R/flexplot_jasp2.R

In dustinfife/flexplot: Graphically Based Data Analysis Using 'flexplot'

#' Flexplot in JASP
#'
#' This function was developed for use in JASP. It takes a dataset as input with user
#' options and returns a flexplot graphic
#'
#' @param jaspResults A JASP  object
#' @param dataset dataset supplied by JASP
#' @param options a list of options to pass to JASP
#'
#' @return a flexplot graphic.
#' @export
#' @importFrom ggsci scale_color_npg scale_color_aaas scale_color_lancet scale_color_tron
flexplot_jasp2 = function(jaspResults, dataset, options) {

  ### check if they've entered anything
  ready <- (options$dependent != "")

  ### read in the dataset
  if (ready) {
    ### read in the dataset
    if (is.null(dataset)){
      dataset = (.readDataSetToEnd(columns=c(options$dependent, options$variables, options$paneledVars)))
    } else {
      return(dataset)
    }
  }

  ### create plots
  .flexPlotRes(jaspResults, formula, dataset, options, ready)
}


.flexPlotRes <- function(jaspResults, formula, dataset, options, ready) {
  #### set up parameters
  flex_Plot <- createJaspPlot(title = "Flexplot",  width = 600, height = 450)
  flex_Plot$dependOn(c("confidence", "dependent", "variables", "paneledVars", "ghostLines", "bw", "nameX", "nameY", "nameLegend", "nameRows", "nameCols"))
  flex_Plot$addCitation("Fife, D. A. (2022). Flexplot: Graphically-based data analysis. Psychological Methods, 27(4):477-496. doi: 10.1037/met0000424.")

  #### pre-populate the jasp object
  jaspResults[["flex_Plot"]] <- flex_Plot

  if (!ready){
    return()
  }

  #### prepare the data for flexplot
  k = data.frame(matrix(nrow=nrow(dataset), ncol=length(options$variables) + length(options$dependent) + length(options$paneledVars)))
  names(k) = c(options$dependent, options$variables, options$paneledVars)
  variables <- unlist(options$variables)
  panels <- unlist(options$paneledVars)
  if (length(panels)>0){
    vars = c(variables, panels)
  } else {
    vars = variables
  }
  k[,1] = dataset[[encodeColNames(options$dependent)]]
  if (length(vars)>0){		#### this statement is necessary to allow histograms
    for (i in 2:(length(vars)+1)){
      k[,i] = dataset[[encodeColNames(vars[i-1])]]
    }
  }
  dv = encodeColNames(unlist(options$dependent));
  v = encodeColNames(unlist(options$variables));
  p = encodeColNames(unlist(options$paneledVars))

  if (length(options$variables)==0){
    formula = as.formula(paste0(dv, "~1"))
  } else if (length(options$paneledVars)>0){
    formula = as.formula(paste0(dv,
                                "~",
                                paste0(v, collapse="+"),
                                " | ",
                                paste0(p, collapse="+")
                                ))
  } else {
    formula = as.formula(paste0(dv, "~", paste0(v, collapse="+")))
  }
  #tst = data.frame(x=1:10, y=1:10)

  #### do a ghost line
  if	(options$ghost & is.numeric(dataset[,v[1]])){
    ghost=rgb(195/255,0,0,.3)
  } else {
    ghost = NULL
  }

  whiskers = list("Quartiles" = "quartiles",
                  "Standard errors" = "sterr",
                  "Standard deviations" = "stdev")

  linetype = tolower(options$type)

  jitter = c(options$jitx,options$jity)
  if (linetype == "regression") linetype = "lm"
  if (linetype == "none") {linetype = "loess"; suppress_smooth = T; } else { suppress_smooth=F}

  plot = flexplot(formula, data=dataset, method=linetype, se=options$confidence, alpha=options$alpha,
                            ghost.line=ghost,
                            suppress_smooth = suppress_smooth,
                            spread=whiskers[[options$intervals]],
                            jitter = jitter)
  plot <- theme_it(plot, options$theme)

