I was working in a tool to detect N+1 queries in Django apps and now I think is in a good state to be used regularly. The tool has two modes of usage, run directly in CLI to analyze a whole Django project, or as an LSP, this can be configured in your editor, and I just made a Visual Studio Code extension to be installed easily. I expect that can be useful for you, specially if you are learning. I'll continuing improving and manage more edge cases, then feedback will be very appreciated.
I solved a specific problem: handling the LSP inside a Docker container without requiring the libraries to be installed on the host. This was focused in Python using Pyright and Ruff, but can be extensible to another language.
Thanks for the feedback! I’ll update the README to clarify these points.
You can view the full Pytest output using the :PytestOutput command. This shows all outputs from failing tests (since Pytest doesn’t display output for passing tests by default).
I also plan to implement support for custom arguments — for example, Pytest's -s flag to show stdout even when tests pass.
If Docker is disabled, all Docker-related features are also disabled.
When Docker Compose is enabled, its configuration takes precedence over the plugin’s direct settings.
The prefix_app setting maps your local directory to the Docker path.
For example, if your current working directory is ~/projects/, and the prefix is app, the Docker path /usr/src/app will map to ~/projects/app.
Docker is responsible only for path mapping and executing the Pytest command inside a running container — the container must already be running.
If you're using Docker Compose, the plugin retrieves the Docker path from the volume configuration.
For example, if your docker-compose.yml contains:
volumes:
- app:/usr/src/app
Then /usr/src/app will be used as the Docker path.
If enable_docker_compose is set to false, the plugin will fall back to the manually configured path instead. (Note: the container itself is not retrieved from Docker Compose at this point.)
I also made a VSCode extension if you want to use it inside the editor: https://github.com/richardhapb/django-check-vscode