Easy to use, do-it-yourself Discord.js mini-framework
You can find the full reference wiki here.
Making Discord.JS bots when you're in a run. You can get started with only 2 lines! (incluiding the import)!
All you need to get started is install it using npm install discordjs-diy and import it into your project.
import { Bot } from "discordjs-diy";
const bot = new Bot("<your Discord API token>", { prefix: "!" });
bot.registerAction("ping", "pong"); //!ping => pongimport { Bot } from "discordjs-diy";
const bot = new Bot("<your Discord API token>", { prefix: "?" });
bot.registerAction(
"rate",
({ args }) =>
`You want me to rate ${args}? Ok... ${Math.floor(Math.random() * 10)}/10`
); //?rate something => The bot replies!import { Bot } from "discordjs-diy";
const bot = new Bot("<your Discord API token>", { prefix: "*" });
bot.registerAction("reactAndReply", {
response: "Hello!",
reaction: () => "🤓",
}); //*reactAndReply => The bot can also react to messages and reply!import { Bot } from "discordjs-diy";
const bot = new Bot("<your Discord API token>", { prefix: "*", ignoreCaps });
bot.registerAction("await", {
response: async ({ msg }) => await waitForResponse(msg.author),
}); //*AWAIT => bot can ignore caps and use async/await!The bot can also easily update presence information
const bot = new Bot("<you know the deal>", { prefix: "!" });
bot.setPresence(["a game", "PLAYING"]);
//or, in case you want it to change every 10 minutes
bot.setPresence([
["a game", "PLAYING"],
["a movie", "WATCHING"],
]);
//you can also set your own time (in ms)
bot.setPresence(
[
["a game", "PLAYING"],
["a movie", "WATCHING"],
],
5 * 60 * 1000
);You can always access the normal client object from Discord.JS
const bot = new Bot("<you know the deal>", { prefix: "!" });
bot.client.on("<some action>", () => "<do something>");Discordjs-diy tries to make using embeds a little easier. You first need to create an Embed object
import { Embed } from "discordjs-diy";
const embed = new Embed({
//you can customize the Embed here, all parameters are optional
color: "#0000FF", //blue
});To use your embeds in your bot you can pass the object to your bot's object
const embed = new Embed({});
const bot = new Bot("<token>", { prefix: "!", embed });
bot.registerAction("test", ({ args, createEmbed }) =>
createEmbed({ desc: args })
);
//!test hello => embed containing hello as a descriptionBots will usually use a collection of images to represent emotions, you can use them easily with the embed.registerImage method
const embed = new Embed({});
const bot = new Bot("<token>", { prefix: "!", embed });
embed.registerImage("happy", "<url to image>");
bot.registerAction("test", ({ args, createEmbed }) =>
createEmbed({ desc: args, sideImage: "happy" })
);
//!test hello => embed containing hello as a description and the image "test"In case your bot requires it, you can set a custom format for the embed description and the footer
const embed = new Embed({
descTransform: (desc: string) => `${desc}, hello!`, //hello => hello, hello!
refTransform: (user: User) => [
`User: ${user.username}`,
user.avatarURL() ?? undefined,
],
});
const bot = new Bot("<token>", { prefix: "!", embed });
bot.registerAction("test", ({ msg, args, createEmbed }) =>
createEmbed({ desc: args, reference: msg.author })
);
//!test hello => embed containing "hello, hello!" as a description and the footer containing "User: <name>"Also your embeds can contain the avatar and name of the bot
const embed = new Embed({
author: bot.user,
});
const bot = new Bot("<token>", { prefix: "!", embed });
bot.registerAction("test", ({ args, createEmbed }) =>
createEmbed({ desc: args })
);Your users can be overwhelmed and confused by your bot's syntax. To aid them in the process, djs-diy offers a way to immediately point out which options they might have meant to type instead.
const bot = new Bot("<token>", { prefix: "!", embed });
bot.on("test", "hi there!");
bot.onTypo(
({ author }, [first, ...rest]) =>
`Hey there, ${
author.username
}! Did you mean to type !${first}? Other options: ${rest.join(", ")}`
);Bot#onTypo can set a callback for an scenario where an user types "tsst" or something similar as any other trigger.
Should be noted that onTypo is available router-wise and will always attempt to fetch a callback from any parent router (incluiding the Bot object's)
onTypo can take a second argument in the form of an object
{
maxDistance: number;
maxSuggestions: number;
}maxDistance: Maximum Levenshtein distance allowed
maxSuggestions: Max amount of suggestions to be provided to the callback
Sometimes you may want a command to contain a subcommand. This is where routers come in. To use them, create a new Router object then assign commands to it. Finally assign it as an action in your main Bot object. Don't worry about the constructor parameters, they'll be filled in for you.
const bot = new Bot("<token>", { prefix: "!" });
const helpRouter = new Router();
helpRouter.on("info", "lorem ipsum");
bot.on("help", helpRouter);
//Bot will now respond to `!help info` with "lorem ipsum"Routers have their own error handling too.
helpRouter.onError("Oh no!");
//if any of the commands under help router fail, "Oh no!" will be sent insteadRouters also have full support for slash commands.
