Process revdeps

[DavisVaughan]

abjutils

Nothing to fix

https://github.com/abjur/abjutils/issues/15

'::' or ':::' import not declared from: ‘lifecycle’

This package Suggests furrr, but then does two things incorrectly:

• pvec() is deprecated, but still calls furrr::future_map() outright with no checks

• The warning, however, comes from this linked line where future_options() are imported from furrr without any checks through the use of a global override. This means that their copy of future_options() is checked by R CMD Check for imports, and both lifecycle and ellipsis get flagged https://github.com/abjur/abjutils/blob/cc3ec0c3474a4d29820e5dde4322799cdb0317b0/R/pvec.R#L91

baguette

https://github.com/HenrikBengtsson/globals/issues/59

disk.frame

Nothing to fix

It looks like disk.frame:::setup_disk.frame() failed to start its workers. I would guess that this is separate from furrr, since that function doesn't use anything from furrr. Might just be a random fluke.

Update: This one went away on a second round of revdeps, so I'll call it a fluke

photosynthesis

Nothing to fix

Issue with dev magrittr and piping into a return() call. Lionel says it is an intended change.

library(magrittr)

x <- 1:5

foobar <- function(x) {
  x %>% return()
}

foobar(x)

simhelpers

Nothing to fix

Running this example with and without parallel doesn't break when I do it locally, so not sure what is happening here

library(simhelpers)
library(furrr)

df <- data.frame(
  n = 3:5,
  lambda = seq(8, 16, 4)
)

#plan(multisession, workers = 2)

evaluate_by_row(df, rpois)

The error that occurs on revdepcheck is:

Error in ...furrr_fn(...) : could not find function "...furrr_fn"

It persists through a second round of revdeps so it is worth looking at again

Update: I can reproduce this with CRAN globals. It is the problem where rpois gets remapped to ...furrr_fn and isn't found. It works with dev globals. I am not sure why revdepcheck didn't use the dev version, since it is in Remotes