Replace prefix string from lines in a file, and put into a bash array

Question

In the file groupAfiles.txt are the lines:

file14
file2
file4
file9

I need a way to convert them to remove file and add /dev/loop and put them all in one line with a space between them.

/dev/loop14 /dev/loop2 /dev/loop4 /dev/loop9

Then I need to put this in an array.

(but the numbers change)

How do I do this?

Answer 1

Read the original names into an array:

readarray -t names <groupAfiles.txt

Replace the initial file prefix with /dev/loop:

names=( "${names[@]/#file//dev/loop}" )

The substitution used here will replace the string file at the start of each element of the names array, if that prefix string is present, with the string /dev/loop.

To display the resulting array with spaces between the elements (assuming the shell variables IFS has its default value):

printf '%s\n' "${names[*]}"

Testing:

$ cat groupAfiles.txt
file14
file2
file4
file9
$ readarray -t names <groupAfiles.txt
$ names=( "${names[@]/#file//dev/loop}" )
$ printf '%s\n' "${names[*]}"
/dev/loop14 /dev/loop2 /dev/loop4 /dev/loop9

Answer 2

To do this with awk you can use command like:

awk '{gsub("file",""); printf "/dev/loop%s ", $0}' input_file

or

awk '{gsub("file","/dev/loop"); printf "%s ", $0}' input_file

It replace file with nothing and then print /dev/loop concatenated with the number/string

To add in array you can use:

array=( $(awk '{gsub("file",""); printf "/dev/loop%s ", $0}' input_file) )

Answer 3

Just read it into the array and then print it space-separated as you like:

$ readarray -t files < <(sed 's:file:/dev/loop:' groupAfiles.txt)

$ declare -p files
declare -a files=([0]="/dev/loop14" [1]="/dev/loop2" [2]="/dev/loop4" [3]="/dev/loop9")

$ printf '%s\n' "${files[*]}"
/dev/loop14 /dev/loop2 /dev/loop4 /dev/loop9