Skip to content

Parse XLSX with Node and create json

Ok so I found this really well documented node_module called js-xlsx

Question: How can I parse an xlsx to output json?

Here is what the excel sheet looks like:

enter image description here

In the end the json should look like this:

[
   {
   "id": 1,
   "Headline": "Team: Sally Pearson",
   "Location": "Austrailia",
   "BodyText": "...",
   "Media: "..."
   },
   {
   "id": 2,
   "Headline": "Team: Rebeca Andrade",
   "Location": "Brazil",
   "BodyText": "...",
   "Media: "..."
   }
]

index.js:

if(typeof require !== 'undefined') {
    console.log('hey');
    XLSX = require('xlsx');
}
var workbook = XLSX.readFile('./assets/visa.xlsx');
var sheet_name_list = workbook.SheetNames;
sheet_name_list.forEach(function(y) { /* iterate through sheets */
  var worksheet = workbook.Sheets[y];
  for (z in worksheet) {
    /* all keys that do not begin with "!" correspond to cell addresses */
    if(z[0] === '!') continue;
    // console.log(y + "!" + z + "=" + JSON.stringify(worksheet[z].v));

  }

});
XLSX.writeFile(workbook, 'out.xlsx');

Answer

Improved Version of “Josh Marinacci” answer , it will read beyond Z column (i.e. AA1).

var XLSX = require('xlsx');
var workbook = XLSX.readFile('test.xlsx');
var sheet_name_list = workbook.SheetNames;
sheet_name_list.forEach(function(y) {
    var worksheet = workbook.Sheets[y];
    var headers = {};
    var data = [];
    for(z in worksheet) {
        if(z[0] === '!') continue;
        //parse out the column, row, and value
        var tt = 0;
        for (var i = 0; i < z.length; i++) {
            if (!isNaN(z[i])) {
                tt = i;
                break;
            }
        };
        var col = z.substring(0,tt);
        var row = parseInt(z.substring(tt));
        var value = worksheet[z].v;

        //store header names
        if(row == 1 && value) {
            headers[col] = value;
            continue;
        }

        if(!data[row]) data[row]={};
        data[row][headers[col]] = value;
    }
    //drop those first two rows which are empty
    data.shift();
    data.shift();
    console.log(data);
});