R/utils_helpers.R

In aassumpcao/ColombiaDashboard: Colombia Diaspora Survey Results

#' helper function
#' @name prepare_data
#' @description function to prepare the variables of interest
#' @param question_analysis the main variable in the analysis
#' @param filter the variable to filter the data by
#' @param extra_filter an additional filter passed onto the dataset
#' @export
prepare_data <- function(
  question_analysis = NULL, filter = c('q3_2' = NULL),
  extra_filter = c(NULL = NULL)
){

  # load the data
  data <- DiasporaSurveyResults::survey_data

  # extra variable and value for filter
  filter_question <- names(filter)
  filter_selected <- unname(filter)

  # extra variable and value for filter
  extrafilter_question <- names(extra_filter)
  extrafilter_selected <- unname(extra_filter)

  # filter data using appropriate question and value
  if (!is.null(filter)) {
    data <- data %>%
      dplyr::filter(!!as.name(filter_question) == filter_selected)
  }

  # select variable
  prepared_data <- data %>%
    dplyr::filter(!is.na(!!as.name(question_analysis)))

  # apply additional fitler
  if (!is.null(extrafilter_selected)){
    prepared_data <- prepared_data %>%
      dplyr::filter(!is.na(!!as.name(extrafilter_question))) %>%
      dplyr::filter(!!as.name(extrafilter_question) == extrafilter_selected)
  }

  # return prepared data
  return(prepared_data)
}

#' helper function
#' @name count_unique
#' @description function to count the values of single-choice questions
#' @param data a dataset
#' @param question_analysis the main variable in the analysis
#' @param group the variable to use as the grouping variable
#' @param sort the variable to use as the sorting variable
#' @importFrom rlang .data
#' @export
count_unique <- function(data, question_analysis, group = NULL, sort = TRUE){

  # count data and create labels
  if (is.null(group)) {
    prepared_data <- data %>%
      dplyr::mutate(question = !!as.name(question_analysis)) %>%
      dplyr::filter(!is.na(.data$question)) %>%
      dplyr::count(.data$question)
  } else {
    prepared_data <- data %>%
      dplyr::mutate(
        question = !!as.name(question_analysis), group = !!as.name(group)
      ) %>%
      dplyr::filter(!is.na(.data$question) & !is.na(.data$group)) %>%
      dplyr::count(.data$question, .data$group)
  }

  # transform counts for graph
  prepared_data <- prepared_data %>%
    dplyr::mutate(
      x_label = .data$question,
      abs_total = .data$n,
      abs_label = .data$n,
      perc_total = (.data$abs_total / sum(.data$abs_total)) * 100,
      perc_label = paste0(round(.data$perc_total, 2), '%')
    ) %>%
    dplyr::select(-.data$n)

  # apply sort if requested by user
  if (sort) {
    prepared_data <- prepared_data %>%
      dplyr::arrange(dplyr::desc(.data$perc_total))
  }

  # return prepared data
  return(prepared_data)
}

#' helper function
#' @name count_multiple
#' @description function to count the values of single-choice questions
#' @param data a dataset
#' @param question_analysis the main variable in the analysis
#' @param group the variable to use as the grouping variable
#' @param sort the variable to use as the sorting variable
#' @importFrom rlang .data
#' @export
count_multiple <- function(data, question_analysis, group = NULL, sort = TRUE){

  # count data and create labels
  if (is.null(group)) {
    prepared_data <- data %>%
      dplyr::mutate(question = !!as.name(question_analysis)) %>%
      dplyr::mutate(question = stringr::str_split(.data$question, ',(?! )')) %>%
      tidyr::unnest(.data$question) %>%
      dplyr::filter(!is.na(.data$question)) %>%
      dplyr::count(.data$question)
  } else {
    prepared_data <- data %>%
      dplyr::mutate(
        question = !!as.name(question_analysis), group = !!as.name(group)
      ) %>%
      dplyr::mutate(question = stringr::str_split(.data$question, ',(?! )')) %>%
      tidyr::unnest(.data$question) %>%
      dplyr::filter(!is.na(.data$question) & !is.na(.data$group)) %>%
      dplyr::count(.data$question, .data$group)
  }

  # transform counts for graph
  prepared_data <- prepared_data %>%
    dplyr::mutate(
      x_label = .data$question,
      abs_total = .data$n,
      abs_label = .data$n,
      perc_total = (.data$abs_total / sum(.data$abs_total)) * 100,
      perc_label = paste0(round(.data$perc_total, 2), '%')
    ) %>%
    dplyr::select(-.data$n)

  # apply sort if requested by user
  if (sort) {
    prepared_data <- prepared_data %>%
      dplyr::arrange(dplyr::desc(.data$perc_total))
  }

  # return prepared data
  return(prepared_data)
}

#' helper function
#' @name extract_labels
#' @description function to count the variables of interest
#' @param data a dataset
#' @param group the variable to use as the grouping variable
#' @importFrom rlang .data
#' @export
extract_labels <- function(data, group = NULL){

  # apply group filter if assigned
  if (!is.null(group)){

    # prepare data
    data <- data %>%
      dplyr::mutate(group = !!as.name(group))
      dplyr::group_by(.data$group) %>%
      dplyr::arrange(dplyr::desc(.data$abs_total), .by_group = TRUE)

    # prepare labels
    labels <- data %>%
      dplyr::pull(.data$group) %>%
      unique() %>%
      rev()

  } else {

    # extract labels
    labels <- data %>%
      dplyr::pull(1) %>%
      unique() %>%
      rev()
  }

  # return data and labels
  return(list(data = data, labels = labels))
}

#' helper function
#' @name load_text
#' @description function to count the variables of interest
#' @param question_analysis a dataset
#' @param group the variable to use as the grouping variable
#' @importFrom rlang .data
#' @export
load_text <- function(question_analysis, group = NULL){

  # load analysis
  analysis <- DiasporaSurveyResults::analysis_text %>%
    dplyr::filter(.data$question == question_analysis) %>%
    as.list()

  # return analysis
  return(
    shiny::tagList(
      shiny::h2(analysis$h2, align = 'left'),
      shiny::p(analysis$p)
    )
  )
}