Complete/expand with additional combinations

[danielsjf]

Is it possible to implement a way to include permutations using more than just the options provided in the data frame?

Now it works like this:

> data.frame(a=c(1,2),b=c(3,4)) %>%
    expand(a,b)
Source: local data frame [4 x 2]
      a     b
  (dbl) (dbl)
1     1     3
2     1     4
3     2     3
4     2     4

But it would be nice to have something like this:

> data.frame(a=c(1,2),b=c(3,4)) %>%
    expand(a,b=c(3:5))
Source: local data frame [6 x 2]
      a     b
  (dbl) (dbl)
1     1     3
2     1     4
3     1     5
4     2     3
5     2     4
6     2     5

The exact same problem holds for complete.

[hadley]

Hmmm, I'm not sure it's possible to make this work in expand() because it currently (for good reason), has select-like semantics, and there's no reliable way to tell the difference between these two calls:

df %>% expand(b = a)
a <- 1:4
df %>% expand(b = a)

I think that suggests we need another variant on expand(). I suspect we also need expand_grid() which doesn't use data at all (I still find myself reaching for expand.grid() when I want to create completely new datasets)