I got sloppy towards then end and start calling sequelize directly in the services. This goes around the whole reason I was passing services as variables into the class. This will be hard to mock in tests and needs to be passed as a class parameter instead
Currently the frontend just takes input from the user and sends it to the backend. We need to be able to view and handle the data the backend retrieves on the frontend side. It should be intuitive for the user to use
The backend needs to fix how it manages database connections. We will be using multiple connections to store repo data. There needs to be a way to pass the database connection around to handle this from different sections of the code at the same time.
A base frontend for the project that can be developed in a Docker container. Should be able to ask the user for a Github OAuth token to be used by the backend. This doesn't need to be pretty, just working well enough to help let the project take shape and be developed Agiley alongside the backend.
Docker container with our Node.js backend working for PoC. Should be able to grab data from the Github API, using a token from the React frontend, and store it in the database.
Need to rewrite git-manager in JS on server and add it to UI. This will probably take some serious refactoring because the server was made with just the dependency viewing in mind.
Right now the frontend literally just passes the userID in the body of the request to the server. This is bad and there needs to be some sort of session token authentication here instead
Database already supports multiple users, but frontend and backend don't. It would be nice to have a login page and the ability to have multiple users that could either see their own dataset OR see a dataset that's shared between other users in a group. Privileges would be a thing here too
We need to get a PoC environment going. This environment would be a shell of our application. We will need the start of a React frontend, Node.js backend, and a MySQL database. These would all be running in Docker containers. This is the starting point to start building these pieces up from there.
The server passes back a list of failures, like dep files that couldn't be scanned and repos that couldn't be scanned. This needs to be displayed to the user.
These should not be in the same list and should be in their own lists according to language type OR maybe it'd be okay if the language version was specified in the RepoList? I don't really like the second solution, but I'm not sure how to cleanly do the first.