​

Guides

​

API

• Overview

​

Data integrations

• Add a new source to the data integration pipeline
• Add a new destination to the data integration pipeline

​

Streaming pipelines

• Add a new source or destination to the streaming pipeline

​

IO classes

• Add a new IO class

​

Style guide

​

Linter

Install flake8 in your IDE to lint the Python code.

To run the linter locally, execute this script:

./scripts/server/lint.sh

​

Testing

​

Unit tests

Add unit tests for the feature in mage_ai/tests directory.

To run the tests locally, execute this script:

./scripts/server/test.sh

It is also possible to run unit tests directly in a live docker instance, as given in the following steps.

1. Find out the backend server container name, with the command docker container ls in a terminal:

 % docker container ls

CONTAINER ID   IMAGE       COMMAND                  CREATED      STATUS          PORTS                    NAMES
8dbcfbe41755   mage/data   "./scripts/install_a…"   5 days ago   Up 38 minutes   0.0.0.0:3000->3000/tcp   mage-ai-app-1
b9d811e1e3e8   mage/data   "python mage_ai/serv…"   5 days ago   Up 38 minutes   0.0.0.0:6789->6789/tcp   mage-ai-server-1

2. Start an interactive bash session with the backend server container:

% docker exec -it mage-ai-server-1 /bin/bash

3. Run unit tests with the following command:

root@b9d811e1e3e8:/home/src# python3 -m unittest discover -s mage_ai.tests --failfast

​

Debugging

Instead of using breakpoint(), add the following line to your code where you want a debug:

import pdb; pdb.set_trace()

Attach to running container to use debugger. To get the container ID, run docker ps and look in the NAMES column.

docker attach [container_id]

​

Snippets

print(self.__dict__, locals()