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
From a discussion with @katamartin , we realized it could be very nice to build a tiny utility (possibly in a standalone package) that validates the structure of a zarr ndpyramid from js.
Key things it would do
try to load the group via zarr-js
extract pyramid metadata
extract coordinates and dimensions
validate everything for internal consistency
raise any issues
Two key use cases
short term, via a CLI or tiny standalone web app, let us inspect all dataset properties before trying to incorporate a dataset into an app (this will be especially useful when it comes to coordinates, and just generally helpful for debugging data formatting issues on the js side)
long term, ideally use the same underlying package to do runtime validation and throw much more helpful error messages when datasets are not formatted correctly
I think this is a great idea. Since we're relying on the array metadata and multiscales spec in consolidated metadata, it may make sense to just write a spec validator for the .zmetadata json. If the spec validates, then a set of functional tests could be run.
FWIW, this would also be a useful test approach in ndpyramid as well. The combination of a validation utility here and a set of unit tests in ndpyramid would be quite a nice combination.
From a discussion with @katamartin , we realized it could be very nice to build a tiny utility (possibly in a standalone package) that validates the structure of a zarr ndpyramid from js.
Key things it would do
zarr-js
Two key use cases
cc @jhamman
The text was updated successfully, but these errors were encountered: