Dplyr: Need silent variant of do()

[hadley]

[romainfrancois]

I can handle the chunking of the data internally to limit the number of copies, i.e. similar to what we use internally with ShrinkableVector but for data frames. This could be useful in do as well.

[hadley]

Oooh, that would be nice!

[hadley]

I think we could either:

• have a do() that invisibly returns the input (so you can use it in a pipe). do_silent()?
• use a naming convention in do() so that e.g. do(mtcars,_= plot(.$displ. $wt) would not modify the output

[hadley]

Since do() is going away, this is no longer needed.

[statwonk]

Curious, what will replace do()?

[hadley]

Mutate + list-cols + purrr

[statwonk]

@hadley I've been thinking about your response "Mutate + list-cols + purrr", I'm not sure I fully understand, would you mind sharing a small example? It'd be appreciated.

[hadley]

This shows how all the pieces fit together: http://r4ds.had.co.nz/many-models.html