When attempting to render_shapes with a color key or a table key that does not exist, spatialdata-plot just plots everything grey. I would expect it to raise a KeyError instead.
Example
# example dataset from tutorial
sdata = sd.read_zarr("visium_brain.zarr/")
(
sdata.pl.render_images(elements="ST8059050_hires_image")
.pl.render_shapes(elements="ST8059050", color="doesntexist")
.pl.show()
)
(
sdata.pl.render_images(elements="ST8059050_hires_image")
.pl.render_shapes(elements="ST8059050", color="mt-Co3", table_name="doesntexist")
.pl.show()
)
Also, somewhat related, the function silently accepts any keyword argument.
For instance, if I mistype table instead of table_name it just uses the default table instead of failing, which can be confusing.
EDIT: again, somewhat related, I'd also expect a failure with corresponding error message that tells me if a region could not be found in the table.
When attempting to
render_shapeswith a color key or a table key that does not exist, spatialdata-plot just plots everything grey. I would expect it to raise a KeyError instead.Example
( sdata.pl.render_images(elements="ST8059050_hires_image") .pl.render_shapes(elements="ST8059050", color="mt-Co3", table_name="doesntexist") .pl.show() )Also, somewhat related, the function silently accepts any keyword argument.
For instance, if I mistype
tableinstead oftable_nameit just uses the default table instead of failing, which can be confusing.EDIT: again, somewhat related, I'd also expect a failure with corresponding error message that tells me if a region could not be found in the
table.