Suppose I have some data that looks like this:

library(tidyverse)

df <- tibble(x = c(0, 2, 5),
             y = c(1000, 1, 1))

df |> 
    ggplot(aes(x,y)) + 
    geom_line()

I want to zoom in on a region based on the x value. But when I try to use coord_cartesian like this

df |> 
    ggplot(aes(x,y)) + 
    geom_line() +
    coord_cartesian(xlim = c(1.5, NA))

the y axis does not automatically adjust. It gives a lot of empty space like this

If I filter the data before plotting, then I’ll lose the meaningful diagonal segment. This image needs to be automatically generated, so I cannot manually specify the ylim in coord_cartesian.