Skip to content

Add Zarr3 support #25

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
sameeul merged 5 commits into from
Feb 5, 2026
Merged

Add Zarr3 support #25

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
943a1f0
Add Zarr3 support - WIP1
sameeul Feb 4, 2026
211df84
Add Zarr v3 write support to TsWriterCPP
sameeul Feb 5, 2026
094db42
Add Zarr v3 write tests
sameeul Feb 5, 2026
813542c
Refactor: Move FileType enum to utilities.h
sameeul Feb 5, 2026
73daa2f
Remove unnecessary tsreader.h include from utilities.cpp
sameeul Feb 5, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
13 changes: 10 additions & 3 deletions src/cpp/interface/interface.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -114,14 +114,21 @@ PYBIND11_MODULE(libbfiocpp, m) {

py::enum_<bfiocpp::FileType>(m, "FileType")
.value("OmeTiff", bfiocpp::FileType::OmeTiff)
.value("OmeZarr", bfiocpp::FileType::OmeZarr)
.value("OmeZarrV2", bfiocpp::FileType::OmeZarrV2)
.value("OmeZarrV3", bfiocpp::FileType::OmeZarrV3)
.export_values();

m.def("get_ome_xml", &bfiocpp::GetOmeXml);


// Writer class
py::class_<bfiocpp::TsWriterCPP, std::shared_ptr<bfiocpp::TsWriterCPP>>(m, "TsWriterCPP")
.def(py::init<const std::string&, const std::vector<std::int64_t>&, const std::vector<std::int64_t>&, const std::string&, const std::string&>())
py::class_<bfiocpp::TsWriterCPP, std::shared_ptr<bfiocpp::TsWriterCPP>>(m, "TsWriterCPP")
.def(py::init<const std::string&, const std::vector<std::int64_t>&, const std::vector<std::int64_t>&, const std::string&, const std::string&, bfiocpp::FileType>(),
py::arg("filename"),
py::arg("image_shape"),
py::arg("chunk_shape"),
py::arg("dtype"),
py::arg("dimension_order"),
py::arg("file_type") = bfiocpp::FileType::OmeZarrV2)
.def("write_image_data", &bfiocpp::TsWriterCPP::WriteImageData);
}
2 changes: 1 addition & 1 deletion src/cpp/reader/tsreader.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -21,7 +21,7 @@ TsReaderCPP::TsReaderCPP(const std::string& fname, FileType ft, const std::strin
if (ft == FileType::OmeTiff){
return GetOmeTiffSpecToRead(fname);
} else {
return GetZarrSpecToRead(fname);
return GetZarrSpecToRead(fname, ft);
}
}();

Expand Down
7 changes: 1 addition & 6 deletions src/cpp/reader/tsreader.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,12 +4,10 @@
#include <memory>
#include <vector>
#include <variant>
#include <iostream>
#include <tuple>
#include <optional>
#include <unordered_map>
#include "tensorstore/tensorstore.h"
#include "../utilities/sequence.h"
#include "../utilities/utilities.h"
using image_data = std::variant<std::vector<std::uint8_t>,
std::vector<std::uint16_t>,
std::vector<std::uint32_t>,
Expand All @@ -26,9 +24,6 @@ using iter_indicies = std::tuple<std::int64_t,std::int64_t,std::int64_t,std::int

namespace bfiocpp{


enum class FileType {OmeTiff, OmeZarr};

class TsReaderCPP{
public:
TsReaderCPP(const std::string& fname, FileType ft, const std::string& axes_list );
Expand Down
92 changes: 68 additions & 24 deletions src/cpp/utilities/utilities.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -25,8 +25,9 @@ tensorstore::Spec GetOmeTiffSpecToRead(const std::string& filename){
}).value();
}

tensorstore::Spec GetZarrSpecToRead(const std::string& filename){
return tensorstore::Spec::FromJson({{"driver", "zarr"},
tensorstore::Spec GetZarrSpecToRead(const std::string& filename, FileType ft){
std::string driver = (ft == FileType::OmeZarrV3) ? "zarr3" : "zarr";
return tensorstore::Spec::FromJson({{"driver", driver},
{"kvstore", {{"driver", "file"},
{"path", filename}}
}
Expand Down Expand Up @@ -152,28 +153,71 @@ std::string GetOmeXml(const std::string& file_path){
return OmeXmlInfo;
}

