Pass dynamic column name to transform() R function

Question

I have a data frame with multiple columns and I would like to pass dynamic column names to the R transform() function while running in loop. Request

For example

df1 <- data.frame(col1=c(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16),
                  col2=c("id1", "id1", "id1", "id1", "id1", "id1", "id2", "id3", "id3", "id4", "id4", "id4", "id5", "id6", "id7", "id7"), 
                  col3=c(1, 1, 1, 2, 3, 4, 2, 4, 2, 1, 4, 5, 1, 1, 2, 3),
                  col4=c(3, 2, 2, 2, 2, 1, 1, 1, 3, 3, 3, 1, 1, 1, 3, 3))


newColN <- "newCol"
colName <- "col3"

transform(df1, newColN = ~colName)

The second argument to the transform() should take dynamic column names for what ever function I pass.

I tried the following

transform(df1, as.name(newColN) = -(as.name(colName)))

transform(df1, !!newColN := -(!!colName)
+ )

transform(df1, !!as.name(newColN) := -(!!as.name(colName)))

but see an error.

Any suggestion to achieve this efficiently would be appreciated.

Answer 1

transform is not really the right tool to use if that is what you want to do. Suggest that, instead, you use any of these

replace(df1, newColN, -df1[[colName]])

df2 <- df1
df2[[newColN]] <- - df2[[colName]]

within(df1, assign(newColN, - df1[[colName]]))

within(df1, assign(newColN, - get(colName)))

If you don't need to dynamically define the variable on the left hand side then transform can still work:

transform(df1, newCol = - df1[[colName]])

transform(df1, newCol = - environment()[[colName]])

transform(df1, newCol = - get(colName))

Answer 2

You might want to change your transform to mutate in dplyr using which it would be simpler to perform such operations.

library(dplyr)
library(rlang)

df1 %>% mutate(!!newColN := get(colName))

#   col1 col2 col3 col4 newCol
#1     1  id1    1    3      1
#2     2  id1    1    2      1
#3     3  id1    1    2      1
#4     4  id1    2    2      2
#5     5  id1    3    2      3
#6     6  id1    4    1      4
#7     7  id2    2    1      2
#8     8  id3    4    1      4
#9     9  id3    2    3      2
#10   10  id4    1    3      1
#11   11  id4    4    3      4
#12   12  id4    5    1      5
#13   13  id5    1    1      1
#14   14  id6    1    1      1
#15   15  id7    2    3      2
#16   16  id7    3    3      3

Or use sym

df1 %>%  mutate(!!newColN := !!sym(colName))