sandbox/join_set.R

In amygoldlist/aggRviz: A package for dealing with data aggregated at many levels

## A is a dataset with feature "a","b","proportion rate"
## B is a dataset with feature "a","c","interesting rate"
## The output will be a dataset with feature "a", "proportion rate", "interesting rate"


#' Join 2 tables on the maximum overlapping set of features
#'
#' This function will pick key column (such as "interesting rate" or proportion rate)
#' and join two lowest datasets by common columns.
#'
#' @param x data.frame
#' @param y data.frame
#'
#' @return data.frame
#'
#'
#'
#'

join_set <- function(x,y){#x,y can be two original datasets
  if (!is.data.frame(x)){
    stop("Error: x should be a dataframe!")
  }
  if (!is.data.frame(y)){
    stop("Error: y should be a dataframe!")
  }

  list_x = c()
  list_y = c()
  #check common columns
  for (i in names(x)){
    if (i %in% names(y) == F){
      list_x = c(list_x,i)
    }
  }
  for (j in names(y)){
    if (j %in% names(x) == F){
      list_y = c(list_y,j)
    }
  }

  #check key column in the first dataset
  for (i in 1:length(list_x)){
    if (T %in% stringr::str_detect(list_x[i],c("measure","rate")) == T){
      list_x = list_x %>% dplyr::setdiff(list_x[i])
    }
  }
  #check key column in the second dataset
  for (i in 1:length(list_y)){
    if (T %in% stringr::str_detect(list_y[i],c("measure","rate")) == T){
      list_y = list_y %>% dplyr::setdiff(list_y[i])
    }
  }

  #use `filter_blanks()` and `aggrViz_filter()` to deal with blanks
  if (length(list_x) != 0){
    x = x %>% aggRviz_filter(list_x)}
  else {x = x %>% filter_blanks()}
  if (length(list_y) != 0){
    y = y %>% aggRviz_filter(list_y)}
  else {y = y %>% filter_blanks()}

  return(dplyr::inner_join(x,y))
}