A Java library for reading OME-Zarr datasets.
This library is currently incomplete with functionality implemented as required by consumers.
JOmeZarr handles the additional OME-Zarr specific .zgroup
layer that specifies the set of underlying Zarr arrays. The actual reading of Zarr data is done by one of the existing Zarr libraries. JOmeZarr currently integrates JZarr, however any library that provides basic chunked reading for Zarr files could be used.
The default compression for many Zarr libraries is Blosc. An implementation of this compressor must be on the library path for any executable using JOmeZarr. On some platforms it is installed by default or can be added with a package manager. For others, it may be necessary to compile or download prebuilt binaries and add the location to the library path.
Several classes and methods related to Zarr specifically will appear to be minimal wrappers around JZarr. This is to insulate callers from any changes to the underlying Zarr implementation.
Some test use the OME-Zarr prototypes samples (v0.4). A TestSampleDir
Java property must be used when running test with a valid location for those tests to be included.