tensorstore::Spec GetZarrSpecToWrite( const std::string& filename,
const std::vector<std::int64_t>& image_shape,
const std::vector<std::int64_t>& chunk_shape,
const std::string& dtype){
std::string GetZarrV3DataType(uint16_t data_type_code) {
// Zarr v3 uses plain dtype names instead of encoded format like "<u2"
switch (data_type_code) {
case 1: return "uint8";
case 2: return "uint16";
case 4: return "uint32";
case 8: return "uint64";
case 16: return "int8";
case 32: return "int16";
case 64: return "int32";
case 128: return "int64";
case 256: return "float32";
case 512: return "float64";
default: return "uint16";
}
}

// valid values for dtype are subset of
// https://google.github.io/tensorstore/spec.html#json-dtype
return tensorstore::Spec::FromJson({{"driver", "zarr"},
{"kvstore", {{"driver", "file"},
{"path", filename}}
},
{"context", {
{"cache_pool", {{"total_bytes_limit", 1000000000}}},
{"data_copy_concurrency", {{"limit", std::thread::hardware_concurrency()}}},
{"file_io_concurrency", {{"limit", std::thread::hardware_concurrency()}}},
}},
{"metadata", {
{"zarr_format", 2},
{"shape", image_shape},
{"chunks", chunk_shape},
{"dtype", dtype},
},
}}).value();
tensorstore::Spec GetZarrSpecToWrite( const std::string& filename,
const std::vector<std::int64_t>& image_shape,
const std::vector<std::int64_t>& chunk_shape,
const std::string& dtype,
FileType ft){

if (ft == FileType::OmeZarrV3) {
// Zarr v3 spec
return tensorstore::Spec::FromJson({{"driver", "zarr3"},
{"kvstore", {{"driver", "file"},
{"path", filename}}
},
{"context", {
{"cache_pool", {{"total_bytes_limit", 1000000000}}},
{"data_copy_concurrency", {{"limit", std::thread::hardware_concurrency()}}},
{"file_io_concurrency", {{"limit", std::thread::hardware_concurrency()}}},
}},
{"metadata", {
{"shape", image_shape},
{"chunk_grid", {
{"name", "regular"},
{"configuration", {{"chunk_shape", chunk_shape}}}
}},
{"chunk_key_encoding", {{"name", "default"}}},
{"data_type", dtype},
{"codecs", {{{"name", "bytes"}, {"configuration", {{"endian", "little"}}}}}}
},
}}).value();
} else {
// Zarr v2 spec (existing)
// valid values for dtype are subset of
// https://google.github.io/tensorstore/spec.html#json-dtype
return tensorstore::Spec::FromJson({{"driver", "zarr"},
{"kvstore", {{"driver", "file"},
{"path", filename}}
},
{"context", {
{"cache_pool", {{"total_bytes_limit", 1000000000}}},
{"data_copy_concurrency", {{"limit", std::thread::hardware_concurrency()}}},
{"file_io_concurrency", {{"limit", std::thread::hardware_concurrency()}}},
}},
{"metadata", {
{"zarr_format", 2},
{"shape", image_shape},
{"chunks", chunk_shape},
{"dtype", dtype},
},
}}).value();
}
}
} // ns bfiocpp
12 changes: 8 additions & 4 deletions src/cpp/utilities/utilities.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -11,16 +11,20 @@

namespace bfiocpp {

enum class FileType {OmeTiff, OmeZarrV2, OmeZarrV3};

tensorstore::Spec GetOmeTiffSpecToRead(const std::string& filename);
tensorstore::Spec GetZarrSpecToRead(const std::string& filename);
tensorstore::Spec GetZarrSpecToRead(const std::string& filename, FileType ft);

