📝 chore(utils.py): add utility function to check if an object is a basic type

📝 chore(loading.py): refactor code to improve readability and maintainability
📝 chore(vector_store.py): refactor code to improve readability and maintainability
📝 chore(run.py): update return type hint for build_sorted_vertices function

src/backend/langflow/graph/vertex/utils.py

@@ -0,0 +1,5 @@
+from langflow.utils.constants import PYTHON_BASIC_TYPES
+
+
+def is_basic_type(obj):
+    return type(obj) in PYTHON_BASIC_TYPES

src/backend/langflow/interface/initialize/loading.py

@@ -1,7 +1,7 @@
 import json
 import orjson
 from typing import Any, Callable, Dict, Sequence, Type, TYPE_CHECKING
-
+from langchain.schema import Document
 from langchain.agents import agent as agent_module
 from langchain.agents.agent import AgentExecutor
 from langchain.agents.agent_toolkits.base import BaseToolkit
@@ -40,12 +40,23 @@ if TYPE_CHECKING:
     from langflow import CustomComponent
 
 
+def build_vertex_in_params(params: Dict) -> Dict:
+    from langflow.graph.vertex.base import Vertex
+
+    # If any of the values in params is a Vertex, we will build it
+    return {
+        key: value.build() if isinstance(value, Vertex) else value
+        for key, value in params.items()
+    }
+
+
 def instantiate_class(
     node_type: str, base_type: str, params: Dict, user_id=None
 ) -> Any:
     """Instantiate class from module type and key, and params"""
     params = convert_params_to_sets(params)
     params = convert_kwargs(params)
+
     if node_type in CUSTOM_NODES:
         if custom_node := CUSTOM_NODES.get(node_type):
             if hasattr(custom_node, "initialize"):
@@ -289,6 +300,13 @@ def instantiate_embedding(node_type, class_object, params: Dict):
 
 def instantiate_vectorstore(class_object: Type[VectorStore], params: Dict):
     search_kwargs = params.pop("search_kwargs", {})
+    # clean up docs or texts to have only documents
+    if "texts" in params:
+        params["documents"] = params.pop("texts")
+    if "documents" in params:
+        params["documents"] = [
+            doc for doc in params["documents"] if isinstance(doc, Document)
+        ]
     if initializer := vecstore_initializer.get(class_object.__name__):
         vecstore = initializer(class_object, params)
     else:

src/backend/langflow/interface/initialize/vector_store.py

@@ -8,7 +8,7 @@ from langchain.vectorstores import (
     SupabaseVectorStore,
     MongoDBAtlasVectorSearch,
 )
-
+from langchain.schema import Document
 import os
 
 import orjson
@@ -201,11 +201,16 @@ def initialize_chroma(class_object: Type[Chroma], params: dict):
         if "texts" in params:
             params["documents"] = params.pop("texts")
         for doc in params["documents"]:
+            if not isinstance(doc, Document):
+                # remove any non-Document objects from the list
+                params["documents"].remove(doc)
+                continue
             if doc.metadata is None:
                 doc.metadata = {}
             for key, value in doc.metadata.items():
                 if value is None:
                     doc.metadata[key] = ""
+
         chromadb = class_object.from_documents(**params)
     if persist:
         chromadb.persist()

src/backend/langflow/interface/run.py

@@ -3,7 +3,7 @@ from langflow.graph import Graph
 from loguru import logger
 
 
-def build_sorted_vertices(data_graph) -> Tuple[Any, Dict]:
+def build_sorted_vertices(data_graph) -> Tuple[Graph, Dict]:
     """
     Build langchain object from data_graph.
     """
@@ -16,7 +16,7 @@ def build_sorted_vertices(data_graph) -> Tuple[Any, Dict]:
         vertex.build()
         if vertex.artifacts:
             artifacts.update(vertex.artifacts)
-    return graph.build(), artifacts
+    return graph, artifacts
 
 
 def build_langchain_object(data_graph):