dput and dget error

On Friday, December 23, 2016 3:24 PM, David Winsemius <dwinsemius at comcast.net> wrote:

 On Dec 23, 2016, at 12:45 PM, Carl Sutton via R-help 

<r-help at r-project.org> wrote:

 Merry Christmas and Happy Holidays

 I am attempting to use dput and or dget to send data along with a help 

request to another list (package specific).   Read the help pages for both and 
they appeared to be fairly simple functions.  Found an example on github and it 
appeared to be an easy task to replicate.   Alas, a copy and paste of the github 
example worked but my toy data example did not.

 #  dput and dget functions
 #  Example from github
 set.seed(1337)
 NN <- 10
 theData <- data.frame(Alpha = rnorm(NN),
                      Beta = rnorm(NN))
 theData$Gamma <- theData$Alpha * 2 + theData$Beta / 2 + rnorm(NN)
 dput(theData, "temporary_file")
 theDataReconstitutedAgain <- dget("temporary_file")
 print(theDataReconstitutedAgain)


 #  try it with my "toy data"library(data.table)

This is a guess based on what I know about data.table-objects and how they are 
represented with dput. You will see an .internal.selfref = <pointer: 
0x7f7fd18f2778>

But if you try to assign the value of the dput output to an R name, it's 
going to fail. You need to remove the .internal.selfref = <pointer: 
0x7f7fd18f2778> and then assign as a data frame and run setDT() on the name.

 library(tidyr)
 #  data table for melt and columns split
 dt1 <- data.table(a_1 = 1:10, b_2 = 20:29,folks = 

c("art","brian","ed",

 "rich","dennis","frank", 

"derrick","paul","fred","numnuts"),

 a_2 = 2:11, b_1 = 21:30)
 melted <- melt(dt1, id = 

"folks")[,c("varType","varIndex") :=

tstrsplit(variable,"_")][,variable:=NULL]

Run this:

dput(melted)

 #  melted has 40 observations from stacking a and b variables
 #  which have lengths of 20 each
 #  here cometh the frustrtion
 dput(melted,"temp_file.txt")
 goGetIt <- dget("temp_file.txt", keep.source = FALSE)
 print(goGetIt)

 goGetIt <- dget("temp_file.txt", keep.source = FALSE)

 Error in parse(file = file, keep.source = keep.source) : 
 temp_file.txt:18:36: unexpected '<'

That's because the regular R parser doesn't recognize the pointer as 
valid R code. It was not designed to recognize data.tables. the tidyr functions 
are designed to return data.frames, but `dget` is not from a package that is 
"data.table-aware"

 17: "varIndex"), row.names = c(NA, -40L), class = 

c("data.table", 

 18: "data.frame"), .internal.selfref = <
 ^

 print(goGetIt)

 Error in print(goGetIt) : object 'goGetIt' not found

Try this instead:

dput(as.data.frame(melted),"temp_file.txt")
goGetIt <- dget("temp_file.txt", keep.source = FALSE)
setDT(goGetIt)
print(goGetIt)

-- 
David.

 the help page states dput just writes an ASCII text representation of an R 

object to a file or connection, or uses one to recreate the object.  Nothing 
there about not allowing a data frame or data table and the github example was a 
data frame.   The line numbers referenced do not appear to refer to my code (but 
maybe they do and I am ignorant of the meaning) and the code works up to the 
dget line.

 Typing dget in the console to see the inner workings of the function was 

not helpful.

 Also, I am unclear as to just how dput and dget work.  If I save a subset 

of actual data to an object, then do dput on that object, can I rely that 
whomever I send my code to (which includes the dput statement) be able to 
retrieve the data?  In other words, dput saves the data in the code file?

 Thanks for your help and may the holidays be wonderful for you and your 

loved one.

 Carl Sutton