sparklyr + rlang: passing strings to rename a tbl_spark object

[hesi]

Hi Everyone and @javierluraschi,

I want to rename columns in a spark table, but I need the flexibility to pass arbitrary names.
Can someone help with Attempt 3 in the reprex below, please?

library(sparklyr)
library(rlang)
#> 
#> Attaching package: 'rlang'
#> The following object is masked from 'package:sparklyr':
#> 
#>     invoke

options(sparklyr.java9 = TRUE)
sc <- spark_connect(master = "local")

data <- dplyr::select(mtcars, mpg, cyl, disp) %>% head(5)
remote_tbl <- dplyr::copy_to(dest = sc, 
                             df = data, 
                             name = "remote_tbl", 
                             overwrite = TRUE)
remote_tbl
#> # Source: spark<remote_tbl> [?? x 3]
#>     mpg   cyl  disp
#>   <dbl> <dbl> <dbl>
#> 1  21       6   160
#> 2  21       6   160
#> 3  22.8     4   108
#> 4  21.4     6   258
#> 5  18.7     8   360

# Attempt 1:  -------------------------------------------------------------
# Standard renaming works fine
x <- c(mpg_version2 = "mpg")
dplyr::rename(remote_tbl, !!!x)
#> # Source: spark<?> [?? x 3]
#>   mpg_version2   cyl  disp
#>          <dbl> <dbl> <dbl>
#> 1         21       6   160
#> 2         21       6   160
#> 3         22.8     4   108
#> 4         21.4     6   258
#> 5         18.7     8   360

# Attempt 2:  -------------------------------------------------------------
# This also works fine
fn <- function(data, new_name) {
  data %>% dplyr::rename(!!new_name := 'mpg')
}
fn(remote_tbl, new_name = "mpg_version2")
#> # Source: spark<?> [?? x 3]
#>   mpg_version2   cyl  disp
#>          <dbl> <dbl> <dbl>
#> 1         21       6   160
#> 2         21       6   160
#> 3         22.8     4   108
#> 4         21.4     6   258
#> 5         18.7     8   360

# Attempt 3:  -------------------------------------------------------------
# Any idea why this fails although Attempt 1 works?
# How can i make it work?
fn <- function(data, new_name) {
  x <- glue::glue("c({new_name} = 'mpg')") %>% 
    rlang::parse_expr() %>% 
    eval()
  
  dplyr::rename(data, !!x)
}

fn(remote_tbl, new_name = "mpg_version2")
#> Error: All arguments must be named

^{Created on 2020-01-17 by the reprex package (v0.3.0)}

Thank you

[hesi]

Just realised the error message was because of a typo in my original post. I had !! instead of !!!

fn <- function(data, new_name) {
  x <- glue::glue("c({new_name} = 'mpg')") %>% 
    rlang::parse_expr() %>% 
    eval()
  
  dplyr::rename(data, !!!x)
}

fn(remote_tbl, new_name = "mpg_version2")
#> # Source: spark<?> [?? x 3]
#>   mpg_version2   cyl  disp
#>          <dbl> <dbl> <dbl>
#> 1         21       6   160
#> 2         21       6   160
#> 3         22.8     4   108
#> 4         21.4     6   258
#> 5         18.7     8   360