How to specify a data frame column using command line arg?

Hi Jim,

Thanks for the quick reply. I was trying to be clear but I think maybe the
example was too simple. What I really have is a csv file with lots of
columns (let's say 26) with a header (say a,b,c,d,e,,,,,z). I want to
generate lots of scatter plots -- 'a' vs. 'b' and 'd' vs 'j' and so on and
so on. I don't want to create a separate program for every combination of
columns -- that would be hundreds of programs. I just want one program that
takes three args --
1) the csv file name
2) the name of the 'x-axis' column (e.g. 'd')
3) the name of the 'y-asix' column (e.g. 'j')

so can run:

scatter_plot.r sample01.csv a b
scatter_plot.r sample01.csv d j
.....

I can't figure out how to get  the column names from the args[] to the
aes(x=?????, y=?????).
There must be some kind of indirect  reference or eval() or substitute()
operator in R but I can't find it.

Anyway, thanks for taking a shot at this.

Best,

doug




On Sun, Jun 5, 2016 at 10:44 PM, Jim Lemon <drjimlemon at gmail.com> wrote:

Hi Doug,
I think this will work for you:

adl1<-read.csv("test.csv")
adl1[,"a"]
[1] 1 4 7

so, adl1[,args[1]] should get you the column that you pass as the
first argument.

Jim


On Mon, Jun 6, 2016 at 5:45 AM, Douglas Johnson <todojo at gmail.com> wrote:

I'm guessing this is trivial but I've spent two hours searching and
reading
FAQ, tutorial, introductory and 'idiom' documents and haven't found a
solution. All I want to do is select a data frame column at run time
using
a command line arg. For example, the "test.csv" file contains:

a,b,c
1,2,3
4,5,6
7,8,9

If I run "test.r a", I get [1,4,7].
If I run "test.r b", I get [2,5,8].

Here's sample code -- how I map args[1] to column/vector 'a' or 'b' or
'c'?

args <- commandArgs(trailingOnly=T)

adl1<-read.csv(file="test.csv",head=T,sep=",")

adl1$SOME-MAGIC-FUNCTION-OR-SYMBOL(args[1])


All I'm really trying to do is generate a bunch of scatter plots of
column
'x' vs. 'y' without writing a separate program for each pair.


Thanks,


Doug



--
http://www.dojopico.org <http://dojopico.org>

        [[alternative HTML version deleted]]

--
http://www.dojopico.org