awk: add column name with filename in loop

Question

I have a question to add a column title during a loop with awk: I currently have this code, which calculates the mean of each line for each csv file :

for i in *.csv; do  
  awk -F"," 'NR==1 { next }
        { T=0
           for(N=1; N<=NF; N++) T+=$N;
           T/=NF
           print T }' $i > Mean$i
done

So I get files MeanName1.csv, MeanName2.csv, ... containing:

2.5                3.5                ...
4.2                3.4
1.2                3.9
1.2                3.7

I would like to get a single file "Mean" containing:

Name1   Name2   ...
2.5     3.5
4.2     3.4
1.2     3.9
1.2     3.7

I failed to use print FILENAME in the loop.

Do you have any suggestions ? Thank you

Answer 1

Joining the output files is easiest done with paste, but I don't think it knows how to add the file names as headers. But we can do that manually to the intermediate output files within the loop, so:

for i in *.csv; do
    out="Mean$i"
    echo "${i%.csv}" > "$out"
    awk -F"," 'NR==1 { next }
        { T=0
           for(N=1; N<=NF; N++) T+=$N;
           T/=NF
           print T }' "$i" >> "$out"
done
paste Mean*.csv > All.csv

${i%.csv} is just the value of $i with the suffix .csv removed, so for foo.csv and bar.csv, that gives an output like this:

bar     foo
4.33333 2.5
3.8     3.5

---

I note you're also skipping the first line of each input line (NR==1 { next }). If they contain useful headers, you could of course use those too by changing that to NR==1 { print; next } and then removing the echo I added above.