[Question]: Duplicate code in tmg and teal.widgets

[donyunardi]

What is your question?

The function variable_type_icons exists in two places:

tmg:

teal.modules.general/R/utils.R

Lines 237 to 251 in a835f3d

	variable_type_icons <- function(var_type) {
	checkmate::assert_character(var_type, any.missing = FALSE)
	class_to_icon <- list(
	numeric = "arrow-up-1-9",
	integer = "arrow-up-1-9",
	logical = "pause",
	Date = "calendar",
	POSIXct = "calendar",
	POSIXlt = "calendar",
	factor = "chart-bar",
	character = "keyboard",
	primary_key = "key",
	unknown = "circle-question"
	)

teal.widgets:
https://github.com/insightsengineering/teal.widgets/blob/d94641a1d7e532bef125e229cfc2407fd0d32d43/R/optionalInput.R#L323-L337

They're identical and I think they should live in teal.widgets.
Should we remove the one in tmg?

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.

[averissimo]

Good catch! We could either export the function in teal.widgets or use getFromNamespace("variable_type_icons", "teal.widgets") in tmg

[llrs-roche]

I would simply access it directly via ::: : If the author of the packages is the same, as is the case, CRAN allows using ::: => variable_type_icons <- teal.widgets:::variable_type_icons

I can't find now the exact reference of were I read it but see for example knitr and xfun written by the same person.

[averissimo]

That's interesting, in the past we were asked to remove them, hence why we use getFromNamespace in some packages (after validating that some tidyverse package was using it)

[llrs-roche]

Maybe they changed the policy recently or it was because on that occasion the maintainers were not the same or simply that the package was not listed as a direct (strong) dependency. It also depends on the reviewer of the package.