View venn.diagram in RStudio viewer (not just write to file) using VennDiagram?

Question

Using the VennDiagram package, we can make a venn diagram like so with the venn.diagram() function like so :

library(tidyverse)
library(hrbrthemes)
library(tm)
library(proustr)

# Load dataset from github
data <- read.table("https://raw.githubusercontent.com/holtzy/data_to_viz/master/Example_dataset/14_SeveralIndepLists.csv", header=TRUE) 
to_remove <- c("_|[0-9]|\\.|function|^id|script|var|div|null|typeof|opts|if|^r$|undefined|false|loaded|true|settimeout|eval|else|artist")
data <- data %>% filter(!grepl(to_remove, word)) %>% filter(!word %in% stopwords('fr')) %>% filter(!word %in% proust_stopwords()$word)

# library
library(VennDiagram)

#Make the plot
venn.diagram(
  x = list(
    data %>% filter(artist=="booba") %>% select(word) %>% unlist() , 
    data %>% filter(artist=="nekfeu") %>% select(word) %>% unlist() , 
    data %>% filter(artist=="georges-brassens") %>% select(word) %>% unlist()
  ),
  category.names = c("Booba (1995)" , "Nekfeu (663)" , "Brassens (471)"),
  filename = 'venn.png',
  output = TRUE ,
  imagetype="png" ,
  height = 480 , 
  width = 480 , 
  resolution = 300,
  compression = "lzw",
  lwd = 1,
  col=c("#440154ff", '#21908dff', '#fde725ff'),
  fill = c(alpha("#440154ff",0.3), alpha('#21908dff',0.3), alpha('#fde725ff',0.3)),
  cex = 0.5,
  fontfamily = "sans",
  cat.cex = 0.3,
  cat.default.pos = "outer",
  cat.pos = c(-27, 27, 135),
  cat.dist = c(0.055, 0.055, 0.085),
  cat.fontfamily = "sans",
  cat.col = c("#440154ff", '#21908dff', '#fde725ff'),
  rotation = 1
)

This results in a .png written to the working directly.

How can it instead be viewed in the RStudio viewer pane, and also used in RMarkdown docs etc (ie just in the same way a regular ggplot or base plots would be viewed)?

Also note, the same question applies to any of the examples found in the ? venn.diagram ? venn.diagram documentation (they all seem to write to file instead of display in the RStudio viewer)

Answer 1

This should also do the job. I deleted the arguments for readability:

...
plt <- venn.diagram(
  filename = NULL,
  cex = 1,
  cat.cex = 1,
  lwd = 2,
  )
grid::grid.draw(plt)

From ?venn.diagram

filename
Filename for image output, or if NULL returns the grid object itself

It seems, you can control almost anything. Again the docs:

... A series of graphical parameters tweaking the plot. See below for details Details

Argument Venn Sizes Class Description
cex 1,2,3,4,5 numeric Vector giving the size for each area label (length = 1/3/7/15 based on set-number)

Thus we need to be able to display grid objects. plot() and print() don't do this job (it seems there is not print.grid() ).

Answer 2

I usually do:

library(VennDiagram)
set.seed(1)
list1 <- list(A=sample(LETTERS, 12), B=sample(LETTERS, 12))
venn1 <- venn.diagram(list1, filename = NULL)
grid.newpage()
grid.draw(venn1)

I think it still writes a log file into the working directory, but not the graph.

You can put two diagrams side by side like this:

library(gridExtra)
set.seed(2)
list2 <- list(A=sample(LETTERS, 16), B=sample(LETTERS, 12))
venn2 <- venn.diagram(list2, filename = NULL)
grid.arrange(gTree(children=venn1),
             gTree(children=venn2),
             ncol=2)

^{Created on 2020-04-23 by the reprex package (v0.3.0)}

Answer 3

I figured out a way - there may be better way(s). This involves writing to tempfile() instead of a file in the working directory and then reading it in with a few extra lines of code

Note: the only changes to the original code are the addition of

• 1 extra line at the start temp_file <- tempfile()
• the rewriting of filename = 'venn.png' into filename = temp_file
• 3 extra lines at the bottom

# Libraries
library(tidyverse)
library(hrbrthemes)
library(tm)
library(proustr)

# Load dataset from github
data <- read.table("https://raw.githubusercontent.com/holtzy/data_to_viz/master/Example_dataset/14_SeveralIndepLists.csv", header=TRUE) 
to_remove <- c("_|[0-9]|\\.|function|^id|script|var|div|null|typeof|opts|if|^r$|undefined|false|loaded|true|settimeout|eval|else|artist")
data <- data %>% filter(!grepl(to_remove, word)) %>% filter(!word %in% stopwords('fr')) %>% filter(!word %in% proust_stopwords()$word)

# library
library(VennDiagram)

temp_file <- tempfile()

#Make the plot
venn.diagram(
  x = list(
    data %>% filter(artist=="booba") %>% select(word) %>% unlist() , 
    data %>% filter(artist=="nekfeu") %>% select(word) %>% unlist() , 
    data %>% filter(artist=="georges-brassens") %>% select(word) %>% unlist()
  ),
  category.names = c("Booba (1995)" , "Nekfeu (663)" , "Brassens (471)"),
  filename = temp_file,
  output = TRUE ,
  imagetype="png" ,
  height = 480 , 
  width = 480 , 
  resolution = 300,
  compression = "lzw",
  lwd = 1,
  col=c("#440154ff", '#21908dff', '#fde725ff'),
  fill = c(alpha("#440154ff",0.3), alpha('#21908dff',0.3), alpha('#fde725ff',0.3)),
  cex = 0.5,
  fontfamily = "sans",
  cat.cex = 0.3,
  cat.default.pos = "outer",
  cat.pos = c(-27, 27, 135),
  cat.dist = c(0.055, 0.055, 0.085),
  cat.fontfamily = "sans",
  cat.col = c("#440154ff", '#21908dff', '#fde725ff'),
  rotation = 1
)


# https://stackoverflow.com/a/20909108/5783745
library(png)
img <- readPNG(temp_file)
grid::grid.raster(img)