Prerequisites

Name Version Link
NodeJS >=10 https://nodejs.org
MySQL >=5.7.27 https://dev.mysql.com/downloads/mysql/
Redis Latest https://redis.io

Any question, see also: FAQ - General

Install Lerna

$ npm i -g lerna

Clone Cabloy Repository

$ mkdir cabloy-lerna
$ cd cabloy-lerna
$ git clone https://github.com/zhennann/cabloy.git

Install Dependencies

$ cd cabloy
$ lerna bootstrap

Config at Front

cp -r src/front/_config src/front/config

Config at Backend

cp -r src/backend/_config src/backend/config

Configure Redis

Cluster now becomes the first class citizen of CabloyJS. In other words, the CabloyJS project is ready to be deployed in a clustered environment

The cluster feature of CabloyJS is designed based on Redis. Therefore, the connection configuration information of Redis needs to be set. Here, take local development environment as an example:

src/backend/config/config.local.js

  // redis
  const __redisConnectionDefault = {
    host: '127.0.0.1',
    port: 6379,
    password: '',
    db: 0,
  };
  const __redisConnectionDefaultCache = Object.assign({}, __redisConnectionDefault, {
    keyPrefix: `cache_${appInfo.name}:`,
  });
  const __redisConnectionDefaultIO = Object.assign({}, __redisConnectionDefault, {
    keyPrefix: `io_${appInfo.name}:`,
  });

  config.redisConnection = {
    default: __redisConnectionDefault,
    cache: __redisConnectionDefaultCache,
    io: __redisConnectionDefaultIO,
  };

  config.redis = {
    clients: {
      redlock: config.redisConnection.default,
      limiter: config.redisConnection.default,
      queue: config.redisConnection.default,
      broadcast: config.redisConnection.default,
      cache: config.redisConnection.cache,
      io: config.redisConnection.io,
    },
  };

CabloyJS implements multiple features based on Redis: limiter, queue, broadcast, cache. These features access the same Redis service by default. When the number of visits increases, separation can be considered

Configure MySQL

EggJS provides three runtime environments, and CabloyJS implements different database access strategies for each runtime environment, see also: Config

Therefore, it is necessary to set up the database connection configuration of the three runtime environments separately

1. unit-test environment

src/backend/config/config.unittest.js

  // mysql
  config.mysql = {
    clients: {
      // donnot change the name
      __ebdb: {
        host: '127.0.0.1',
        port: '3306',
        user: 'root',
        password: '',
        database: 'sys', // donnot change the name
        hook: {
          meta: {
            long_query_time: 200,
          },
        },
      },
    },
  };
Name Description
__ebdb the system’s built-in MySQL node name, which is not allowed to change
database specified as sys. Every time service started, the old test database is always deleted and a new test database is created
hook.meta.long_query_time if the execution time of SQL statement exceeds long_query_time(ms), it will be outputed to the console

2. local development environment

src/backend/config/config.local.js

  // mysql
  config.mysql = {
    clients: {
      // donnot change the name
      __ebdb: {
        host: '127.0.0.1',
        port: '3306',
        user: 'root',
        password: '',
        database: 'sys', // recommended
        hook: {
          meta: {
            long_query_time: 0,
          },
        },
      },
    },
  };
Name Description
database recommendeded as sys. Every time service started, the available test database is always looked up, and if not found, a new one is created
If specified as the other specific database name, then use this one directly
hook.meta.long_query_time if long_query_time set to 0，all of SQL statements will be outputed to the console

3. production environment

src/backend/config/config.prod.js

  // mysql
  config.mysql = {
    clients: {
      // donnot change the name
      __ebdb: {
        host: '127.0.0.1',
        port: '3306',
        user: 'root',
        password: '',
        database: '{{name}}',
        hook: {
          meta: {
            long_query_time: 500,
          },
        },
      },
    },
  };
Name Description
database The database need to be created manually
hook.meta.long_query_time if the execution time of SQL statement exceeds long_query_time(ms), it will be outputed to the console

Unit-Test

$ npm run test:backend

CabloyJS recommends a test-driven development mode. By test-driven, you can lock in the results of development, and when there are code changes, you can find problems as soon as possible, so as to solve problems

In addition, the database architecture changes frequently during the development stage, when only one unit test command is needed to take effect, which means that the old test database is always deleted and a new test database is created automatically

Run

Start backend service

$ npm run dev:backend

Start frontend service

$ npm run dev:front

Enjoy

Docker Compose

There is a docker-compose.yml configuration file in the root directory of the project. If you have installed the docker compose environment, you can start all services of CabloyJS with only one command, including Redis, MySQL, Nginx and CabloyJS backend service

$ sudo docker-compose up