00:07:36 Olivier Leroy: r$> (if (FALSE) "bob")
NULL
00:07:48 Olivier Leroy: (dangerous still)
00:08:42 Eamon Caddigan: I was really surprised he didn’t suggest isTRUE to prevent `if` gotchas. I bet there’s a good reason and I haven’t figured out why.
00:10:29 Eamon Caddigan: Weird case_when thing I found: since it (reasonably) insists that all the outputs have the same type, you’ll need to specify the type of NA , e.g., with NA_character_ (at least this was true for earlier versions of dplyr, I haven’t poked at that in a while).
00:10:37 Olivier Leroy: Reacted to "I was really surpris..." with 👍
00:10:53 Stephan Koenig: Usually case_when I order from more specific first to general last.
00:11:02 Olivier Leroy: Reacted to "Usually case_when I ..." with 👍
00:11:04 Nicholas Giangreco: Reacted to "Usually case_when I ..." with 👍
00:11:57 Olivier Leroy: is quaro "Caution" svg is VLC Icon?
00:12:04 Olivier Leroy: quarto
00:13:03 Jon Harmon (jonthegeek): Replying to "Weird case_when th..."
I'm PRETTY sure this isn't required anymore. That's how I learned about NA types, though 😄
00:14:13 Eamon Caddigan: Reacted to "I'm PRETTY sure this..." with ➕
00:19:19 Olivier Leroy: for some reasons I always have an hard time remember seq_along ...
00:19:29 Retselisitsoe Monyake: Reacted to "for some reasons I a..." with 😅
00:25:37 Olivier Leroy: r$> ifelse(c(TRUE, FALSE), 1, "no")
[1] "1" "no"
00:26:12 Olivier Leroy: (doing the wrong stuff)
00:26:48 Jon Harmon (jonthegeek): Yeah, it coerces to the "lowest" common form (character can't go to numeric, but numeric can go to character).
00:27:39 Martin: numeric() = NULL?
00:27:39 Olivier Leroy: convert and recycle
00:27:56 Olivier Leroy: I think numeric 0
00:28:10 Olivier Leroy: r$> numeric()
numeric(0)
00:28:18 Martin: nice
00:28:46 Eamon Caddigan: Thanks Jacob this was great!
00:29:06 Stephan Koenig: You can also test with the %||% Operator to check if something is NULL I believe (R 4.4)
00:29:12 Olivier Leroy: Reacted to "Thanks Jacob this wa..." with 👋
00:29:17 Martin: Reacted to "You can also test wi..." with ❤️
00:29:57 Jon Harmon (jonthegeek): Reacted to "You can also test wi..." with ☝️
00:37:11 Martin Venegas: thank u!
00:37:39 Nicholas Giangreco: Reacted to "You can also test wi..." with ❤️
00:40:27 Martin Venegas: that's good to know