reproducible (version 3.2.1)

prepopulateCacheAsync: Pre-populate the in-memory showCache cache for a given cachePath

Description

Forks a background process that runs showCache() against cachePath; subsequent showCache() / Cache()->showSimilar() calls in the same R session can then harvest the result instead of re-scanning the cache directory synchronously. Useful for very large caches (tens of thousands of entries) where the cold first scan can take a minute or more.

Idempotent: a second call with the same cachePath reuses the existing job. Skipped silently on Windows (forking-based) and when the parallel package isn't available.

This helper is called automatically the first time Cache() or showCache() is invoked against a given cachePath, so most users do not need to call it explicitly. It is exported for workflows that want to kick off the spawn early (e.g. inside setupProject()) so the fork has more wall-clock time to complete before the first manual showCache() call.

Usage

prepopulateCacheAsync(cachePath = getOption("reproducible.cachePath"))

Value

Invisibly returns the spawn job handle, or NULL if the spawn was skipped.

Arguments

cachePath

A character path. Defaults to getOption("reproducible.cachePath").

Examples

if (FALSE) { # identical(Sys.getenv("NOT_CRAN"), "true")
# Scanning a large flat-file cache is slow, so this warms it in the
# background; a later showCache()/Cache(showSimilar = TRUE) then returns
# without waiting. Guarded on interactive() because it spawns a process.
cachePath <- file.path(tempdir(), "prepopulateExample")
dir.create(cachePath, showWarnings = FALSE)

prepopulateCacheAsync(cachePath = cachePath)

unlink(cachePath, recursive = TRUE)
}