The code to generate sample dataset is as below:

z = data.frame(a = c(1:4),b = as.character(c(1:4)),c = c('s,','d','g','y'))

zz = data.frame(a = c(5:8),b = as.numeric(c(5:8)),e = c(2,4,5,6))

zzz = data.frame(a = c(9:12),b = as.numeric(c(9:12)),r = c(4,8,6,5))

z4 = list(z,zz,zzz)

lapply(z4,function(x)x %>% select(a,b)) %>% bind_rows

The output is

> lapply(z4,function(x)x %>% select(a,b)) %>% bind_rows
Error in `bind_rows()`:
! Can't combine `..1$b` <character> and `..2$b` <double>.
Run `rlang::last_error()` to see where the error occurred.

Is there anyway to use select and as.numeric at the same time?

CodePudding user response：

We may automatically convert the list elements to its appropriate type with type.convert

library(purrr)
library(dplyr)
map_dfr(type.convert(z4, as.is = TRUE), ~ .x %>%
          select(a, b))

-output

    a  b
1   1  1
2   2  2
3   3  3
4   4  4
5   5  5
6   6  6
7   7  7
8   8  8
9   9  9
10 10 10
11 11 11
12 12 12

This may not work in all the conditions i.e. if there is a column with non-numeric element, it remains as character. In that case, instead of converting the columns to numeric (which could return NA for non-numeric), convert all the columns to character, bind them and later apply type.convert

map_dfr(z4, ~ .x %>%
            select(a, b) %>%
            mutate(across(everything(), as.character))) %>%
    type.convert(as.is = TRUE)