xd-makeFiles

Native file format generators for Node.js. Produces .zip, .docx, .xlsx, and .pdf files using only built-in modules — no external dependencies.

Why

	xd-makeFiles	docx (npm)	exceljs	pdfkit	archiver
Install size	0 B	~800 KB	~2.1 MB	~1.5 MB	~600 KB
Dependencies	0	3	8	12	7
Lines of code	~1000	~12K	~45K	~30K	~8K
Formats	zip, docx, xlsx, pdf	docx	xlsx	pdf	zip/tar

Four modules, ~1000 lines total, zero node_modules.

Architecture

xd-makeFiles/
├── index.js         Unified API — makeFile('format', opts) router
├── xd-zip/          ZIP archive engine (foundation layer)
│   └── zip.js       CRC32, ZIP64, store method — used by docx & xlsx
├── xd-docx/         Word document generator
│   └── docx.js      OOXML WordprocessingML → ZIP container
├── xd-xls/          Excel spreadsheet generator
│   └── xls.js       OOXML SpreadsheetML → ZIP container
└── xd-pdf/          PDF generator from Markdown
    └── pdf.js       Markdown parser → layout engine → PDF 1.4 binary

Dependency graph:

xd-docx ──→ xd-zip ←── xd-xls
xd-pdf (standalone, uses node:zlib)

Quick Start

ZIP — create archives

import { createZip, generateZip } from './xd-zip/zip.js';

const zip = createZip();
zip.addFile('hello.txt', 'Hello World');
zip.addFile('data.json', JSON.stringify({ version: 1 }));
fs.writeFileSync('output.zip', zip.build());

DOCX — generate Word documents

import { generateDocx } from './xd-docx/docx.js';

generateDocx({
 outputPath: 'report.docx',
 title: 'Q1 Report',
 sections: [
  { type: 'heading', text: 'Summary', level: 2 },
  { type: 'text', text: 'Revenue increased 23% year-over-year.' },
  { type: 'table', rows: [['Metric', 'Value'], ['Revenue', '$1.2M'], ['Users', '45K']] },
  { type: 'list', items: ['Expand to EU', 'Launch mobile app'], ordered: true }
 ]
});

XLSX — generate Excel spreadsheets

import { generateXlsx } from './xd-xls/xls.js';

generateXlsx({
 outputPath: 'data.xlsx',
 sheets: [{
  name: 'Sales',
  data: [
   ['Product', 'Q1', 'Q2', 'Total'],
   ['Widget A', 150, 200, '=B2+C2'],
   ['Widget B', 300, 180, '=B3+C3']
  ]
 }]
});

PDF — Markdown to PDF

import { generatePdfFromMarkdown } from './xd-pdf/pdf.js';

const markdown = `# Project Status
## Overview
Development is **on track** with *minor* delays in testing.

| Task       | Status    |
|------------|-----------|
| Backend    | Complete  |
| Frontend   | In Review |

> Next milestone: v2.0 release
`;

generatePdfFromMarkdown(markdown, 'status.pdf', { title: 'Project Status' });

Single Import

Use index.js for unified access to all generators:

import makeFile from './index.js';

// Route by format string
makeFile('docx', {
 outputPath: 'report.docx',
 title: 'Q1 Report',
 sections: [{ type: 'text', text: 'Revenue grew 23%.' }]
});

makeFile('xlsx', {
 outputPath: 'data.xlsx',
 data: [['Name', 'Score'], ['Alice', 95], ['Bob', 87]]
});

makeFile('pdf', {
 outputPath: 'notes.pdf',
 markdown: '# Meeting Notes\n\nAction items:\n- Ship v2\n- Hire 3 devs'
});

makeFile('zip', {
 outputPath: 'bundle.zip',
 files: [{ name: 'readme.txt', content: 'Hello' }]
});

Or import individual generators directly:

import { generateDocx, generateXlsx, generatePdfFromMarkdown, createZip } from './index.js';

Format aliases: doc → docx, xls/excel → xlsx. Auto-detection from outputPath extension when format is omitted.

Module Details

Each module has its own README with full API reference and internals:

Module	Output	LOC	Node.js deps	Docs
xd-zip	.zip	~200	node:fs	README
xd-docx	.docx	~185	node:fs + xd-zip	README
xd-xls	.xlsx	~127	node:fs + xd-zip	README
xd-pdf	.pdf	~500	node:fs, node:zlib	README

Requirements

Node.js >= 18 (uses Buffer.writeBigUInt64LE, ESM imports)

Design Principles

Zero dependencies — every byte of output is constructed from raw binary operations and XML string assembly. No npm packages, no transpilers, no build step.

Object literals over classes — all internal structures use factory functions and Object.create() prototypal inheritance. No class keyword anywhere.

Format-compliant output — generated files open correctly in Microsoft Office, LibreOffice, macOS Preview, and all major PDF readers. OOXML structures follow the ECMA-376 specification; PDF output conforms to PDF 1.4.

Known Limitations

• PDF: Uses WinAnsi encoding with Type1 fonts — non-Latin characters (including Polish diacritics) are transliterated to ASCII approximations. This is a deliberate tradeoff: embedding Unicode font subsets would add ~500 LOC and require bundling .ttf files.
• XLSX: No chart generation, conditional formatting, or pivot tables. Supports the data layer: cell types, formulas, dates, multi-sheet.
• DOCX: No headers/footers, table of contents, or page numbers. Covers the content layer: headings, paragraphs, tables, lists, images.
• ZIP: Store method only (no compression). Files within OOXML containers are XML text, so the size overhead is minimal for typical document sizes.

License

MIT — Jakub Śledzikowski