Published February 9, 2026 | Version 1.12.0
Software Open

ropensci/targets: tar_igraph()

  • 1. Eli Lilly and Company @EliLillyCo
  • 2. @sunholo-data
  • 3. Stanford University
  • 4. Hakai Institute
  • 5. Chiba University
  • 6. @wildlifeevoeco
  • 7. @cynkra
  • 8. University of Cambridge
  • 9. Scarlatti
  • 10. @posit-pbc
  • 11. Gilead Sciences, Inc.
  • 12. @jumpingrivers
  • 13. University of Leeds
  • 14. devXY
  • 15. Integrated Digital East Antarctica, Australian Antarctic Division
  • 16. @ropensci
  • 17. Human Predictions LLC

Description

targets 1.12.0

  • Avoid loading non-targets in tar_load(everything()) (#1529, @pitakakariki).
  • For internal consistency, return NA_character_ from tar_path_target() when format is "file" (#1532, @noamross).
  • Add tar_option_unset() (#1521, @noamross).
  • Use a new TAR_ACTIVE environment variable instead of a field of the tar_runtime object for tar_runtime() (https://github.com/ropensci/tarchetypes/issues/232, @lgaborini).
  • Add tar_igraph() and recommend it along with igraph::find_cycle() for debugging (#1562, @tylermorganwall).
  • Avoid the unclean shutdown message in the clustermq multi-process scheduler by calling cleanup() repeatedly until it returns TRUE. Uses exponential backoff to avoid excessive CPU load.
  • Use sequential controller in covr.
  • Ensure hash stability of vector slices in R >= 4.6.0 (#1566).
  • Use a dummy send instead of send_wait() to work around https://github.com/mschubert/clustermq/issues/340 (#1566).
  • Set strict = TRUE by default in tidyselect::eval_select() to make tidyselect interfaces consistent with dplyr::select() (except the exclude argument of graph functions, which need to allow names to not exist in the list of choices) (#1563, @statzhero).

Files

ropensci/targets-1.12.0.zip

Files (1.3 MB)

Name Size Download all
md5:eeef62776b8b5f560884a8c0f5709258
1.3 MB Preview Download

Additional details

Related works

Is supplement to
Software: https://github.com/ropensci/targets/tree/1.12.0 (URL)

Software