Patterns Don't Appear in Legend

[jvanbelzen]

Bug description

Patterns appear in the graph, but not the legends. I noticed others described the same issue, but their problem seemed to be limited to exported image files. I have the issue in the image displayed in RStudio.

UPDATE The issue is specific to RStudio! If I run these same codes in R, the legend correctly displays the patterns.

Minimal, reproducible example

Two examples from independent posts (Stackoverflow and Posit). The responses indicate that the code produces the pattern in the graph and legends, as you would expect. However, when run on my end the same code produces the pattern in the graph but not the legend.

# Example 1
library(ggplot2)
library(ggpattern)
library(magrittr)

dat <- data.frame(drv = c("4", "4", "4", "4", "4", "f", "f", "f", "f", "r", "r", "r"), 
           class = c("compact", "midsize", "pickup", "subcompact", "suv", 
                     "compact", "midsize", "minivan", "subcompact", 
                     "2seater", "subcompact", "suv"), 
           y = c(12L, 3L, 33L, 4L, 51L, 35L, 38L, 11L, 22L, 5L, 9L, 11L)) 

dat 

# Give the same name for scale_fill_manual() and scale_pattern_manual(). Also, specify the group (or groups) you would like to show the pattern (or patterns).

dat %>% ggplot(aes(x = class, y = y, fill = drv, pattern = drv)) +
  #  geom_col()+   # not necessary
  geom_col_pattern() +
  coord_flip() +
  theme_minimal() +
  scale_fill_manual(
    name = "Drive",
    values = c(
      "4" = "#DA4943",
      "f" = "#D058EC",
      "r" = "#FF27D5"
    )
  ) +
  scale_pattern_manual(
    name = "Drive",
    values = c(
      "4" = "none",
      "f" = "stripe", 
      "r" = "none"
    )
  )

# Example 2
cData <- structure(list(Individual = c("Indv 1", "Indv 2", "Indv 3", "Indv 1", "Indv 2", "Indv 3", "Indv 1", "Indv 2", "Indv 3",
                                       "Indv 1", "Indv 2", "Indv 3", "Indv 1", "Indv 2", "Indv 3", "Indv 1", "Indv 2", "Indv 3"),
                        Change = c("Gain", "Gain", "Gain", "Gain", "Gain", "Gain", "Gain", "Gain", "Gain",
                                   "Loss", "Loss", "Loss", "Loss", "Loss", "Loss", "Loss", "Loss", "Loss"),
                        Type = c("Type A", "Type A", "Type A", "Type B", "Type B", "Type B", "Type C", "Type C", "Type C",
                                 "Type A", "Type A", "Type A", "Type B", "Type B", "Type B", "Type C", "Type C", "Type C"),
                        Value = c(39L, 5L, 12L, 28L, 21L, 7L, 21L, 9L, 116L, 28L, 108L, 30L, 0L, 3L, 25L, 13L, 4L, 26L),
                        Stripe_Angle = c(25L, 25L, 25L, 135L, 135L, 135L, 90L, 90L, 90L, 25L, 25L, 25L, 135L, 135L,
                                         135L, 90L, 90L, 90L)), row.names = c(NA, -18L), class = "data.frame")

library(ggplot2)
library(dplyr)
library(ggpattern)

colr <- c("orange", "steelblue", "darkred")

cData %>% 
  ggplot(aes(x=Change, y=Value), pattern="stripe") +
  geom_col_pattern(aes(pattern_angle=Type, color=Type, fill=Type),
                   pattern_fill="white", pattern_color="white", pattern_spacing=.05) +
  scale_color_manual(values = colr) +
  scale_fill_manual(values = colr) +
  scale_pattern_angle_manual(values = c(0, -45, 45)) +
  scale_y_continuous(breaks = seq(0, 140, 10)) +
  labs(x="", y="Amount") +
  facet_grid(~Individual, switch = "x") +
  theme_classic()

Session info

R version 4.3.2 (2023-10-31)
Platform: aarch64-apple-darwin20 (64-bit)
Running under: macOS Sonoma 14.4.1, RStudio 2023.12.1.402

Locale: en_US.UTF-8 / en_US.UTF-8 / en_US.UTF-8 / C / en_US.UTF-8 / en_US.UTF-8

Package version:
  cachem_1.0.8       class_7.3.22       classInt_0.4.10    cli_3.6.2          colorspace_2.1.0  
  DBI_1.2.2          e1071_1.7.14       fansi_1.0.6        farver_2.1.1       fastmap_1.1.1     
  ggpattern_1.0.1    ggplot2_3.5.1      glue_1.7.0         graphics_4.3.2     grDevices_4.3.2   
  grid_4.3.2         gridpattern_1.2.1  gtable_0.3.5       isoband_0.2.7      KernSmooth_2.23.22
  labeling_0.4.3     lattice_0.22.6     lifecycle_1.0.4    magrittr_2.0.3     MASS_7.3.60.0.1   
  Matrix_1.6.5       memoise_2.0.1      methods_4.3.2      mgcv_1.9.1         munsell_0.5.1     
  nlme_3.1.164       pillar_1.9.0       pkgconfig_2.0.3    png_0.1.8          proxy_0.4.27      
  R6_2.5.1           RColorBrewer_1.1.3 Rcpp_1.0.12        rlang_1.1.3        s2_1.1.6          
  scales_1.3.0       sf_1.0.16          splines_4.3.2      stats_4.3.2        tibble_3.2.1      
  tools_4.3.2        units_0.8.5        utf8_1.2.4         utils_4.3.2        vctrs_0.6.5       
  viridisLite_0.4.2  withr_3.0.0        wk_0.9.1          

[trevorld]

When I run the first example in my terminal I get:

and the second example looks like:

What graphic device are you drawing with? What does names(dev.cur()) give you?

[trevorld]

This is a duplicate of #86 which remains open. I still need a minimal, reproducible example...

[jvanbelzen]

What graphic device are you drawing with? What does names(dev.cur()) give you?

[1] "RStudioGD"

And it does only seem to be an issue for me when working in RStudio. Is ggpatterns() not compatible with RStudio?

[trevorld]

• I don't often use RStudio myself but I think people do use {ggpattern} in RStudio.
• I vaguely recall you can customize which graphics device you use within RStudio. In particular the "cairo" devices seem to work well for people. I think I've seen in other issues in other packages that setting options(bitmapType="cairo") solved {ggplot2} bugs in RStudio before. I think there is a way to customize what device you use in the GUI as well.
• Sometimes users report that turning on/off R 4.1 feature support in {ggpattern} fixes things.
• Sometimes users report that making the legend windows a bit bigger fixes things.
• Sometimes users report that using pattern_key_scale_factor to adjust size of patterns in legends fixes things.

[trevorld]

And it does only seem to be an issue for me when working in RStudio. Is {ggpatterns} not compatible with RStudio?

I just installed RStudio on my Linux laptop and the charts work fine for me in RStudio.

I wildly speculate that maybe it is defaulting to the "quartz" device and maybe there is something weird going on the with the quartz png device which the "RStudioGD" device maybe wrapping around that doesn't affect the cairo png device?

You may want to click on "Tools" -> "Global Options" -> "Graphics" and try different graphics backends...