Code intelligence for Python (beta)

This extension provides Python code intelligence on Sourcegraph. Ensure this extension is enabled and then try it on some examples:

• Tensorflow: tensorflow_ranking/examples/tf_ranking_libsvm.py
• Flask: flask/views.py
• Zulip: corporate/lib/stripe.py
• Trivial sample: pkg0/m0.py

Usage with private Sourcegraph instances

This extension is configured to talk to a language server over WebSockets. If you are running a private Sourcegraph instance, you should run your own language server. The server is available as a Docker image sourcegraph/lang-python from Docker Hub.

🔐 Secure deployment 🔐

If you have private code, we recommend deploying the language server behind an auth proxy (such as the example below using HTTP basic authentication in NGINX), a firewall, or a VPN.

HTTP basic authentication

You can prevent unauthorized access to the language server by enforcing HTTP basic authentication in nginx, which comes with the sourcegraph/server image. At a high level, you'll create a secret then put it in both the nginx config and in your Sourcegraph global settings so that logged-in users are authenticated when their browser makes requests to the Python language server.

Here's how to set it up:

Create an .htpasswd file in the Sourcegraph config directory with one entry:

$ htpasswd -c ~/.sourcegraph/config/.htpasswd langserveruser
New password:
Re-type new password:
Adding password for user langserveruser

Add a location directive the nginx.conf that will route requests to the Python language server:

...
http {
    ...
    server {
        ...
        location / {
            ...
        }

        location /python {
            proxy_pass http://host.docker.internal:4288;
            proxy_http_version 1.1;
            proxy_set_header Upgrade $http_upgrade;
            proxy_set_header Connection "Upgrade";

            auth_basic "basic authentication is required to access the language server";
            auth_basic_user_file /etc/sourcegraph/.htpasswd;
        }
    }
}

• If you're running the quickstart on Linux, change host.docker.internal to the output of ip addr show docker0 | grep -Po 'inet \K[\d.]+'.
• If you're using Kubernetes (e.g. deploy-sourcegraph), change host.docker.internal to lang-python.

Add these to your Sourcegraph global settings:

  "python.serverUrl": "ws://langserveruser:PASSWORD@host.docker.internal:7080/python",
  "python.sourcegraphUrl": "http://host.docker.internal:7080",

Fill in the PASSWORD that you created above.

• If you're running the quickstart on Linux, change host.docker.internal to the output of ip addr show docker0 | grep -Po 'inet \K[\d.]+'.
• If you're using Kubernetes (e.g. deploy-sourcegraph):
  • python.serverUrl is the address of the Python language server from the perspective of a user's browser (e.g. https://sourcegraph.example.com/python)
  • python.sourcegraphUrl is the address of the Sourcegraph instance from the perspective of the Python language server (e.g. http://sourcegraph-frontend:30080)

Finally, restart the sourcegraph/server container (or nginx deployment if deployed to Kubernetes) to pick up the configuration change.

After deploying the language server, unauthenticated access to http://localhost:7080/python (or https://sourcegraph.example.com/python) should be blocked, but code intelligence should work when you're logged in.

You can always revoke the PASSWORD by deleting the .htpasswd file and restarting nginx.

Using Docker

1. Run the Python language server Docker container:

   docker run --rm -p 4288:4288 sourcegraph/lang-python:insiders

2. Enable this extension on your Sourcegraph https://sourcegraph.example.com/extensions/sourcegraph/lang-python

3. Add these to your Sourcegraph settings:

     "python.serverUrl": "ws://localhost:4288",
     "python.sourcegraphUrl": "http://host.docker.internal:7080",

   If you're running on Linux, change python.sourcegraphUrl to the IP given by:

   ip addr show docker0 | grep -Po 'inet \K[\d.]+'

Known issues

• Dependencies are not installed. Hovers, definitions, and references only work for sources checked into the single repository you are viewing.
• Cross-repository definitions and references are not yet supported.
• Hangs on very large repositories (e.g. django/django)