• load thumbnails as in https://github.com/will-moore/ome-ngff-validator rather than using IDR URLs
• possibly load table data dynamically from the Zarr itself
• create a new template repo under github.com/ome
• strip out IDR specific columns like "View in IDR"
• slightly unrelated:
  • create an action button to solidify copy/open/validate/...

This will mostly be CC-BY but a few studies are either under CC0 or under more restrictive CC licenses.
Adding the license to the table should make the conditions of re-usage unambiguous

e.g.

aws s3 --no-sign-request --endpoint-url https://uk1s3.embassy.ebi.ac.uk cp --recursive s3://idr/zarr/v0.1/1884807.zarr .

see: https://imagesc.zulipchat.com/#narrow/stream/329366-.5B2022-09.5D-Fiji.2BNGFF-Prague-hackathon/topic/sample.20download

cc: @will-moore

Hi

I'd like to access https://uk1s3.embassy.ebi.ac.uk/idr/zarr/v0.4/idr0052A/5514375.zarr, but only get a NoSuchKey error:

How can I access this data for local viewing/processing?

Thanks

This could likely be pulled from the generator metadata, but ordering the file by date added would help users to find the most recent additions. (We might also be able to auto-generate a changelog from that)

To download via aws, you need e.g. --endpoint-url:

$ aws s3 sync --no-sign-request --endpoint-url https://uk1s3.embassy.ebi.ac.uk s3://idr/zarr/v0.4/idr0062A/6001240.zarr .

It would be useful to have this info on the samples page itself. Maybe even an option to copy the above command for a given sample?

In the current version of the published catalog, thumbnails are broken for all HCS OME-NGFF datasets.

The current table generation logic extracts the ID from the S3 URL to construct a webclient/render_thumbnail URL. This relies on the assumption that the URL ends with <image_id>.zarr and thus applies exclusively to individually converted images.

The simplest solution would be to capture either ID of a representative image for the dataset or the full URL to a representative thumbnail as an extra column.

Possibly, this points out the fact that images and plates should be maintained in separate CSV files and/or represented as separate pages in the generated page.