sum of rows for every numeric column and sum of rows contain a certain string for every numeric column as two rows in the dataset?

Please see sample dataset below.

mydata = data.frame (type =c ("fixed A", "fixed B", "fixed C", "fixed D", "fixed E", "variable A", "variable B", "variable C", "variable D", "variable E"),
                     cost =c(234,34,53,256,45,345,24,26,78,56))

Im trying to do the following:

create "TOTAL" row across all numeric columns summing all rows.
Create another "Total Variable" row across all numeric column summing only the rows that contain the word "variable" in them.

For number 1, I have used the following code:

mydata <- mydata %>%
  adorn_totals("row") %>%
  mutate(across(where(is.numeric),(abs)))

For number 2, I tried to use the following code to filter out the "Variable" strings, create a sum row and then merge it back to the original "mydata" dataset, but keep getting the following error:

x <- mydata %>%
  filter(grepl("variable",type))%>%
    adorn_totals("row")

error: Error in adorn_totals(., "row") : trying to re-add a totals dimension that is already been added

I would ideally like the following output:

     type     cost
    fixed A        234
    fixed B        34
    fixed C        53
    fixed D        256
    fixed E        45
    TOTAL          1151

 variable A        345
 variable B        24
 variable C        26
 variable D        78
 variable E        56
Total Variable     529

Any help would be really appreciated - thank you!

Solution

The following seems to work. I don't love the idea of creating multiple different datasets, but this seems to solve the error I was getting (Error in adorn_totals(., "row", name = "Total Variable") : trying to re-add a totals dimension that is already been added)


x <- mydata %>%
  adorn_totals("row") %>%
  mutate(across(where(is.numeric),(abs))) %>%
  filter(grepl("fixed|Total",type))

x1 <- mydata %>%
  filter(grepl("variable",type))%>%
  adorn_totals("row", name = "Total Variable") 

res <- rbind(x,x1)