tests/testthat/test_calc_limits_per_groups.R

In datacleanr: Interactive and Reproducible Data Cleaning

test_that("calculated limits are equal to manual approach (grouped iris)", {
  x <- "Sepal.Length"
  y <- "Sepal.Width"


  testdf <- dplyr::group_by(iris, Species) %>%
    dplyr::mutate(group_index = dplyr::cur_group_id()) %>%
    dplyr::filter(group_index %in% c(1, 3))

  xrange <- range(testdf[, x])
  yrange <- range(testdf[, y])

  compare_list <- list(
    xlim = xrange,
    ylim = yrange
  )



  lim_func <- calc_limits_per_groups(
    dframe = dplyr::group_by(iris, Species),
    group_index = c(1, 3),
    xvar = x,
    yvar = y,
    scaling = 0
  )


  expect_identical(lim_func, compare_list)
})

test_that("calculated limits are equal to manual approach (ungrouped iris)", {
  x <- "Sepal.Length"
  y <- "Sepal.Width"


  testdf <- iris %>%
    dplyr::mutate(group_index = dplyr::cur_group_id())

  xrange <- range(testdf[, x])
  yrange <- range(testdf[, y])

  compare_list <- list(
    xlim = xrange,
    ylim = yrange
  )



  lim_func <- calc_limits_per_groups(
    dframe = iris,
    group_index = c(1),
    xvar = x,
    yvar = y,
    scaling = 0
  )
  expect_identical(lim_func, compare_list)
})

test_that("calculated limits are equal to manual approach (ungrouped iris) with scaling !=0", {
  x <- "Sepal.Length"
  y <- "Sepal.Width"

  scl <- 0.1

  testdf <- iris %>%
    dplyr::mutate(group_index = dplyr::cur_group_id())



  ylim <- (diff(range(testdf[, y])) * scl) * c(-1, 1) +
    range(testdf[, y])

  xlim <- (diff(range(testdf[, x])) * scl) * c(-1, 1) +
    range(testdf[, x])

  compare_list <- list(
    xlim = xlim,
    ylim = ylim
  )



  lim_func <- calc_limits_per_groups(
    dframe = iris,
    group_index = c(1),
    xvar = x,
    yvar = y,
    scaling = scl
  )
  expect_identical(lim_func, compare_list)
})


test_that("calculated limits are equal to manual approach on time series df)", {
  tsdf <- data.frame(
    x = seq(as.POSIXct("2000-01-01", tz = "UTC"),
      as.POSIXct("2000-02-01", tz = "UTC"),
      length.out = 25
    ),
    y = rnorm(25)
  ) %>%
    dplyr::mutate(group_index = dplyr::cur_group_id())


  x <- "x"
  y <- "y"




  xrange <- range(tsdf[, x])
  yrange <- range(tsdf[, y])


  total_range <- as.numeric(range(tsdf[, x, drop = TRUE], na.rm = TRUE))
  # offset <- (diff(total_range)) * c(-1,1) + scaling
  offset <- (diff(total_range) * 0.01) * c(-1, 1)


  lim <- range(tsdf[, x, drop = TRUE], na.rm = TRUE) +
    offset

  compare_list <- list(
    xlim = lim,
    ylim = yrange
  )






  lim_func <- calc_limits_per_groups(
    dframe = tsdf,
    group_index = 1,
    xvar = x,
    yvar = y,
    scaling = 0
  )
  expect_identical(lim_func, compare_list)
})



test_that("calculated limits are equal to manual approach on time series df)", {
  tsdf <- data.frame(
    y = seq(as.POSIXct("2000-01-01", tz = "UTC"),
      as.POSIXct("2000-02-01", tz = "UTC"),
      length.out = 25
    ),
    x = rnorm(25)
  ) %>%
    dplyr::mutate(group_index = dplyr::cur_group_id())


  x <- "x"
  y <- "y"




  xrange <- range(tsdf[, x])
  yrange <- range(tsdf[, y])


  total_range <- as.numeric(range(tsdf[, y, drop = TRUE], na.rm = TRUE))
  # offset <- (diff(total_range)) * c(-1,1) + scaling
  offset <- (diff(total_range) * 0.01) * c(-1, 1)


  lim <- range(tsdf[, y, drop = TRUE], na.rm = TRUE) +
    offset

  compare_list <- list(
    xlim = xrange,
    ylim = lim
  )






  lim_func <- calc_limits_per_groups(
    dframe = tsdf,
    group_index = 1,
    xvar = x,
    yvar = y,
    scaling = 0
  )
  expect_identical(lim_func, compare_list)
})