CODEWITHSUNDEEP
r
NelsonGon
NelsonGon

Reputation: 13319

Prevent Factor Conversion to Levels

I'm trying to use lapply and do.call to combine lists into a data.frame object. However, how do I get to prevent factor conversion to levels within the resulting object? Illustration:

dummy<-list(list(data.frame(A=c("Boy","He","Is","Cool"),
                            stringsAsFactors = T)),
                  list(data.frame(B=c("Boy","Not","So","Cool"),
                                  stringsAsFactors = T)))

test1<-lapply(dummy,"[")
as.data.frame(do.call(cbind,test1))

Issue:

       V1         V2
1 1, 3, 4, 2 1, 3, 4, 2

Ignoring the nature of the output, how can I keep the original "values" ie Boy,cool,etc? Thanks! Desired Output: 

A           B
contents    contents

I cannot know the number of lists beforehand. I've tried this but it returns NAs:

plyr::ldply(test1,function(ele) do.call(cbind,ele))

Upvotes: 0

Views: 63

Answers (1)

Anders Ellern Bilgrau
Anders Ellern Bilgrau

Reputation: 10223

Absolutely no conversion has been done:

out <- as.data.frame(do.call(cbind,test1))
out[[1]]
#[[1]]
#     A
#1  Boy
#2   He
#3   Is
#4 Cool

But I find your question unclear since you write "Ignoring the nature of the output". And you explicitly create factors in to data.frame, so I'm unsure exactly what you expected. Remember that factors in R are integer vectors with labels. That is what you are seeing in your print.

Edit:

test1 <- lapply(dummy, "[[", 1)
as.data.frame(do.call(cbind, test1))
#     A    B
#1  Boy  Boy
#2   He  Not
#3   Is   So
#4 Cool Cool

Edit2 From discussion in the comments, does any of these work?

test1 <- lapply(dummy, unlist)
as.data.frame(do.call(cbind, test1))
# Same output as above

Alternatively:

data.frame(unlist(test1, recursive = FALSE))
# Same output as above

Upvotes: 2

Related Questions