run-screen

run-screen is tool to run multiple process in parallel in different screen in order to switch from one output to the other.

For example, run-screen "watch uptime" htop will start watch uptime (refresh uptime every 2 sec) and htop in parallel. Pressing the numeric key 2, of your keyboard, will switch to the htop screen. Pressing the numeric key 1, of your keyboard, will switch to the watch uptime screen.

You can have up to 10 process in parallel, switching from one screen to the other by the numeric key of your keyboard, from 1 to 10.

To use it:

run-screen "command 1" "command 2" "command 3" "... bis 10"

To exit, press key combination ctrl+c

To stop a process, press key combination ., start again the process by pressing again .

Action keys

• To exit, press key combination ctrl+c
• Stop/start process, press key .
• Next screen, press key >
• Previous screen, press key <
• Dashboard, press key tab, and press tab again to come back to the previous screen

Dashboard

The dashboard give you an overview of all the screens. If a new error happen on a screen, it will be shown.

Advance features

Instead to give the command directly as arguments, it is possible to pass them from a config file.

run-screen example.config.js

example.config.js:

module.exports = {
    screens: [
        {
            cmd: 'yarn foo',
        },
        {
            cmd: 'command 2',
        },
    ],
}

This will allow you to access some advance feature, like executing a function before or after the command run:

module.exports = {
    screens: [
        {
            before: (id, screenConfig) => {
                console.log(`\nLet's wait 3 sec to run cmd "${screenConfig.cmd}"\n`);
                return new Promise(resolve => setTimeout(resolve, 3000));
            },
            cmd: 'yarn foo',
            after: (screen, runScreenInstance) => {
                screen.proc.on('close', () => {
                    console.log(`\nRestart process after exit. "${screen.config.cmd}"\n`);
                    runScreenInstance.startScreen(screen);
                });
            },
        },
        ...,
    ],
}

before and after don't get the same parameters. This is because when calling the before function the screen and the process are not yet created in contrast to the after function.

ScreenConfig {
    before?: (id: number, screenConfig: ScreenConfig, runScreen: RunScreen) => Promise<void> | void;
    after?: (screen: Screen, runScreen: RunScreen) => Promise<void> | void;
    cmd: string;
}

The config file give you as well the possibility to change the key binding:

module.exports = {
    keys: {
        TOGGLE_PROCESS: 'key', // e.g '\u0000' for ctrl+space
        KILL_PROCESS: 'key', // e.g 'k'
        TOGGLE_DASHBOARD: 'key', // e.g '#'
        NEXT_SCREEN: 'key', // e.g '<'
        PREV_SCREEN: 'key', // ...
    },
    screens: [...],
}

Wait for port

Another example, waiting for a port to open, before to start the command:

First install wait-on library

yarn add wait-on --dev

example-wait-on.config.js

const waitOn = require('wait-on');
 
module.exports = {
    screens: [
        {
            before: () => waitOn({ resources: ['tcp:3000']}),
            cmd: 'yarn foo',
        },
        // ...
    ],
}

run example

run-screen example-wait-on.config.js