Reading a FITS file

The Node path uses NodeFileReader. All the readers, the parser, and the image decoder come from the same entry, @fits-js/core.

import { NodeFileReader, openFits, readImage } from "@fits-js/core";

const reader = await NodeFileReader.open("data/image.fits");
try {
  const { hdus, warnings } = await openFits(reader);
  for (const w of warnings) console.warn(w);

  const primary = hdus[0];
  const { data, shape, bitpix } = await readImage(primary, reader);
  console.log(`${shape.join("x")} BITPIX=${bitpix}`, data.length);
} finally {
  await reader.close();
}

NodeFileReader.open accepts a string path or a file: URL. The handle stays open until close() runs, so wrap a try/finally around any work that throws.

Reading a small cutout

readImage accepts an optional region in FITS axis order. The reader pulls only the bytes that cutout covers.

const region = { start: [100, 200], shape: [16, 16] };
const { data, shape } = await readImage(primary, reader, { region });
// data.length === 16 * 16, shape === [16, 16]

region.start and region.shape have one entry per axis. NAXIS1 first, so on a 2D image start[0] is the X origin and start[1] is the Y origin.

Choosing an extension

The HDU at hdus[0] is the primary HDU. Extensions follow in file order. findHdu looks them up by EXTNAME (and optionally EXTVER).

import { findHdu } from "@fits-js/core";

const sci = findHdu(hdus, "SCI");
if (sci?.type === "image") {
  const { data } = await readImage(sci, reader);
}