2

I have a sample data below 

data <- data.frame(yr=c(1999,2000,2001,2002,2003,2004,2005,2006,2007,2009,2010,2011,2012), 
                   ntemp =c(11,12,13,14,15,16,17,18,19,20,21,12,23))

When I try running this function, to access the variable inside a ggplot function.

FUN<-function(data, fun.y,yr) {
  fun.data <- data     
  ggplot(fun.data,aes(yr, fun.y))+geom_point()+scale_y_continuous(fun.y)    
}

FUN(data, "ntemp", yr)

I get an Error in eval(expr, envir, enclos) : object 'fun.y' not found

How can I solve this on R3.02?

tonytonov
  • 24,116
  • 16
  • 77
  • 94
Keniajin
  • 1,579
  • 2
  • 21
  • 41
  • You are trying to map `fun.y`, which is supposed to be a variable in `fun.data`. Since it is not, you get an error. – tonytonov Jan 20 '14 at 09:47
  • I have passed fun.y in the function as ntemp – Keniajin Jan 20 '14 at 09:50
  • ```aes_string()``` is the way. If you would be interested in adding facets, here is a [question that migth help](http://stackoverflow.com/questions/19016007/metaprogramming-with-ggplot2). – marbel Jan 20 '14 at 10:06
  • If the proposed solution works for you, please do not forget to mark question as accepted. – tonytonov Jan 20 '14 at 10:23

1 Answers1

2

aes only looks at the variables in data argument. If you would like to pass variable as an argument to FUN by its character name, use aes_string:

FUN <- function(data, x, y) {
  ggplot(data, aes_string(x=x, y=y)) + geom_point()
}

FUN(data, y="ntemp", x="yr")

A small correction: variable inside aes call should be defined in the scope where the ggplot object is evaluated, so technically a variable is looked up in data first, then in global environment (by default). See this and this questions.

Community
  • 1
  • 1
tonytonov
  • 24,116
  • 16
  • 77
  • 94