`table_name` not automatically inferred

[LucaMarconato]

In this example created for a user I am annotating the polygons from. blobs with 2 tables. The first table has the categorical column cell_type, while the second has the column cluster_id.

When plotting this into the shapes, in the first case table_name is not required and the plot can be performed anyway, in the second case instead I need to specify table_name otherwise the coloring is not applied.

The expected behavior would be for both plots to be possible without having to specify table_name, since in this case there is no ambiguity.

@melonora I wonder if this is automatically fixed by the new "per-element" parameter refactoring.

##
import spatialdata_plot
import matplotlib.pyplot as plt

from spatialdata.datasets import blobs_annotating_element
from spatialdata.models import TableModel
from anndata import AnnData
import pandas as pd

sdata = blobs_annotating_element("blobs_polygons").subset(["blobs_polygons"])
print(sdata)

##
# plotting the data without annotations
sdata.pl.render_shapes("blobs_polygons").pl.show()
plt.show()

##
# defining the annotations
celltype_dataframe = pd.DataFrame(
    {
        "cell_type": ["A", "B", "C", "A", "B"],
        "cell_id": sdata["blobs_polygons"].index,
    }
)

##
# option 1: merge the dataframe with the existing table
sdata["table"].obs = pd.merge(
    sdata["table"].obs, celltype_dataframe, left_on="instance_id", right_on="cell_id", how="left"
)

##
sdata.pl.render_shapes("blobs_polygons", color='cell_type').pl.show()
plt.show()

##
# option 2: add a new table
celltype_dataframe["region"] = ["blobs_polygons"] * 5
celltype_dataframe['cluster_id'] = [1, 1, 3, 2, 2]

annotations = AnnData(obs=celltype_dataframe)
annotations = TableModel.parse(annotations, region="blobs_polygons", region_key="region", instance_key="cell_id")
sdata["annotations"] = annotations
print(sdata)

##
sdata.pl.render_shapes("blobs_polygons", color="cluster_id", table_name='annotations').pl.show()
plt.show()

[melonora]

It will be yes, the refactoring now has been performed only for images and needs some minor fixing still, but am also working on refactoring of the other elements where the parameters will be validated per element. So if you specify a color it will be checked whether there is a table annotating that element and having the color column. If not table and color is set to None for that element.

[LucaMarconato]

Thanks for the update. One note: I think that we should throw a warning when the color can't be found automatically, as otherwise the user would silently see a gray plot.

[melonora]

The issue described has been fixed now the per element refactor is completed.