Dunitz 1685 hosted annotations (#1789)

* save tiledb array to s3, dont cache user annotations

* Add option to disable annotation filename prompt (#1787)

Co-authored-by: Madison Dunitz <dunitzm@gmail.com>

* set tiledb default context in cxg_adaptor

Co-authored-by: maniarathi <arathi.mani@chanzuckerberg.com>
Co-authored-by: Severiano Badajoz <sbadajoz@chanzuckerberg.com>

server/test/__init__.py

@@ -33,7 +33,8 @@ def data_with_tmp_tiledb_annotations(ext: MatrixDataType):
     data_locator = DataLocator(fname)
     config = AppConfig()
     config.update_server_config(
-        multi_dataset__dataroot=data_locator.path, authentication__type="test"
+        multi_dataset__dataroot=data_locator.path,
+        authentication__type="test",
     )
     config.update_default_dataset_config(
         embeddings__names=["umap"],
@@ -49,7 +50,7 @@ def data_with_tmp_tiledb_annotations(ext: MatrixDataType):
     data = MatrixDataLoader(data_locator.abspath()).open(config)
     annotations = AnnotationsHostedTileDB(
         tmp_dir,
-        DbUtils("postgresql://postgres:test_pw@localhost:5432")
+        DbUtils("postgresql://postgres:test_pw@localhost:5432"),
     )
     return data, tmp_dir, annotations
 

server/test/unit/common/test_writable_annotation.py

@@ -21,6 +21,9 @@ class auth(object):
     def get_user_id():
         return "1234"
 
+    def get_user_name():
+        return "person name"
+
 
 class WritableTileDBStoredAnnotationTest(unittest.TestCase):
     def setUp(self):
@@ -70,7 +73,7 @@ class WritableTileDBStoredAnnotationTest(unittest.TestCase):
             self.assertEqual(type(df), tiledb.array.SparseArray)
 
             # convert to pandas df
-            pandas_df = self.annotations.convert_to_pandas_df(df)
+            pandas_df = self.annotations.convert_to_pandas_df(df, annotation.schema_hints)
             self.assertEqual(type(pandas_df), pd.DataFrame)
 
     def test_write_labels_creates_a_dataset_if_it_doesnt_exist(self):
@@ -111,7 +114,9 @@ class WritableTileDBStoredAnnotationTest(unittest.TestCase):
 
             self.assertEqual(pandas_df.shape, (self.n_rows, 2))
             self.assertEqual(set(pandas_df.columns), {"cat_A", "cat_B"})
+
             self.assertTrue(self.data.original_obs_index.equals(pandas_df.index))
+
             self.assertTrue(np.all(pandas_df["cat_A"] == ["label_A"] * self.n_rows))
             self.assertTrue(np.all(pandas_df["cat_B"] == ["label_B"] * self.n_rows))