Skip to content

cran-comments.md

Latest commit

 

History

History
73 lines (53 loc) · 2.86 KB

cran-comments.md

File metadata and controls

73 lines (53 loc) · 2.86 KB

Resubmission (0.6.2)

This release addresses the check failure reported by CRAN on 2026-05-18 on r-devel-linux-x86_64-fedora-clang (the only platform that failed; all 12 other CRAN check platforms passed):

test-neb-equivalence.R:458:3: error Nh method (arithmetic mean) is faster than harmonic mean speedup_vs_harmonic > 0.5 is not TRUE

That assertion compared wall-clock times for two micro-benchmarks (1,000 iterations of weighted.mean() on a 10,000-element vector). The ratio is sensitive to check-machine load and was unreliable on the failing platform.

Fix: the two performance-benchmark tests in tests/testthat/test-neb-equivalence.R (Test 13: "Nh method (arithmetic mean) is faster than harmonic mean"; Test 14: "Nh method scales well with dataset size") now call skip_on_cran(). This matches the established pattern already used elsewhere in the suite (test-ecosystem-integration.R, test-visualizations.R, test-zenodo-*.R, test-jss-vignette.R). The underlying correctness assertions (1e-10 agreement between the Nh and harmonic-mean methods, non-zero error from the arithmetic-mean-of-burden method) remain in the tests and run on local development and CI.

No other source changes.

Test environments

  • Local: Linux Mint 22 (Linux 6.8.0-86-generic), R 4.3.3
  • GitHub Actions (on pull request and push to main):
    • macOS-latest (release)
    • Windows-latest (release)
    • Ubuntu-latest (devel)
    • Ubuntu-latest (release)
    • Ubuntu-latest (oldrel-1)

R CMD check results

0 errors | 0 warnings | 1 note

Note

  • checking for future file timestamps ... NOTE
    • unable to verify current time
    • This is a system-level timing issue and does not affect package functionality.

Previously resolved issues

  • PDF manual generation: Fixed by using --no-manual flag (manual not required for CRAN)
  • Non-standard top-level files: Fixed by updating .Rbuildignore
  • Vignette PDF size: Optimized with --compact-vignettes=both (reduced from 392KB to 113KB)

Expected CRAN NOTEs (first submission)

  • checking CRAN incoming feasibility ... NOTE

    • Maintainer: 'Eric Scheier eric@scheier.org'
    • New submission
    • This is expected for first CRAN submission

  • checking installed package size ... NOTE (if present)

    • sub-directories of 1Mb or more: data
    • The package includes sample census tract data (nc_sample.rda, orange_county_sample.rda)
    • These datasets are necessary for vignettes and examples to run without external data dependencies

Submission notes

This is a first submission to CRAN.

The package provides tools for calculating and analyzing household energy burden using the Net Energy Return (Nh) aggregation methodology, based on peer-reviewed research published in Nature Energy.

All tests pass on all platforms (614 tests across 7 platform configurations in GitHub Actions CI).