How can I convert a data-masked variable name to a string (in R)-CodePudding

I am writing a function (in R) that I'd like to use data masking on so that the variable name can be passed in easily. I have read Programming with dplyr and understand how to use my variable inside the function by embracing it. However, I also want to do a join, which requires a string to be passed to it.

The following code is a MWE of what I'm trying to do, and it works. However, for the join, I have this hard-coded, and won't work if I want to join by a different variable.

How can I use data-masking to get at the variables easily, but then convert it to a string so it can be used in the join?

Thank you!

dat1 <- tibble::tibble(dat1.v1 = 1:10, dat1.v2 = 101:110)
dat2 <- tibble::tibble(dat2.var1 = 1:10, dat2.var2 = 1001:1010)
my.func <- function(df1, df2, my.var){
  df1 <- df1 %>%
    mutate("{{my.var}}.plus.one" := {{my.var}}   1)
  left_join(df2, df1, by=c("dat2.var1" = "dat1.v1"))
}
my.func(dat1, dat2, dat1.v1)

CodePudding user response：

I don't know of a clean way to do it natively using the by argument in a join. This approach works for me.

library(tidyverse)
dat1 <- tibble::tibble(dat1.v1 = 1:10, dat1.v2 = 101:110)
dat2 <- tibble::tibble(dat2.var1 = 1:10, dat2.var2 = 1001:1010)

my_func <- function(df1, df2, my_var){
  df2 <- df2 %>% 
    rename({{ my_var }} := dat2.var1)
  df1 <- df1 %>% 
    mutate("{{ my_var }}.plus.one" := {{ my_var }}   1) %>% 
    right_join(df2)
  df1
}
my_func(dat1, dat2, dat1.v1)
#> Joining, by = "dat1.v1"
#> # A tibble: 10 × 4
#>    dat1.v1 dat1.v2 dat1.v1.plus.one dat2.var2
#>      <int>   <int>            <dbl>     <int>
#>  1       1     101                2      1001
#>  2       2     102                3      1002
#>  3       3     103                4      1003
#>  4       4     104                5      1004
#>  5       5     105                6      1005
#>  6       6     106                7      1006
#>  7       7     107                8      1007
#>  8       8     108                9      1008
#>  9       9     109               10      1009
#> 10      10     110               11      1010