"Samples" in a VariantData file normally refer to diploids (i.e. tskit "individuals"). However, vdata.num_samples currently returns the number of sample nodes (i.e. twice the expected number of VariantData samples, in the diploid case). I presume this is to match the SampleData format, but it seems very confusing to me.
I wonder if it would be a good idea to define vdata.num_sample_nodes instead, and raise an error when calling vdata.num_samples that says "this is no longer supported: either use .num_sample_nodes if you want the skit-version. or num_individuals if you want the number of sampled individuals (e.g. diploids)"
"Samples" in a VariantData file normally refer to diploids (i.e. tskit "individuals"). However,
vdata.num_samplescurrently returns the number of sample nodes (i.e. twice the expected number of VariantData samples, in the diploid case). I presume this is to match the SampleData format, but it seems very confusing to me.I wonder if it would be a good idea to define
vdata.num_sample_nodesinstead, and raise an error when callingvdata.num_samplesthat says "this is no longer supported: either use .num_sample_nodes if you want the skit-version. or num_individuals if you want the number of sampled individuals (e.g. diploids)"