Select a specific line of an array in a for loop - arrays

Question

I have an array with multiple lines, and I would like to select a specific line of the array in a for loop. And then use the data in the selected line for some analysis in the for loop.
ARRAY=(A1,A2
B1,B2
C1,C2)
When I applied the below code, I can assign "B1" to Trait1 and "B2" to Trait2. This works for j=1 or j=3 as well.
j=2
echo ${ARRAY[$j-1]}
Trait1="$(echo ${ARRAY[$j-1]} | cut -d',' -f1)"
Trait2="$(echo ${ARRAY[$j-1]} | cut -d',' -f2)"
echo $Trait1
echo $Trait2
Ultimately, I want to put the above code in a for loop. But this failed.
nLine=3
for j in $(eval echo "{1..$nLine}")
do
Trait1="$(echo ${ARRAY[$j-1]} | cut -d',' -f1)"
Trait2="$(echo ${ARRAY[$j-1]} | cut -d',' -f2)"
echo $Trait1
echo $Trait2
done
-bash: 1 2 3-1: syntax error in expression (error token is "2 3-1")
-bash: 1 2 3-1: syntax error in expression (error token is "2 3-1")
Thank you.

Answer 1

Keeping the current code design I'd probably want to use parameter expansion to break the array element into its 2 components ...
Data setup:
$ ARRAY=(A1,A2
B1,B2
C1,C2)
$ typeset -p ARRAY
declare -a ARRAY=([0]="A1,A2" [1]="B1,B2" [2]="C1,C2")
Quick demo of parameter expansion to break array element into parts:
$ x=${ARRAY[0]}
$ echo ${x}
A1,A2
$ echo ${x%,*}
A1
$ echo ${x#*,}
A2
Updating the PO's for loop to use parameter expansion, and replacing the eval/echo/j-1 with something a bit easier to read:
nLine=3
for (( j=0 ; j<nLine ; j++ ))
do
Trait1="${ARRAY[${j}]%,*}"
Trait2="${ARRAY[${j}]#*,}"
echo "${Trait1}"
echo "${Trait2}"
done
Generates the output:
A1
A2
B1
B2
C1
C2

Answer 2

The way you're using it, you should have a single string, not an array.
ARRAY='A1,A2
B1,B2
C1,C2'
Then when you use it as a here-string input to sed, it will receive multiple lines of input, and you can select the line you want with ${j}p.
As mentioned in the comments, this seems like a convoluted way to access array elements. You can simply use the shell's built-in array indexing operator.

Answer 3

Array assignments
ARRAY=(A1,A2
B1,B2
C1,C2)
and
ARRAY=(A1,A2 B1,B2 C1,C2)
are equivalent and both define an array with 3 elements: A1,A2, B1,B2, and C1,C2. Space(s), tab(s) and new-line(s) are used to separate array elements in compound assignments.
A for loop can be used to traverse an array:
# "${ARRAY[#]}" expands each element of ARRAY to a separate word
for elm in "${ARRAY[#]}"; do
# The value of the array element accessed is "$elm"
done
Or, using the alternate form of the for loop:
# ${#ARRAY[*]} expands to the number of elements in the ARRAY
for ((j = 0; j < ${#ARRAY[*]}; ++j)); do
# Array element can be referenced using ${ARRAY[j]}
done
assuming there is no gap between indices (bash allows an array with "holes").
Here strings, a variant of here documents, and read builtin with IFS variable can be used to parse the elements of the array:
IFS=, read Trait1 Trait2 <<< ${ARRAY[j]}
Here the IFS is used with the value , to cut the array element into Trait1 and Trait2 when there is a ,.
So, a sample program to demonstrate all of these may be:
#!/bin/bash
ARRAY=(A1,A2
B1,B2
C1,C2)
for elem in "${ARRAY[#]}"; do
IFS=, read Trait1 Trait2 <<< $elem
printf "%s\n%s\n" "$Trait1" "$Trait2"
done
or,
#!/bin/bash
ARRAY=(A1,A2
B1,B2
C1,C2)
for ((j = 0; j < ${#ARRAY[*]}; ++j)); do
IFS=, read Trait1 Trait2 <<< ${ARRAY[j]}
printf "%s\n%s\n" "$Trait1" "$Trait2"
done