feat: check for "units" field

[lubianat]

Why?

The "unit" field is optional, so it is not validated.

This checks if the "units" field is used, which is almost certainly a mistake.

Noted on IDR/ome-ngff-samples#36 in a sample. This may be happening silently to other data out there.

tested locally with

{
  "attributes": {
    "ome": {
      "version": "0.5",
      "multiscales": [
...
          "axes": [
            {
              "name": "c",
              "type": "channel"
            },
            {
              "name": "z",
              "type": "space"
            },
            {
              "name": "y",
              "type": "space"
            },
            {
              "name": "x",
              "type": "space",
              "units": "micrometer"
            }
          ]
        }
      ],
      ...
    }
  },
  "zarr_format": 3,
  "

and for

• Validate unit, e.g. "micrometer" not "µm" #52

[netlify]

✅ Deploy Preview for ome-ngff-validator ready!

Name	Link
🔨 Latest commit	20b81ad
🔍 Latest deploy log	https://app.netlify.com/projects/ome-ngff-validator/deploys/6a286c1e12290b000971f690
😎 Deploy Preview	https://deploy-preview-76--ome-ngff-validator.netlify.app
📱 Preview on mobile	Toggle QR Code... Use your smartphone camera to open QR code link.

---

To edit notification comments on pull requests, go to your Netlify project configuration.

[will-moore]

I was thinking myself that this could be a useful check to add. 👍

However, I realised in checking v0.6dev4 data that we miss all the axes checks because it's only looking for multiscale.axes above at const { axes, datasets } = multiscale;

If there's no axes there, then we should check at least the first coordinateSystem axes. E.g. axes = multiscale.coordinateSystems?.[0]?.axes.

[lubianat]

added the checks @will-moore ; for disclosure: I used the brand new Fable 5 model for support, but reviewed/edited line by line. It does look like it did a fine job.

I am still to get a 0.6dev4 dataset and try out

[will-moore]

Code looks good.

You could consider adding an invalid example to this repo, to test the error display.
E.g. like https://ome.github.io/ome-ngff-validator/?source=https://raw.githubusercontent.com/ome/ome-ngff-validator/refs/heads/main/test_samples/invalid/invalid_version.zarr/