na.rm appears not to propagate to the transform in stat_dots

[gdbassett]

chunk %>%
    filter(!is.na(tdiff)) %>%
    mutate(tdiff = if_else(tdiff == 0, 1, tdiff)) %>%
  ggplot() + 
    ggdist::stat_dots(aes(x=tdiff, y=0), quantiles=100,
                      na.rm=TRUE
    ) +
    scale_x_log10() +
    NULL

results in the error:

Warning message in self$trans$transform(x):
“NaNs produced”
Warning message:
“Transformation introduced infinite values in continuous x-axis”

It appears as if the na.rm=TRUE is not getting correctly propagated to the transform. The error doesn't occur when either quantiles is not set or scale_x_log10() is not used.

[gdbassett]

I've found I also run into what appears to be a similar issue setting limits on the x scale:

Warning message:
“Computation failed in `stat_dots()`:
missing values and NaN's not allowed if 'na.rm' is FALSE”
Warning message in rep(yes, length.out = len):
“'x' is NULL so the result will be NULL”
ERROR while rich displaying an object: Error in ans[ypos] <- rep(yes, length.out = len)[ypos]: replacement has length zero

[mjskay]

@gdbassett this should be fixed now (sadly not on the cran version; I didn't get to it in time, but will be in the next release!)

thanks again for all the feedback!

[gdbassett]

You're welcome. Thank you for fixing it!

Gabe