How do I take a string, something simple like "Hello World!" and split it into its individual characters?
Using the example above, I want an array with a single character put into each value. So the array's internals would end up looking like:
{"H", "e", "l", "l", "o", " ", "W", "o", "r", "l", "d", "!"}
str="Hello world!"
for (( i=0 ; i < ${#str} ; i++ )) {
arr[$i]=${str:i:1}
}
#print
printf "=%s=\n" "${arr[@]}"
output
=H=
=e=
=l=
=l=
=o=
= =
=w=
=o=
=r=
=l=
=d=
=!=
You can assign into array the result of any command using the
mapfile -t array < <(command args)
Unfortunately, defining the custom delimiter -d needs bash 4.4.+. Let say, want break the above string into 2 char pieces - using grep
mapfile -t -d '' a2 < <(grep -zo .. <<<"$str")
printf "=%s=\n" "${a2[@]}"
output:
=He=
=ll=
=o =
=wo=
=rl=
=d!=
It is a fairly simple thing to do in bash using array indexing. Simply loop over all chars and pick then off into an array, e.g.
#!/bin/bash
a="Hello World!"
for ((i = 0; i < ${#a}; i++)); do
array+=("${a:i:1}") ## use array indexing for individual chars
done
printf "%s\n" "${array[@]}" ## output individual chars
Example Use/Output
$ sh bashchar.sh
H
e
l
l
o
W
o
r
l
d
!
"${array[@]}", which swallowed the space ;)
:)${array[@]} unquoted gets split up and the space disappears.:)I can see two ways to do that. In pure Bash, iterating over the string character by character and adding each character to an array:
$ str='Hello World!'
# for (( i = 0; i < ${#str}; ++i )); do myarr+=("${str:i:1}"); done
$ declare -p myarr
declare -a myarr='([0]="H" [1]="e" [2]="l" [3]="l" [4]="o" [5]=" " [6]="W" [7]="o" [8]="r" [9]="l" [10]="d" [11]="!")'
The key element is the substring expansion, "${str:i:1}", which expands to the substring of str that starts at index i and has length 1. Notice that this is one of the few times where you don't have to prepend a variable with $ to get its contents, because the i is in an arithmetic context here.
Using an external tool, fold:
$ readarray -t arr <<< "$(fold -w 1 <<< "$str")"
$ declare -p arr
declare -a arr='([0]="H" [1]="e" [2]="l" [3]="l" [4]="o" [5]=" " [6]="W" [7]="o" [8]="r" [9]="l" [10]="d" [11]="!")'
fold -w 1 wraps the input string to one character per line, and the readarray command read its input into an array, line by line (-t removes the newline characters from each element).
Notice that readarray requires Bash 4.0 or newer.
Pure Bash approach - walk through the string one character at a time and grab the substring:
#!/bin/bash
declare -a arr
string="Hello World!"
for ((i = 0; i < ${#string}; i++)); do
# append i'th character to the array as a new element
# double quotes around the substring make sure whitespace characters are protected
arr+=("${string:i:1}")
done
declare -p arr
# output: declare -a arr=([0]="xy" [1]="y" [2]="H" [3]="e" [4]="l" [5]="l" [6]="o" [7]="W" [8]="o" [9]="r" [10]="l" [11]="d" [12]="!")
Try this -
$v="Hello World!"
$awk '{n=split($0,a,""); for(i=1;i<=n;i++) {print a[i]}}' <<<"$v"
H
e
l
l
o
W
o
r
l
d
!
awk '{ for ( i=1;i<=length($0);i++ ) printf substr($0,i,1)"\n" }' <<< $str
mapfile arry1 < <(echo "$str1")
If you are trying to generate JSON, use square instead of curly brackets and jq instead of Bash:
jq -c -R 'explode|map([.]|implode)' <<<'Hello World!'
["H","e","l","l","o"," ","W","o","r","l","d","!"]
For the sake of variety, a solution in pure Bash without array indexing:
string="Hello world"
split=( )
while read -N 1; do
split+=( "$REPLY" )
done < <( printf '%s' "$string" )
The last line does process substitution to pass the output of printf to a loop. The loop uses read -N 1 to read exactly one character at a time.
awk -v FS="" ' {printf"{";for(i=1;i<=NF;i++) if(i!=NF)printf "\""$i"\",";print $i"}"}' inputfilethis has flaws so not putting it as answer.