skoestlmeier / icss Goto Github PK

Hello!

I was playing around with some view count data and found that if the data set is too flat, the process will not converge. However, instead of returning an NA. An unhelpful error is raise. The data here is the difference in view counts over time.

test <- c(0, 0, 2, 1, 0, -1, 1, 0, 1, 0, -1, 1, 1, 5, -2, -1, -1, -1, 2, -1, -1, 0, 0, 0, 0, 0, 0, 0, 0, -1, 1, -3, 1, -1, -1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, -1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, -1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, -1, 0, 0, 0, 0, 0, 0, 0)

> ICSS(test)
Error in while (exceeds) { : missing value where TRUE/FALSE needed

This appears to be happening because on line 116-117 as CenteredCusumValues returns a list of NAs at some point in the while loop due to all the values being equal, as in this case (0 0 0 0 0 0 0). Then, check_critical_value(Dk_step2b) set tmp$exceeds = NA which breaks the while(exceed) loop on line 114 and line 99.

I'll open up a pull request once I fix the bug. A quick is_na(tmp$exceeds) would fix this issue.