Convert nested list to data frame while preserving column types

Question

Given a list, b01 , of lists: tp01 , tp02 , tp03 , where each tp stores a specific type of data, how do I extract the class of each tp ?

The goal is to return a single data frame that preserves the original variable's type (eg int, char). I was thinking of extracting the classes of each tp and storing them, then overwriting the data frame's classes with those stored classes.

In the example below, the resulting data frame has 3 variables that are considered Factor . I would like them to be considered int , char , int .

I tried the following without luck:

str(as.data.frame(lapply(b01,unlist)))
str(sapply(as.data.frame(lapply(b01,unlist)),class))
lapply(b01,unlist)
sapply(str(lapply(b01,unlist)),unlist)

MWE:

set.seed(1)
tp01 <- list(sample(seq(0,100),10))
set.seed(2)
tp02 <- list(sample(seq(0,100),10))
tp03 <- list(letters[c(sample(seq(0,26),10))])
b01 <- list(tp01, tp02, tp03)

df <- as.data.frame(matrix(unlist(lapply(all,unlist)),
    nrow=length(unlist(b01[1]))*length(all)))
> df
   V1 V2 V3
1  26  q 18
2  37  j 70
3  56  t 56
4  89  c 16
5  19  g 91
6  86  y 90
7  97  w 12
8  62  v 78
9  58  r 43
10  5  b 50
11 18 26  q
12 70 37  j
13 56 56  t
14 16 89  c
15 91 19  g
16 90 86  y
17 12 97  w
18 78 62  v
19 43 58  r
20 50  5  b

str(df)
'data.frame':   20 obs. of  3 variables:
 $ V1: Factor w/ 19 levels "12","16","18",..: 5 6 10 16 4 15 19 12 11 8 ...
 $ V2: Factor w/ 20 levels "19","26","37",..: 15 14 17 12 13 20 19 18 16 11 ...
 $ V3: Factor w/ 20 levels "12","16","18",..: 3 7 6 2 10 9 1 8 4 5 ...

Answer 1

You can do this with lapply and as.data.frame:

names(b01) <- c("tp01","tp02","tp03")  #optional
first.step <- lapply(b01, unlist) 
second.step <- as.data.frame(first.step, stringsAsFactors = F) 

See unlist() documentation for info on coercion with that step. The stringsAsFactors = F argument will preserve the character vector in the second step.