retire make_afun usage

[shajoezhu]

Feature description

retire make_afun usage

Batch 1

• R/prop_diff.R
• R/prop_diff_test.R
• R/summarize_num_patients.R
• R/count_cumulative.R
• R/count_missed_doses.R

Batch 2

• R/summarize_ancova.R
• R/summarize_glm_count.R

Batch 3

• R/abnormal.R
• R/abnormal_by_baseline.R
• R/abnormal_by_marked.R
• R/abnormal_by_worst_grade.R
• R/abnormal_by_worst_grade_worsen.R

Batch 4

• R/estimate_multinomial_rsp.R
• R/estimate_proportion.R
• R/odds_ratio.R
• R/survival_coxph_pairwise.R
• R/survival_timepoint.R

Code of Conduct

• I agree to follow this project's Code of Conduct.

Contribution Guidelines

• I agree to follow this project's Contribution Guidelines.

Security Policy

• I agree to follow this project's Security Policy.

[Melkiades]

This would be amazing

[danielinteractive]

Interesting, what is the rationale and replacement strategy?
(@iaugusty I found it!)

[Melkiades]

Interesting, what is the rationale and replacement strategy? (@iaugusty I found it!)

make_a_fun is a bit of a black-box function that is very hard to debug. It is also challenging to track down how a variable is sent to the analysis function from the top level and in which levels (a_ or s_) it will be available. In practice, we had many issues with unclear definitions of default values through different nested calls of make_a_fun. For these reasons, we are moving towards directly only {rtables} code. This makes it easier to read for future users or people that start working directly from {rtables}