How do I split a string on a delimiter in Bash?

You can set the internal field separator (IFS) variable, and then let it parse into an array. When this happens in a command, then the assignment to IFS only takes place to that single command's environment (to read ). It then parses the input according to the IFS variable value into an array, which we can then iterate over.

This example will parse one line of items separated by ;, pushing it into an array:

IFS=';' read -ra ADDR <<< "$IN"
for i in "${ADDR[@]}"; do
  # process "$i"
done

This other example is for processing the whole content of $IN, each time one line of input separated by ;:

while IFS=';' read -ra ADDR; do
  for i in "${ADDR[@]}"; do
    # process "$i"
  done
done <<< "$IN"

If you don't mind processing them immediately, I like to do this:

for i in $(echo $IN | tr ";" "\n")
do
  # process
done

You could use this kind of loop to initialize an array, but there's probably an easier way to do it.

How about this approach:

IN="bla@some.com;john@home.com" 
set -- "$IN" 
IFS=";"; declare -a Array=($*) 
echo "${Array[@]}" 
echo "${Array[0]}" 
echo "${Array[1]}" 

Source

echo "bla@some.com;john@home.com" | sed -e 's/;/\n/g'
bla@some.com
john@home.com

you can apply awk to many situations

echo "bla@some.com;john@home.com"|awk -F';' '{printf "%s\n%s\n", $1, $2}'

also you can use this

echo "bla@some.com;john@home.com"|awk -F';' '{print $1,$2}' OFS="\n"

IN="bla@some.com;john@home.com"
IFS=';'
read -a IN_arr <<< "${IN}"
for entry in "${IN_arr[@]}"
do
    echo $entry
done

Output

bla@some.com
john@home.com

System : Ubuntu 12.04.1

ADDR1=${IN%%;*}
ADDR2=${IN##*;}

Edit: I'm sorry, I had read somewhere on SO that perl is required by POSIX, so I thought it would be legitimate to use it. But on unix.stackexchange.com, several users state perl is NOT part of the POSIX specification.

My solution: A function that uses perl's split to do the work.

With detailed comments:

#!/bin/bash

# This function is a wrapper for Perl's split.\
# \
# Since we cannot return an array like in Perl,
# it takes the name of the resulting array as last
# argument.\
# \
# See https://perldoc.perl.org/functions/split for usage info
# and examples.\
# \
# If you provide a Perl regexp that contains e. g. an escaped token like \b,
# space(s) and/or capture group(s), it must be quoted, and e. g. /\b/ must
# be single-quoted.\
# Thus, it's best to generally single-quote a Perl regexp.
function split # Args: <Element separator regexp> <string> <array name>
{
    (($# != 3)) && echo "${FUNCNAME[0]}: Wrong number of arguments, returning." && return 1

    local elementSepRE=$1
    local string=$2
    local -n array=$3

    local element i=0

    # Attention! read does Word Splitting on each line!
    # I must admit I didn't know that so far.
    # This removes leading and trailing spaces, exactly
    # what we don't want.
    # Thus, we set IFS locally to newline only.
    local IFS=$'\n'

    while read element; do
        # As opposed to array+=($element),
        # this preserves leading and trailing spaces.
        array[i++]=$element
    done <<<$(_perl_split)
}

# This function calls Perl's split function and prints the elements of the
# resulting array on separate lines.\
# It uses the caller's $elementSepRE and $string.
function _perl_split
{
    # A heredoc is a great way of embedding a Perl script.
    # N.B.: - Shell variables get expanded.
    #         - Thus:
    #           - They must be quoted.
    #           - Perl scalar variables must be escaped.
    #       - The backslash of \n must be escaped to protect it.
    #       - Instead of redirecting a single heredoc to perl, we may
    #         use multiple heredocs with cat within a command group and
    #         pipe the result to perl.
    #         This enables us to conditionally add certain lines of code.

    {
        cat <<-END
            my \$elementSepRE=q($elementSepRE);
        END

        # If $elementSepRE is a literal Perl regexp, qr must be applied
        # to it in order to use it.
        # N.B.: We cannot write this condition in Perl because when perl
        # compiles the script, all statements are checked for validity,
        # no matter if they will actually be executed or not.
        # And if $elementSepRE was e. g. == ', the line below – although
        # not to be executed – would give an error because of an unterminated
        # single-quoted string.
        [[ $elementSepRE =~ ^m?/ && $elementSepRE =~ /[msixpodualn]*$ ]] && cat <<-END
            \$elementSepRE=qr$elementSepRE;
        END

        cat <<-END
            my @array=split(\$elementSepRE, q($string));

            print(\$_ . "\\n") for (@array);
        END
    } | perl
}

And the same without comments for those who see at a glance what's going on ;)

#!/bin/bash

# This function is a wrapper for Perl's split.\
# \
# Since we cannot return an array like in Perl,
# it takes the name of the resulting array as last
# argument.\
# \
# See https://perldoc.perl.org/functions/split for usage info
# and examples.\
# \
# If you provide a Perl regexp that contains e. g. an escaped token like \b,
# space(s) and/or capture group(s), it must be quoted, and e. g. /\b/ must
# be single-quoted.\
# Thus, it's best to generally single-quote a Perl regexp.
function split # Args: <Element separator regexp> <string> <array name>
{
    (($# != 3)) && echo "${FUNCNAME[0]}: Wrong number of arguments, returning." && return 1

    local elementSepRE=$1
    local string=$2
    local -n array=$3

    local element i=0

    local IFS=$'\n'

    while read element; do
        array[i++]=$element
    done <<<$(_perl_split)
}

function _perl_split
{
    {
        cat <<-END
            my \$elementSepRE=q($elementSepRE);
        END

        [[ $elementSepRE =~ ^m?/ && $elementSepRE =~ /[msixpodualn]*$ ]] && cat <<-END
            \$elementSepRE=qr$elementSepRE;
        END

        cat <<-END
            my @array=split(\$elementSepRE, q($string));

            print(\$_ . "\\n") for (@array);
        END
    } | perl
}

Yet another late answer... If you are java minded, here is the bashj (https://sourceforge.net/projects/bashj/) solution:

#!/usr/bin/bashj

#!java

private static String[] cuts;
private static int cnt=0;
public static void split(String words,String regexp) {cuts=words.split(regexp);}
public static String next() {return(cnt<cuts.length ? cuts[cnt++] : "null");}

#!bash

IN="bla@some.com;john@home.com"

: j.split($IN,";")    # java method call

while true
do
    NAME=j.next()     # java method call
    if [ $NAME != null ] ; then echo $NAME ; else exit ; fi
done