library(vdiffr)
library(ggplot2)
library(distributional)

df <- data.frame(x1 = 2.62, x2 = 3.57, 
                 y1 = 21.0, y2 = 15.0)
uncertain_df <- data.frame(x1 = dist_normal(2.62, 0.1), 
                           x2 = dist_normal(3.57,0.1), 
                           y1 = dist_normal(21.0, 0.1), 
                           y2 = dist_normal(15.0,0.1))


a <- ggplot(uncertain_mtcars, aes(wt, mpg)) +
  geom_point_sample(size=0.1) +
  scale_x_continuous_distribution(limits = c(1,6)) +
  scale_y_continuous_distribution(limits = c(8,36))

# You can also use geom_segment to recreate plot(type = "h") :
set.seed(1)
# deterministic data
counts <- as.data.frame(table(x = rpois(100,5)))
counts$x <- as.numeric(as.character(counts$x))
# random data
uncertain_counts <- counts
uncertain_counts$Freq <- dist_binomial(counts$Freq, 0.9)

test_that("geom_segment_sample tests", {
  
  set.seed(876)
  
  p1 <- a +
    geom_segment_sample(aes(x = x1, y = y1, xend = x2, yend = y2, colour = "segment"), 
                        data = uncertain_df, alpha=0.5)
  
  expect_doppelganger("Example 1", p1)
  
  p2 <- ggplot(uncertain_counts, aes(x, Freq)) +
    geom_segment_sample(aes(xend = x, yend = 0), linewidth = 10, lineend = "butt", alpha=0.1) +
    scale_x_continuous_distribution(limits = c(0,12)) +
    scale_y_continuous_distribution(limits = c(0,25)) 
  
  expect_doppelganger("Example 2", p2)
  
}
)