Converting a list into a single line string in bash

Question

I have a file that contains a list:

line1
line2
line3
.
.
.

I want to join everything in a string in bash like so:

"line1", "line2", "line3", ...... (no comma at the end)

How do I do that?

Edit: This is not a duplicate of How to join multiple lines of file names into one with custom delimiter? because I am trying to get every item in double quotes while joining.

Do you want to merge the lines in another file or in a variable in your script? — Fred, Jan 30 '17 at 15:07
Yours is still a duplicate; if you need help figuring out how to wrap each line in double quotes before joining, that's a separate question (but also [a duplicate](http://stackoverflow.com/questions/6554066/how-to-enclose-every-line-in-a-file-in-double-quotes-with-sed)). — tripleee, Jan 31 '17 at 04:48

Benjamin W. · Answer 1 · 2020-10-27T13:27:12.207

For an input file like

line1
line2
line3

I would use sed and tr as follows:

$ sed 's/.*/"&"/;$!s/$/, /' infile | tr -d '\n'
"line1", "line2", "line3"

The first sed command quotes every line; the second one adds , at the end of every line except the last (the $! address: "not the last line").

tr then removes all linebreaks.

If you want to do it completely in Bash:

#!/usr/bin/env bash

mapfile -t arr < infile               # Read file into array
arr=("${arr[@]/#/\"}")                # Prepend " to each element
arr=("${arr[@]/%/\"}")                # Append " to each element
IFS=,                                 # Set separator to comma
str="${arr[*]}"                       # Comma separated string
printf '%s\n' "${str//\",\"/\", \"}"  # Insert space after commas

score 3 · Answer 2 · answered Jan 30 '17 at 17:21

3

Using awk:

$ awk '{ printf "%s\"%s\"", (NR==1?"":", "), $0 } END{ print "" }' file
"line1", "line2", "line3"

answered Jan 30 '17 at 17:21

James Brown

36,089
7
43
59

score 2 · Accepted Answer · edited Jan 30 '17 at 16:01

2

Lets say that you file name is test then this will do the trick

  while IFS= read -r i; do echo " \"$i\""; done < test | paste -sd "," |sed 's/ //1'

edited Jan 30 '17 at 16:01

chepner

497,756
71
530
681

answered Jan 30 '17 at 15:04

Mitesh Pant

524
2
15

I figured this out but it does not include the quotes. – Iokanaan Iokan Jan 30 '17 at 15:06
now try again, I have updated the answer – Mitesh Pant Jan 30 '17 at 15:11
Looks better but still no spaces after commas. – Iokanaan Iokan Jan 30 '17 at 15:14
1

Use `while read` instead of your `for` loop. – Chem-man17 Jan 30 '17 at 15:18
I have now added the space as well – Mitesh Pant Jan 30 '17 at 15:19
`for i in $(cat xyz); do echo " \"$i\""; done |paste -sd ","|sed 's/ //1'` – Ronak Patel Jan 30 '17 at 15:20
You can use @RonakPatel sed to remove the first space that is added – Mitesh Pant Jan 30 '17 at 15:22
Can the -ve voter please comment ?? – Mitesh Pant Jan 30 '17 at 15:35
I didn't vote, but [don't read lines with `for`](http://mywiki.wooledge.org/DontReadLinesWithFor). – Benjamin W. Jan 30 '17 at 15:42
This is not guaranteed to work in its current form. Your loop does not iterate over the lines of the file; it iterates over words resulting from the command substitution. – chepner Jan 30 '17 at 15:42
cat reads each line in the file, and the for loop is over the output of cat, so how is it not iterating . Plus I have verified it to be working – Mitesh Pant Jan 30 '17 at 15:47
2

Try it with an input file that contains more than one word on a single line. – Benjamin W. Jan 30 '17 at 15:49
Also try it with a line like `*`. – chepner Jan 30 '17 at 15:51

score 1 · Answer 4 · answered Jan 30 '17 at 23:21

Building a general solution to most problems of this kind.
Using a general start, middle and end string

bash A solution with only bash:

#!/bin/bash

infile="infile"
start='"'
middle='", "'
end='"'

res="$start"                             # start the result with str "$start".
while  IFS=$'\n' read -r line            # for each line in the file.
do     res="${res}${line}${middle}"      # add the line and the middle str.
done   <"$infile"                        # for file "$infile"
res="${res%"$middle"}${end}"             # remove "$middle" on the last line.
printf '%s\n' "${res}"                   # print result.

awk And a solution for larger files with awk:

#!/bin/bash

infile="infile"
start='"'
middle='", "'
end='"'

awk -vs="$start" -vm="$middle" -ve="$end" '
BEGIN{printf("%s",s)}
{
if(ll){printf("%s%s",ll,m)}
ll=$0
}
END{printf("%s%s%s",ll,e,"\n")}
' "$infile"

Converting a list into a single line string in bash

4 Answers4