Merge branch 'main' into docs-deployment-folder

Author: mendonk

.github/changes-filter.yaml

@@ -45,6 +45,7 @@ components:
   - "src/frontend/tests/extended/integrations/**"
   - "src/frontend/tests/extended/features/**"
   - "src/frontend/tests/extended/regression/**"
+  - "src/backend/base/langflow/custom/**"
 
 workspace:
   - "src/backend/base/langflow/inputs/**"

src/backend/base/langflow/api/v1/chat.py

@@ -69,7 +69,7 @@ async def try_running_celery_task(vertex, user_id):
     return vertex
 
 
-@router.post("/build/{flow_id}/vertices")
+@router.post("/build/{flow_id}/vertices", deprecated=True)
 async def retrieve_vertices_order(
     *,
     flow_id: uuid.UUID,
@@ -471,7 +471,7 @@ async def listen_for_disconnect(self, receive: Receive) -> None:
                 break
 
 
-@router.post("/build/{flow_id}/vertices/{vertex_id}")
+@router.post("/build/{flow_id}/vertices/{vertex_id}", deprecated=True)
 async def build_vertex(
     *,
     flow_id: uuid.UUID,
@@ -713,7 +713,7 @@ async def _stream_vertex(flow_id: str, vertex_id: str, chat_service: ChatService
         yield str(StreamData(event="close", data={"message": "Stream closed"}))
 
 
-@router.get("/build/{flow_id}/{vertex_id}/stream", response_class=StreamingResponse)
+@router.get("/build/{flow_id}/{vertex_id}/stream", response_class=StreamingResponse, deprecated=True)
 async def build_vertex_stream(
     flow_id: uuid.UUID,
     vertex_id: str,

src/backend/base/langflow/base/agents/agent.py

@@ -69,7 +69,7 @@ class LCAgentComponent(Component):
     ]
 
     outputs = [
-        Output(display_name="Agent", name="agent", method="build_agent", hidden=True),
+        Output(display_name="Agent", name="agent", method="build_agent", hidden=True, tool_mode=False),
         Output(display_name="Response", name="response", method="message_response"),
     ]
 

src/backend/base/langflow/base/tools/component_tool.py

@@ -169,6 +169,24 @@ def __init__(self, component: Component, metadata: pd.DataFrame | None = None):
         self.component = component
         self.metadata = metadata
 
+    def _should_skip_output(self, output: Output) -> bool:
+        """Determines if an output should be skipped when creating tools.
+
+        Args:
+            output (Output): The output to check.
+
+        Returns:
+            bool: True if the output should be skipped, False otherwise.
+
+        The output will be skipped if:
+        - tool_mode is False (output is not meant to be used as a tool)
+        - output name matches TOOL_OUTPUT_NAME
+        - output types contain any of the tool types in TOOL_TYPES_SET
+        """
+        return not output.tool_mode or (
+            output.name == TOOL_OUTPUT_NAME or any(tool_type in output.types for tool_type in TOOL_TYPES_SET)
+        )
+
     def get_tools(
         self,
         tool_name: str | None = None,
@@ -178,7 +196,7 @@ def get_tools(
     ) -> list[BaseTool]:
         tools = []
         for output in self.component.outputs:
-            if output.name == TOOL_OUTPUT_NAME or any(tool_type in output.types for tool_type in TOOL_TYPES_SET):
+            if self._should_skip_output(output):
                 continue
 
             if not output.method:

src/backend/base/langflow/components/agents/agent.py

@@ -10,7 +10,9 @@
 from langflow.components.helpers import CurrentDateComponent
 from langflow.components.helpers.memory import MemoryComponent
 from langflow.components.langchain_utilities.tool_calling import ToolCallingAgentComponent
+from langflow.custom.custom_component.component import _get_component_toolkit
 from langflow.custom.utils import update_component_build_config
+from langflow.field_typing import Tool
 from langflow.io import BoolInput, DropdownInput, MultilineInput, Output
 from langflow.logging import logger
 from langflow.schema.dotdict import dotdict
@@ -84,8 +86,7 @@ async def message_response(self) -> Message:
                 if not isinstance(self.tools, list):  # type: ignore[has-type]
                     self.tools = []
                 # Convert CurrentDateComponent to a StructuredTool
-                current_date_tool = (await CurrentDateComponent().to_toolkit()).pop(0)
-                # current_date_tool = CurrentDateComponent().to_toolkit()[0]
+                current_date_tool = (await CurrentDateComponent(**self.get_base_args()).to_toolkit()).pop(0)
                 if isinstance(current_date_tool, StructuredTool):
                     self.tools.append(current_date_tool)
                 else:
@@ -122,7 +123,7 @@ async def get_memory_data(self):
         # filter out empty values
         memory_kwargs = {k: v for k, v in memory_kwargs.items() if v}
 
