I have an array tokens which contains tokens=( first one two three last ). How do I obtain the values ( one two three ) if I do not know how many numbers are in the array? I want to access everything between first and last (exclusive).
echo ${tokens[*]:1:3}
will give one two three but if I do not know the length of the array how can I get all the elements after first and before last? I am looking for something similar to using negative indices in Python such as tokens[1:-1]
If the array is not sparse, you can do:
bash-5.2$ tokens=( {1..10} )
bash-5.2$ printf ' - %s\n' "${tokens[@]:1:${#tokens[@]}-2}"
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
If the array may be sparse, you'd need to determine the index of the second element (here assuming it has at least 2 elements):
bash-5.2$ tokens=([12]=a [15]=b [23]=c [123]=d)
bash-5.2$ ind=( "${!tokens[@]}" )
bash-5.2$ printf ' - %s\n' "${tokens[@]:ind[1]:${#tokens[@]}-2}"
- b
- c
In zsh (which has normal arrays, not sparse arrays), it's just $tokens[2,-2].
Depending on your version of Bash, the following should work:
tokens=( first one two three last )
echo "${tokens[@]:1:${#tokens[@]}-2}"
Result
one two three
