         form = request.form
         group_id = uuid.UUID(form["group_id"])
-        dataset_id = form["dataset_id"]
+        dataset_ids = form.getlist("dataset_ids")
         dataset_type = form.get("dataset_type")
         if dataset_type not in ("mrna", "genotype", "phenotype"):
             raise InvalidData("Unexpected dataset type requested!")
@@ -221,7 +221,7 @@ def link_data() -> Response:
             group = group_by_id(conn, group_id)
             with gn3dbutils.database_connection() as gn3conn:
                 return link_data_to_group(
-                    conn, gn3conn, dataset_type, dataset_id, group)
+                    conn, gn3conn, dataset_type, dataset_ids, group)
 
         return jsonify(with_db_connection(__link__))