CODEWITHSUNDEEP
arrayslinuxbashshellfor-loop
Lynchburg
Lynchburg

Reputation: 1

storing a string with separated space into one index bash

Here is my code

# text file has: column1:column2:column3:column4:column5:column6:column7
# such as: tang:orange juice

while read line ; do
fields=($(printf "%s" "$line" | cut -d':' --output-delimiter=' ' -f1-))
     for i in "${fields[@]}" ; do
            echo $i  #i also tried echo "$i" but same result
     done
done < file.txt

my output would be like:

tang
orange
juice

but it should be like:

tang
orange juice

Upvotes: 0

Views: 62

Answers (3)

Marco
Marco

Reputation: 902

Besides the very short version from @Barmar this would do it in a slightly longer way, which might be more intuitive for a beginner.

while IFS=: read -r -a fields
do
    printf "%s\n" "${fields[@]}"
done < file.txt

Upvotes: 0

Barmar
Barmar

Reputation: 782755

Use IFS to specify the field separator for the shell. Then you can split the line using : as the separator.

IFS=: fields=($line)

Upvotes: 1

paxdiablo
paxdiablo

Reputation: 882756

Your problem is that you turn the colons into spaces to handle the things as words, not taking into account the fact that some things already have spaces. That means they'll be split into separate words as well.

If your intent is just to take each line of input and place each thing on its own line, you can use something like sed 's/:/\n/g' as in the following transcript:

pax:~> printf "tang:orange juice:coke\nwater:pale ale\n" | sed 's/:/\n/g'
tang
orange juice
coke
water
pale ale

In your particular case, that would be the relatively simple:

sed 's/:/\n/g' file.txt

Upvotes: 0

Related Questions