48

Is there a simple way I can echo a file, skipping the first and last lines? I was looking at piping from head into tail, but for those it seems like I would have to know the total lines from the outset. I was also looking at split, but I don't see a way to do it with that either.

91

Just with sed, without any pipes :

sed '1d;$d' file.txt

NOTE

  • 1 mean first line
  • d mean delete
  • ; is the separator for 2 commands
  • $ mean last line
33

You don't need to know the number of lines in advance. tail and head can take an offset from the beginning or end of the file respectively.

This pipe starts at the second line of the file (skipping the first line) and stops at the last but one (skipping the final line). To skip more than one line at the beginning or end, adjust the numbers accordingly.

tail -n +2 file.txt | head -n -1

doing it the other way round, works the same, of course:

head -n -1 file.txt | tail -n +2
  • I don't know why, but head -n -1 removes the first AND the last line of my .txt file, on Ubuntu 14.04.2LTS. – Sopalajo de Arrierez Dec 22 '15 at 23:20
8

Here is how to do it with awk:

awk 'NR>2 {print t} {t=$0}'

Also another way for sed:

sed '1d;x' file.txt

x is advanced sed command, it switches current line with the previous one: current goes into the buffer and previous goes to the screen and so on while sed processing stream line by line (this is why the first line will be blank).

awk solution on each step (line) puts current line into the variable and starts printing it out only after the second line is passed by. Thus, we got shitfed sequence of lines on the screen from the second to the last but one. Last line is omitted becasue the line is in the variable and should be printed only on the next step, but all steps already run out and we never see the line on the screen.

Same idea in the perl:

perl -ne 'print $t if $.>2 ; $t=$_' file.txt

$. stands for line number and $_ for current line.
perl -n is shortcut for while(<..>) {..} structure and -e is for inline script.

1

In python i would do like this.

#!/usr/bin/python3
import re
import sys
file = sys.argv[1]
with open(file, 'r') as f:
    L = []
    for line in f:
        line = re.sub(r'\n', r'', line)
        L.append(line)
    print('\n'.join(L[1:-1]))

Paste the above code into a file and name it as script.py. Run the script against the file you want to check with.

python3 script.py /path/to/the/file

Example:

$ cat file
foo
apple
banana
bar
$ python3 script.py file
apple
banana
1

If you want to solve this problem, only using "head" and "tail" commands, then the solution provided by Olaf Dietsche, for this question works perfectly fine. But, the head -n -1 option does not work on Mac.

For Mac Users:

On Mac, head -n -1 wont work. And, I was trying to find a simple solution [ without worrying about processing time ] to solve this problem only using "head" and/or "tail" commands.

I tried the following sequence of commands and was happy that I could solve it just using "tail" command [ with the options available on Mac ]. So, if you are on Mac, and want to use only "tail" to solve this problem, you can use this command :

cat file.txt | tail -r | tail -n +2 | tail -r

Explanation:

  1. tail -r : simply reverses the order of lines in its input

  2. tail -n +2 : this prints all the lines starting from the second line in its input

You can combine this with the solution provided by Olaf Dietsche, to solve your question.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.