Bash's extension to string equality tests for pattern matching is useful, but it can sometimes be more convenient to use the older POSIX-specified case construct, which is shell script's analogue of the C switch statement.

The case statement allows us to run commands based on the outcome of matching a string against glob patterns. For example, to check whether the command variable matches help, we might do this:

case $command in
    help) printf 'Command help:\n...' ;;
esac

Note the following details of this syntax:

• You don't have to double-quote $command just after the case statement.
• The pattern to match has a closing ), but does not require an opening one.
• The closing keyword is esac, which is case spelled backwards, just as the closing keyword for  if is fi.
• Each option is terminated with two semicolons.

We can add other options according to the same pattern:

case $command in
    help) printf 'Command help:\n...' ;;
    version) printf 'Command version: 0.1.0\n' ;;
esac

We can specify alternative values for any of these options for the same outcome by separating the patterns with a pipe character:

case $command in
    help|h|usage) printf 'Command help:\n...' ;;
esac

Perhaps most usefully, we can include glob-matching characters: *, ?, and [...] to test for partial matches. This is a good way to have a "catch all" or default option if nothing else matches:

case $command in
    debug|verb*)
        printf 'Running in verbose mode.\n'
        verbose=1
        ;;
    *) printf 'Unknown command\n' ;;
esac

Remember that only the first condition that matches will be run; the others will be ignored, as long as they're correctly separated with two semicolons, ;;.