sed - Multiple replacements in single command - sed

Question

EDIT
My use case does require substitution of a URL pattern.
thanks #glenn jackman for response but there may be a problem escaping the // characters in the URL. That was why I was using | for the delimiter.
sh-4.4$ EC2_ID="http://test.com"
sh-4.4$ sed -E '
s|\\$\{instance-hostname\}|\ ${EC2_HOSTNAME}|
s|\\$\{instance-id\}|\ ${EC2_ID}|
' test.txt
AWS EC2 Instance DNS: ${instance-hostname}
AWS EC2 Instance ID: ${instance-id}
sh-4.4$ sed -E '
s/\$\{instance-hostname\}/'"${EC2_HOSTNAME}"'/
s/\$\{instance-id\}/'"${EC2_ID}"'/
' test.txt
sed: -e expression #1, char 65: unknown option to `s'
============================================
I would appreciate assistance with my sed script.
sh-4.4$ EC2_HOSTNAME="A"
sh-4.4$ EC2_ID="B"
sh-4.4$ sed -r "s|\\$\{instance-hostname\}|\ ${EC2_HOSTNAME}|" -r "s|\\$\{instance-id\}|\ ${EC2_ID}|" test.txt
sed: can't read s|\$\{instance-id\}|\ B|: No such file or directory
AWS EC2 Instance DNS: A
AWS EC2 Instance ID: ${instance-id}
sh-4.4$ cat test.txt
AWS EC2 Instance DNS: ${instance-hostname}
AWS EC2 Instance ID: ${instance-id}
sh-4.4$
Substitution works if only one replacement instruction is provided - on either replacement token.
sh-4.4$ sed -r "s|\\$\{instance-hostname\}|\ ${EC2_HOSTNAME}|" test.txt
AWS EC2 Instance DNS: A
AWS EC2 Instance ID: ${instance-id}
sh-4.4$ sed -r "s|\\$\{instance-id\}|\ ${EC2_ID}|" test.txt
AWS EC2 Instance DNS: ${instance-hostname}
AWS EC2 Instance ID: B
sh-4.4$
FYI
sh-4.4$ cat /etc/os-release
NAME="Red Hat Enterprise Linux"
VERSION="8.5 (Ootpa)"
ID="rhel"
ID_LIKE="fedora"
VERSION_ID="8.5"
sed version
sh-4.4$ sed --version
sed (GNU sed) 4.5
Copyright (C) 2018 Free Software Foundation, Inc.
Thanks

Answer 1

The sed body can contain newlines, so this is a readable way to write it:
sed -E '
s/\$\{instance-hostname\}/'"${EC2_HOSTNAME}"'/
s/\$\{instance-id\}/'"${EC2_ID}"'/
' test.txt
AWS EC2 Instance DNS: A
AWS EC2 Instance ID: B
If the shell variables contain the / character, sed will throw errors. There are (ugly) workarounds if this is an issue.
perl allows a more dynamic approach, assuming the "instance_*" and "EC2_*" variable names align
export EC2_HOSTNAME EC2_ID
perl -pe 's<\$\{instance-(\w+)\}>< $key = "EC2_" . uc($1); $ENV{$key} >e' test.txt

Answer 2

You really don't want to use sed for this. If the original author of test.txt had used reasonable names, this would be trivial. Fortunately, your current input is easily fixable:
$ cat test.txt
AWS EC2 Instance DNS: ${instance-hostname}
AWS EC2 Instance ID: ${instance-id}
$ tr - _ < test.txt | { instance_hostname=A instance_id=B envsubst; }
AWS EC2 Instance DNS: A
AWS EC2 Instance ID: B