Closed fruce-ki closed 7 years ago
Actually the NA
values are explicitly coded in case of no results, to ensure that the named fields are created and to prevent the return value from being an empty list
or data.frame
. Instead, this should have been achieved by assigning empty character vectors instead of NA
values.
The isoform switching summary has an off-by-one error when there are no results to report. The ID lists, instead of being empty vectors, contain a
NA
value. As a result, the tallies report values of 1 instead of 0. When there are results to report, theNA
s and the off-by-one error do not apply.This is likely a
data.table
subsetting error. When nothing matches the subset criteria,data.table
returns a row ofNA
. I need to catch this.A similar issue used to happen with the DTU summary, so that should serve as a template for the solution here too.