Minimal reproducible example:
I have a dataframe df with 3 columns, x, y, and z. I want to plot x and y by melting the first two columns of df and passing the melted dataframe to ggplot. I want to scale x by 10 to plot it on a secondary axis. Currently I do the following:

df <- data. Frame(Date = seq(as.Date("2000/1/1"), by = "month", length. Out = 6),
                    x = rep(c(0,1), 3),
                    y = runif(6, 0,10 ),
                    z = 1:6)
df_tmp <- df[, c("Date", "x", "y")]
df_tmp[["x"]] <- df_tmp[["x"]] * 10
df_melt <- df_tmp |>
            reshape2::melt(id.vars = "Date", 
                 variable.name = "Variable")

This works but seems clumsy – I really don’t want to create the intermediate dataframe df_tmp. How can I rewrite this using the Base R (4.3.3) pipe so that I go from df to df_melt with x scaled as I want but without modifying df?

I’m looking for something that looks like:

df_melt <- df[, c("Date", "x", "y")] |>
            ???[["x"]] * 10 |>
            reshape2::melt(id.vars = "Date", 
                 variable.name = "Variable")

Many thanks in advance for your help

Thomas Philips