How to convert JSON array to CSV using Node.js?

Do it yourself like this:

'use strict';

var fs = require('fs');

let myObj = {
  "rows": [
    [
      "New , Visitor",
      "(not set)",
      "(not set)",
      "0"
    ],
    [
      "New Visitor",
      "(not set)",
      "(not set)",
      "mobile"
    ],
    [
      "New Visitor",
      "(not set)",
      "(not set)",
      "mobile"
    ],
    [
      "New Visitor",
      "(not set)",
      "(not set)",
      "mobile",
    ]
  ]
}

// 1. One way - if you want the results to be in double quotes and you have comas inside

// choose another string to temporally replace commas if necessary
let stringToReplaceComas = '!!!!';

myObj.rows.map((singleRow) => {
  singleRow.map((value, index) => {
    singleRow[index] = value.replace(/,/g, stringToReplaceComas);
  })
})

let csv = `"${myObj.rows.join('"\n"').replace(/,/g, '","')}"`;
// // or like this
// let csv = `"${myObj.rows.join('"\n"').split(',').join('","')}"`;

csv = csv.replace(new RegExp(`${stringToReplaceComas}`, 'g'), ',');

// // 2. Another way - if you don't need the double quotes in the generated csv and you don't have comas in rows' values
// let csv = myObj.rows.join('\n')

fs.writeFile('name.csv', csv, 'utf8', function(err) {
  if (err) {
    console.log('Some error occured - file either not saved or corrupted file saved.');
  } else {
    console.log('It\'s saved!');
  }
});

Use libraries

ex. https://github.com/mrodrig/json-2-csv, https://github.com/wdavidw/node-csv, https://github.com/wdavidw/node-csv-stringify

an example using json-2-csv (https://github.com/mrodrig/json-2-csv)

'use strict';

const converter = require('json-2-csv');

let myObj = {
  "rows": [
    {
      value1: "New Visitor",
      value2: "(not set)",
      value3: "(not set)",
      value4: "0"
    },
    {
      value1: "New Visitor",
      value2: "(not set)",
      value3: "(not set)",
      value4: "mobile"
    },
    {
      value1: "New Visitor",
      value2: "(not set)",
      value3: "(not set)",
      value4: "mobile"
    },
    {
      value1: "New Visitor",
      value2: "(not set)",
      value3: "(not set)",
      value4: "mobile",
    }
  ]
}

let json2csvCallback = function (err, csv) {
    if (err) throw err;
    fs.writeFile('name.csv', csv, 'utf8', function(err) {
      if (err) {
        console.log('Some error occured - file either not saved or corrupted file saved.');
      } else {
        console.log('It\'s saved!');
      }
    });
};

converter.json2csv(myObj.rows, json2csvCallback, {
  prependHeader: false      // removes the generated header of "value1,value2,value3,value4" (in case you don't want it)
});

an example using csv-stringify (https://github.com/wdavidw/node-csv-stringify)

'use strict';

var stringify = require('csv-stringify');
var fs = require('fs');

let myObj = {
  "rows": [
    [
      "New Visitor",
      "(not set)",
      "(not set)",
      "0"
    ],
    [
      "New Visitor",
      "(not set)",
      "(not set)",
      "mobile"
    ],
    [
      "New Visitor",
      "(not set)",
      "(not set)",
      "mobile"
    ],
    [
      "New Visitor",
      "(not set)",
      "(not set)",
      "mobile",
    ]
  ]
}

stringify(myObj.rows, function(err, output) {
  fs.writeFile('name.csv', output, 'utf8', function(err) {
    if (err) {
      console.log('Some error occured - file either not saved or corrupted file saved.');
    } else {
      console.log('It\'s saved!');
    }
  });
});

Three easy steps: Read. Convert. Write.

Step 1: Read.

If you need to read the JSON from a file (as indicated by your inclusion of the filename response.json in your post), you will require the Node.js FileSystem API:

const fs = require('fs');                          // Require Node.js FileSystem API.
const JSONFile = fs.readFileSync('response.json'); // Read the file synchronously.

Note: If you prefer, you can read the file asynchronously with fs.readFile() and perform the conversion in a callback function.

Step 2: Convert.

Whether you read your JSON from a local file or GET it from a server, you will need to parse it into a Plain Old JavaScript Object first using the JSON.parse method:

const JSONasPOJO = JSON.parse(JSONFile); // Parse JSON into POJO.

Then perform a series of joins on the child arrays and parent array:
SEE EDIT BELOW

/* THIS IS UNNECESSARY FOR "COMMA" SEPARATED VALUES
const CSVString = JSONasPOJO
    .rows                    // Get `rows`, which is an array.
    .map(                    // Map returns a new array.
        row => row.join(',') // Each child array becomes a comma-separated string.  
     )                    
    .join('\n');             // Parent array becomes a newline-separated string...
                             // ...of comma-separated strings.
                             // It is now a single CSV string!
*/

EDIT:

While the previous code certainly works, it is unnecessary to use .map and .join on the child arrays. As @Relu demonstrates, a single .join on the parent array is sufficient because JavaScript will automatically convert the child arrays into comma-separated strings by default since .join must return a string and cannot contain any child arrays.

You could use the above pattern if you want to join the child arrays with something other than a comma.

Otherwise:

var CSVString = JSONasPOJO.rows.join('\n'); // Array becomes a newline-separated...
                                            // ...string of comma-separated strings.
                                            // It is now a single CSV string!

Here, we can see that conversion in action:

const JSONasPOJO = {
  "rows": [
    [
      "New Visitor",
      "(not set)",
      "(not set)",      
      "0"
    ],
    [
      "New Visitor",
      "(not set)",
      "(not set)",
      "mobile"      
    ],
    [
      "New Visitor",
      "(not set)",
      "(not set)",
      "mobile"    
    ],
    [
      "New Visitor",
      "(not set)",
      "(not set)",
      "mobile" // NOTE: Here I removed a trailing comma,
               // ...which is invalid JSON!
    ]
  ]
}

const CSVString = JSONasPOJO.rows.join('\n');

console.log(CSVString);

Step 3: Write.

Using the FileSystem API again, write to a file, and log an error or a success message:

fs.writeFile('name.csv', CSVString, err => {
    if (err) return console.log(err);
    console.log('FILE SUCCESSFULLY WRITTEN!\n');
});

Note: Here, I demonstrate the asynchronous pattern using a callback to log my error and success messages. If you prefer, you can write the file synchronously with fs.writeFileSync().

Putting it all together

I like to add plenty of console.log() messages to my Node.js scripts.

const fs = require('fs');

const inFilename  = 'response.json',
      outFilename = 'name.csv';

console.log(`Preparing to read from ${inFilename} …`);

const JSONContents = fs.readFileSync(inFilename);

console.log(`READ:\n${JSONContents}`);
console.log('Preparing to parse as JSON …');

const JSONasPOJO = JSON.parse(JSONContents);

console.log(`PARSED:\n${JSONasPOJO}`);
console.log('Preparing to convert into CSV …');

const CSVString = JSONasPOJO.rows.join('\n');

console.log(`CONVERTED:\n${CSVString}`);
console.log(`Preparing to write to ${outFilename} …`);

fs.writeFile(outFilename, CSVString, err => {
    if (err) return console.error(err);
    console.log('FILE SUCCESSFULLY WRITTEN!');
});