Skip to content

Guard NSE against S4Vectors parallelVectorNames() regression (#140) - #129

Closed
sa-lee wants to merge 1 commit into
tidyomics:develfrom
sa-lee:worktree-debug-filter-nsbs
Closed

Guard NSE against S4Vectors parallelVectorNames() regression (#140)#129
sa-lee wants to merge 1 commit into
tidyomics:develfrom
sa-lee:worktree-debug-filter-nsbs

Conversation

@sa-lee

@sa-lee sa-lee commented Jul 7, 2026

Copy link
Copy Markdown
Collaborator

Fixes plyranges NSE on S4Vectors devel, where parallelVectorNames() now lists metadata columns (Bioconductor/S4Vectors#140). This behaviour changes how mcols columns bind to the masking env i.e. named like functions (e.g. gcbase::gc(); baseMean → error), which breaks filter/mutate/summarise/select and the vignette build.

  • overscope_env() (new) builds the core env from an mcols-stripped copy, then splices the correctly-bound metadata env back as parent.
  • select_rng() core names.
  • select.DeferredGenomicRanges() exclude names.

Note: needs a DESCRIPTION version bump before release.

@sa-lee

sa-lee commented Jul 7, 2026

Copy link
Copy Markdown
Collaborator Author

Closing: fixed upstream in S4Vectors 8b236a8 (parallelVectorNames() now reads the class prototype instead of x[0L]). Verified the vanilla package passes the suite and builds the vignette on the patched S4Vectors, so the plyranges-side guard is unnecessary. Keeping the regression tests only.

@sa-lee sa-lee closed this Jul 7, 2026
@sa-lee
sa-lee deleted the worktree-debug-filter-nsbs branch July 7, 2026 01:26
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant