R help - Mar 2009 - Using vectors of names in calls for functions

Dear R people,

it is a pleasure to try and use functionnalities that are in ggplot,
even with my poor capacities in R. so far, i'dlike to be able to use
these in my own function, and to

"be able to use it repeatedly from
vectors where I store the names of my columns."

I have understood that this is a very discussed topic, that many post
have been exhanged but I cannot understand what is the good way to
handle this and not to use this vectors of names ...

below is the idea of what I would like to have:

? ? ? ? ? ?V1_P2_plot<-function(df, x, y, z) {
? ? ? ? ? ? ? ?pp<-ggplot(df, aes(x = x, y= y , z = z ))
? ? ? ? ? ? ? ?pp<-pp + geom_tile(aes(fill= z))
? ? ? ? ? ? ? ?return(pp)
? ? ? ? ? ?}

that would allow this type of call :
j<-c("k", "l")
i<-"m"
m<-rnorm(25)
vsim<-cbind(expand.grid(k=1:5, l=1:5*10), m)

V1_P2_plot(df=vsim, x=j[1], y=j[2], z=i)

This is obviously not working, and I understand from the numerous discussion on
the R help list that this is related to references and calls.
I spend the whole afternoon on these topics and it did not allowed
me to understand the major principles. And most of my trials with
deparse(substitute()) and eval(parse()) where inefficient.


I was only able to have the following functionning :

qplot(get(j[1]),get(i),data=vsim, geom=c("line", "point"),
group=get(j[2]), colour=get(j[2]), xlab=j[1],ylab=i )

but this strategy fails as soon as I want to call it from my function.
and I also understood that the "get()" way and the vector of names are
not the good way to adress this.
but I was not able to use/generalise the examples so far discussed on
this list :

mytable <- function(x,y) eval.parent(substitute(table(x,y)))

logplot <- function(y, ylab = deparse(substitute(y))) {
                 ylab
                 y <- log(y)
                 plot(y, ylab = ylab)
                 }

to achieve my problem.

but i I would be very happy if you could help me on this.

Best regards

Pascal Boisson

Hi Pascal,

You need the aes_string function:

V1_P2_plot<-function(df, x, y, z) {
   ggplot(df, aes_string(x = x, y= y, fill = z)) +
    geom_tile()
}
j<-c("k", "l")
i<-"m"
m<-rnorm(25)
vsim<-cbind(expand.grid(k=1:5, l=1:5*10), m)

V1_P2_plot(df=vsim, x=j[1], y=j[2], z=i)

Hadley

On Thu, Mar 5, 2009 at 3:16 PM, Pascal Boisson <hoothootprod at gmail.com>
wrote:
> Dear R people,
>
> it is a pleasure to try and use functionnalities that are in ggplot,
> even with my poor capacities in R. so far, i'dlike to be able to use
> these in my own function, and to
>
> "be able to use it repeatedly from
> vectors where I store the names of my columns."
>
> I have understood that this is a very discussed topic, that many post
> have been exhanged but I cannot understand what is the good way to
> handle this and not to use this vectors of names ...
>
> below is the idea of what I would like to have:
>
> ? ? ? ? ? ?V1_P2_plot<-function(df, x, y, z) {
> ? ? ? ? ? ? ? ?pp<-ggplot(df, aes(x = x, y= y , z = z ))
> ? ? ? ? ? ? ? ?pp<-pp + geom_tile(aes(fill= z))
> ? ? ? ? ? ? ? ?return(pp)
> ? ? ? ? ? ?}
>
> that would allow this type of call :
> j<-c("k", "l")
> i<-"m"
> m<-rnorm(25)
> vsim<-cbind(expand.grid(k=1:5, l=1:5*10), m)
>
> V1_P2_plot(df=vsim, x=j[1], y=j[2], z=i)
>
> This is obviously not working, and I understand from the numerous
discussion on
> the R help list that this is related to references and calls.
> I spend the whole afternoon on these topics and it did not allowed
> me to understand the major principles. And most of my trials with
> deparse(substitute()) and eval(parse()) where inefficient.
>
>
> I was only able to have the following functionning :
>
> qplot(get(j[1]),get(i),data=vsim, geom=c("line",
"point"),
> group=get(j[2]), colour=get(j[2]), xlab=j[1],ylab=i )
>
> but this strategy fails as soon as I want to call it from my function.
> and I also understood that the "get()" way and the vector of
names are
> not the good way to adress this.
> but I was not able to use/generalise the examples so far discussed on
> this list :
>
> mytable <- function(x,y) eval.parent(substitute(table(x,y)))
>
> logplot <- function(y, ylab = deparse(substitute(y))) {
> ? ? ? ? ? ? ? ? ylab
> ? ? ? ? ? ? ? ? y <- log(y)
> ? ? ? ? ? ? ? ? plot(y, ylab = ylab)
> ? ? ? ? ? ? ? ? }
>
> to achieve my problem.
>
> but i I would be very happy if you could help me on this.
>
> Best regards
>
> Pascal Boisson
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>


-- 
http://had.co.nz/