DIF to JSONDIF

Data Interchange Format.

A plain-text interchange format from the VisiCalc era, still offered as a save format by Excel and LibreOffice Calc and found in archived data. TabularJS reads the header, the optional label section and the data values.

A DIF file starts with a header (TABLE, VECTORS, TUPLES) followed by a DATA section in which every cell takes two lines: a type and value pair such as 0,709000 for a number, then a second line, which holds the text for a string cell. Each row starts with -1,0 and BOT, and the file ends with -1,0 and EOD.

TabularJS parses that layout and returns a single worksheet. DIF stores values only, so there are no formulas, styles or sheet names to extract.

Feature support

What TabularJS extracts from DIF files

• One worksheet. Numeric cells (type 0) become numbers, string cells (type 1) become strings, and short rows are padded with empty strings.
• Column titles from a LABEL section when the file has one; otherwise A, B, C...
• The title string of the TABLE header (Excel writes "EXCEL") in meta.version, quotes included, and lines of a COMMENT section in meta.comments.
• The text encoding is detected by trying common codepages (cp850, cp437, Latin-1, UTF-8, UTF-16) unless you pass encoding.

Convert DIF to JSON in Node.js

import tabularjs from 'tabularjs';

const result = await tabularjs('cities.dif');
const sheet = result.worksheets[0];

console.log(sheet.meta);    // { version: '"EXCEL"' }
console.log(sheet.data[1]); // [ 'Oslo', 709000 ]

// Skip the detection when you know the encoding
const latin = await tabularjs('cities.dif', { encoding: 'latin1' });

Read DIF files in the browser

import tabularjs from 'tabularjs';

const input = document.querySelector('input[type=file]');

input.addEventListener('change', async () => {
  // In the browser, pass the encoding: detection works best in Node.js
  const result = await tabularjs(input.files[0], { encoding: 'windows-1252' });
  console.log(result.worksheets[0].data);
});

JSON output

{
  "worksheets": [
    {
      "data": [
        ["City", "Population"],
        ["Oslo", 709000],
        ["Bergen", 291000]
      ],
      "columns": [{ "title": "A", "width": 100 }, { "title": "B", "width": 100 }],
      "meta": { "version": "\"EXCEL\"" }
    }
  ]
}

Shape of the result for a small DIF file saved by Excel.

Options

Limitations

• No formulas: DIF stores calculated values only.
• Encoding detection is a heuristic. It uses the optional chardet and iconv-lite packages, which are available in Node.js only; in the browser, pass encoding when the file is not UTF-8 or Latin-1.

Frequently asked questions

Why TabularJS for DIF?

• No external spreadsheet parser dependencies: faster installs, smaller bundles.
• One API in Node.js and modern browsers.
• Output is shape-compatible withJspreadsheet: render parsed data in a live workbook with one call.