CODEWITHSUNDEEP
bashsedawkcut
Olga
Olga

Reputation: 1022

how to cut several lines from file bash unix

I have a text file:

Hello
1
2
3
(unknown number of lines)
Hello
(unknown number of lines)
Hello
(unknow number of lines)
Hello

How to cut lines between two first "hello" and save it to a file?
So the output will be

1
2
3
(unknown number of lines)

Upvotes: 2

Views: 5706

Answers (2)

Iskren
Iskren

Reputation: 1341

Here's a simple bash script that worked for me:

#!/bin/bash
WORD="$1" # Word we look for, in this case 'Hello'
COUNT=0 # Internal counter for words
let MAXCOUNT="$2" # How many words to encounter before we stop
OUTPUT="$3" # Output filename
FILENAME="$4" # The file to read from
while read -r; do # read the file line by line
    [ "$MAXCOUNT" -le "$COUNT" ] && break; # if we reached the max number of occurances, stop
    if [[ "$WORD" = "$REPLY" ]]; then # current line holds our word
        let COUNT=$COUNT+1; # increment counter
        continue; # continue reading
    else # if current line is not holding our word
        echo "$REPLY" >> "$OUTPUT"; # print to output file
    fi
done <"$FILENAME" # this feeds the while with our file's contents

Worked like this:

$./test.sh "Hello" 2 output.txt test.txt # Read test.txt, look for "Hello" and print all lines between the first two occurances into output.txt

This is what I've got:

$cat output.txt 
1
2
3
(unknown number of lines)

And test.txt contains:

Hello
1
2
3
(unknown number of lines)
Hello
(unknown number of lines)
Hello
(unknow number of lines)
Hello

Upvotes: 0

Gilles Qu&#233;not
Gilles Qu&#233;not

Reputation: 185025

Using awk :

awk '$1=="Hello"{c++;next} c==1' oldfile | tee newfile

To have the Nth occurence, change the count variable : 

awk -v count=1 '$1=="Hello"{c++;next} c==count' oldfile | tee newfile

Upvotes: 3

Related Questions