We have detected issues with helda when checking the next release of ggplot2, due in two weeks

helda

• Version: 1.1.1
• Source code: https://github.com/cran/helda
• URL: https://www.github.com/Redcart/helda
• BugReports: https://github.com/Redcart/helda/issues
• Date/Publication: 2020-05-14 11:40:03 UTC
• Number of recursive dependencies: 102

Run cloud_details(, "helda") for more info

Newly broken

• checking tests ... ERROR

    Running ‘testthat.R’
  Running the tests in ‘tests/testthat.R’ failed.
  Last 13 lines of output:
    Component "layers": Component 3: Component 3: Component 2: target, current do not match when deparsed
    Component "plot_env": Component "plot_2": Component "layers": Component 3: Component 3: Component 2: target, current do not match when deparsed
    
    ── 2. Failure: lift effect for titanic data set (@test-lift_effect.R#18)  ──────
    `result` not equivalent to `lift_effect_test`.
    Component "layers": Component 2: Component 3: Component 2: target, current do not match when deparsed
    Component "plot_env": Component "plot_1": Component "layers": Component 2: Component 3: Component 2: target, current do not match when deparsed
    
    ══ testthat results  ═══════════════════════════════════════════════════════════
    [ OK: 5 | SKIPPED: 0 | WARNINGS: 0 | FAILED: 2 ]
    1. Failure: lift curve for titanic data set (@test-lift_curve.R#18) 
    2. Failure: lift effect for titanic data set (@test-lift_effect.R#18) 
    
    Error: testthat unit tests failed
    Execution halted
  

I've looked into your unit test, but since you simply compare two very complex plot objects and see if they are the same it is hard to figure out where the issue lies. Note that we may from time to time change some internals of ggplot2 which will be reflected in the resulting plot object, so comparing the plot object wholesale to a cached one will always be very fragile and hard to maintain. I'd suggest you instead compare specific values you know are important and related to your package code.

We will submit ggplot2 in two weeks so I hope you'll have time to fix the unit test by then. Please reach out if you have any further questions