CREATE INDEX IF NOT EXISTS is broken. (#1337)

This PR fixes an issue in CREATE INDEX IF NOT EXISTS command wherein if 'IF NOT EXISTS' is passed, we had an unreferenced variable issue. Added Unit Tests to check the correctness of both the cases. Also reverted the index changes while merging dataframes after vector scan, as it's failing for some cases where indexes can be undefined.
georgia-tech-db · Nov 3, 2023 · c296318 · c296318
1 parent f56ef82
commit c296318
Show file tree

Hide file tree

Showing 3 changed files with 18 additions and 12 deletions.
diff --git a/evadb/executor/insert_executor.py b/evadb/executor/insert_executor.py
@@ -59,8 +59,10 @@ def exec(self, *args, **kwargs):
             for column in table_catalog_entry.columns:
                 if column == index.feat_column:
                     is_index_on_current_table = True
+                    break
             if is_index_on_current_table:
-                create_index_query_list = index.index_def.split(" ")
+                create_index_query = index.index_def
+                create_index_query_list = create_index_query.split(" ")
                 if_not_exists = " ".join(create_index_query_list[2:5]).lower()
                 if if_not_exists != "if not exists":
                     create_index_query = (

diff --git a/evadb/executor/vector_index_scan_executor.py b/evadb/executor/vector_index_scan_executor.py
@@ -142,16 +142,13 @@ def _evadb_vector_index_scan(self, *args, **kwargs):
                 res_data_list.append(row_dict)
 
         result_df = pd.DataFrame(res_data_list)
-        result_df.set_index(row_num_col_name, inplace=True)
-        result_df = result_df.reindex(row_num_np)
-        row_num_df.set_index(pd.Index(row_num_np), inplace=True)
 
         final_df = pd.merge(
             row_num_df,
             result_df,
-            left_index=True,
-            right_index=True,
-            how="left",
+            left_on="row_num_np",
+            right_on=row_num_col_name,
+            how="inner",
         )
 
         if "row_num_np" in final_df:

diff --git a/test/integration_tests/long/test_similarity.py b/test/integration_tests/long/test_similarity.py
@@ -428,17 +428,15 @@ def test_end_to_end_index_scan_should_work_correctly_on_image_dataset_faiss(self
             drop_query = "DROP INDEX testFaissIndexImageDataset"
             execute_query_fetch_all(self.evadb, drop_query)
 
-    def test_index_auto_update_on_structured_table_during_insertion_with_faiss(self):
-        create_query = "CREATE TABLE testIndexAutoUpdate (img_path TEXT(100))"
-        execute_query_fetch_all(self.evadb, create_query)
-
+    def _helper_for_auto_update_during_insertion_with_faiss(self, if_exists: bool):
         for i, img_path in enumerate(self.img_path_list):
             insert_query = (
                 f"INSERT INTO testIndexAutoUpdate (img_path) VALUES ('{img_path}')"
             )
             execute_query_fetch_all(self.evadb, insert_query)
             if i == 0:
-                create_index_query = "CREATE INDEX testIndex ON testIndexAutoUpdate(DummyFeatureExtractor(Open(img_path))) USING FAISS"
+                if_exists_str = "IF NOT EXISTS " if if_exists else ""
+                create_index_query = f"CREATE INDEX {if_exists_str}testIndex ON testIndexAutoUpdate(DummyFeatureExtractor(Open(img_path))) USING FAISS"
                 execute_query_fetch_all(self.evadb, create_index_query)
 
         select_query = """SELECT _row_id FROM testIndexAutoUpdate
@@ -452,6 +450,15 @@ def test_index_auto_update_on_structured_table_during_insertion_with_faiss(self)
         res_batch = execute_query_fetch_all(self.evadb, select_query)
         self.assertEqual(res_batch.frames["testindexautoupdate._row_id"][0], 5)
 
+    def test_index_auto_update_on_structured_table_during_insertion_with_faiss(self):
+        create_query = "CREATE TABLE testIndexAutoUpdate (img_path TEXT(100))"
+        drop_query = "DROP TABLE testIndexAutoUpdate"
+        execute_query_fetch_all(self.evadb, create_query)
+        self._helper_for_auto_update_during_insertion_with_faiss(False)
+        execute_query_fetch_all(self.evadb, drop_query)
+        execute_query_fetch_all(self.evadb, create_query)
+        self._helper_for_auto_update_during_insertion_with_faiss(True)
+
     @qdrant_skip_marker
     def test_end_to_end_index_scan_should_work_correctly_on_image_dataset_qdrant(self):
         for _ in range(2):