Using method = "auto" fails if there is missing data due to the type checking

[jhorzek]

The following currently fails:

correlation::correlation(data = data.frame(a = sample(c(NA, 1:5), 1000, replace = TRUE),
                                           b = sample(c(NA, 1:5), 1000, replace = TRUE)),
                         method = "auto",
                         missing = "keep_pairwise")

The issue arises because the utils_find_correlationtype function assumes no missing data in all(x %% 1 == 0). I've replaced this with a basic check that ensures that NAs are ignored.

[Copilot]

Pull request overview

This pull request fixes a bug where using method = "auto" fails when the dataset contains missing values. The issue occurred in the .vartype() function which performs type checking to determine the appropriate correlation method.

Changes:

• Updated the is_count check to properly handle NA values by adding any(!is.na(x)) guard and na.rm = TRUE parameter to the all() function

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

There appears to be trailing whitespace at the end of these lines. Please remove the trailing spaces to maintain code consistency.

Suggested change

	if (any(!is.na(x)) && all(x %% 1 == 0, na.rm = TRUE)) {
	out$is_count <- TRUE
	if (any(!is.na(x)) && all(x %% 1 == 0, na.rm = TRUE)) {
	out$is_count <- TRUE

[jhorzek]

I reformatted the code with air.

[Copilot]

Consider adding a test case for this specific bug fix to ensure that method = "auto" works correctly with missing data. For example, test the scenario mentioned in the PR description where correlation is called with method = "auto" and missing = "keep_pairwise" on data with NA values.

[jhorzek]

I've added a very basic test for missing data combined with method = "auto" on continuous data.