Error in `$<-.data.frame`(x, name, value) : replacement has 1 row, data has 0

Question

When running this code

projectile <- data.table()
projectile$angle<- c(15, 30, 45, 60, 75)  
projectile$distance <- c(5.1, 8, 10,  8.5, 4.8)
ggplot() + geom_point( aes(projectile$angle, projectile$distance)  ) # Works
ggplot(projectile) + geom_point( aes(angle,distance) ) # Does not work

I get this error:

Error in `$<-.data.frame`(x, name, value) : 
  replacement has 1 row, data has 0

Can anyone explain what this error means, why I have it and how to fix it, please?

UPDATE: the same error occurs at earlier step if I use data.frame:

projectile <- data.frame()
projectile$angle<- c(15, 30, 45, 60, 75) 

Error:

Error in `$<-.data.frame`(`*tmp*`, angle, value = c(15, 30, 45, 60, 75 : 
  replacement has 5 rows, data has 0

The reason why I dont create data.frame(angle=c(..), distance=c(..) right away is obviously that these values are NOT known until the user enters them later.

Answer 1

It could be related to the way the dataset was created. If we reconvert it to data.table, it would work

ggplot(setDT(setDF(projectile))) + 
       geom_point( aes(angle,distance) )

---

The data.table was created as a NULL object and then assigned columns. Instead, it can be done

projectile <- data.table(angle = c(15, 30, 45, 60, 75) , distance =  c(5.1, 8, 10,  8.5, 4.8)) 

---

As we are assigning on a NULL object, the attributes for row.names are of length 0

row.names = integer(0)

When we construct the dataset in a proper way, it would show

row.names = c(NA, -5L)

in the dput output

---

If we want to start with a NULL object, use a list, then at the end, convert it to data.frame or data.table. In that way, it will add the attributes for data.frame/data.table. Note that data.frame is a list with elements (i.e. columns) of equal length

projectile <- list()
projectile$angle<- c(15, 30, 45, 60, 75) 
projectile$distance <- c(5.1, 8, 10,  8.5, 4.8)
projectile <- data.frame(projectile)

Answer 2

Try this:

#Empty data table
projectile <- data.table()
#Add variables
projectile <- projectile[,list(angle=c(15, 30, 45, 60, 75),
                            distance=c(5.1, 8, 10,  8.5, 4.8)),]
#Plot                                 
ggplot(plotdat) + geom_point( aes(angle,distance) )

Output: