ps-chronicle

A logger package which is eGain PS specific and uses Winston internally. It is used to create logs in 2 formats -> JSON or Simple. The default format is JSON. The log levels allowed are : error, wspayload, info, warn, debug. The timezone is GMT

What's New

Version 0.1.6 (March 25, 2024)

• Introduced support for advanced log filtering tailored for multi-tenant applications. Refer to the usage documentation for implementation details.

installation

npm i ps-chronicle

usage

Step 1) Import/require package

const Logger = require('ps-chronicle');

Step 2) Initialize your logger - once per file/class inside your lambda handler . All fields are optional If outputFormat is not given then the default format will be "json", If loglevel is not set during initialization, then by default it will "debug"

var outputFormat = 'json' or 'simple'
var logger = new Logger('example.js', 'json', 'info'); // (fileName, outputFormat, logLevel)

Step 3) Set requestID per lambda. This request id can be obtained from context object of the lambda. For more information please visit https://docs.aws.amazon.com/lambda/latest/dg/nodejs-context.html Please set request Id as the format ==> LoggerClassName.setRequestId(context.awsRequestId);

Logger.setRequestId(context.awsRequestId); // ClassName.methodName as it is a static method

Step 4) Set customer name. This method is static, so it can be called just once directly on the class. You can use it in index.js or the initial file of your program.

Logger.setCustomerName('TMXXXX'); // connected app
Logger.setCustomerName('RSA'); // customer

Step 5) Also, set method name for each method.

logger.setMethodName('testMethod()');

Step 6) If the log level defined in the log() function is not in the scope then the default log level ,i.e info is set. log() function parameters : i. log level ii. message iii. after above two parameters, n number of parameters can be given. All will be the part of key: xadditionalinfo , which is an array.

function sum(a, b) {
	logger.setMethodName('sum()');
	logger.log('INFO', ' testing  info logger ');
	logger.log('silly', 'testing silly logger '); //silly is not in scope so default log level is set -> info

	logger.log(
		'error',
		' testing error logger ',
		{ tags: 'HEADERS', headers: event.headers },
		{ tags: 'BODY', body: event.body }
	);
	logger.log(
		'wspayload',
		'testing simple wspayload logger ',
		{ tags: 'HEADERS', headers: event.headers },
		{ tags: 'BODY', body: event.body }
	);
	logger.log(
		'warn',
		'testing simple warn logger ',
		{ tags: 'HEADERS', headers: event.headers },
		{ tags: 'BODY', body: event.body }
	);
	logger.log(
		'debug',
		'testing simple debug logger ',
		{ tags: 'HEADERS', headers: event.headers },
		{ tags: 'BODY', body: event.body }
	);
	//rest of the code
}

Step 7) Log filtering for multi-tenant apps

• In multi-tenant apps, we've added the ability to filter logs for specific tenants. This helps pinpoint and debug issues for individual tenants more effectively.

var logger = new Logger('example.js', 'json', 'info');
Logger.setCustomerName('TMXXXX');
Logger.setCustomerToFilterLogs('TMXXXX');

//Real-world eg:
Logger.setCustomerName(event.headers['tenant-id']);
Logger.setCustomerToFilterLogs(process.env.CUSTOMER_NAME_TO_FILTER);

• Both methods are static. You can use them in index.js or the initial file of your program.
• CUSTOMER_NAME_TO_FILTER can be dynamic and updated via the config API.
• setCustomerToFilterLogs() is an optional setting. If not provided, the logger functions normally without any filtering.
• Log statements will only print if the customer's name matches between the setCustomerName method and the logger initialization.

Important : If loggers are not getting printed for async functions then please use the below code at the end of your lambda function:

await logger.waitForLogger(Logger);
`