Unable to run igneous view on OME-Zarr data #197
Description
I am using the dataset that I described in this issue #196
Using the following code (please comment out the cloudpath = ... lines to manually run the three cases), I get the following errors:
from igneous_cli import view
from click.testing import CliRunner
# case 1 and 2: please replace the absolute paths with your own
# case 3: please run http-server in the directory where the zarr file is located
cloudpath = "zarr:///Users/macbook/ssd/biodata/ome-zarr/20200812-CardiomyocyteDifferentiation14-Cycle1.zarr/B/03/0"
cloudpath = "/Users/macbook/ssd/biodata/ome-zarr/20200812-CardiomyocyteDifferentiation14-Cycle1.zarr/B/03/0"
cloudpath = "http://127.0.0.1:8080"
runner = CliRunner()
result = runner.invoke(
view,
[cloudpath],
)Case 1 error
Neuroglancer opens with this error
Case 2 error
Neuroglancer opens with this error
Case 3 error
Neuroglancer opens with this error
Workaround / hack
As a proof of concept solution / hack, if I replace the single line (the commented one with the commented one) in igneous_cli.cli as follows:
config = {
"dimensions": dimensions,
"layers": [
{
"type": cv.layer_type,
# "source": f"{cv.meta.path.format}://{cloudpath}",
"source": f"zarr://{cloudpath}",
"tab": "source",
"name": layer_name
}
],The case 2 and 3 will work.
Solution
I believe an easy solution can be implemented by slightly modifying the function view() in igneous_cli.cli and/or the logic in the function extract_format_protocol() in cloudvolume.paths.