The example project provides the ability clients to manage projects. New clients can sign up which creates an account and a user with role admin. Two basic roles are available for users: admins and users. Admins have the ability to perform all CRUD actions on projects and users. Those users with role user only have the ability to view projects and users.
All business logic should be contained as a package outside both the web app and web API. This enables both the web app and web API to use the same API (Golang packages) with the only main difference between them is their response, HTML or JSON.
Navigate to the root of the project and first set your AWS configs. Copy `sample.env_docker_compose` to `.env_docker_compose` and update the credentials for docker-compose.
Use the `docker-compose.yaml` to run all of the services, including the 3rd party services. The first time to run this command, Docker will download the required images for the 3rd party services.
Default configuration is set which should be valid for most systems. Use the `docker-compose.yaml` file to configure the services differently using environment variables when necessary.
You can hit <ctrl>C in the terminal window running `docker-compose up`. Once that shutdown sequence is complete, it is important to run the `docker-compose down` command.
Responsive web application that renders HTML using the `html/template` package from the standard library to enable direct interaction with clients and their users. It allows clients to sign up new accounts and provides user authentication with HTTP sessions. The web app relies on the Golang business logic packages developed to provide an API for internal development.
REST API available to clients for supporting deeper integrations. The API implements JWT authentication that renders results as JSON to clients. This API is not directly used by the web app to prevent locking the functionally needed internally for development of the web app to the same functionality exposed to clients. It’s believed that in the beginning, having to define an additional API for internal purposes is worth at additional effort as the internal API can handle more flexible updates. The API exposed to clients can then be maintained in a more rigid/structured process to manage client expectations.
To make a request to the service you must have an authenticated user. Users can be created with the API but an initial admin user must first be created. While the Web App service is running, signup to create a new account. The email and password used to create the initial account can be used to make API requests.
Before any authenticated requests can be sent you must acquire an auth token. Make a request using HTTP Basic auth with your email and password to get the token.
Schema is a minimalistic database migration helper that can manually be invoked via CLI. It provides schema versioning and migration rollback. To support POD architecture, the schema for the entire project is defined globally and is located inside internal: [internal/schema](https://gitlab.com/geeks-accelerator/oss/saas-starter-kit/tree/master/example-project/internal/schema)
Keeping a global schema helps ensure business logic then can be decoupled across multiple packages. It’s a firm belief that data models should not be part of feature functionality. Globally defined structs are dangerous as they create large code dependencies. Structs for the same database table can be defined by package to help mitigate large code dependencies.
The example schema package provides two separate methods for handling schema migration.
List of direct SQL statements for each migration with defined version ID. A database table is created to persist executed migrations. Upon run of each schema migration run, the migraction logic checks the migration database table to check if it’s already been executed. Thus, schema migrations are only ever executed once. Migrations are defined as a function to enable complex migrations so results from query manipulated before being piped to the next query.
If you have a lot of migrations, it can be a pain to run all them, as example, when you are deploying a new instance of the app, in a clean database. To prevent this, use the initSchema function that will run if no migration was run before (in a new clean database). Remember to create everything here, all tables, foreign keys and what more you need in your app.
Another bonus with the globally defined schema allows testing to spin up database containers on demand include all the migrations. The testing package enables unit tests to programmatically execute schema migrations before running any unit tests.
Postgres is only supported based on its dependency of sqlxmigrate. MySQL should be easy to add to sqlxmigrate after determining better method for abstracting the create table and other SQL statements from the main testing logic.
