This repository contains an early preview of a Java implementation of the Zarr specification. It is intended for collecting feedback from the community and not for use. The API is subject to changes.
Refer to JZarr for a stable implementation of Zarr version 2.
import dev.zarr.zarrjava.store.FilesystemStore;
import dev.zarr.zarrjava.store.HttpStore;
import dev.zarr.zarrjava.v3.Array;
import dev.zarr.zarrjava.v3.DataType;
import dev.zarr.zarrjava.v3.Group;
Group hierarchy = Group.open(
new HttpStore("https://static.webknossos.org/data/zarr_v3")
.resolve("l4_sample")
);
Array array = hierarchy.get("color").get("1");
ucar.ma2.Array outArray = array.read(
new long[]{0, 3073, 3073, 513}, // offset
new int[]{1, 64, 64, 64} // shape
);
Array array = Array.create(
new FilesystemStore("/path/to/zarr").resolve("array"),
Array.metadataBuilder()
.withShape(1, 4096, 4096, 1536)
.withDataType(DataType.UINT32)
.withChunkShape(1, 1024, 1024, 1024)
.withFillValue(0)
.withCodecs(c -> c.withSharding(new int[]{1, 32, 32, 32}, c1 -> c1.withBlosc()))
.build();
);
array.write(
new long[]{0, 0, 0, 0}, // offset
ucar.ma2.Array.factory(ucar.ma2.DataType.UINT, new int[]{1, 1024, 1024, 1024})
);
To be able to run the tests locally, make sure to have python3.11
installed.
Also, you need to set up a venv for zarrita at the root of the project:
python3.11 -m venv venv_zarrita
.
Then install zarrita there with venv_zarrita/Scripts/pip install zarrita
for Windows and venv_zarrita/bin/pip install zarrita
for Linux.
Furthermore, you will need the l4_sample
test data:
curl https://static.webknossos.org/data/zarr_v3/l4_sample.zip -o testdata/l4_sample.zip && cd testdata && unzip l4_sample.zip