  # modify color palette
  if (options$palette!="GGplot Default") {
    ops = c("Nature", "AAAS", "Lancet", "JCO", "Tron")
    fun = c(ggsci::scale_color_npg(), ggsci::scale_color_aaas(), ggsci::scale_color_lancet(), ggsci::scale_color_jco(), ggsci::scale_color_tron())
    plot = plot + fun[ops == options$palette]
  }
    
  if (options$bw) {
    plot = convert_to_grayscale(plot)
  }
  colsnstuff = ifelse(length(options$variables)>1, options$variables[2], "")
  plot = fancifyMyLabels_strings(plot, options)

  # #### create flexplot object
  flex_Plot$plotObject <- plot
  return()
}



#### create a table
.printedResults = function(jaspResults, dataset, options, ready){

  if (!is.null(jaspResults[["resultsTable"]])) return()

  # Create Table
  resultsTable <- createJaspTable(title="Flexplot Table")
  resultsTable$dependOn(c("variables", "dependent", "confidence", "type"))
  resultsTable$addCitation("Fife, Dustin A. (2019). Flexplot (Version 0.9.2) [Computer software].")

  resultsTable$showSpecifiedColumnsOnly <- TRUE

  ### add columns to table
  resultsTable$addColumnInfo(name = "var",   title = "Variable",   type = "string", combine = TRUE)
  resultsTable$addColumnInfo(name = "mean",   title = "Mean",   type = "string", combine = TRUE)
  resultsTable$addColumnInfo(name = "median",   title = "Median",   type = "string", combine = TRUE)
  resultsTable$addColumnInfo(name = "typ",   title = "Type",   type = "string", combine = TRUE)

  #### tell jasp how many rows we expect
  if (ready)
    resultsTable$setExpectedSize(length(options$variables))

  for (variable in options$variables){
    #row <- c(variable, options$dependent, options$confidence, options$type)
    resultsTable$addRows(list(
      "var" = variable,
      "mean" = mean(dataset[[encodeColNames(variable)]]),
      "median" = median(dataset[[encodeColNames(variable)]]),
      "typ" = mode(dataset[[encodeColNames(variable)]])))
  }
  #f <- paste0(options$dependent, "~", paste0(options$variables, collapse="+"))
  #message <- options$variables
  resultsTable$addFootnote(message="hello", symbol="<em>Note.</em>")

  jaspResults[["resultsTable"]] <- resultsTable
}

#### read in data
.flexReadData <- function(dataset, options) {
  if (!is.null(dataset))
    return(dataset)
  else
    return(.readDataSetToEnd(columns=c(options$dependent, options$variables)))
}




themeJasp = function(graph,
                     xAxis = TRUE,
                     yAxis = TRUE,
                     sides = "bl",
                     axis.title.cex = jaspGraphs::getGraphOption("axis.title.cex"),
                     bty = jaspGraphs::getGraphOption("bty"),
                     fontsize = jaspGraphs::getGraphOption("fontsize"),
                     family = jaspGraphs::getGraphOption("family"),
                     horizontal = FALSE,
                     legend.position = "right",
                     legend.justification = "top",
                     axisTickLength = jaspGraphs::getGraphOption("axisTickLength"),
                     axisTickWidth = jaspGraphs::getGraphOption("axisTickWidth")) {

    if (is.list(bty) && bty[["type"]] == "n")
    graph <- graph + jaspGraphs::geom_rangeframe(sides = sides)

  if (horizontal)
    graph <- graph + coord_flip()

  graph <- graph +
                  jaspGraphs::themeJaspRaw(legend.position = legend.position,
                                axis.title.cex = axis.title.cex, family = family,
                                fontsize = fontsize, legend.justification = legend.justification,
                                axisTickLength = axisTickLength, axisTickWidth = axisTickWidth) +
            theme(panel.spacing = unit(.2, "cm"))


  return(graph)

}