You can easily do a 2 part command, expecting a reply from the same user
For example, a simple conversation could go like
(user) => "!wakeMeUp";
(bot) => "When should I wake you up?";
(user) => "When September ends";
(bot) => "Ok, I'll wake you up When September ends";This can be easily achieved with the expectReply method the action toolkit provides
bot.registerAction("wakeMeUp", async ({ expectReply }) => {
//built in Promise handling!
const reply = await expectReply("When should I wake you up?", true); //You can choose if the bot should delete this message or not by setting the second parameter
return `Ok, I'll wake you up ${reply?.content}`;
});The expectReply() promise will resolve to undefined if there's a timeout
Discordjs-DIY comes with integrated slash command generation. When you execute the .registerAction method, the library automatically generates a slash command JSON to be sent to the API.
The default is a command with no parameters and a description of "A command"
To specify parameters manually, add a parameters property or a third parameter to .registerAction.
Example parameters array:
[
{
name: "key",
type: "STRING" /*optional, defaults to STRING*/,
description: "Hello" /*optional, defaults to "A command"*/,
},
{ name: "value" },
];The Bot object now provides a Bot#commands.register method. On its own, it'll register the commands globally (Read about the implications here). You can pass an array of strings if you only want to register the commands on certain guild IDs (for development, etc).
bot.on("debug_register", async ({ guild }) =>
(await bot.commands.register(guild.id)) ? "Done" : "Something went wrong"
); //registers all the available slash commands , in the guild this message was sent inBe sure to call this method AFTER all of your commands and routers have been registered. For troubleshooting, Bot#compileCommands will return the objects that will then be passed to the Discord API (via Bot#commands.overwriteCommands)
DJS-diy will automatically create subcommands in the case of routers. Be mindful of the nesting limitations.
Routers (and the Bot object) support having a loading action be executed for any slash command interactions that may take longer than 2.5s to execute (Discord's official timeout is 3s).
This action is a sendable message, meaning it can be a string or a function which receives ActionParameters. Do note that for performance reasons, this action can't be async
bot.onLoading(({ author }) => `I'm working on it, ${author.username}!`);
router1.onLoading("Hold tight!!!");DJS-diy offers a small addon for running jobs after a response has been issued.
You can enqueue these jobs by calling asyncEffect from the ActionParameters received by an action.
Should be noted that msg in the passed ActionParameters object will contain the newly created response.
bot.registerAction("image", async ({ createEmbed, asyncEffect }) => {
asyncEffect(async ({ msg }) => {
await msg.edit({
content: "Hello there",
});
});
return "This message will change";
});ActionParameters include a runAction callback which can be used to manually invoke an action through the DJS pipeline, thus fully supporting asyncEffect and message component subscriptions.
You need a ResponseAction and ActionParameters to call this. You may reuse the action's ActionParameters, be aware that reusing asyncEffect or subscribe's ActionParameters will have the side-effect of making it seem like the Bot is the one calling the action (ergo, making components unclickable to users)
//test action can include asyncEffect calls and components!
const testAction = () => "hello!";
bot.on("coolAction", (params) => {
const { subscribe, createEmbed } = params;
/* `runAction` is also available here! ^*/
const buttonRow = subscribe(
{
label: "Press me",
style: "PRIMARY",
},
({ runAction }) => {
runAction(testAction, params);
}
);
return createEmbed({
desc: "Press the button to be greeted",
components: [buttonRow],
});
});DJS-diy offers per-action and global approaches to error handling.
To handle any exception using the same action:
bot.setErrorAction({
reaction: "😭",
response({ args }) {
//args will contain the value of `e.message`
return `Error ocurred => ${args}`;
},
});To handle an exception for a specific action, pass in an object and include the onError ActionObject:
bot.registerAction("hello", {
response() {
throw new Error("No hellos for you today");
},
onError: {
reaction: "😭",
},
});Discordjs-diy provides support for custom middleware.
interface MyCustomMW{
myMW: {
currentTime: number
}
}
const bot = new Bot("<token>", { prefix: "!", embed });
function myMiddleware(params: ActionParameters): ActionParameters<MyCustomMW>{
return {
...params,
middleware: {
...params.middleware,
myMW: {
currentTime: Date.now()
}
}
}
}
bot.useMiddleware<MyCustomMW>(myMiddleware)
bot.registerAction(({middleware}) => //now you can use middleware.myMW in every action execution)Discordjs-diy provides a solution for sessions.