uint16_t GetDataTypeCode (std::string_view type_name);
std::string GetEncodedType(uint16_t data_type_code);
std::string GetUTCString();
std::string GetOmeXml(const std::string& file_path);
std::tuple<std::optional<int>, std::optional<int>, std::optional<int>>ParseMultiscaleMetadata(const std::string& axes_list, int len);
tensorstore::Spec GetZarrSpecToWrite(const std::string& filename,
const std::vector<std::int64_t>& image_shape,
tensorstore::Spec GetZarrSpecToWrite(const std::string& filename,
const std::vector<std::int64_t>& image_shape,
const std::vector<std::int64_t>& chunk_shape,
const std::string& dtype);
const std::string& dtype,
FileType ft);
std::string GetZarrV3DataType(uint16_t data_type_code);
} // ns bfiocpp
11 changes: 8 additions & 3 deletions src/cpp/writer/tswriter.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -15,15 +15,20 @@ TsWriterCPP::TsWriterCPP(
const std::vector<std::int64_t>& image_shape,
const std::vector<std::int64_t>& chunk_shape,
const std::string& dtype_str,
const std::string& dimension_order
const std::string& dimension_order,
FileType file_type
): _filename(fname),
_image_shape(image_shape),
_chunk_shape(chunk_shape),
_dtype_code(GetDataTypeCode(dtype_str)) {


// Use appropriate dtype encoding based on file type
std::string encoded_dtype = (file_type == FileType::OmeZarrV3)
? GetZarrV3DataType(_dtype_code)
: GetEncodedType(_dtype_code);

TENSORSTORE_CHECK_OK_AND_ASSIGN(_source, tensorstore::Open(
GetZarrSpecToWrite(_filename, _image_shape, _chunk_shape, GetEncodedType(_dtype_code)),
GetZarrSpecToWrite(_filename, _image_shape, _chunk_shape, encoded_dtype, file_type),
tensorstore::OpenMode::create |
tensorstore::OpenMode::delete_existing,
tensorstore::ReadWriteMode::write).result()
Expand Down
9 changes: 6 additions & 3 deletions src/cpp/writer/tswriter.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,8 +2,10 @@

#include <string>
#include <vector>
#include <optional>
#include "tensorstore/tensorstore.h"
#include "../utilities/sequence.h"
#include "../utilities/utilities.h"
#include <pybind11/numpy.h>

namespace py = pybind11;
Expand All @@ -13,11 +15,12 @@ namespace bfiocpp{
class TsWriterCPP{
public:
TsWriterCPP (
const std::string& fname,
const std::vector<std::int64_t>& image_shape,
const std::string& fname,
const std::vector<std::int64_t>& image_shape,
const std::vector<std::int64_t>& chunk_shape,
const std::string& dtype_str,
const std::string& dimension_order
const std::string& dimension_order,
FileType file_type = FileType::OmeZarrV2
);

void WriteImageData (
Expand Down
10 changes: 8 additions & 2 deletions src/python/bfiocpp/tswriter.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
import numpy as np
from typing import Optional
from .libbfiocpp import TsWriterCPP, Seq
from .libbfiocpp import TsWriterCPP, Seq, FileType


class TSWriter:
Expand All @@ -12,14 +12,20 @@ def __init__(
chunk_shape: list,
dtype: np.dtype,
dimension_order: str,
file_type: FileType = FileType.OmeZarrV2,
):
"""Initialize tensorstore Zarr writer

file_name: Path to write file to
image_shape: Shape of the image [T, C, Z, Y, X]
chunk_shape: Shape of chunks [T, C, Z, Y, X]
dtype: Data type of the image
dimension_order: Order of dimensions (e.g., "TCZYX")
file_type: FileType.OmeZarrV2 (default) or FileType.OmeZarrV3
"""

self._image_writer: TsWriterCPP = TsWriterCPP(
file_name, image_shape, chunk_shape, str(dtype), dimension_order
file_name, image_shape, chunk_shape, str(dtype), dimension_order, file_type
)

def write_image_data(
Expand Down
4 changes: 2 additions & 2 deletions tests/test_read.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -211,7 +211,7 @@ def test_read_zarr_2d_slice(self):
"""test_read_zarr_2d_slice - Read tiff using TSReader"""
br = TSReader(
str(TEST_DIR.joinpath("5025551.zarr/0")),
FileType.OmeZarr,
FileType.OmeZarrV2,
"",
)
assert br._X == 2702
Expand All @@ -235,7 +235,7 @@ def test_read_zarr_4d_slice(self):
"""test_read_zarr_4d_slice - Read tiff using TSReader"""
br = TSReader(
str(TEST_DIR.joinpath("5025551.zarr/0")),
FileType.OmeZarr,
FileType.OmeZarrV2,
"",
)

Expand Down
Loading
Loading