Awesome

<div align="center"> <h3>sunday-driver</h3> <a href="https://npmjs.org/package/sunday-driver"> <img src="https://img.shields.io/npm/v/sunday-driver.svg?style=flat-square" /> </a> <a href="https://www.codacy.com/app/spencerkelly86/sunday-driver"> <img src="https://api.codacy.com/project/badge/grade/1b0f3874f43f4b8c87ac855bb69bca8f" /> </a> <div>process a large file, at a steady cruise</div> </div> <p></p> <div align="center"> <h2>🕶️</h2> <div><sup> - slow is smooth, smooth is fast -</sup></div> </div>

sunday-driver works through a large file at a responsible pace - it pauses to let you consider the data, at given points, and waits to resume working once that's all done.

this allows processing a large file, by sizable chunks, without any race-conditions or memory leaking.

(heavily) inspired by line-by-line, by Markus Ostertag🙏

<div align="center"> <code>npm i sunday-driver</code> </div>

const sundayDriver = require('sunday-driver')

let options= {
	file: './my/large/file.tsv',
	splitter: '\n',
	start: '80%', //as percentages, or in bytes
	end: '100%',
	//do your thing, for each segment
	each: (chunk, resume) => {
		console.log(chunk)//do your thing..
		resume()
	}
	//log progress-based events
	atPercent: {
		50: (status) => {
			console.log('50%!')
		},
		75: () => {
			console.log('75%!')
		},
	},
	//log time-based events
	atInterval: {
		'1min': (status) => {
		console.log('1 minute')
		},
		'2mins': () => {
		console.log('2 minutes')
		},
	}
}

sundayDriver(options).then((status)=>{
	console.log('done!')
})

any events/intervals will provide you with all the details of the current reader's status:

/*{
  chunksDone:: 10,     // how many times we've called the 'each' function
	bytesDone:: 20480,   // how many bytes we've processed so far
	filesize:: 61440,    // size of the whole file
	position: 34.42,     // where, in percentage, we are in the file. (if we didn't start at the top!)
	progress: 68.84      // how far, in percentage, we are to being complete
}*/

it was built to support unleashing multiple workers on the same file, and letting them run safely and responsibly, without blowing any fuses.

MIT