Plotting well log interval data in ggplot2-CodePudding

I'd like to be able to produce a simple well log graph with ggplot2. Test data is simple:

From    To  Lith
0   25  A
25  50  B
50  75  C
75  100 B
100 125 C
125 150 A
150 175 B
175 200 C

The plot should be in a form of a strip, colored according to lith column with correctly displayed depths and should look something like this ("SLF" used instead of "Lith"):

As you can see, certain units repeat down the column and I need them to be in their correct place. I was hoping there is a simple way to force geom_bar or geom_col to simply draw the data as it is, without summarising it in any way, but it seems I was wrong.

How should I go about doing this? Which *geom" should I use?

TIA

CodePudding user response：

It's fairly straighforward to do this using geom_rect instead of geom_bar

library(ggplot2)

ggplot(df, aes(x = 1, y = From, fill = Lith))  
  geom_rect(aes(xmin = 0.5, xmax = 1.5, ymax = From, ymin = To))  
  scale_y_reverse(name = 'Depth')  
  scale_x_continuous(limits = c(0, 2), name = NULL)  
  theme_minimal(base_size = 16)  
  theme(axis.text.x = element_blank())  
  scale_fill_brewer(palette = 'OrRd')

Data taken from question in reproducible format

df <- structure(list(From = c(0L, 25L, 50L, 75L, 100L, 125L, 150L, 
175L), To = c(25L, 50L, 75L, 100L, 125L, 150L, 175L, 200L), Lith = c("A", 
"B", "C", "B", "C", "A", "B", "C")), class = "data.frame", 
row.names = c(NA, -8L))

CodePudding user response：

Use geom_rect. You have the min and max heights so it's fairly easy.

x <- 'From    To  Lith
0   25  A
25  50  B
50  75  C
75  100 B
100 125 C
125 150 A
150 175 B
175 200 C'
df1 <- read.table(textConnection(x), header = TRUE)

library(ggplot2)

ggplot(df1, aes(fill = Lith))  
  geom_rect(aes(xmin = 0, xmax = 1, ymin = From, ymax = To))  
  scale_x_continuous(breaks = NULL)  
  theme(aspect.ratio = 8)

^{Created on 2022-09-27 with reprex v2.0.2}