How do I match rows within a group and filter the rows that do not agree?

In this example dataset, I have 3 groups with 2 different views and 2 different types. Here is what I want to achieve:

1. Determine whether the two types within a group match (Y or N)
2. If the types do not match, I want to select the row where Type = Y
3. If the types DO match, remove the duplicate so that I end up with a data frame with only one
   entry per group (Type = Y or Type = N)

Here is my example data:

structure(list(grp = c("1", "1", "2", "2", "3", "3"), view = c("A", 
"B", "A", "B", "A", "B"), type = c("Y", "N", "Y", "Y", "N", "N"
)), class = "data.frame", row.names = c(NA, -6L))

I want the resulting dataset to look like this:

structure(list(grp = c("1", "2", "3"), view = c("A", "A", "A"
), type = c("Y", "Y", "N")), class = "data.frame", row.names = c(NA, 
-3L))

Any help would be greatly appreciated.

Thank you so so much!

>Solution :

Using dplyr:

d1 %>% 
  group_by(grp) %>% 
  filter(
    n_distinct(type) == 1 & row_number() == 1 | 
    n_distinct(type) == 2 & type == 'Y'
  ) %>% ungroup()

Gives:

# A tibble: 3 × 3
  grp   view  type 
  <chr> <chr> <chr>
1 1     A     Y    
2 2     A     Y    
3 3     A     N 

Read as, within each group, keep the rows that satisfy either of these criteria:

• There is one type in this group, and this row is the first row (this is arbitrary, you didn’t specify which row to keep in this case)
• There are two types in this group, and this row has type Y