ngff icon indicating copy to clipboard operation
ngff copied to clipboard

Published 20 hours ago • verified publisher icon ome

Constrain dtypes & shape for multiscale arrays

Open joshmoore opened this issue 3 years ago • 4 comments

see: https://github.com/tischi/i2k-2020-s3-zarr-workshop/issues/9#issuecomment-732369669

The current spec allows for each resolution level to be a different dtype. This should be restricted with a clause that they "MUST be the same".

joshmoore avatar Nov 23 '20 19:11 joshmoore

I think this is still relevant

constantinpape avatar Feb 02 '22 22:02 constantinpape

I was wondering today whether the number of dims len(shape) MUST be the same for each resolution level? I assume this is the case but I didn't see anything in the spec?

will-moore avatar Feb 02 '22 22:02 will-moore

:+1: for both constraints (title of issue updated)

joshmoore avatar Feb 03 '22 08:02 joshmoore

@constantinpape Please go ahead and open a PR against the spec to state that dytpes must be the same for each Dataset resolution. You might get conflicts with other PRs (e.g. #138 changes quite a lot) but at least this will push this change forward and check there are no objections etc.

will-moore avatar Oct 21 '22 12:10 will-moore