I'm going through R CMD check WARNINGS and currently grappling with the below WARNING message:
* checking for missing documentation entries ... WARNING
Undocumented code objects:
‘diagnoses’
Undocumented data sets:
‘diagnoses’
Right now, I have an .RData file in my data/ folder in my R package skeleton that has 2 - not just 1 - R objects that get loaded when you load the RData contents. However, I cannot find any documentation that tells me how to declare both R objects so that the man pages pick up on it when I roxgenise() and perform a subsequent R CMD check.
So far I'm trying to document both R objects in a dataset man page using the "@format" tag. I have also tried the "@return" tag instead of "@format", but the WARNING persists. I've noticed it likes when the name of the R object matches the name of the RData file. Thus, the Miller2015 R object isn't reported as undocumented, but the diagnoses data.frame is.
#' @title Untargeted metabolomic analysis for the clinical screening of inborn errors
#' of metabolism
#'
#' @description Global metabolic profiling obtained by untargeted mass spectrometry-based
#' metabolomic platform for the detection of novel and known inborn errors of
#' metabolism.
#'
#' @name Miller2015
#' @aliases Miller2015
#' @docType data
#' @usage data(Miller2015)
#' @format Miller2015 - The data frame with 1203 metabolite features as rows, and 186 untargeted metabolomics patient samples as columns, alongside 16 metabolite annotations.
#' @format diagnoses - A data.frame where all patient IDs are mapped to their given biochemical diagnosis.
#' @keywords datasets
#' @references Miller et al. (2015) J Inherit Metab Dis. 2015; 38: 1029–1039
#' (\href{https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4626538/}{PubMed})
#' @source \href{https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4626538/bin/10545_2015_9843_MOESM1_ESM.xls}{Dataset}
#'
#' @examples
#' require(CTD)
#' data(Miller2015)
"Miller2015"
What can I do to resolve this WARNING without having to break up the R objects into separate RData files? What is best practice?
