Custom effect size rules were not being properly reported as such, showing malformed text with a blank space instead of indicating they were custom rules.

Issue

When using custom rules with effectsize::rules(), the report output would show:

Effect sizes were labelled following  recommendations.

Notice the blank space between "following" and "recommendations."

Root Cause

The .text_effectsize() function in R/report_effectsize.R used a switch() statement to map known rule names like "cohen1988", "gignac2016", etc. When custom rules with unknown names (like "Unknown") were passed, the switch returned NULL for effsize_name, resulting in the malformed text.

Solution

Modified the logic in .text_effectsize() to check if the rule name is recognized. If the switch() statement returns NULL (indicating an unknown rule name), the function now correctly identifies it as a custom rule.

Before and After

Before:

library(report)
library(effectsize)

custom_rules <- effectsize::rules(1, c("tiny", "yeah okay"), name = "Unknown")
t.test(formula = extra ~ group, data = sleep) |>
  report(rules = custom_rules)
#> Effect sizes were labelled following  recommendations.

After:

library(report)
library(effectsize)

custom_rules <- effectsize::rules(1, c("tiny", "yeah okay"), name = "Unknown")
t.test(formula = extra ~ group, data = sleep) |>
  report(rules = custom_rules)
#> Effect sizes were labelled following a custom set of rules.

Testing

• Added comprehensive unit tests for .text_effectsize() covering all predefined rules and custom rule scenarios
• Enhanced integration tests for t-test reporting with custom rules
• Verified no regression in existing functionality with built-in rules
• All existing tests continue to pass

Fixes #458.

✨ Let Copilot coding agent set things up for you — coding agent works faster and does higher quality work when set up for your repo.