add_total function for dplyr-pipes

add_total.R

add_total <- function(df, grp_vars, sum_vars, label='Total', ...) {
  
  if(missing(grp_vars) & !missing(sum_vars)) {
    total =
      df %>%
      group_by() %>%
      summarise(across({{sum_vars}}, sum),
                across(!{{sum_vars}} & where(is.factor), ~factor(label)),
                across(!{{sum_vars}} & where(is.character), ~label),
                .groups='drop')
    res = df %>%
      mutate(across(!{{sum_vars}} & where(is.factor), ~forcats::fct_expand(., label))) %>%
      bind_rows(total)
    
    
  } else if(!missing(grp_vars) & !missing(sum_vars)) {
    total =
      df %>%
      group_by(across({{grp_vars}})) %>%
      summarise(across({{sum_vars}}, sum),
                across(!{{sum_vars}} & where(is.factor), ~factor(label)),
                across(!{{sum_vars}} & where(is.character), ~label),
                .groups='drop')
    
    res = df %>%
      mutate(across(c(!{{grp_vars}}, !{{sum_vars}}) & where(is.factor), ~forcats::fct_expand(., label))) %>%
      bind_rows(total)
    
  } else if(missing(grp_vars) & missing(sum_vars)) {
    smry_funs = enquos(...)
    
    total =
      df %>%
      summarise(!!!smry_funs,
                across(where(is.factor), ~factor(label)),
                across(where(is.character), ~label),
                .groups='drop')
    
    res = df %>% bind_rows(total)
    
  } else if(!missing(grp_vars) & missing(sum_vars)) {
    smry_funs = enquos(...)
    
    total =
      df %>%
      group_by(across({{grp_vars}})) %>%
      summarise(!!!smry_funs,
                across(where(is.factor), ~factor(label)),
                across(where(is.character), ~label),
                .groups='drop')

    res = df %>% bind_rows(total)
    
  }
  
  return(res)
}

# Tests
df = 
  mtcars %>% 
  as_tibble() %>% 
  mutate(across(c(cyl, vs), factor)) %>% 
  group_by(cyl, vs) %>% 
  summarise(n = n(), mmpg = mean(mpg), .groups = 'drop')
  
df %>% add_total(sum_vars = n)

df %>% add_total(grp_vars = cyl, sum_vars = n)

df %>% add_total(mmpg = weighted.mean(mmpg, n), n = sum(n))

df %>% add_total(grp_vars = cyl, mmpg = weighted.mean(mmpg, n), n = sum(n))