fix graphy bug

GraphScope · Nov 25, 2024 · 6f3b759 · 6f3b759
1 parent c239625
commit 6f3b759
Show file tree

Hide file tree

Showing 4 changed files with 27 additions and 8 deletions.
diff --git a/python/graphy/apps/demo_app.py b/python/graphy/apps/demo_app.py
@@ -805,11 +805,11 @@ def graphy_dataset():
                         download_name=zip_filename,
                     )
 
-                node_names = self.get_workflow_node_names(dataset_id)
+                # node_names = self.get_workflow_node_names(dataset_id)
                 graph_builder = GraphBuilder(
                     data_path, self.get_persist_store(dataset_id)
                 )
-                graph_builder.extract_fact_data(node_names)
+                graph_builder.extract_fact_data()
 
                 graph_builder.build_graph()
                 graph_path = os.path.join(data_path, "_graph")

diff --git a/python/graphy/db/base_store.py b/python/graphy/db/base_store.py
@@ -106,7 +106,9 @@ def get_total_data(self) -> List[str]:
                 visible_items = [
                     item
                     for item in items
-                    if not item.startswith((".", "_")) and item != "navigator"
+                    if not item.startswith((".", "_"))
+                    and item != "navigator"
+                    and os.path.isdir(os.path.join(self.output_folder, item))
                 ]
                 return visible_items
             except Exception as e:

diff --git a/python/graphy/tests/app/demo_app_test.py b/python/graphy/tests/app/demo_app_test.py
@@ -162,13 +162,13 @@ def test_get_workflow_config(setup_dataset):
 def test_extract(setup_dataset):
     response = requests.post(
         f"{BASE_URL}/api/dataset/extract",
-        json={"dataset_id": "8547eb64-a106-5d09-8950-8a47fb9292dc", "thread_num": 1},
+        json={"dataset_id": DATASET_ID, "thread_num": 1},
     )
     print(response.json())
     assert response.status_code == 200
 
 
-# @pytest.mark.skip(reason="requires LLM connection")
+@pytest.mark.skip(reason="requires LLM connection")
 def test_get_extracted_data(setup_dataset):
     response = requests.get(
         f"{BASE_URL}/api/dataset/extract", params={"dataset_id": DATASET_ID}
@@ -184,6 +184,24 @@ def test_get_extracted_data(setup_dataset):
     assert response.status_code == 200
 
 
+@pytest.mark.skip(reason="requires LLM connection")
+def test_graphy_data(setup_dataset):
+    response = requests.post(
+        f"{BASE_URL}/api/dataset/graphy", json={"dataset_id": DATASET_ID}
+    )
+    print(response)
+    assert response.status_code == 200
+
+
+@pytest.mark.skip(reason="requires LLM connection")
+def test_get_graphy_data(setup_dataset):
+    response = requests.get(
+        f"{BASE_URL}/api/dataset/graphy", params={"dataset_id": DATASET_ID}
+    )
+    print(response)
+    assert response.status_code == 200
+
+
 def test_delete_dataset(setup_dataset):
     # Clean up by deleting the dataset
     response = requests.delete(f"{BASE_URL}/api/dataset/{DATASET_ID}")

diff --git a/python/graphy/utils/data_extractor.py b/python/graphy/utils/data_extractor.py
@@ -153,9 +153,8 @@ def extract_fact_data(self, dimension_node_names=[]):
                         dimension_node_names.remove("Paper")
                     except ValueError:
                         pass  # Do nothing if "Paper" is not in the list
-                else:
-                    for node_name in dimension_node_names:
-                        self._extract_dimension_data(paper_id, folder, node_name)
+                for node_name in dimension_node_names:
+                    self._extract_dimension_data(paper_id, folder, node_name)
             if edge_data:
                 for edge_name, edge_pairs in edge_data.items():
                     formatted_edges = [