blob: eaf560d48b250136d4e666f84ab4dfe4a73c3c50 [file] [log] [blame]
# which column
if [ $# -ne 2 ]; then
echo "tell me file name and col number"
exit 1
fi
k=$2
file=$1
# sort it
sort --key $k -n $file > $file-sort
# now we generate CDF
totalLines=$(cat $file-sort | wc -l)
echo "total line numbers: $totalLines"
rm ./xx
for i in `seq 1 100`;
do
lx=$(( $i * $totalLines ))
lxx=$(( $lx / 100 )) # you get the line number here
sed -n "${lxx}p" $file-sort >> ./xx
done
cat ./xx | awk '{print $1,"\t",NR}' > cdf-$file