Fix tmp dir not exists (#401)

* Fix tmp dir not exists

* Update tests to make it more clear

* Add check if save path is not None

autogen/retrieve_utils.py

@@ -254,7 +254,10 @@ def get_files_from_dir(dir_path: Union[str, List[str]], types: list = TEXT_FORMA
 def get_file_from_url(url: str, save_path: str = None):
     """Download a file from a URL."""
     if save_path is None:
+        os.makedirs("/tmp/chromadb", exist_ok=True)
         save_path = os.path.join("/tmp/chromadb", os.path.basename(url))
+    else:
+        os.makedirs(os.path.dirname(save_path), exist_ok=True)
     with requests.get(url, stream=True) as r:
         r.raise_for_status()
         with open(save_path, "wb") as f:

test/test_retrieve_utils.py

@@ -140,7 +140,7 @@ class TestRetrieveUtils:
                 db = lancedb.connect(db_path)
                 table = db.open_table("my_table")
                 query = table.search(vector).where(f"documents LIKE '%{search_string}%'").limit(n_results).to_df()
-                return {"ids": query["id"].tolist(), "documents": query["documents"].tolist()}
+                return {"ids": [query["id"].tolist()], "documents": [query["documents"].tolist()]}
 
             def retrieve_docs(self, problem: str, n_results: int = 20, search_string: str = ""):
                 results = self.query_vector_db(
@@ -166,7 +166,7 @@ class TestRetrieveUtils:
 
         create_lancedb()
         ragragproxyagent.retrieve_docs("This is a test document spark", n_results=10, search_string="spark")
-        assert ragragproxyagent._results["ids"] == [3, 1, 5]
+        assert ragragproxyagent._results["ids"] == [[3, 1, 5]]
 
     def test_custom_text_split_function(self):
         def custom_text_split_function(text):