Parallelize a function

Usage

parallelize_fun(
  x,
  fun,
  cores = 1,
  export_fun = NULL,
  clean_result = FALSE,
  throw_error = TRUE,
  progress_bar_width = 10L,
  timestamp_format = paste0("[", format(Sys.time(), "%Y-%m-%d %H:%M:%S"), "] "),
  verbose = TRUE,
  backend = c("auto", "fork", "psock"),
  timeout = Inf,
  total_timeout = Inf,
  seed = NULL
)

Arguments

x: A vector or list to apply over.
fun: The function to be applied to each element.
cores: The number of worker processes to use for parallelization. Default is 1.
export_fun: The functions to export the function to workers.
clean_result: Whether to remove failed results from output. If FALSE, failed results are kept as error objects. Default is FALSE.
throw_error: Whether to print detailed error information for failed results. Default is TRUE.
progress_bar_width: Width of the verbose progress bar in characters. Default is 10L.
timestamp_format: Format string for timestamp display. Default is "%Y-%m-%d %H:%M:%S".
verbose: Whether to print the message. Default is TRUE.
backend: Parallel backend. "auto" uses PSOCK on every platform. Use "fork" to opt in to forked workers on supported systems. "fork" is unavailable on Windows. Prefer PSOCK in long-lived sessions that also use child-process managers such as callr or processx, because they can compete with R's fork-worker signal handler.
timeout: Maximum number of seconds that a parallel worker task may run. Inf disables task timeouts. This is ignored when execution uses one core.
total_timeout: Maximum number of seconds allowed for the complete call. Inf disables the overall deadline. In single-core mode the deadline is checked between inputs but cannot interrupt a function that is already running.
seed: Optional integer seed. When supplied, every input receives a deterministic independent L'Ecuyer-CMRG random-number stream, making results reproducible across worker counts and scheduling order. The caller's random number state is restored when the call finishes.

Value

A list of computed results. If clean_result = FALSE, failed results are included as error objects. If clean_result = TRUE, only successful results are returned.

Examples

parallelize_fun(1:3, function(x) {
  Sys.sleep(0.2)
  x^2
})
#> ℹ [2026-08-04 13:51:58] Using 1 core
#> ⠙ [2026-08-04 13:51:58] Running for 1 [1/3] ■■■         33% | ETA:  0s
#> ⠹ [2026-08-04 13:51:58] Running for 2 [2/3] ■■■■■■      67% | ETA:  0s
#> ✔ [2026-08-04 13:51:58] Completed 3 tasks in 643ms
#> 
#> ℹ [2026-08-04 13:51:58] Building results
#> $`1`
#> [1] 1
#> 
#> $`2`
#> [1] 4
#> 
#> $`3`
#> [1] 9
#> 

parallelize_fun(list(1, 2, 3), function(x) {
  Sys.sleep(0.2)
  x^2
}, cores = 2)
#> ℹ [2026-08-04 13:51:59] Using 2 cores
#> ℹ [2026-08-04 13:51:59] Building results
#> [[1]]
#> [1] 1
#> 
#> [[2]]
#> [1] 4
#> 
#> [[3]]
#> [1] 9
#> 

# Examples with error handling
parallelize_fun(1:5, function(x) {
  if (x == 3) stop("Error on element 3")
  x^2
}, clean_result = FALSE)
#> ℹ [2026-08-04 13:52:00] Using 1 core
#> ℹ [2026-08-04 13:52:00] Building results
#> ! [2026-08-04 13:52:00] Found 1 failed result
#> Warning: Found 1 failed result
#> ℹ [2026-08-04 13:52:00] ✖ Error details:
#> ℹ                       ✖ Error on element 3 (1): "3"
#> $`1`
#> [1] 1
#> 
#> $`2`
#> [1] 4
#> 
#> $`3`
#> $error
#> [1] "Error on element 3"
#> 
#> $index
#> [1] 3
#> 
#> $input
#> [1] 3
#> 
#> attr(,"class")
#> [1] "parallelize_error"
#> 
#> $`4`
#> [1] 16
#> 
#> $`5`
#> [1] 25
#> 

parallelize_fun(1:5, function(x) {
  if (x == 3) stop("Error on element 3")
  x^2
}, clean_result = TRUE)
#> ℹ [2026-08-04 13:52:01] Using 1 core
#> ℹ [2026-08-04 13:52:01] Building results
#> ! [2026-08-04 13:52:01] Found 1 failed result
#> Warning: Found 1 failed result
#> ℹ [2026-08-04 13:52:01] ✖ Error details:
#> ℹ                       ✖ Error on element 3 (1): "3"
#> ℹ [2026-08-04 13:52:01] Removed 1 failed result
#> $`1`
#> [1] 1
#> 
#> $`2`
#> [1] 4
#> 
#> $`4`
#> [1] 16
#> 
#> $`5`
#> [1] 25
#> 

# Control error printing
parallelize_fun(1:5, function(x) {
  if (x == 2) stop("Error on element 3")
  if (x == 4) stop("Error on element 4")
  x^2
})
#> ℹ [2026-08-04 13:52:01] Using 1 core
#> ℹ [2026-08-04 13:52:01] Building results
#> ! [2026-08-04 13:52:01] Found 2 failed results
#> Warning: Found 2 failed results
#> ℹ [2026-08-04 13:52:01] ✖ Error details:
#> ℹ                       ✖ Error on element 3 (1): "2"
#> ℹ                       ✖ Error on element 4 (1): "4"
#> $`1`
#> [1] 1
#> 
#> $`2`
#> $error
#> [1] "Error on element 3"
#> 
#> $index
#> [1] 2
#> 
#> $input
#> [1] 2
#> 
#> attr(,"class")
#> [1] "parallelize_error"
#> 
#> $`3`
#> [1] 9
#> 
#> $`4`
#> $error
#> [1] "Error on element 4"
#> 
#> $index
#> [1] 4
#> 
#> $input
#> [1] 4
#> 
#> attr(,"class")
#> [1] "parallelize_error"
#> 
#> $`5`
#> [1] 25
#> 

parallelize_fun(1:5, function(x) {
  if (x == 3) stop("Error on element 3")
  x^2
}, throw_error = FALSE)
#> ℹ [2026-08-04 13:52:01] Using 1 core
#> ℹ [2026-08-04 13:52:01] Building results
#> ! [2026-08-04 13:52:01] Found 1 failed result
#> Warning: Found 1 failed result
#> $`1`
#> [1] 1
#> 
#> $`2`
#> [1] 4
#> 
#> $`3`
#> $error
#> [1] "Error on element 3"
#> 
#> $index
#> [1] 3
#> 
#> $input
#> [1] 3
#> 
#> attr(,"class")
#> [1] "parallelize_error"
#> 
#> $`4`
#> [1] 16
#> 
#> $`5`
#> [1] 25
#>