Process Substitution in Practice (diff <(...) <(...))

#!/bin/bash
diff <(printf "a\nb\nc\n") <(printf "a\nb\nc\n") && echo "identical output, no differences"

#!/bin/bash
if diff <(printf "a\nb\n") <(printf "a\nc\n") > /dev/null; then
    echo "no differences"
else
    echo "differences were found"
fi

#!/bin/bash
list_a="banana\napple\ncherry"
list_b="cherry\nbanana\napple"
if diff <(printf "$list_a" | sort) <(printf "$list_b" | sort) > /dev/null; then
    echo "both lists contain the same elements"
fi

1. Trying to use process substitution with commands that require a real seekable file rather than a pipe-like descriptor; some tools don't support it.
2. Forgetting diff returns a non-zero exit status when differences are found, which is expected behavior, not an error.
3. Using process substitution in a POSIX sh script and being confused why it fails; it's a bash (and a few other shells') extension.

• diff <(cmd1) <(cmd2) compares two commands' live output without intermediate temp files.
• diff exits 0 if inputs are identical and 1 if they differ, which scripts can check directly.
• Process substitution can be used anywhere a filename is expected, not just with diff.
• Each <(...) runs its command in a subshell, so it cannot leak variables back to the calling script.