Trying to read nth line from file and split in to array based on delimiter
HEAD_START=4
IFS='|' read -r -a headers < sed "${HEAD_START}q;d" "/FILE_UPLOADS/Checklist-Relationship (4).txt"
The above gives "sed: cannot open [No such file or directory]"
But when i run just sed "${HEAD_START}q;d" "/FILE_UPLOADS/Checklist-Relationship (4).txt" in the prompt it works fine
read -r -a headers < sed ... is trying to open a file named sed for reading.
In bash, to run sed as a command and make its output available on the standard input stream, you can use a process substitution:
IFS='|' read -r -a headers < <(sed "${HEAD_START}q;d" "/FILE_UPLOADS/Checklist-Relationship (4).txt")
IFS='|' read -r -a headers < <(sed "${HEAD_START}q;d" "/FILE_UPLOADS/Checklist-Relationship (4).txt") and the error i am getting now is read: -a: unknown option Usage: read [-ACprsSv] [-d delim] [-u fd] [-t timeout] [-n count] [-N count] [var?prompt] [var ...]
ksh rather than bash?
-A in place of -a in the builtin read command - see for example Read first line split based on delimiter and store it in array (in ksh)
If the file is not too big, I'd skip sed:
mapfile -t lines < filename
IFS='|' read -ra headers <<< "${lines[HEAD_START - 1]}"
