How to iterate over text file having multiple-words-per-line using shell script?

Question

I know how to iterate over lines of text when the text file has contents as below:

abc  
pqr  
xyz

However, what if the contents of my text fil

Answer 1

If the delimiter is a space then you can do:

#!/bin/bash

ALLVALUES=()
while read line
do
    ALLVALUES+=( $line )
done < "/path/to/your/file"

So after, you can just reference an element by ${ALLVALUES[0]} or ${ALLVALUES[1]} etc

Answer 2

read splits the line by $IFS as many times as you pass variables to it:

while read var1 var2 ; do
    echo "var1: ${var1} var2: ${var2}"
done

You see, if you pass var1 and var2 both columns go to separate variables. But note that if the line would contain more columns var2 would contain the whole remaining line, not just column2.

Type help read for more info.

Answer 3

If you want to read every word in a file into a single array you can do it like this:

arr=()
while read -r -a _a; do
   arr+=("${a[@]}")
done < infile

Which uses -r to avoid read from interpreting backslashes in the input and -a to have it split the words (splitting on $IFS) into an array. It then appends all the elements of that array to the accumulating array while being safe for globbing and other metacharacters.

Answer 4

This awk command reads the input word by word:

awk -v RS='[[:space:]]+' '1' file

abc
xyz
cdf
pqr
lmn
rst

To populate a shell array use awk command in process substitution:

arr=()
while read -r w; do
   arr+=("$w")
done < <(awk -v RS='[[:space:]]+' '1' file)

And print the array content:

declare -p arr
declare -a arr='([0]="abc" [1]="xyz" [2]="cdf" [3]="pqr" [4]="lmn" [5]="rst")'