-        return await MemoryComponent().set(**memory_kwargs).retrieve_messages()
+        return await MemoryComponent(**self.get_base_args()).set(**memory_kwargs).retrieve_messages()
 
     def get_llm(self):
         if isinstance(self.agent_llm, str):
@@ -265,3 +266,16 @@ async def update_build_config(
                         component_class, build_config, field_value, "model_name"
                     )
         return dotdict({k: v.to_dict() if hasattr(v, "to_dict") else v for k, v in build_config.items()})
+
+    async def to_toolkit(self) -> list[Tool]:
+        component_toolkit = _get_component_toolkit()
+        tools_names = self._build_tools_names()
+        agent_description = self.get_tool_description()
+        # TODO: Agent Description Depreciated Feature to be removed
+        description = f"{agent_description}{tools_names}"
+        tools = component_toolkit(component=self).get_tools(
+            tool_name=self.get_tool_name(), tool_description=description, callbacks=self.get_langchain_callbacks()
+        )
+        if hasattr(self, "tools_metadata"):
+            tools = component_toolkit(component=self, metadata=self.tools_metadata).update_tools_metadata(tools=tools)
+        return tools

src/backend/base/langflow/components/embeddings/text_embedder.py

@@ -38,46 +38,26 @@ def generate_embeddings(self) -> Data:
             embedding_model: Embeddings = self.embedding_model
             message: Message = self.message
 
-            # Validate embedding model
-            if not embedding_model:
-                msg = "Embedding model not provided"
+            # Combine validation checks to reduce nesting
+            if not embedding_model or not hasattr(embedding_model, "embed_documents"):
+                msg = "Invalid or incompatible embedding model"
                 raise ValueError(msg)
 
-            # Extract the text content from the message
             text_content = message.text if message and message.text else ""
             if not text_content:
                 msg = "No text content found in message"
                 raise ValueError(msg)
 
-            # Check if the embedding model has the required attributes
-            if not hasattr(embedding_model, "client") or not embedding_model.client:
-                msg = "Embedding model client not properly initialized"
-                raise ValueError(msg)
-
-            # Ensure the base URL has proper protocol
-            if hasattr(embedding_model.client, "base_url"):
-                base_url = embedding_model.client.base_url
-                if not base_url.startswith(("http://", "https://")):
-                    embedding_model.client.base_url = f"https://{base_url}"
-
-            # Generate embeddings using the provided embedding model
             embeddings = embedding_model.embed_documents([text_content])
-
-            # Validate embeddings output
             if not embeddings or not isinstance(embeddings, list):
                 msg = "Invalid embeddings generated"
                 raise ValueError(msg)
 
             embedding_vector = embeddings[0]
-
+            self.status = {"text": text_content, "embeddings": embedding_vector}
+            return Data(data={"text": text_content, "embeddings": embedding_vector})
         except Exception as e:
             logging.exception("Error generating embeddings")
-            # Return empty data with error status
             error_data = Data(data={"text": "", "embeddings": [], "error": str(e)})
             self.status = {"error": str(e)}
             return error_data
-
-        # Create a Data object to encapsulate the results
-        result_data = Data(data={"text": text_content, "embeddings": embedding_vector})
-        self.status = {"text": text_content, "embeddings": embedding_vector}
-        return result_data

src/backend/base/langflow/components/langchain_utilities/csv_agent.py

@@ -56,7 +56,7 @@ class CSVAgentComponent(LCAgentComponent):
 
     outputs = [
         Output(display_name="Response", name="response", method="build_agent_response"),
-        Output(display_name="Agent", name="agent", method="build_agent", hidden=True),
+        Output(display_name="Agent", name="agent", method="build_agent", hidden=True, tool_mode=False),
     ]
 
     def _path(self) -> str:

src/backend/base/langflow/components/langchain_utilities/tool_calling.py

@@ -2,6 +2,8 @@
 from langchain_core.prompts import ChatPromptTemplate
 
 from langflow.base.agents.agent import LCToolsAgentComponent
+from langflow.custom.custom_component.component import _get_component_toolkit
+from langflow.field_typing import Tool
 from langflow.inputs import MessageTextInput
 from langflow.inputs.inputs import DataInput, HandleInput
 from langflow.schema import Data
@@ -54,3 +56,11 @@ def create_agent_runnable(self):
         except NotImplementedError as e:
             message = f"{self.display_name} does not support tool calling. Please try using a compatible model."
             raise NotImplementedError(message) from e
+
+    async def to_toolkit(self) -> list[Tool]:
+        component_toolkit = _get_component_toolkit()
+        toolkit = component_toolkit(component=self)
+        tools = toolkit.get_tools(callbacks=self.get_langchain_callbacks())
+        if hasattr(self, "tools_metadata"):
+            tools = toolkit.update_tools_metadata(tools=tools)
+        return tools