You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
It would also be nice (but definitely not a must-have for scipy) to see some sort of information about the contents of the zarr dataset. Some nice inspiration for that:
I believe this is a good short term solution to ship a much needed feature quickly. This is the same approach used in https://catalog.pangeo.io/.
However, in the long term, I don't think this architecture is optimal for the following reasons:
Opening the dataset with python on the backend is expensive and ultimately unnecessary, since the Zarr dataset is well described by a compact json file.
Xarray's html repr is unnecessarily large, with lots of inline custom CSS. That will prevent us from customizing the look and feel of the display.
Instead, I think we should aim to develop a native javascript solution to generating the dataset preview. Ideally an npm package providing react components that we can simply drop in instead of the API call. Maybe something like that already exists? Some inspiration for how this would look can be found at https://idr.github.io/ome-ngff-samples/.
If we go with this solution, we should open a placeholder issue to keep track of that for a future refactor.
idr.github.io/ome-ngff-samples
ome.github.io/ome-ngff-validator/?source=https://uk1s3.embassy.ebi.ac.uk/idr/zarr/v0.4/idr0076A/10501752.zarr
Originally posted by @rabernat in #93 (comment)
The text was updated successfully, but these errors were encountered: