Skip to content

To Do's for visR CRAN release

Jump to bottom
Charlotta edited this page Mar 3, 2021 · 39 revisions

To Dos visR on CRAN

  • Finalize API discussion [done]

  • Tables to new API: (vr_create_tableone, vr_render_table, get_pvalue) [done]

    • Merge render_tableone and render_table code to single S3 table rendering function [done]
    • Split risktable to create table df for individual rendering + add_risktable fct for plotting [done]
    • Table 1 (vr_create_tableone.R):
      • Implement as S3 [done]
      • rename group_cols to strata [done]
      • implement get_pvalue.tableone to add column with p value to table 1 data frame (wait for Steven and Alexandra)
      • implement unit test (min 1): Create simple data frame and test the basic treatment of factors/integers/characters [done - not commited]
      • repair summary function for integer [done]
    • Table rendering (vr_render_table)
      • simplify <title>/ arguments to a single argument [done]
      • append N to column name in table1 function ([done], but only added to the variable names now? Different styles?)
      • add footnote to tables [done]
      • function to add footnotes on statistical tests performed (esp. for table1; give an additional argument? Wait for implementation of p-value function?)
    • fix bug in DT table caption (not sure how/if possible)
    • Postprocessing (after refactoring is approved)
      • Check all descriptions [done]
      • Check all examples [done]
      • Check/Rewrite vignettes

  • Port to new API:

    • Attrition code
      • table
      • barchart with labels
      • flowchart
        • fix bug in how to integrate into report (vignette)
        • evaluate if there is a way to implement in ggplot)
    • KM Model
      • vr_KM_est --> visR::estimate_KM
        • Add CNSR and AVAL to argument lists + adjust the code/documentation for this
        • write unit testing
      • vr_plot --> visR::plot
        • Test transformations
        • Write unit tests
      • get_quantiles
        • Verify
        • Write unit tests
      • Trend test + Equality across strata [Steven]
        • Verify
        • Finish function
        • Write unit testing
      • Risktable
        • When min_at_risk = 0 show all information up until last tick mark
        • How to write unit test?
    • Cox PH model integration
      • Summary table
      • Add annotation to KM plot

  • Unit tests for all functions

  • Expand vignettes

    • Provide example data set
    • One for each major feature
    • Update example TTE analysis

  • Not for Release

    • Get CI/CD Workflow to successfully build package

Link to function overview and sprint tasks:

To Do List for final push to CRAN (from Hadley's R Packages book):

  • Pick a version number.
  • Run and document R CMD check.
  • Check that you’re aligned with CRAN policies.
  • Update README.md and NEWS.md.
  • Submit the package to CRAN.
  • Prepare for the next version by updating version numbers. Publicise the new version.
Clone this wiki locally