SparseAssays -> SimpleList coercion

PeteHaitch / SparseSummarizedExperiment

An experimental R package that defines S4 classes and methods for sparse genomic data

Other

2 stars 0 forks source link

Open PeteHaitch opened 8 years ago

PeteHaitch commented 8 years ago

Currently (dbaca193986ae61ec498dbeac8c41bd672b0d8c9) it is documented that a concrete subclass of the SparseAssays virtual class must have lossless coercion to the SimpleList class (https://github.com/PeteHaitch/SparseSummarizedExperiment/blob/dbaca193986ae61ec498dbeac8c41bd672b0d8c9/R/SparseAssays-class.R#L33). Is this strictly necessary?

PeteHaitch commented 8 years ago

[ ] Identify all places in the code where this is explicitly required and see if this could be safely removed. It might be better to rely on explicit getters/setters for each concrete subclass rather than relying on these having a consistent SimpleList representation.