Cannot enlarge LLM output blocks in UI

[konradbjk]

Bug Description

When I get a response from the LLM I cannot enrage the window to see the whole response

Reproduction

Use some LLm output block

Expected behavior

I am able to enlarge LLM output rsponse preview to see it in full

Who can help?

@anovazzi1 @Cristhianzl @lucaseduoli @igorrCarvalho

Operating System

Ubuntu 24 LTS on Azure VM

Langflow Version

1.0.15

Python Version

3.12

Screenshot

Flow File

No response

[dosubot]

Hello @konradbjk! I'm here to help you with any issues or questions you have. I can assist with bug fixes, answer your queries, and guide you on becoming a contributor. Let's work together to resolve the issue you're facing.

The issue you're experiencing with enlarging the LLM output blocks in the UI might be related to the current implementation of the resizing functionality. Langflow uses the ReactFlow component from the react-flow library to handle the resizing and movement of nodes within the flow. However, it seems that the specific resizing of the LLM output blocks might not be fully supported or implemented.

To address this, you might want to look into the ReactFlow component's handlers and callbacks, such as onNodeDragStart, onNodeDragStop, onSelectionDragStart, and onMoveEnd, which manage the resizing and movement of nodes. Here is a snippet of how these handlers are set up:

<ReactFlow
  nodes={nodes}
  edges={edges}
  onNodesChange={onNodesChange}
  onEdgesChange={onEdgesChange}
  onConnect={onConnectMod}
  disableKeyboardA11y={true}
  onInit={setReactFlowInstance}
  nodeTypes={nodeTypes}
  onEdgeUpdate={onEdgeUpdate}
  onEdgeUpdateStart={onEdgeUpdateStart}
  onEdgeUpdateEnd={onEdgeUpdateEnd}
  onNodeDragStart={onNodeDragStart}
  onNodeDragStop={onNodeDragStop}
  onSelectionDragStart={onSelectionDragStart}
  onMoveEnd={onMoveEnd}
  connectionLineComponent={ConnectionLineComponent}
  onDragOver={onDragOver}
  onDrop={onDrop}
  onSelectionChange={onSelectionChange}
  deleteKeyCode={[]}
  className="theme-attribution"
  minZoom={0.01}
  maxZoom={8}
  zoomOnScroll={!view}
  zoomOnPinch={!view}
  panOnDrag={!view}
  panActivationKeyCode={""}
  proOptions={{ hideAttribution: true }}
  onPaneClick={onPaneClick}
>
  <Background className="" />
  {!view && (
    <Controls className="fill-foreground stroke-foreground text-primary [&>button]:border-b-border [&>button]:bg-muted hover:[&>button]:bg-border"></Controls>
  )}
  <SelectionMenu
    lastSelection={lastSelection}
    isVisible={selectionMenuVisible}
    nodes={lastSelection?.nodes}
    onClick={() => {
      handleGroupNode();
    }}
  />
</ReactFlow>

This setup allows for the dynamic resizing and manipulation of UI elements within the Langflow interface ^[1].

Additionally, you might want to check if there are any specific configurations or properties for the genericNode type that could be adjusted to enable resizing for the LLM output blocks. If the issue persists, it might require further investigation or a potential update to the Langflow codebase to support this functionality.

To continue talking to Dosu, mention @dosu.

[anovazzi1]

hello @konradbjk, thanks for the feedback, are you sure you are using the v1.0.15? I just tried this release on my linux machine with chrome and it worked:

[konradbjk]

I had to downgrade from v1.0.15 to v1.0.14 due to not working authentication in v1.0.15. My langflow instance has a public IP. I trust you will fix the authentication issue with 1.0.16 and then I can test the fix

[anovazzi1]

Oh, that's make sense, can you please link here the issue related to the authentication ? if there is no issue yet can you please create one? we will fix asap

[konradbjk]

I do not have 1.0.15 running anymore. I thought someone already would raise it

[anovazzi1]

ok, I will close this issue because it's already fixed. We will track the auth related issue on it's own issue

[codenprogressive]

@anovazzi1 there is an issue related to not be able to sign up new users. it is open here: #3313
can you please take a look?

[konradbjk]

#3438 - in my case it is superuser