Convert multiple files

File converter for various types, tested on linux and windows.

Requirements

• Node.js >= 18
• Linux or Windows (a LibreOffice bundle is downloaded on install). On other platforms document conversion is unavailable, but convertWordFileToHTML and convertToBase64 still work.

Supported file types

Accepted file formats	Output
doc	docx
doc	pdf
doc	odt
docx	doc
docx	odt
docx	pdf
odt	doc
odt	docx
odt	pdf
docx	html
all	base64

Install

npm i convert-multiple-files

Convert Word files

import { convertWordFiles } from 'convert-multiple-files';
import * as path from 'path';

async test() {
  // Return promise => convertWordFiles(path of the file to be converted, convertTo, outputDir)
  const pathOutput = await convertWordFiles(path.resolve(__dirname, 'teste.doc'), 'pdf', path.resolve(__dirname));
  console.log(pathOutput);
}

test();

The returned path is verified to exist; if LibreOffice fails to produce it, the promise rejects instead of returning a path to a missing file.

An optional fourth argument controls the conversion timeout (default 60000 ms):

await convertWordFiles(input, 'pdf', outputDir, { timeout: 120000 });

Conversions run against a private LibreOffice profile cloned from a shared template kept in the system temp directory. Set CMF_LO_PROFILE_DIR to relocate it if the temp directory is not writable.

Convert many files at once

LibreOffice startup dominates conversion time, so converting a list in one call is far faster than calling convertWordFiles in a loop.

import { convertWordFilesBatch } from 'convert-multiple-files';

async test() {
  // Return promise => convertWordFilesBatch(paths of the files, convertTo, outputDir)
  const pathsOutput = await convertWordFilesBatch(['a.docx', 'b.docx', 'c.docx'], 'pdf', outputDir);
  console.log(pathsOutput);
}

test();

Convert DOCX to HTML

import { convertWordFileToHTML } from 'convert-multiple-files';
import * as path from 'path';

async test() {
  // Return promise => convertWordFileToHTML(path of the file to be converted, outputDir, outputPrefix)
  const infoOutput = await convertWordFileToHTML(path.resolve(__dirname, 'file2.docx'), path.resolve(__dirname), 'filehtml-151412');
  console.log(infoOutput);
}

test();

Resolves undefined (and writes no file) when the document produces no content.

Convert all files to Base64

import { convertToBase64 } from 'convert-multiple-files';
import * as path from 'path';

async test() {
  // Return promise => convertToBase64(path of the file to be converted)
  const base64 = await convertToBase64(path.resolve(__dirname, 'file2.docx'));
  console.log(base64);
}

test();

Upgrading to 2.0.0

The three original functions keep their names and positional arguments, so existing calls continue to work. What changed:

• Node >= 18 is now required, and the DOCX-to-HTML backend moved from the unmaintained mammoth-style fork to mammoth, which can render slightly different HTML for the same document.
• Conversion no longer blocks the event loop. It previously ran through execSync, which froze the whole process for the duration of the conversion.
• Arguments are passed to LibreOffice as an argument array instead of being interpolated into a shell command string.
• Each conversion runs with a private LibreOffice profile, so concurrent conversions no longer collide.
• Unsupported platforms now raise a clear error instead of silently returning a path to a file that was never created.
• The output format is validated strictly: values like xdoc used to slip through and now raise Format to be converted not accepted.
• Errors keep their original messages and now carry the underlying failure in error.cause.