export-lib

Export tabular data (header + data) to CSV, JSON, XLSX, or PDF through one unified API.

const { exportData } = require('export-lib');

const buffer = await exportData({
  header: ['Name', 'Role', 'Years'],
  data: [
    ['Ada Lovelace', 'Mathematician', 3],
    ['Grace Hopper', 'Computer Scientist', 44],
  ],
  type: 'pdf', // 'csv' | 'json' | 'xlsx' | 'pdf'
});

// buffer is a Buffer — write it to disk, stream it in an HTTP response,
// or turn it into a Blob in the browser. This library doesn't force any I/O.

Why

Most projects end up writing the same "export as CSV / export as PDF" logic over and over, each format pulling in a different library with a different API. This wraps them behind one consistent interface so callers don't need to care what's underneath.

Install

npm install tabulate-export

Usage

Write directly to a file (Node)

const { exportToFile } = require('export-lib');

await exportToFile(
  { header: ['Name', 'Age'], data: [['Ada', 28]], type: 'xlsx' },
  './report.xlsx'
);

Rows as objects instead of arrays

await exportData({
  header: ['Name', 'Age'],
  data: [{ Name: 'Ada', Age: 28 }], // keyed by header name — order doesn't matter
  type: 'csv',
});

Serve as an HTTP download (Express example)

app.get('/export.pdf', async (req, res) => {
  const buffer = await exportData({ header, data, type: 'pdf' });
  res.set('Content-Type', 'application/pdf');
  res.set('Content-Disposition', 'attachment; filename="report.pdf"');
  res.send(buffer);
});

API

exportData({ header, data, type, options }) => Promise<Buffer>

| Param | Type | Description | |---------|--------------------------|----------------------------------------------------------------------| | header | string[] | Column names, in order. | | data | Array<Array \| Object> | Rows. Arrays are positional; objects are matched by header name. | | type | string | 'csv', 'json', 'xlsx', 'pdf' (leading . is optional). | | options | object | Format-specific, see below. |

exportToFile(params, filePath) => Promise<string>

Same as exportData, but writes the result straight to disk and returns the path. Node-only.

registerFormat(name, adapter)

Add support for a new format without touching library internals:

const { registerFormat } = require('export-lib');

registerFormat('txt', {
  generate: async (header, rows, options) =>
    Buffer.from(rows.map((r) => r.join('\t')).join('\n')),
});

Any adapter just needs to implement generate(header, rows, options) => Buffer | Promise<Buffer>.

Format-specific options

| Type | Option | Default | Notes | |------|---------------|-----------|-----------------------------------------| | csv | delimiter | , | Change to ; for European locale CSVs | | json | pretty | true | Set false for minified output | | xlsx | sheetName | Sheet1 | | | pdf | title | none | Optional heading above the table | | pdf | pageSize | A4 | Any pdfkit page size |

Architecture

Each format is an isolated adapter behind a shared interface (generate(header, rows, options) => Buffer). The core (exportData) only validates input and dispatches to the right adapter via a registry — it has no format-specific logic itself. This keeps formats swappable and makes it trivial to add new ones (see registerFormat above) without touching existing code.

src/
├── index.js          # public API
├── registry.js        # format name -> adapter lookup
├── adapters/
│   ├── csv.js          # hand-rolled, RFC-4180-style escaping
│   ├── json.js          # rows -> array of keyed objects
│   ├── xlsx.js          # wraps exceljs
│   └── pdf.js            # wraps pdfkit, hand-laid-out table w/ pagination
└── utils/
    └── validate.js     # shape validation + array/object row normalization

Development

npm install
npm test

License

MIT