R/align.r

In mojson: A Serialization-Style Flattening and Description for JSON

#' JSON Lists Alignment
#'
#' @description Align the two JSON lists by specifying the primary path(keys), to support
#'     the left/inner/right-join style comparison.
#'
#' @param json_new \code{list}. The new JSON list.
#' @param json_old \code{list}. The old JSON list.
#' @param sep \code{character}. A character/string passed to \code{\link{flattenj}}.
#'     Defaults to @ to avoid the occasional overriding. Not recommended to use some risky characters like . and \.
#' @param primary \code{character}. The primary path(keys) for identifying a unique JSON object.
#'     The value provided should contain the \code{sep} value to specify the nesting information unless it is an outermost key.
#'
#' @return \code{list}. The result list contains the alignment information of three types: the primary paths only in the new JSON,
#'     only in the old JSON, and in both.
#' @details The function borrows the idea from the data set operation, and the result contains:
#'     \itemize{
#'     \item{`new`, contains the flattening result of `json_new`.}
#'     \item{`old`, contains the flattening result of `json_old`.}
#'     \item{`common_primary`, contains the primary paths both in `json_new` and `json_old`.}
#'     \item{`new_primary`, contains the primary paths only in `json_new`.}
#'     \item{`old_primary`, contains the primary paths only in `json_old`.}}
#' @importFrom magrittr %>%
#' @export
#'
#' @examples
#'
#' library(mojson)
#' j1 <- list(list(id = list(x = 1 ,y = 2), gender = 'M'),
#'            list(id = list(x = 2 ,y = 2), gender = 'M'))
#' j2 <- list(list(id = list(x = 2 ,y = 2), gender = 'F'),
#'            list(id = list(x = 3 ,y = 2), gender = 'F'))
#' alignj(j1, j2, primary = 'id@x')
alignj <- function(json_new, json_old, sep = '@', primary) {
  new <- flattenj(json_new, sep = sep)
  old <- flattenj(json_old, sep = sep)
  warning('Please make sure the primary value uses the sep value or @ to specify the nesting information,
          unless the primary value is an outermost key.')
  primary_new <- subset(new, paths == primary)$values
  primary_old <- subset(old, paths == primary)$values
  common_part <- intersect(primary_new, primary_old)
  only_new <- setdiff(primary_new, primary_old)
  only_old <- setdiff(primary_old, primary_new)
  return(list(new = new,
              old = old,
              common_primary = common_part,
              new_primary = only_new,
              old_primary = only_old))

}