minimal voronoi tree

voronoi_tree_03.R

library(flametree)  # github.com/djnavarro/flametree
library(voronoise)  # github.com/djnavarro/voronoise
library(dplyr)

# set seed
seed <- 1
set.seed(seed)

# the "flametree" itself
ftree <- flametree_grow(
  seed = seed,
  angle = c(-2:4) * 10,
  scale = c(.6, .8, .9)
)

# "leaf" coordinates are at terminal locations (id_step = 2) 
# on the terminal branches (id_leaf == TRUE) in the tree
vleaf <- ftree %>% filter(id_leaf == TRUE, id_step == 2)

# a simple "perturb" function: drop each leaf to the ground
leaf_fall <- function(data) {
  data %>% 
    group_by(group) %>% 
    mutate(y = y - min(y)) %>% 
    ungroup()
}


# create the plot...
p <- ggplot() +
  
  # tree trunk is drawn using geom_bezier from the 
  # ggforce package (loaded by voronoise)
  geom_bezier(
    data = ftree, 
    mapping = aes(
      x = coord_x,
      y = coord_y,
      group = id_path,
      size = seg_wid
    ), 
    lineend = "round", 
    show.legend = FALSE
  ) +
  
  # add points drawn at the leaves
  geom_point(
    data = vleaf,
    mapping = aes(
      x = coord_x,
      y = coord_y
    ), 
    size = 8
  ) + 
  
  # add voronoi tiles with no perturbation
  geom_voronoise(
    data = vleaf,
    mapping = aes(
      x = coord_x,
      y = coord_y
    ),
    max.radius = .2,
    fill = "#ffffffcc",
    colour = "black",
    size = 2
  ) +

  # add voronoi tiles for falling leaves
  geom_voronoise(
    data = vleaf,
    mapping = aes(
      x = coord_x,
      y = coord_y
    ),
    max.radius = .2,
    fill = "#ffffffcc",
    colour = "black",
    size = 2,
    perturb = leaf_fall
  ) +
  
  # styling
  theme_void() +
  coord_equal()


# save the file
ggsave(
  filename = "~/Desktop/vtree_03.png", 
  plot = p,
  width = 100/3,
  height = 100/3,
  dpi = 150
)

Forgot to add, the art you make with data is great.

I think this issue is caused by a weird interaction. It looks like you're running R 3.6 and the version of ggforce (which voronoise depends upon) was built under 3.6.2. Normally that would just throw a warning rather than an error, but install_github() automatically converts warnings to errors (I vaguely recall that there is a good reason for this, but it's frustrating). So I think you should be able to solve the issue by installing ggforce, ggplot2 and dplyr from CRAN (or, more likely, I'm guessing you already have these!), and then installing voronoise from github without dependencies:

install_github("djnavarro/voronoise", dependencies = FALSE)

That said, I'm just guessing!

I tried both. By putting "dependencies = FALSE" and with out it. But it didn't work.
By the way my R Studio version is 3.6.1.