graphql-engine/frontend/docs/generic-info.md
Nicolas Beaussart 8b447a5880 frontend: switch to yarn berry
PR-URL: https://github.com/hasura/graphql-engine-mono/pull/9052
Co-authored-by: Samir Talwar <47582+SamirTalwar@users.noreply.github.com>
GitOrigin-RevId: d9c8bc29f9309247944d130d545c6dcaccff366a
2023-05-05 08:57:54 +00:00

11 KiB

Hasura Console - Generic info

The Hasura console is an admin dashboard to manage the connected database and to try out GraphQL APIs. It is a React application bundled with webpack and the state is managed using Redux. Served by:

  1. Hasura GraphQL Engine: The console is served by GraphQL Engine at /console endpoint (when --enable-console flag is used). Typically runs in No Migration Mode which means that actions on the console are not spitting out migration “yaml” files automatically. Most users will be using the Hasura console in this mode.
  2. Hasura CLI: Served by the Hasura CLI using hasura console command, typically runs with migration mode enabled. All the changes to schema/hasura metadata will be tracked and spit out on the filesystem as migration yaml files and a metadata yaml file. This allows for easy version controlling of the schema/hasura metadata.

    To set up Hasura PRO CLI development environment, follow the steps mentioned here.

Contributing to Hasura console

This guide is for setting-up the console for development on your own machine, and how to contribute.

Console issues in the repo

Issues in the repo for the console UI are labelled as c/console(see list). Issues also labelled as good first issue are aimed at those making their first contribution to the repo (see list). Others marked as help wanted are those requiring community contributions on priority (see list). Please note that some of these issues, labelled with both c/console and c/server, are part of a change/task that requires modifications in both the server and the console. Feel free to open pull requests to address these issues or to add/fix console features, even if a corresponding issue doesn't exist. If you are unsure about whether to go ahead and work on something like the latter, please get in touch with the maintainers in the GraphQL Engine->contrib channel in the community Discord.

Prerequisites

Set up and install dependencies

  • Fork the repo on GitHub.
  • Clone your forked repo: git clone https://github.com/<your-username>/graphql-engine
git clone https://github.com/<your-user-name>/graphql-engine
cd graphql-engine
cd console
nvm use
yarn install
yarn nx build-server-assets console-ce

At this point you need to compile and run the graphql-engine (follow the getting started guide), run it, and the console will be served at http://localhost:8080. Alternatively, run the Console development server as described below.

Run console development server

Hasura console can be developed in two modes, server or cli mode. If you are looking to add/tweak functionality related to migrations, check out Develop with Hasura CLI, otherwise check out Develop with Hasura GraphQL engine.

Both modes require a running instance of GraphQL Engine. The easiest way to get Hasura GraphQL engine instance is by Heroku. You can get it by following the steps given in this link. Other methods to install Hasura GraphQL engine are documented here.

Dotenv is used for setting environment variables for development. In production, these environment variables are templated by the server or CLI.

Develop with Hasura GraphQL engine (server mode)

In server mode, migrations will be disabled and the corresponding functionality on the console will be hidden.

Set up .env file

Environment variables accepted in server mode:

  • NODE_ENV: Console build environment (development/production)
  • NX_CDN_ASSETS: Should assets be loaded from CDN (true/false)
  • NX_ASSETS_PATH: Path to console assets
  • NX_ASSETS_VERSION: Version of console assets being served
  • NX_ENABLE_TELEMETRY: Whether to enable telemetry (true/false)
  • NX_URL_PREFIX: Path at which the console is running
  • NX_DATA_API_URL: The Hasura GraphQL engine url. (If you are running it on Heroku, it will look like https://<app-name>.herokuapp.com, if you are running locally, it will look like http://localhost:<port>)
  • NX_SERVER_VERSION: Hasura GraphQL Engine server version
  • NX_CONSOLE_MODE: In server mode, it should be server
  • NX_IS_ADMIN_SECRET_SET: Is GraphQl engine configured with an admin secret (true/false)
  • NX_HASURA_CONSOLE_TYPE: The environment where the console is running, this could be oss, pro or cloud

Here's an example .env file for server mode:

NODE_ENV=development
NX_CDN_ASSETS=true
NX_ASSETS_PATH=https://graphql-engine-cdn.hasura.io/console/assets
NX_ASSETS_VERSION=channel/stable/v1.0
NX_ENABLE_TELEMETRY=true
NX_URL_PREFIX=/
NX_DATA_API_URL=http://localhost:8080
NX_SERVER_VERSION=v1.0.0
NX_CONSOLE_MODE=server
NX_HASURA_CONSOLE_TYPE=oss
NX_IS_ADMIN_SECRET_SET=true

The .env file can be placed both at the root of the /frontend directory or on a per-app basis. If you put the .env files on a per-app-basis, please remember that also the E2E tests apps (console-ce-e2e, for instance) need a dedicated .env file because they internally launch the web server of the frontend application before launching Cypress. FYI: here is the order Nx follows to read the .env files.

