Combine two data by every 3rd row

I have the following two data frames:

df<- structure(t(data.frame(x=1,1,2,2,3,3,4,4,5,5,6,6,7,7,8,8,9,9,10,10)))
df2<- structure(data.frame(x= rep(30, times=10)))

I am looking to combine the two data frames so that one row of df2 is inserted into every third spot on df. So the final data frame should look like:

df3<- structure(t(data.frame(x=1,1,30,2,2,30,3,3,30,4,4,30,5,5,30,6,6,30,7,7,30,8,8,30,9,9,30,10, 10,30)))

This could be done by making a grouping variable for each data frame. Like:

df$group<- rep(1:10, times=2) %>% arrange(group)

df2$group<- rep(1:10)

And then merging based on the group, but the order of my rows in df is important and cannot be switched around.

Any help would be appreciated

Solution

We can do this by joining the two data frames together with a customer sort order.

Something like the following:

df<- structure(t(data.frame(x=1,1,2,2,3,3,4,4,5,5,6,6,7,7,8,8,9,9,10,10)))
df2<- structure(data.frame(x= rep(30, times=10)))

# current row orders
df = df %>%
  mutate(row_number = 1:nrow(df),
         source = "df")
df2 = df2 %>%
  mutate(row_number = 1:nrow(df2),
         source = "df2")

# prep customer sorters
df2 = df2 %>% mutate(row_number = 2 * row_number + 0.5)

# attend and sort
df3 = rbind(df, df2) %>%
  arrange(row_number)

# remove extra columns
df3 = df3 %>%
  select(-row_number, -source)

Most of the cleverness happens in the customer sorter: row_number = 2 * row_number + 0.5 of df2.

Adding 0.5 ensure that the df2 rows occur between df rows.
Multiplying by 2 means there will be two rows from df between each row in df2
Getting the current row orders means that row orders for df and df2 are preserved.