Merge pull request #1 from joshmoore/writer

Looks good, let me merge this and add tests and such

ome_zarr/data.py

@@ -12,6 +12,7 @@
 from skimage.segmentation import clear_border
 
 from .scale import Scaler
+from .writer import write_multiscale
 
 CHANNEL_DIMENSION = 1
 
@@ -92,17 +93,6 @@ def rgb_to_5d(pixels: np.ndarray) -> List:
     return video
 
 
-def write_multiscale(pyramid: List, group: zarr.Group) -> None:
-    """Write a pyramid with multiscale metadata to disk."""
-    paths = []
-    for path, dataset in enumerate(pyramid):
-        group.create_dataset(str(path), data=pyramid[path])
-        paths.append({"path": str(path)})
-
-    multiscales = [{"version": "0.1", "datasets": paths}]
-    group.attrs["multiscales"] = multiscales
-
-
 def create_zarr(
     zarr_directory: str,
     method: Callable[..., Tuple[List, List]] = coins,

ome_zarr/writer.py

@@ -1,67 +1,61 @@
 """Image writer utility
 
 """
-import json
 import logging
-from pathlib import Path
 from typing import Any, List, Tuple, Union
 
 import dask.array as da
 import numpy as np
 import zarr
 
-from .io import parse_url
-from .reader import Node
 from .types import JSONDict
 
 LOGGER = logging.getLogger("ome_zarr.writer")
 
 
+def write_multiscale(
+    pyramid: List, group: zarr.Group, chunks: Union[Tuple[int], int] = None,
+) -> None:
+    """Write a pyramid with multiscale metadata to disk."""
+    paths = []
+    for path, dataset in enumerate(pyramid):
+        # TODO: chunks here could be different per layer
+        group.create_dataset(str(path), data=pyramid[path], chunks=chunks)
+        paths.append({"path": str(path)})
+
+    multiscales = [{"version": "0.1", "datasets": paths}]
+    group.attrs["multiscales"] = multiscales
+
+
 def write_image(
-    path: str,
     image: np.ndarray,
-    name: str = "0",
-    group: str = None,
+    group: zarr.Group,
     chunks: Union[Tuple[int], int] = None,
     byte_order: Union[str, List[str]] = "tczyx",
     **metadata: JSONDict,
-) -> zarr.hierarchy.Group:
+) -> None:
     """Writes an image to the zarr store according to ome-zarr specification
 
     Parameters
     ----------
-    path: str,
-      a path to the zarr store location
     image: np.ndarray
       the image to save
-    group: str, optional
+    group: zarr.Group
       the group within the zarr store to store the data in
     chunks: int or tuple of ints,
       size of the saved chunks to store the image
     byte_order: str or list of str, default "tczyx"
       combination of the letters defining the order
       in which the dimensions are saved
-
-    Return
-    ------
-    Zarr Group which contains the image.
     """
 
-    zarr_location = parse_url(path, "w")
-    if zarr_location is None:
-        raise ValueError
-
-    node = Node(zarr=zarr_location, root=[])
-
     if image.ndim > 5:
         raise ValueError("Only images of 5D or less are supported")
 
     shape_5d: Tuple[Any, ...] = (*(1,) * (5 - image.ndim), *image.shape)
     image = image.reshape(shape_5d)
 
-    if chunks is None:
-        image = da.from_array(image)
-    else:
+    if chunks is not None:
         _chunks = _retuple(chunks, shape_5d)
         image = da.from_array(image, chunks=_chunks)
 
@@ -94,11 +88,8 @@ def write_image(
             omero["rdefs"] = {"model": "color"}
 
     metadata["omero"] = omero
-    da.to_zarr(arr=image, url=node.zarr.subpath(name))
-    with open(Path(node.zarr.subpath(name)) / ".zattrs", "w") as za:
-        json.dump(metadata, za)
-
-    return node
+    write_multiscale([image], group)  # TODO: downsample
+    group.attrs.update(metadata)
 
 
 def _retuple(chunks: Union[Tuple[int], int], shape: Tuple[Any, ...]) -> Tuple[int, ...]: