Refactor add labels #167
Refactor add labels #167
Conversation
|
@michaelmcd18 This is ready for your review. |
R/class-yspec.R
Outdated
| col_labels <- map_chr(spec,fun) | ||
| for(i in seq_along(data)) { | ||
| attr(data[[i]],"label") <- col_labels[[i]] | ||
| col_labels <- map_chr(spec, fun) |
There was a problem hiding this comment.
nit-pick of no consequence: since col_labels isn't used until after the guard on 446-448, to my eyes it'd be nicer to move it below so it's the line before col_labels <- col_labels[names(col_labels) %in% names(data)].
There was a problem hiding this comment.
IC; yep, totally agree ... I like doing this in the code. Thanks for picking it up.
R/class-yspec.R
Outdated
| #' @param strict if `TRUE`, generate an error when `data` names do not match | ||
| #' the `spec` names; otherwise, only data items in common between `data` | ||
| #' and `spec` will be labeled. |
There was a problem hiding this comment.
From reading this, I'm not sure if it'd be obvious to me that "match" means identical rather than setequal (i.e. that order matters). May be worth explicitly mentioning order in this description.
There was a problem hiding this comment.
I clarified the language to include / exclude order. This makes it seem very strict, but the use case for this function is really labeling the data set corresponding to the spec ... formally writing the data set to .xpt; but now there's a workaround for the fringe cases where we want to label whatever happens to be there.
Relax the requirement that names must be identical between
dataandspec. The update adds an argument that bypass the requirement for identical names betweendataandspec. If no columns are found in common betweendataandspec, the data set is returned unlabeled with a warning.Also, some incremental updates to Rd file format are implemented here, limited to the code relevant to this PR.
See #158