I noticed that in forte_table_metadata.csv, missing values are sometimes left blank and sometimes explicitly coded as NA. We should definitely be consistent in how we present these in the functions, and should probably be consistent in how we're storing them as well.
A temporary fix is to amend the read_csv_file function to call read.csv(..., na.strings = c("", "NA")). The more robust fix is to also go through all the CSV files and remove explicit NA (or replace all blank entries with NA -- whatever we think is best. I personally prefer blanks, but either should work).
I noticed that in
forte_table_metadata.csv, missing values are sometimes left blank and sometimes explicitly coded asNA. We should definitely be consistent in how we present these in the functions, and should probably be consistent in how we're storing them as well.A temporary fix is to amend the
read_csv_filefunction to callread.csv(..., na.strings = c("", "NA")). The more robust fix is to also go through all the CSV files and remove explicitNA(or replace all blank entries withNA-- whatever we think is best. I personally prefer blanks, but either should work).