Streams

XLSX.stream turns a worksheet into a Node Readable stream. Use it to pipe a big sheet to a file or a network response without building the whole output string first.

Function	Stream output
stream.to_csv(ws, opts)	CSV text; options as for utils.sheet_to_csv
stream.to_html(ws, opts)	an HTML document
stream.to_json(ws, opts)	object mode: one row object per chunk; options as for utils.sheet_to_json
stream.set_readable(Readable)	give the library a Readable constructor (Node does this for you)

const XLSX = require("@agent-sheet/wasm");
const fs = require("fs");
const { pipeline } = require("node:stream/promises");

async function main() {
  const ws = XLSX.utils.aoa_to_sheet([["a", "b"], [1, 2], [3, 4]]);
  await pipeline(XLSX.stream.to_csv(ws), fs.createWriteStream("out.csv"));
  console.log(JSON.stringify(fs.readFileSync("out.csv", "utf8")));

  const rows = [];
  for await (const row of XLSX.stream.to_json(ws)) rows.push(row);
  console.log(rows);
}
main().catch((error) => { console.error(error); process.exitCode = 1; });

Output

"a,b\n1,2\n3,4"
[ { a: 1, b: 2 }, { a: 3, b: 4 } ]

The streams read the worksheet object as they run. The library does not read a file in pieces. To work with a huge file, read it with readFile first (see Large files and performance), then stream the output.

Note

The CSV stream starts with a byte order mark, like the CSV that write produces.

In the browser

fs and stream are not available in browsers. The stream functions need a Readable class. In a bundler, either do not use them or call XLSX.stream.set_readable(Readable) with a class from a stream package that you bundle.

Empty sheets

stream.to_html without !ref infers a range from existing cells; an empty worksheet produces a table with a blank A1 cell. The original raises an unhandled stream error that can terminate the process. See Known differences.