Move `add_column_to_layer_tabular_data` upstream

[haesleinhuepf]

I'm a big fan of this method:

add_column_to_layer_tabular_data

Would anyone mind if we move it to napari-skimage-regionprops? For backwards compatibility we could then import it in utilities like this:

from napari_skimage_regionprops import add_column_to_layer_tabular_data

With this import, no code would have to be changed immediately.

I'm also thinking if this function (or a similar function) would resolve haesleinhuepf/napari-skimage-regionprops#16

Maybe @jo-mueller has an opinion?

[Cryaaa]

@haesleinhuepf I'm all for it! It would fit there better anyway since it is mainly a table specific function! And regarding the issue raised I think this function would circumvent what @jo-mueller was talking about.

[jo-mueller]

Hi @haesleinhuepf @Cryaaa ,

sounds like a good idea to me - another step of taking infrastructure code out of the clusters plotter and making it a bit more lightweight :)

[jo-mueller]

Since we are already discussing about it here - would it be desirable to have this function throw a warning if, for instance, the passed column name was already in the target table? I created a simple implementation here.

[haesleinhuepf]

would it be desirable to have this function throw a warning if, for instance, the passed column name was already in the target table

Yes, I think this is a good idea. Furthermore, how about adding a parameter overwrite_existing with default value False?