r - documenting a list of data frames with roxygen2 -

i trying document data append in package.

i have list 3 elements (data.frames) , each of data frames has columns. wonder how document hierarchical structure (list -> data.frame -> column) within roxygen2. examples show 1 layer, data.frame , column names. here minimal example 2 data frames in list.

list(   names=data.frame(id=1:10, a=letters[1:10]),   values=data.frame(id=rep(1:10, each=10), values=runif(100)) ) 

in list, 3 data.frames connected id column, kind of related, don't want put them 1 data frame, because of saving memory.

any suggestions welcome.

edit

i tried add segments documentation, not seem work

#' list group information test data set #' #' dataset list 2 data.tables (proteins, timepoint).  #' has common id column in data.tables. #' #' \strong{proteins} #' @format data frame 5458 rows , 3 columns #' \describe{ #'   \item{id}{unique group id} #'   \item{names}{individual names} #'   \item{other names}{other names} #' } #' #' \strong{timepoint} #' @format data frame 80248 rows , 5 columns #' \describe{ #'   \item{id}{unique group id} #'   \item{timepoint}{individual timepoint} #'   \item{imputed_mean}{mean value including imputed values} #'   \item{measured_mean}{mean value without imputing (contains nas)} #'   \item{value_count}{number of measured values within replicates} #' } 'pg_test' 

the output looks can handle 1 @format parameter.

any other suggestion?

actually, figured out solution, long nobody else has better suggestion...

#' list group information test data set #' #' dataset list 2 data.tables (proteins, timepoint).  #' has common id column in data.tables. #' #' @format  #' \enumerate{ #' \item \strong{proteins} data frame 5458 rows , 3 columns #' \describe{ #'   \item{id}{unique group id} #'   \item{names}{individual names} #'   \item{other names}{other names} #' } #' #' \item \strong{timepoint} data frame 80248 rows , 5 columns #' \describe{ #'   \item{id}{unique group id} #'   \item{timepoint}{individual timepoint} #'   \item{imputed_mean}{mean value including imputed values} #'   \item{measured_mean}{mean value without imputing (contains nas)} #'   \item{value_count}{number of measured values within replicates} #' } #' } 'pg_test' 

could solved

#' \describe{ #'   \item{one}{first item} .... 

instead of \enumerate guess. solution works now.