Dplyr/tidyverse: rename_at `.funs` must contain one renaming function, not 4 - r

Question

I have the following test data:
library(tidyverse)
df <- tibble(
g1 = c(1, 1, 2, 2, 2),
g2 = c(a, a, a, b, b),
a = sample(5),
b = sample(5)
)
I would like to write a function that summarises grouped columns with a mean and I wish I could have the resulting columns prefixed with "mean_"
my_summarise1 <- function(df, group_var, summarise_var) {
df %>%
group_by_at(.vars = group_var) %>%
summarise_at(.vars = summarise_var, .funs= mean) %>%
rename_at(.vars= summarise_var, .funs=paste('mean_', .))
}
Without rename_at line it works fine, but with it throws error:
my_summarise1(df, vars(g1,g2),vars(a,b))
R responds with
Error: `.funs` must contain one renaming function, not 4
How should I effectively prefix the new column names?
Smaller question: is it possible to avoid vars() or quotes arount parameters
column names when calling a function?
Knowing these two small things would greatly enhance my code, thank you all very much in advance for help.

Answer 1

While the earlier answer by #docendodiscimus is more succinct, for what it's worth, there are two issues with your code:
You need to wrap the paste (better: paste0) function within funs.
You need to ungroup prior to renaming (see e.g. this post).
A working version of your code looks like this:
my_summarise1 <- function(df, group_var, summarise_var) {
df %>%
group_by_at(group_var) %>%
summarise_at(summarise_var, mean) %>%
ungroup() %>%
rename_at(summarise_var, funs(paste0('mean_', .)))
}
my_summarise1(df, vars(g1, g2), vars(a, b))
## A tibble: 3 x 4
# g1 g2 mean_a mean_b
# <dbl> <chr> <dbl> <dbl>
#1 1. a 2.50 2.50
#2 2. a 4.00 5.00
#3 2. b 3.00 2.50

Answer 2

If you want to take a simple route, you can use dplyr's way of adding suffixes to the summarised columns:
my_summarise1 <- function(df, group_var, summarise_var) {
df %>%
group_by_at(.vars = group_var) %>%
summarise_at(.vars = summarise_var, funs(mean=mean))
}
my_summarise1(df, vars(g1,g2), vars(a,b))
# A tibble: 3 x 4
# Groups: g1 [?]
g1 g2 a_mean b_mean
<dbl> <chr> <dbl> <dbl>
1 1. a 3.50 4.50
2 2. a 4.00 1.00
3 2. b 2.00 2.50
In this case, funs(mean=mean) tells dplyr to use the suffix mean and apply the function mean. For clarity, you could use funs(mysuffix = mean) to use any different suffix and apply the mean function.
Re OP's question in comment: you can use the following modification which doesn't require the use of vars when calling the function.
my_summarise2 <- function(df, group_var, summarise_var) {
df %>%
group_by_at(.vars = group_var) %>%
summarise_at(.vars = summarise_var, funs(mean=mean))
}
my_summarise2(df, c("g1","g2"), c("a","b"))