CODEWITHSUNDEEP
bashawk
StatsSorceress
StatsSorceress

Reputation: 3099

Bash and Awk: write results of sort to a new file line by line

I have a file myinputfile.txt that looks like this:

2.34 1.25 ./RUN1 ./RUN1/mod/file1.csv
3.34 1.25 ./RUN1 ./RUN1/mod/file2.csv
1.32 1.25 ./RUN2 ./RUN2/mod/file1.csv
4.35 1.25 ./RUN2 ./RUN2/mod/file2.csv

I want a new file that contains the line that has the maximum value of column 1 for each value of RUN*. Desired contents of new file (myresultsfile.txt):

3.34 1.25 ./RUN1 ./RUN1/mod/file2.csv
4.35 1.25 ./RUN2 ./RUN2/mod/file2.csv

I have a script that mostly gets me there, but doesn't put things on new lines:

#!/bin/bash
resultsfile="myresults.txt"
inputfile="myinputfile.txt"
res1=$(sort -k3,3r $inputfile | awk '{split($nF,d,"/")} !a[d[3]]++')
echo $res1 >> $resultsfile

Upvotes: 0

Views: 41

Answers (2)

karakfa
karakfa

Reputation: 67537

with sort assisted awk

$ sort -k1nr file | awk '!a[$3]++'

3.34 1.25 ./RUN1 ./RUN1/mod/file2.csv
4.35 1.25 ./RUN2 ./RUN2/mod/file2.csv

simply redirect to a new file

$ sort -k1nr file | awk '!a[$3]++' > sorted_file

Upvotes: 1

anubhava
anubhava

Reputation: 785731

You can use awk like this:

awk '$1 > max[$3] { max[$3] = $1; rec[$3] = $0 } END { for (i in rec) print rec[i] }' file

3.34 1.25 ./RUN1 ./RUN1/mod/file2.csv
4.35 1.25 ./RUN2 ./RUN2/mod/file2.csv
  • max is an associative array that holds max value of first column by the key as column 3
  • rec is an associative array that holds full line by the key as column 3

Upvotes: 0

Related Questions