R/get_node_ids.R

In DiagrammeR: Graph/Network Visualization

#' Get a vector of node ID values
#'
#' @description
#'
#' Obtain a vector of node ID values from a graph object. An optional filter by
#' node attribute can limit the set of node ID values returned.
#'
#' @inheritParams render_graph
#' @param conditions An option to use filtering conditions for the retrieval of
#'   nodes.
#'
#' @return A vector of node ID values.
#'
#' @examples
#' # Create a node data
#' # frame (ndf)
#' ndf <-
#'   create_node_df(
#'     n = 4,
#'     type = "letter",
#'     color = c(
#'       "red", "green",
#'       "blue", "blue"),
#'     value = c(
#'       3.5, 2.6, 9.4, 2.7))
#'
#' # Create a graph using
#' # the ndf
#' graph <-
#'   create_graph(
#'     nodes_df = ndf)
#'
#' # Get a vector of all nodes in a graph
#' graph %>% get_node_ids()
#'
#' # Get a vector of node ID values using a
#' # numeric comparison (i.e., all nodes with
#' # `value` attribute greater than 3)
#' graph %>%
#'   get_node_ids(
#'     conditions = value > 3)
#'
#' # Get a vector of node ID values using
#' # a match pattern (i.e., all nodes with
#' # `color` attribute of `green`)
#' graph %>%
#'   get_node_ids(
#'     conditions = color == "green")
#'
#' # Use multiple conditions to return nodes
#' # with the desired attribute values
#' graph %>%
#'   get_node_ids(
#'     conditions =
#'       color == "blue" &
#'       value > 5)
#'
#' @import rlang
#' @export
get_node_ids <- function(
    graph,
    conditions = NULL
) {

  # Get the name of the function
  fcn_name <- get_calling_fcn()

  # Capture provided conditions
  conditions <- rlang::enquo(conditions)

  if (is_graph_empty(graph)) {
    return(NA)
  } else {
    nodes_df <- graph$nodes_df
  }

  # If conditions are provided then
  # pass in those conditions and filter the
  # data frame of `nodes_df`
  if (!is.null(
    rlang::enquo(conditions) %>%
    rlang::get_expr())) {

    nodes_df <- dplyr::filter(.data = nodes_df, !!conditions)
  }

  # If no nodes remain then return NA
  if (nrow(nodes_df) == 0) {
    return(NA)
  }

  nodes_df %>% dplyr::pull(id)
}