furrr with carrier::crate and globals = FALSE has stopped working

[alping]

Following the article here https://furrr.futureverse.org/articles/carrier.html, I cannot get furrr to work with carrier::crate and globals = FALSE. This has worked previously, but unfortunately I don't know what has changed (e.g. updated packages). Running the crated function by itself or through furrr with globals = TRUE works. Setting globals = FALSE gives an error.

Tested with R=4.4.1/R=4.5.0, furrr=0.3.1, and carrier=0.1.1.

Code

# From the example in the article

constant <- 1.67

crt5 <- crate(
  constant = constant,
  function(x) {
    x + constant
  }
)

crt5(2:5) # Works

plan(multisession, workers = 2)

opts <- furrr_options(globals = FALSE)

x <- list(1:10, 11:20)

future_map(x, crt5, .options = opts) # Gives error

opts <- furrr_options(globals = TRUE)

future_map(x, crt5, .options = opts) # Works

Error

Error in (function (.x, .f, ..., .progress = FALSE)  :
  ℹ In index: 1.
Caused by error in `...furrr_fn()`:
! object 'constant' not found

Session info

> sessionInfo()
R version 4.5.0 (2025-04-11 ucrt)
Platform: x86_64-w64-mingw32/x64
Running under: Windows 10 x64 (build 19045)

Matrix products: default
  LAPACK version 3.12.1

locale:
[1] LC_COLLATE=English_United Kingdom.utf8 
[2] LC_CTYPE=English_United Kingdom.utf8
[3] LC_MONETARY=English_United Kingdom.utf8
[4] LC_NUMERIC=C
[5] LC_TIME=English_United Kingdom.utf8

time zone: Europe/Stockholm
tzcode source: internal

attached base packages:
[1] stats     graphics  grDevices datasets  utils     methods   base

other attached packages:
[1] furrr_0.3.1   future_1.40.0 carrier_0.1.1

loaded via a namespace (and not attached):
 [1] digest_0.6.37     codetools_0.2-20  magrittr_2.0.3    parallel_4.5.0
 [5] lifecycle_1.0.4   cli_3.6.5         parallelly_1.43.0 vctrs_0.6.5
 [9] renv_1.1.4        compiler_4.5.0    purrr_1.0.4       globals_0.18.0
[13] tools_4.5.0       listenv_0.9.1     rlang_1.1.6

[alping]

After some additional testing, it seems that this bug is related to something happening in the future package between versions 1.34.0 and 1.40.0. Reverting back to future=1.34.0 fixes the problem.

[alping]

Works again after the release of future=1.49.0.