tibble() should silently drop NULL columns

[hadley]

To be consistent with tidyverse/design#24

And it would also make tibble(x = 1, y = NULL) return the same as:

df <- tibble(x = 1)
df$y <- NULL

[krlmlr]

What should tibble(a = 1, a = NULL) do?

[hadley]

I think the current behaviour of throwing an error about duplicated columns is probably correct.

[krlmlr]

According to the principles a = NULL would be treated as absent.?

[hadley]

The principles aren't set in stone yet.

[hadley]

The question is really about which principle (names must be unique; NULL columns are silently dropped) should be applied first. I don't have a strong argument either way, so do whatever is easiest to implement.

[github-actions]

This old thread has been automatically locked. If you think you have found something related to this, please open a new issue and link to this old issue if necessary.