Skip to content

Add package anchors to Rd links (Rd cross-references NOTE/WARNING) - #808

Merged
joshua-d-campbell merged 6 commits into
develfrom
fix/rd-xref-anchors
Oct 9, 2026
Merged

joshua-d-campbell merged 6 commits into
develfrom
fix/rd-xref-anchors

Conversation

@joshua-d-campbell

Copy link
Copy Markdown
Collaborator

Depends on #807. This branch builds on fix/ci-bioc-devel, so its
diff also shows #807's commits until #807 merges.

What changed and why

R CMD check notes, and the Bioconductor devel builders report as a
WARNING, that 138 help pages link to other packages' topics without a
package anchor:

checking Rd cross-references ... NOTE
Found the following Rd file(s) with Rd \link{} targets missing package anchors

The roxygen sources in R/ now name the package for each target, and
man/ was regenerated with make docs:

Was Now
\linkS4class{SingleCellExperiment} (189×) \link[SingleCellExperiment:SingleCellExperiment-class]{SingleCellExperiment}
\linkS4class{GeneSetCollection}, {BiocParallelParam}, {SummarizedExperiment} same pattern with GSEABase, BiocParallel, SummarizedExperiment
\link{colData}, \link{assay} \link[SummarizedExperiment]{...}
\link{metadata} \link[S4Vectors]{metadata}
\link{altExp}, \link{reducedDim} \link[SingleCellExperiment]{...}
\link{DelayedArray}, \link{readMM}, \link{Matrix} DelayedArray, Matrix
reticulate, GSEABase, scRNAseq, msigdbr, igraph topics anchored to their packages
  • Every target was checked to exist as a help alias in its package.
  • This changes documentation only. Version 2.23.6.

How it was tested

  • make check (local, R 4.6.1): checking Rd cross-references ... OK,
    where it was a NOTE before. 0 errors.
  • Only R/*.R roxygen lines, man/*.Rd, DESCRIPTION, and NEWS changed.

Checklist

  • Tests added or updated, make test passes, and make coverage
    didn't drop. N/A: documentation only.
  • make check-full and make bioccheck pass with no new errors or
    warnings. make check was run; the Rd cross-references check is OK.
  • make docs run, if roxygen comments changed
  • New exports added to _pkgdown.yml, and make site-check passes.
    N/A.
  • NEWS.md updated for user-facing changes
  • Version bumped (z) if this will be pushed to Bioconductor
  • Plan review and /code-review run; findings fixed or answered. A
    mechanical change, checked by R CMD check.
  • Related issue linked. N/A.
  • Shiny only: verified with a screenshot of the running app. N/A.

ADR

N/A.

Scientific correctness

Generated content

Claude Code (Opus 5.5) wrote all of this PR, as part of the plan the
maintainer approved for getting CI and the Bioconductor build clean.

🤖 Generated with Claude Code

joshua-d-campbell and others added 6 commits October 8, 2026 12:45
batchelor 1.28's fastMNN() and reducedMNN(), like mnnCorrect(), still call
deprecated scuttle functions, which fails R CMD check on the runFastMNN
example. A shared helper, .muffleUpstreamDeprecations(), muffles only
deprecatedWarnings raised inside those calls and documents why. Version
2.23.5.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Set R_BIOC_VERSION to the Bioconductor devel version so CI installs the
same dependencies as the Bioconductor builders, link Homebrew's OpenSSL
for alabaster.base's source build on macOS, run BiocCheck on Ubuntu with
BiocCheck installed from devel, and use actions/checkout@v4. ADR 0007
records the decision; the release checklist gains the step to update
R_BIOC_VERSION.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
From /code-review: the alabaster.base source build may also need
OpenSSL headers, and runFastMNN(useReducedDim = ) was untested.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
R CMD check notes, and the Bioconductor devel builders warn, that 138 help
pages link to other packages' topics without a package anchor. The roxygen
sources now give the package for each target: \linkS4class{X} becomes
\link[pkg:X-class]{X}, and \link{topic} becomes \link[pkg]{topic}, for
SingleCellExperiment, SummarizedExperiment, S4Vectors, DelayedArray,
Matrix, GSEABase, BiocParallel, reticulate, scRNAseq, msigdbr, and igraph.
Every target was checked to exist as a help alias in its package.
R CMD check now reports 'checking Rd cross-references ... OK'.
Version 2.23.6.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
BiocCheck 1.49 (Bioc 3.24) errors on Git-tracked .claude/ files, and the
lab standards commit .claude/settings.json. The built tarball excludes it
through .Rbuildignore, and the Bioconductor builders also check tarballs.
ADR 0007 records this; the standards question goes to
campbio/r-bioc-dev-standards.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
@joshua-d-campbell
joshua-d-campbell merged commit 3a4e70b into devel Oct 9, 2026
4 checks passed
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