使用转换 dplyr 使用变量的字符串内容来引用数据帧中的列

Question

I would like to use the variable "second_column" as a variable to refer to the column "test1$b". I tried different things, but couldn't find any solutions yet. test1 and test2 should be the same in the end...

test1 <- data.frame(a = 1:5, b = 6:10)

second_column = "b"

test2 <- test1 %>% 
  transmute(variable1 = a,
         variable2 = second_column
  )

test1
test2

Answer 1

I would suggest using !! from rlang and as.name() . These two elements are useful for the kind of evaluation you want in order to create your variables:

#Data
test1 <- data.frame(a = 1:5, b = 6:10)
#Var name
second_column = "b"
#Data 2
test2 <- test1 %>% 
  transmute(variable1 = a,
            variable2 = !!as.name(second_column)
  )
#Dataframes
test1
test2

Output:

test1

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

test2

  variable1 variable2
1         1         6
2         2         7
3         3         8
4         4         9
5         5        10

Answer 2

You can use .data pronoun :

library(dplyr)

second_column = 'b'
test2 <- test1 %>% 
          transmute(variable1 = a,
                    variable2 = .data[[second_column]])
test2

#  variable1 variable2
#1         1         6
#2         2         7
#3         3         8
#4         4         9
#5         5        10

Answer 3

We can just use base R for this

data.frame(variable1 = test1[[1]], variable2 = test1[[second_column]])
#   variable1 variable2
#1         1         6
#2         2         7
#3         3         8
#4         4         9
#5         5        10

---

The operation showed in the OP's post needs only select and select can take both quoted or unquoted

library(dplyr)
test1 %>% 
    select(variable1 = a, variable2 = second_column)
#    variable1 variable2
#1         1         6
#2         2         7
#3         3         8
#4         4         9
#5         5        10

Or using transmute with across

library(stringr)
test1 %>%
    transmute(across(c(a, second_column))) %>%
    rename_with(~ str_c('variable', seq_along(.)))
#  variable1 variable2
#1         1         6
#2         2         7
#3         3         8
#4         4         9
#5         5        10