debug-env

debug-env

A wrapper for switching logging libraries depending on environment.

Works like a drop in place replacement of Debug but uses Pino in production environment.

Debug-env:

• Can be used just like debug
• Adds levels to Debug
• Adds namespaces to Pino
• Swaps between them depending on NODE_ENV or DEBUGGER var
• Adds a silent logger
• Can be configured via code.

Installation

 npm install debug-env --save

Usage

You can use exactly like Debug. Once you want to separate messages by level just call by method.

Example:

const debug = require('debug-env')('namespace:namespace');
debug("Default behaviour"); // level debug

// Calling it with levels
debug.warn("You want to look at this at runtime");
debug.trace("SQL for something", sql); // You don't want to see this often

// Check available levels (Same as in Pino)
console.log(require('debug-env').levels);

['fatal', 'error', 'warn', 'info', 'debug', 'trace']

Displaying debug info

Works by reading the following environment variables:

• DEBUG or NS: namespace (from Debug)*
• DEBUG_LEVEL: (default level: debug) (from Pino)
• DEBUGGER: Force overrides debugger to use

* Many packages use Debug, if you activate DEBUG=* all packages will log directly. If you have that issue use NS instead.

Example:

DEBUG=myapp:* DEBUG_LEVEL=debug npm start

Changing the debugger / other options

You can change nearly all configuration using force:

let debug =  require('debug-env');
let options = {
  loggers: {
    production: 'pino',
    development: 'debug',
    test: 'debug'
  },
  level: 'warn',
  env: 'development',
  namespaces: 'test:msg'
};
debug.force(options);
process.env.DEBUG = 'test:msg'; // we don't override the real var in the package
debug = test('test:msg');

// Or force a debbuger from the command line
DEBUGGER=pino npm start

Using in production mode

In production mode the logger is changed to Pino without you making any changes to the code. Namespaces are added to the output as ns property. You can use Pino transports and other external libraries as usual.

Example Using pm2 to execute and control the log rotation. Check pm2 ecosystem file.

pm2 ecosystem.conf.js

module.exports = {
  apps : [{
    name      : 'myapp',
    script    : 'node index.js',
    env: {
      NODE_ENV: 'development',
      DEBUG:'boot:*',
      DEBUG_LEVEL:'debug'
    },
    env_production : {
      NODE_ENV: 'production',
      DEBUG:'myapp:*',
      DEBUG_LEVEL:'info'
    }
  }]
};

Advanced use

It is possible to use libraries specific functionality but it will break compatibility. So you need to do it conditionally.

if (process.env.NODE_ENV === 'production') {
    // do stuff for PINO
} else {
    // do stuff for DEBUG
}

Example with debug

Debug actually returns debug, and it's also available under the logger property.

// Example taken and adapted from Debug site

const createDebug = require('debug-env');
createDebug.logger.debug.formatters.h = (v) => {
  return v.toString('hex');
}
const debug = createDebug('foo');

Example with Pino

Pino receives parameters as an object when creating loggers. Pass this object as a second parameter after namespace and it will be directly passed to pino.

// Example from pino-noir
var pino = require('pino')({
  serializers: redaction
});

// To do the same with debug-env:
const pinoOptions = {
  serializers: redaction
};
const createDebug = require('debug-env')(namespace, pinoOptions));

Debug & Pino advanced functionality

Both libraries are far more powerful. If you require advance functionality, you can add conditionally or use the library directly.