graphql-engine/console
Vijay Prasanna e4fc6a868e feat (console): setup a simple data abstraction layer
PR-URL: https://github.com/hasura/graphql-engine-mono/pull/4640
GitOrigin-RevId: 003cbce502f196deed03960af139617876f5e76f
2022-06-09 09:01:00 +00:00
..
.storybook remove font awesome and fix the setting gears 2022-06-02 09:22:54 +00:00
bin console: migrate HTML helpers, route hooks helpers and version utils to TypeScript (#4641) 2020-05-14 19:30:05 +05:30
cypress console: add support for application/x-www-form-urlencoded in rest connectors 2022-05-30 13:51:15 +00:00
docs Console/permissions tab component 2022-03-08 07:46:19 +00:00
exports Add Feature Flags section in Settings 2022-03-23 10:57:17 +00:00
scripts workflow: add lint and format action for console (#4501) 2020-05-29 16:45:43 +05:30
src feat (console): setup a simple data abstraction layer 2022-06-09 09:01:00 +00:00
static console: add storybook support 2021-09-20 16:31:25 +00:00
webpack remove font awesome and fix the setting gears 2022-06-02 09:22:54 +00:00
.babelrc console: add storybook support 2021-09-20 16:31:25 +00:00
.eslintignore Console/permissions tab component 2022-03-08 07:46:19 +00:00
.eslintrc fix(console): actions search case insensitive (close #6351) 2022-04-06 09:21:21 +00:00
.gitignore console: use react-query for data fetching with custom hook 2021-11-04 10:42:17 +00:00
.npmignore console: package modules from oss 2021-03-17 18:59:18 +00:00
.prettierignore console: fix linter config (#4137) 2020-04-06 16:53:04 +05:30
cypress-local.json console/tests: add manage db tests for mssql 2021-08-09 14:04:54 +00:00
cypress.json console/tests: add manage db tests for mssql 2021-08-09 14:04:54 +00:00
hasuraconfig.js add console to graphql-engine 2018-06-28 10:27:37 +05:30
jest.config.js console: add form library components 2021-10-07 09:51:10 +00:00
Makefile console: Automatically install node_modules. 2022-06-03 12:40:27 +00:00
package-lock.json feat (console): setup a simple data abstraction layer 2022-06-09 09:01:00 +00:00
package.json feat (console): setup a simple data abstraction layer 2022-06-09 09:01:00 +00:00
postcss.config.js console: add tailwindcss for styles 2021-08-10 13:46:52 +00:00
README.md console: document console development tooling [CON-90] 2022-03-23 05:39:29 +00:00
tailwind.config.js Con 229 migrate remote schemas to tailwind 2022-06-07 09:57:22 +00:00
tsconfig.json console: add form library components 2021-10-07 09:51:10 +00:00
tsconfig.test.json console: add schema sharing 2021-07-08 08:18:25 +00:00

Hasura Console

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
npm ci

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)
  • PORT: The port where Hasura console will run locally
  • CDN_ASSETS: Should assets be loaded from CDN (true/false)
  • ASSETS_PATH: Path to console assets
  • ASSETS_VERSION: Version of console assets being served
  • ENABLE_TELEMETRY: Whether to enable telemetry (true/false)
  • URL_PREFIX: Path at which the console is running
  • 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>)
  • SERVER_VERSION: Hasura GraphQL Engine server version
  • CONSOLE_MODE: In server mode, it should be server
  • IS_ADMIN_SECRET_SET: Is GraphQl engine configured with an admin secret (true/false)

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

NODE_ENV=development
PORT=3000
CDN_ASSETS=true
ASSETS_PATH=https://graphql-engine-cdn.hasura.io/console/assets
ASSETS_VERSION=channel/stable/v1.0
ENABLE_TELEMETRY=true
URL_PREFIX=/
DATA_API_URL=http://localhost:8080
SERVER_VERSION=v1.0.0
CONSOLE_MODE=server
IS_ADMIN_SECRET_SET=true

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:
 npm run dev

Develop with Hasura CLI (cli mode)

Set up .env file

Environment variables accepted in cli mode:

  • NODE_ENV: Console build environment (development/production)
  • PORT: The port where Hasura console will run locally
  • API_HOST: Hasura CLI host. Hasura CLI runs on http://localhost by default.
  • API_PORT: Hasura CLI port. Hasura CLI exposes the API at 9693 by default
  • CDN_ASSETS: Should assets be loaded from CDN (true/false)
  • ASSETS_PATH: Path to console assets
  • ASSETS_VERSION: Version of console assets being served
  • ENABLE_TELEMETRY: Whether to enable telemetry (true/false)
  • URL_PREFIX: Path at which the console is running
  • 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>)
  • SERVER_VERSION: Hasura GraphQL Engine server version
  • CONSOLE_MODE: In cli mode, it should be cli
  • ADMIN_SECRET: the admin secret passed via the CLI
  • 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
API_HOST=http://localhost
API_PORT=9693
CDN_ASSETS=true
ASSETS_PATH=https://graphql-engine-cdn.hasura.io/console/assets
ASSETS_VERSION=channel/stable/v1.0
ENABLE_TELEMETRY=true
URL_PREFIX=/
DATA_API_URL=http://localhost:8080
SERVER_VERSION=v1.0.0
CONSOLE_MODE=cli
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
npm run dev
Start development storybook
npm run storybook

Check out the console

Visit http://localhost:3000 to confirm the setup.

Testing Development Server

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.

By default redux-logger is enabled to assist in development. You can disable it if you wish by commenting out the createLogger line in src/client.js

E2E and Unit Tests

  • E2E Tests are written using Cypress .

    • Run tests: npm run cypress
    • Write your tests in the cypress/integration directory.
  • Unit tests are written using Jest Framework

    • Tests are written inside __test__ folder of respective services.
    • Unit testing is mostly applied on the utility functions.
    • If you want to run tests, execute npm run jest or npm run jest-watch (for watch mode)
    • Snapshot testing is also used with unit tests.
    • when you want to update the snapshot, you can run npm run jest -- -u or press u if you are in jest-watch mode.

Linter and formatter

Pre-commit git-hook that runs linter and formatter is by default disabled. You can enable it by adding HUSKY_PRE_COMMIT=true to your .env file.

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

npm run lint

To format all files, you can run:

npm run format

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