Problems creating plot with boxcox using rapache module function

Question

I'm trying to create a plot using boxcox function from package MASS.

but it's creating an rapache error.

The r code:

<%
  csvDF<- read.csv(GET$name1, header=TRUE)
  a<-lm(csvDF[,GET$col_variable]~1)
  require(MASS)

  filename1 <- paste(tempfile(tmpdir='/var/www/images'), '.png', sep='')
  png(filename1)
  bx<-boxcox(a)
  dev.off() 
%>

**GET$name1 is the csv data file address. **GET$col_variable is the variables column.

When I lose the "bx<-boxcox(a)" line the error disappear, so I guess that the boxcox causes the error.

Here are the rapache errors:

RApache Warning/Error!!!

Error in eval(expr, envir, enclos) : object 'csvDF' not found
RApache Warning/Error!!!

In addition:
RApache Warning/Error!!!

Warning messages:
RApache Warning/Error!!!

1: In readLines(icon, 1) : incomplete final line found on '/var/www/brew/sampleplan/step5_box_cox.php'
RApache Warning/Error!!!

2: In readLines(icon, 1) : incomplete final line found on '/var/www/brew/sampleplan/step5_box_cox.php'
RApache Warning/Error!!!

3: In readLines(icon, 1) : incomplete final line found on '/var/www/brew/sampleplan/step5_box_cox.php'
RApache Warning/Error!!!

4: In readLines(icon, 1) : incomplete final line found on '/var/www/brew/sampleplan/step5_box_cox.php'
RApache Warning/Error!!!

5: In readLines(icon, 1) : incomplete final line found on '/var/www/brew/sampleplan/step5_box_cox.php'
RApache Warning/Error!!!

Function brew returned an object of 'try-error'. Returning HTTP response code 500. 

I will be gratefull for any suggestion.

Answer 1

It is very difficult to give a complete answer because your whole setup isn't available. The error message (as opposed to the warnings; worry about them later) is the variable csvDF isn't found. It is unclear whether this error happens before or after you call read.csv. Either way, the problem isn't the call to boxcox.

Also note that lm has a data argument that could make your code clearer. Try something like

lm_formula <- as.formula(paste(col_variable, "1", sep = "~"))
a <- lm(lm_formula, data = csvDF)

You would also benefit from separating out code that reads data, calculates statistics, creates plots and writes plots to file.