A couple of bugs with missing_plot

[nathansam]

Hi there,

I just thought I would flag a few bugs I have come across whilst using the missing_plot function. The vlabels variable is created here in a conditional statement and is used later when creating the ggplot object. Therefore if the function is called with use_labels = FALSE, the function will error.

finalfit/R/missing_plot.R

Lines 39 to 42 in b1fa0f0

	# Labels
	if(use_labels){
	vlabels = extract_labels(df.in)$vfill
	}

Having

else {
 vlabels = NULL
}

fixes the problem. Secondly, the code here:

finalfit/R/missing_plot.R

Line 51 in b1fa0f0

df.in$.id = rownames(df.in) %>% as.numeric()

only works if the row names are the row numbers. If you use the function in a context where the row names aren't the row numbers (such as after subsetting the data frame), then the plot is bugged (as shown in the screenshot below).

Replacing the above line with the below line should stop this behaviour.

 df.in$.id = seq(1, nrow(df))

I would be happy to open a PR, if you would like the bugs fixed, but it might just be easier for you to make the changes yourselves.

Kind regards

[nathansam]

[ewenharrison]

Sorry I missed this. Many thanks for posting.
It is fixed in 1.04.

[ewenharrison]

(I should say that this function also fails if there are too many rows in the dataset. It may be a ggplot issue.)