syncfusion-javascript-docx

Syncfusion JavaScript Word Library (@syncfusion/ej2-docx)

Non-UI library for creating, reading, editing and saving .docx files in Node.js and browsers. No Microsoft Word needed. Unsupported content (unknown XML, parts, relationships) is preserved on open → save round-trips.

Install

npm install @syncfusion/ej2-docx

Import from the package root: import { WordDocument } from '@syncfusion/ej2-docx';.

Navigation guide

Read only the reference needed for the task:

Task	Read
Create / open / save, base64, browser vs Node I/O, errors	references/getting-started.md
Paragraphs, text, character & paragraph formatting, lists	references/text-and-formatting.md
Tables, cells, merge, table formatting	references/tables.md
Images, hyperlinks, fields, content controls	references/inline-content.md
Sections, page setup, headers/footers, page numbers	references/sections-headers-footers.md
Limits, gotchas, internal/unsupported APIs	references/troubleshooting.md

Quick start

import { WordDocument } from '@syncfusion/ej2-docx';

const document = WordDocument.create();                 // new blank document
const section = document.sections[0];                   // add content on the section body

const paragraph = section.body.paragraphs[0];           // default empty paragraph
paragraph.appendText('Bold intro. ', { bold: true });
paragraph.appendText('Normal text.');

const title = section.body.appendParagraph({ text: 'Quarterly Report' });
title.items; // paragraph items (text ranges, tabs, ...)

const table = section.body.appendTable();               // 1x1 table
table.resetCells(2, 2);                                 // resize to 2 rows x 2 cols
table.rows[0].cells[0].text = 'Name';
table.rows[0].cells[1].text = 'Qty';

const bytes: Uint8Array = doc.saveSync();               // OPC zip bytes
// Node:   fs.writeFileSync('out.docx', bytes)
// Browser: await doc.save('out.docx')  -> triggers a download

Open an existing file:

import { readFileSync } from 'node:fs';
const doc = WordDocument.open(new Uint8Array(readFileSync('in.docx')));
console.log([...doc.sections[0].body.paragraphs].map(p => p.text));

Core object model

• WordDocument → sections[] → Section → body (Body), pageSetup, headersFooters.
• Body / table cells / headers / footers hold ordered blocks (items): Paragraph, Table, block content controls. body.paragraphs is a typed view.
• Paragraph → items (TextRange, Tab, fields, hyperlinks, pictures, ...), paragraphFormat, listFormat, text (read-only plain text).
• TextRange.characterFormat holds run formatting (bold, italic, fontName, fontSize, textColor, ...).

Rules of thumb

1. File paths are not read by open; pass bytes (Uint8Array/ArrayBuffer), a File/Blob via openAsync, or base64 with { format: 'base64' }.
2. Node file writing is done by you (fs.writeFileSync) using saveSync() or await save(). save('name.docx') is a browser download only.
3. Add paragraphs and tables on the section body (document.sections[0].body.appendParagraph / appendTable), not on WordDocument. Objects must be attached to a document to be mutated (detached ones throw DetachedObjectError).
4. Verify any API against the .d.ts of the installed version before using symbols not shown in these references; some members are @internal and are trimmed from published typings.
5. Do not execute or rely on VBA/OLE/ActiveX content; the library only preserves it.