CODEWITHSUNDEEP
bashcsvawk
data princess
data princess

Reputation: 1158

Awk append existing CSV column of variable number

I know how to pass variables to awk, but I haven't been able to combine that with the column number functionality when working with a CSV. I'm also not sure how to perform a string concatenation this way.

I'm expecting some columns in my CSVs to contain dates in MM-dd-yyyy or yyyy/MM/dd format, or some permutation of those, and would like to append a default hh:mm:ss onto it, so a column of 1/10/2017 becomes 1/10/2017 00:00:00. I've tried doing something along these lines in awk, but it doesn't seem to recognize the column I'm trying to reference. And I don't think this is how string concatenation is done.

for file in *.csv; do
  headline=$(head -n 1 $file)
  byline=$(sed '2q;d' $file)
  IFS=', ' read -r -a headers <<< $headline  
  IFS=', ' read -r -a entries <<< $byline

  for i in "${!headers[@]}"; do                                 
    date='^[0-9]+[/-][0-9]+[/-][0-9]+$'

    if [[ ${entries[$i]} =~ $date ]] ; then                 
      awk -F, -v col='$i' '{$($col)+="00:00:00";}1' OFS=, $file  
    fi
  done
done

Upvotes: 0

Views: 245

Answers (1)

tobiasegli_te
tobiasegli_te

Reputation: 1463

It would be easier to answer your question if you provided a minimal reproducible example of your input data.

Without knowing the data, I assume that you should get what you're looking by replacing your awk command with the following:

awk -v col=$(($i + 1)) 'BEGIN{FS=", "} {print $col " 00:00:00"}' "$file"

Upvotes: 1

Related Questions