Feature Request adorn_subtotals

[annlaurawalker]

Feature requests

I've used adorn_totals for months now but I'm wondering if it's possible to have an adorn_subtotals function to get subtotals of groups?

[ggrothendieck]

Subtotals can be done like this; however, one feature that would make it easier would be if the totals excluded total rows. Note that we had to use filter to exclude the junk rows since it would otherwise total the total rows. Also sometimes one wants the totals at the top rather than the bottom of the group. That would also be a nice feature addition.

Data <-
  structure(list(CT = c("1", "1", "1", "1", "1", "1", "1", "1", 
"1", "1", "1", "1", "2", "2", "2", "2", "2", "2", "2", "2", "2", 
"2"), SCT = c("1", "1", "1", "1", "1", "1", "2", "2", "2", "2", 
"2", "2", "1", "1", "1", "1", "2", "2", "2", "2", "2", "2"), 
    SSCT = c("1", "2", "3", "1", "2", "3", "1", "2", "1", "2", 
    "1", "2", "1", "2", "1", "2", "1", "2", "3", "1", "2", "3"
    ), Category = c(111L, 112L, 113L, 111L, 112L, 113L, 121L, 
    122L, 121L, 122L, 121L, 122L, 211L, 212L, 211L, 212L, 221L, 
    222L, 223L, 221L, 222L, 223L), Y = c(10L, 12L, 15L, 11L, 
    10L, 13L, 21L, 22L, 20L, 25L, 23L, 24L, 31L, 33L, 36L, 39L, 
    32L, 31L, 36L, 41L, 44L, 45L)), row.names = c(NA, -22L), class = "data.frame")

library(dplyr)
library(janitor)

Data %>% 
  group_by(CT, SCT, SSCT) %>%
  summarize(Y = sum(Y), .groups = "drop") %>%
  adorn_totals %>%
  group_by(CT) %>%
  group_modify(~ adorn_totals(.)) %>%
  group_by(CT, SCT) %>%
  group_modify(~ adorn_totals(.)) %>%
  ungroup %>%
  filter(rowSums(across() == "Total") <= 1)

## # A tibble: 17 x 4
##    CT    SCT   SSCT      Y
##    <chr> <chr> <chr> <int>
##  1 1     1     1        21
##  2 1     1     2        22
##  3 1     1     3        28
##  4 1     1     Total    71
##  5 1     2     1        64
##  6 1     2     2        71
##  7 1     2     Total   135
##  8 1     Total -       206
##  9 2     1     1        67
## 10 2     1     2        72
## 11 2     1     Total   139
## 12 2     2     1        73
## 13 2     2     2        75
## 14 2     2     3        81
## 15 2     2     Total   229
## 16 2     Total -       368
## 17 Total -     -       574

[masterofzion]

I'm interested in this as well. It'd be great if we were able to do subtotals and grand totals by group, though the example here is useful.

[sfirke]

It seems in-scope to me for tabyl and there seems to be enough interest in it. If someone wants to mock up an example for discussion, then create a pull request for it, I'd review.

[Chanelle280]

Hi, I'm also interested in using sub-totals. I'd like to help with a mock up, but would need to know what exactly this would entail as I've never done one.

I find I always use tabyl() combined with kable() and within the kable package it has pack_rows(). I wonder if this is something that could be reproduced for tabyl()?

Thanks.