[R] sourcearg function is there a better way already built into R

[Bert Gunter]

?pmatch
?switch

_may_ be of use...

But if the file calls it Fe and you want to label it "Iron," then
you'll need some sort of explicit dictionary.

-- Bert

On Tue, Mar 6, 2012 at 12:57 PM, Robert Sandefur <RSandefur at cam-llc.com> wrote:
> List and Jim:
>
> My solution works to my satisfaction so perhaps I should not have not ask. The problem was basically this:
> 1. I receive a CSV file containing assays on various chemical elements. The names of these fields are inconsistent in terms of spelling, length and case.
> 2. As an example I would like to do a cumulative frequency on each of the fields and have the plot labeled with the name of the field as well is operating on that field.
> 3. The reason I wrote sourcearg was to simultaneously have a command that would do the cum freq and label the plot
> i.e. sourcearg  generates and sources
> vuse=(TCU+0.01-DUPTCU+0.01)/(TCU+0.01+DUPTCU+0.01)/2
> qqnorm(vuse,main=\"Redacted TCUCumFreq
> ((TCU+0.01-DUPTCU+0.01)/(TCU+0.01+DUPTCU+0.01)/2)")
>
> Which gives me what I want but I can't embed the picture because of the restrictions on the mailing list.
>
> Thanx
>
> Bob
>
>
>
>
>
>
>
>
> -----Original Message-----
> From: jim holtman [mailto:jholtman at gmail.com]
> Sent: Tuesday, March 06, 2012 09:47
> To: Robert Sandefur
> Cc: r-help at r-project.org
> Subject: Re: [R] sourcearg function is there a better way already built into R
>
> The basic question is what do you want the function to do.  It seems
> convoluted and I am not sure exactly what you are giving it as input
> and what you expect as output. Not exactly sure of what the 'sink' and
> 'source' are supposed to do.  So "tell me what you want to do, not how
> you want to do it".
>
> On Tue, Mar 6, 2012 at 10:29 AM, Robert Sandefur <RSandefur at cam-llc.com> wrote:
>> Hi list:
>>
>>
>>
>> I work with a lot of laboratory analytical data and I often have
>> inconsistent names of files and variables within those files so I wrote
>> this sourcearg function to facilitate handling file and variable names
>> as both character and R names. The source of the function is given below
>>
>>
>>
>> sourcearg=function(arg){
>>
>> su=sprintf("%s\n",arg)
>>
>> sink("TMPTMPRarg")
>>
>> printf("%s\n",arg) # printf = cat(sprintf(....))
>>
>> sink()
>>
>> source("TMPTMPRarg")
>>
>> return(su)
>>
>> }
>>
>>
>>
>> A trival invocation is given below but the next time I get data the
>> names of assays might be Mg,MN,and Fe
>>
>> asyl=c("MG","MN")
>>
>> for(asy in asyl){
>>
>> j=read.csv(sprintf("%sDUPS.csv",asy),head=T)
>>
>> attach(j)
>>
>> printf("plot(%s,DUP%s,log=\"xy\")",asy,asy)
>>
>> sourcearg(sprintf("plot(%s+0.01,DUP%s+0.01,log=\"xy\",main=\"Redacted
>> %s\nOrg Log+0.01 vs Dup Log+0.01 Plot\" )",asy,asy,asy))
>>
>> print(names(j))
>>
>> ipsend()
>>
>> detach(j)
>>
>> }
>>
>>
>>
>> A more typical invocation would use tapply to get the names of all the
>> variables in the CSV file into asyl and then do the for loop.
>>
>>
>>
>> sourcearg works fine for my purposes but my question is:  is there a
>> better way to do this in R that is already built-in?
>>
>>
>>
>> Thanx
>>
>>
>>
>> Robert (Bob) L. Sandefur PE
>>
>>
>>
>> Senior Geostatistician / Reserve Analyst
>>
>>
>>
>> CAM
>>
>>
>>
>>
>>
>>
>>        [[alternative HTML version deleted]]
>>
>> ______________________________________________
>> 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.
>
>
>
> --
> Jim Holtman
> Data Munger Guru
>
> What is the problem that you are trying to solve?
> Tell me what you want to do, not how you want to do it.
>
> ______________________________________________
> 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.



-- 

Bert Gunter
Genentech Nonclinical Biostatistics

Internal Contact Info:
Phone: 467-7374
Website:
http://pharmadevelopment.roche.com/index/pdb/pdb-functional-groups/pdb-biostatistics/pdb-ncb-home.htm