Add multiple line into one in shell script linux

Question

I have a file in linux called 1.txt with multiple lines and need to merge every three lines

into one line.I have mentioned the output below and need to know how to do that

[xxx@c0040229 ~]$ cat 1.txt
c0043726.xxx.xx
50G,3.6G,44G,8%,/opt/tibco    
200K,/opt/tibco/scripts
c0043725.xxx.xx
50G,9.5G,38G,21%,/opt/tibco
196K,/opt/tibco/scripts
c0043724.xxx.xx
50G,1.6G,46G,4%,/opt/tibco
92K,/opt/tibco/scripts

Out put should be

c0043726.xxx.xx,50G,3.6G,44G,8%,/opt/tibco,200K,/opt/tibco/scripts
c0043725.xxx.xx,50G,9.5G,38G,21%,/opt/tibco,196K,/opt/tibco/scripts
c0043724.xxx.xx,50G,1.6G,46G,4%,/opt/tibco,92K,/opt/tibco/scripts

Can somebody help me ?

First ask a question. The only question in what you've typed is "can somebody help me". Post the input and your expected output. Then mention what you've tried so that people can offer constructive advice. — Noufal Ibrahim
– Noufal Ibrahim, Commented Aug 1, 2014 at 6:51
I have edited the question with input as in "cat 1.txt" and out put as c0043726.xxx.xx,50G,3.6G,44G,8%,/opt/tibco,200K,/opt/tibco/scripts c0043725.xxx.xx,50G,9.5G,38G,21%,/opt/tibco,196K,/opt/tibco/scripts c0043724.xxx.xx,50G,1.6G,46G,4%,/opt/tibco,92K,/opt/tibco/scripts — user3311501
– user3311501, Commented Aug 1, 2014 at 6:57

darkgrin · Accepted Answer · 2014-08-01 07:02:39Z

3

I think you can achieve what you want with the paste command:

cat 1.txt | paste -d',' - - -

answered Aug 1, 2014 at 7:02

darkgrin

5803 silver badges8 bronze badges

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

4 Comments

user3311501 Over a year ago

[xxx@c0040229 ~]$ cat 1.txt | paste -d',' - - - ,200K,/opt/tibco/scriptsbco ,196K,/opt/tibco/scriptsibco ,92K,/opt/tibco/scriptsibco

user3311501 Over a year ago

it is different output

darkgrin Over a year ago

Your whole command should be just: cat 1.txt | paste -d',' - - -

darkgrin Over a year ago

It looks like your file might have Windows' line endings. In that case, could you try this? cat 1.txt | sed 's/\r//g' | paste -d',' - - -

damienfrancois · Accepted Answer · 2014-08-01 07:12:23Z

1

Use the paste command and give it stdin (-) as input three times like this:

$ cat 1.txt 
c0043726.xxx.xx
50G,3.6G,44G,8%,/opt/tibco
200K,/opt/tibco/scripts
c0043725.xxx.xx
50G,9.5G,38G,21%,/opt/tibco
196K,/opt/tibco/scripts
c0043724.xxx.xx
50G,1.6G,46G,4%,/opt/tibco
92K,/opt/tibco/scripts
$ paste -d, - - - < 1.txt 
c0043726.xxx.xx,50G,3.6G,44G,8%,/opt/tibco,200K,/opt/tibco/scripts
c0043725.xxx.xx,50G,9.5G,38G,21%,/opt/tibco,196K,/opt/tibco/scripts
c0043724.xxx.xx,50G,1.6G,46G,4%,/opt/tibco,92K,/opt/tibco/scripts

Paste is normally used to 'paste -- merge corresponding or subsequent lines of files', and if you feed it with stdin three times, you get the desired effet.

answered Aug 1, 2014 at 7:12

damienfrancois

60.5k9 gold badges116 silver badges128 bronze badges

4 Comments

user3311501 Over a year ago

for me output is always below . where is the problem ? [xxx@c0040229 ~]$ paste -d, - - - < 1.txt ,200K,/opt/tibco/scriptsbco ,196K,/opt/tibco/scriptsibco ,92K,/opt/tibco/scriptsibco

damienfrancois Over a year ago

is the file being transferred from a Windows computer to a UNIX computer or vice versa?

damienfrancois Over a year ago

what does cat -e 1.txt output?

user3311501 Over a year ago

Thanks For all you help .There is strange character ^M comming in 1.txt , which preventing me to slove this issue. [xxx@c0040229 ~]$ cat -v 1.txt | xargs -L 1| sed -e 's/ /,/g' c0043726.itcs.hp.com^M 50G,3.6G,44G,8%,/opt/tibco^M 200K,/opt/tibco/scripts^M c0043725.itcs.hp.com^M 50G,9.5G,38G,21%,/opt/tibco^M 196K,/opt/tibco/scripts^M c0043724.itcs.hp.com^M 50G,1.6G,46G,4%,/opt/tibco^M 92K,/opt/tibco/scripts^M Now it is ok after removing after ^M with sed command .This creature only visible by cat -v command. Thanks Again for your help

David C. Rankin · Accepted Answer · 2014-08-01 08:55:57Z

A simple script can also accomplish what you want. Reading your datafile and then merging 3 lines into one. Below the output is just echoed to the screen for illustration, but you can simply redirect each echo statement to a file using the >> append operator. This uses a case statement and simple counter. You can save a line or two using $cnt % 3 modulo operator to test for the 3rd line. Here is a short example:

#!/bin/bash

declare -i cnt=0

while IFS=$'\n' read -a line || test -n "$line"; do 
    ((cnt++))
    case "$cnt" in
      [1-2] ) echo -n "$line,";;
      3     ) echo "$line"; cnt=0;;
    esac
done < "$1"

exit 0

output:

$ ./3lines.sh dat/3line.dat
c0043726.xxx.xx,50G,3.6G,44G,8%,/opt/tibco,200K,/opt/tibco/scripts
c0043725.xxx.xx,50G,9.5G,38G,21%,/opt/tibco,196K,/opt/tibco/scripts
c0043724.xxx.xx,50G,1.6G,46G,4%,/opt/tibco,92K,/opt/tibco/scripts

Damanjeet Singh · Accepted Answer · 2014-08-01 10:35:33Z

0

You can use awk 'NR % 3 !=0 {printf $0;printf " "} NR % 3 ==0 {print " "}' fileName.txt

answered Aug 1, 2014 at 10:35

Damanjeet Singh

579 bronze badges

Collectives™ on Stack Overflow

Add multiple line into one in shell script linux

4 Answers 4

4 Comments

4 Comments

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

4 Answers 4

4 Comments

4 Comments

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related