The server also templates consolePath in window.__env which is the relative path of the current page (something like /console/data/schema/public). Using this path, the console determines the DATA_API_URL in production. You do not need to worry about this in development since you are hardcoding the value of DATA_API_URL in .env.

If you're contributing to team-console i.e. console for Hasura Cloud or Hasura EE, refer to this doc.

Run console development server:

Switch to the correct version of Node with nvm (it's strongly suggested to activate deeper shell integration)

nvm install
nvm use

Run the development server

npx nx run console-ce:serve

Develop with Hasura CLI (cli mode)

Set up .env file in apps/console-ce

Environment variables accepted in cli mode:

  • NODE_ENV: Console build environment (development/production)
  • NX_API_HOST: Hasura CLI host. Hasura CLI runs on http://localhost by default.
  • NX_API_PORT: Hasura CLI port. Hasura CLI exposes the API at 9693 by default
  • NX_CDN_ASSETS: Should assets be loaded from CDN (true/false)
  • NX_ASSETS_PATH: Path to console assets
  • NX_ASSETS_VERSION: Version of console assets being served
  • NX_ENABLE_TELEMETRY: Whether to enable telemetry (true/false)
  • NX_URL_PREFIX: Path at which the console is running
  • NX_DATA_API_URL: The Hasura GraphQL engine url. (If you are running it on Heroku, it will look like <app-name>.herokuapp.com, if you are running locally, it will look like http://localhost:<port>)
  • NX_SERVER_VERSION: Hasura GraphQL Engine server version
  • NX_CONSOLE_MODE: In cli mode, it should be cli
  • NX_ADMIN_SECRET: the admin secret passed via the CLI
  • NX_HASURA_CLOUD_ROOT_DOMAIN: cloud root domain, used to simulate and test Hasura Pro CLI with PAT mode Eg. lux-dev.hasura.me for local lux setup

Here's an example .env file for cli mode:

NODE_ENV=development
PORT=3000
NX_API_HOST=http://localhost
NX_API_PORT=9693
NX_CDN_ASSETS=true
NX_ASSETS_PATH=https://graphql-engine-cdn.hasura.io/console/assets
NX_ASSETS_VERSION=channel/stable/v1.0
NX_ENABLE_TELEMETRY=true
NX_URL_PREFIX=/
NX_DATA_API_URL=http://localhost:8080
NX_SERVER_VERSION=v1.0.0
NX_CONSOLE_MODE=cli
NX_ADMIN_SECRET=my-admin-secret
Run console development server:

This setup requires a Hasura CLI console server to be running.

Start Hasura CLI console server

Start Hasura CLI console with the same Hasura GraphQL engine url as configured for DATA_API_URL.

hasura console --endpoint <DATA_API_URL> --admin-secret <your-admin-secret> (optional)
Start development console server in community edition mode
npx nx run console-ce:serve
Start development storybook
npx nx run console-legacy-ce:storybook

Check out the console

Visit http://localhost:4200 to confirm the setup.

Make changes to the code

Make changes to the code and the console will reload automatically to reflect the new changes. Keep iterating. When adding a new feature, it is recommended to add corresponding tests too.

You can use the Redux DevTools Extension to inspect and debug the Redux store. It should automatically connect to the Redux store when started in development mode.

Linter and formatter

If you want to run a linter for all files, you can do:

nx run-many --target=lint

To format all files, you can run:

npx nx format:write

More tooling is listed here (storybook).

Submitting a pull request

  • All the development work happens in your own fork of the graphql-engine.
  • Make sure your commit messages meet the guidelines.
  • Once the changes are done, create a pull request.
  • CI configured for PR will run the test suite.
  • Once everything goes well, it will generate a preview Heroku app.
  • The source code and the preview app will be reviewed by maintainers.

Working with PRO console

Wiki page: https://github.com/hasura/graphql-engine-internal/wiki/Console:-Code-Sharing-Between-OSS-and-PRO