langflow-ai / langflow

Langflow is a low-code app builder for RAG and multi-agent AI applications. It’s Python-based and agnostic to any model, API, or database.
http://www.langflow.org
MIT License
35.41k stars 4.22k forks source link

Running the forked repo with all the changes in Docker #4711

Closed Prat011 closed 5 days ago

Prat011 commented 5 days ago

Bug Description

So ive forked the repo and have made changes, but i cannot figure out how to run this version of langflow on docker. The docker example folder allows me to run only the latest release versions. How can i get the image of the local repository?

Reproduction

There's no error as such but I'm trying to understand how I can modify the dockerfile

Expected behavior

I should be able to test Langflow with all changes ive made, on Docker

Who can help?

No response

Operating System

MacOS

Langflow Version

1.1.0

Python Version

None

Screenshot

No response

Flow File

No response

dosubot[bot] commented 5 days ago

To run your forked version of Langflow with all your changes on Docker, you need to modify the Docker setup to build the image from your local repository. Here’s how you can do it:

  1. Clone your forked repository:

    git clone <your-forked-repo-url>
  2. Navigate to the docker_example directory in your cloned repository:

    cd <your-repo-name>/docker_example
  3. Modify the Docker Compose file to build the image from your local Dockerfile. In the docker-compose.yml file, replace the image field with a build context pointing to the directory containing your Dockerfile. For example:

    services:
     langflow:
       build: .
       ports:
         - "7860:7860"
       depends_on:
         - postgres
  4. Run Docker Compose to build and start your services:

    docker compose up --build

This setup will build the Docker image from your local changes and run it, allowing you to test your forked version of Langflow [1].

To continue talking to Dosu, mention @dosu.

italojohnny commented 5 days ago

The make docker_build command will create a Docker image locally with the changes in the directory.

Let me know if this solves your problem

Prat011 commented 5 days ago

that fixed it, thanks!