Designing a discord bot commands module

I am currently designing a Discord bot. I am working on a module to easily make commands.

This is currently my commands module.

module.exports = {
add: function(name, help, permissions, callback) {
    var command = {};
    command['name'] = name;
    command['help'] = help;
    command['permissions'] = permissions;
    command['callback'] = callback;

    list.push(command);
},

parse: function(message, data) {
    var args = message.split(' ');

    for (var i = 0; i < list.length; i++) {
        var command = list[i];

        if (command.name == args[0]) {
            isAuthorized(command.permissions, data, function() {
                console.log("RAN");
                command.callback(data, args);
            });
        }
    }
},

get: function() {
    return list;
}
}

and here is how I would go about using it.

commands.add('ping', 'Send a ping-pong request.', null, function(data, args) {
if (args[1] == "-p") {
    discord.sendMessage({to: data.channelID, message: "pong"});
} else {
    discord.sendMessage({to: data.channelID, message: "ding"});
}
});

It all good, however I am worry about the design. I am not liking it too much as it can be buggy and annoying. For instance, the above would simply parse a command by a whitespace. Like so:

!ping -d

It will call the ping function and check the first element of args to see if it contain a "-d" flag. I want this to be as similar to Linux terminal as I can get. The above is simple, but what about a command that takes more than one flag? For example:

!ping -d -help

Which will obviously print out "ding" and if implemented it should also print something that would describe the command and it arguments.

But what if I write something like

!ping -help -d -f -m

How would that work out? I rather not have a bunch of conditional statement.

Permissions

Another concern I thought about is permission levels. I have each role setup in a database.

+----+-----------+
| id | name      | 
+----+-----------+
|  1 | guest     |
|  2 | member    |
|  3 | mod       |
|  4 | developer |
|  5 | admin     |
|  6 | master    |
|  8 | owner     |
+----+-----------+

These roles were created using this command.

roles.createRole = function(name) {
database.insert('roles', ['name'], ['"' + name + '"']);
}

commands.add('createrole', 'Create a new role.', ['master'], function(data, args) {
var role = args[1];

if (role) {
    roles.createRole(role);
}
 });

And if they have the permission called "master" they are able to create a role.

This is how I am storing the users information.

-------------------+ 
| id | username      | discord_id         | permissions                                           
|
+----+---------------+--------------------+-------------------------------------------------------+
|  1 | kurieita      | 000000000 |   {"test":true,"master":true,"admin":true,"guest":true} |

Obviously I have the role and can execute the command. But what if I set someone to "admin", I also want them to have moderator powers. I would have to give them moderator powers too.

Ideas

Here are my ideas I came up with (while I was writing this)

• Add "value" field for database.roles and check if user permission level is greater than the required permission field.

• I thought about doing a more OOP approach. Something along these lines.

  var command = new Command(name, description); command.setCallback(function() {}); command.setFlag(name, function() {});

  I hadn't really thought much about that or how it would work tho.

• Another idea I had which is similar to the way I am doing it now is this

  commands.add(name, description, permissions, function(data, args, flags) {}

where name and description are string, permission is a index array. Data is an array based wit Discord information (userID, channelID, etc), args is an array with values. For example: !message text "text" would be an arg and flags would contain anything prepended with "-". For instance !ping -d, "d" would be in the flags array.

Questions

• How does Linux parse it commands and flags?
• How does Linux take care of permission levels for each command?
• How would I go about implementing a Linux like parser for my discord bot commands module?
• What would the best way to deal with permission levels be?

Any help is appreciated. Thank you in advanced.

via Kurieita