Convenience functions and exposed metadata

[maximedion2]

Hey there! This is a really cool library! I was considering using it in a project I'm working on, but I have a couple questions, I skimmed through the docs but I'm not sure about 2 functionalities that would be quite useful on my end.

• are there "convenience" functions on top of the manual chunk writes? What I mean is e.g. once a store is set up, with chunk and array shapes, codecs, etc... you'd pass such a function an array (maybe nested vectors) and it handles compression, chunking, sharding, etc... and just writes to the appropriate files in the store?
• I see that the metadata can be returned as a json object, but is there any way to access the various things encoded in the metadata as actual objects (maybe enums, vectors for shapes, etc...) that can be checked in code, used in branching statements, etc...? I assume those objects exist inside the code, but they might not be publicly exposed?

[LDeakin]

I think you might be looking for the following methods:

• Array::store_array_subset
• Array::store_array_subset_elements
• Array::store_array_subset_ndarray

These can store an arbitrary subset of an array, you don't have to think about chunks. They (and other Array::store_* methods, excluding store_encoded_chunk) take care of everything for you, compression, sharding, etc.

You can grab individual components of the metadata: see https://docs.rs/zarrs/latest/zarrs/array/struct.Array.html#immutable-array-metadata--properties and below. EDIT: also Array::metadata() returns an enum ArrayMetadata that can represent V2/V3 metadata. The variants are completely pub.

[maximedion2]

Great, that seems like what I need, thanks!