go to a specific line of a text file by shell script

Question

I have two text file. The structure of those two file is same. I have a while loop which reads from those two text file at the same time. But both of those two file , first 4 line is not needed. so what i need is, at the very starting of my program the two head will jump to line number 4 of both of those files and then start to read through my while loop. Here is my current code

while read compareFile1 <&3 && read compareFile2 <&4; do
echo $compareFile1
echo $compareFile2 
done 3<test1.txt 4<test2.txt

Here is my sample file ..

=== Predictions on test data ===

    inst#     actual  predicted error prediction (id)
        1 1:positive 1:positive       0.774 (10001996.txt)
        2 1:positive 2:negative   +   0.889 (10003432.txt)
        3 1:positive 1:positive       1 (10003865.txt)
        4 1:positive 1:positive       0.999 (10004065.txt)
        5 1:positive 1:positive       0.991 (10004266.txt)
        6 1:positive 1:positive       0.999 (10006157.txt)
        7 1:positive 1:positive       0.869 (10007003.txt)
        8 1:positive 2:negative   +   1 (10008447.txt)
        9 1:positive 1:positive       0.998 (10009702.txt)
       10 1:positive 1:positive       0.994 (10011072.txt)

How can i do this by bash? I am using mac. Thanks.

Derlin · Accepted Answer · 2016-02-15 18:32:10Z

2

You could use the tail utility. By default, it outputs the last 10 lines of a file, but it also has some quite useful parameters. To skip the first X lines, use -n+X.

Example:

 tail -n+5 myfile.txt

will output the whole file from the 5th line (skipping the first 4).

But in your case, you could simply increment a variable to start the processing on line 4. Example:

l=0 
while read compareFile1 <&3 && read compareFile2 <&4; do
  if [[ $l < 4 ]]; then 
     l=$((l+1)); 
  else
     # do your processing here
     echo $compareFile1
     echo $compareFile2 
  fi  
done 3<test1.txt 4<test2.txt

edited Feb 15, 2016 at 18:32

answered Feb 15, 2016 at 18:02

Derlin

9,9212 gold badges34 silver badges57 bronze badges

Sign up to request clarification or add additional context in comments.

2 Comments

sovon Over a year ago

please see my edited question. I added my code and file. " tail -n+5 myfile.txt " command is returning the part of the file ..not shifting the pointer head (as far as i understood)

sovon Over a year ago

Thanks man, that's a very quick and easy solution. I am wondering, why i did not think about it

Lars Fischer · Accepted Answer · 2016-02-15 18:28:10Z

0

Here is another alternative, it avoids reading both files in a shell loop:

nl File1.txt | paste - File2.txt | sed -n '5,$ p'

It produces in the first column the line number (via nl), then there is the content of first file, followed by the content of the second file. The sed command removes the first four lines.

Instead of sed you could plug in awk and access the fields from your files.

Depending on the files size this could be faster.

answered Feb 15, 2016 at 18:28

Lars Fischer

10.4k3 gold badges31 silver badges38 bronze badges

Comments

tbroberg · Accepted Answer · 2016-02-15 18:03:38Z

-1

To read in all but the first 4 lines of myfile: tail -n +4 myfile

answered Feb 15, 2016 at 18:03

tbroberg

6453 silver badges13 bronze badges

Collectives™ on Stack Overflow

go to a specific line of a text file by shell script

3 Answers 3

2 Comments

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

3 Answers 3

2 Comments

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related