[This comment](https://github.com/ilan-gold/zarrs-python/pull/87#discussion_r1959293178) by @flying-sheep made me realise we are missing a potentially large perf optimisation for `zarrs-python`. We should reuse partial decoders for batches of `chunk_descriptions` that refer to the same chunk in `retrieve_chunks_and_apply_index`. https://github.com/ilan-gold/zarrs-python/blob/53af191c7ca5215c31221520da1ee1c634852b78/src/lib.rs#L312-L329 The `ChunkRepresentation` can be reused for each batch too. Relevant: https://book.zarrs.dev/arrays/array_read.html#reading-multiple-chunk-subsets
Activity