Discord bots. Simplified and boosted · Docs & Guide · Contribute

---

WARNING

THIS LIBRARY IS UNDER DEVELOPMENT! Parts of the stuff described here is not done yet, and we are just planning to implement it. You can find out more information about the development in our Discord.

Getting Started

See installation guide or API reference.

Features

• Smart Caching
• • Policies — cache policies for guilds, channels, users, etc. disable any cache!
• • Sharding Ready — operate with cache from any sharding instance to any sharding instance
• • Any Storage — use any storage for cache (memory, redis, etc.) via cache providers
• • Async Cache — operate with cache asynchronously, without blocking the event loop
• • DB-like Operations — operate with cache via database like operations (set, get, delete, count, etc.)
• • Editable in runtime — enable and disable caching for different entities in runtime without any breakage
• Smart Sharding
• • Process — spawn shards in separate processes
• • Worker — spawn shards in separate workers
• • Cluster — spawn shards in separate clusters
• • Gateway Shards — spawn multiple gateway shards in one sharding instance (process, worker, cluster) and save host resources
• • Fast and Lightweight IPC — fast and lightweight IPC between sharding instances via Unix/Windows sockets or TCP/UDP
• • ShardingManager — spawn, restart, destroy shards, execute code on shards, get statistics, etc.
• • Hello/Identify — hello/identify algorithm for shards
• • Load Balancing — reconfigure which of the sharing instances should be responsible for which shards with restarting only changed shards (will be implemented in v0.14.0, now it is done by restarting all shards)
• Smart Rest
• • Global Rate Limit — global rate limit for all requests synced between all sharding instances & block protection
• • Invalid Request Limit — invalid request limit for all requests synced between all sharding instances & block protection
• • Rest Provider — rest provider allows you to customize the rest client as you wish, for example, to use a proxy or load balancer
• • Load Statistics — capture how many requests were made to the Discord API every second or less
• • Errors Handler — handle errors from the Discord API in a convenient way, not with .catch everywhere (will be implemented in v0.13.0)
• Smart Gateway
• • Custom Events — create custom events and emit them from anywhere via gateway provider + typings!
• • Super Scale — scale your application with the power of events queues (RabbitMQ, Redis, etc.) via gateway provider
• • Zero-Downtime — zero-downtime restarts of the application via gateway provider
• Smart Application
• • Rest-Only — create a rest-only application without gateway, for example, http interactions!
• • Cache-Only — create a cache-only application without gateway and rest
• • Cache Independent — everything you need to do anything is ID, no cache needed
• • Performance — consumes about 40% less RAM on big bots than djs, without any optimizations. Up to 90% less RAM with optimizations.
• • Extendable Entities — extend entities with your own methods and properties, with breakage protection
• • Editable Entities — choose which properties of entities you want to cache, with breakage protection

Planned features

• Waifoo — a framework for creating discord bots based on Discordoo (commands, other features).
• Kawaioo — a library / nestjs microservice that will allow you to interact with the Discordoo ShardingManager over TCP and UDP, to create your own scalable Rest API for the bot.
• Voice support — at this moment, the library does not support voice.
• Microservices — injection of microservices into the DiscordApplication that can communicate with each other.

Release milestone

A complete description of everything that must be in Discordoo to be released

Release deadline: When it's ready.

Sharding

• Implement sharding
• Shards spawning
  • use processes for shards
  • use workers for shards
  • use clusters for shards (unfortunately, windows not supported)
• Shards communication
  • Shards must be able to communicate in fast and scalable way
  • Shards must use hello/identify algorithm
  • Shards must use heartbeat algorithm
  • Shards must be able to send, receive and handle CACHE_OPERATE messages
  • Shards must be able to send errors to ShardingManager, manager must handle these errors
• User-land APIs
  • User must be able to spawn shards
  • User must be able to restart specified shard(s)
  • User must be able to destroy shard(s)
  • User must be able to get shard(s) statistics
    • v8 statistics
    • custom statistics
    • events per gateway shard per second
    • common statistics (guilds in cache, users in cache, channels in cache, etc.)
  • User-friendly sharding APIs in sharding instances

Gateway

• Implement gateway
  • Implement gateway provider
• Connecting to gateway
• Reconnecting/resuming in common/emergency cases
• Receive and send etf encoded messages
• Receive and decode zlib-compressed messages
• Multi gateway shards support in one WebSocketManager instance
• Processing events/s smoothing/limiting
• Handling gateway rate limits
• Send events to WebSocketManager, and then to GatewayProvider
• Gateway must be able to restart only specified shard(s)

Rest

• Implement rest
  • Implement rest provider - rest request (like constructor) & rest provider (performs requests and handles rate-limits optionally)
• Requests
  • Sending requests to Discord
• Responses
  • Unified response form for library internals
• Rate-limits
  • Rest must smartly-handle rate-limits using remaining header
  • Rest must synchronize global-rate-limit between shards on one machine

Cache

• Implement cache
  • Implement cache provider
• Library must be able to store something
• Library must be able to delete something from cache
• Library must be able to check cache size
• Library must be able to sweep cache
• Library must be able to iterate cache
• Library must be able to operate with cache in a different shards from one shard
• Caching policies
  • Cache must handle GlobalCachingPolicy
  • Cache must handle MessagesCachingPolicy
  • Cache must handle GuildsCachingPolicy
  • Cache must handle GuildMembersCachingPolicy
  • Cache must handle ThreadMembersCachingPolicy
  • Cache must handle ChannelsCachingPolicy
  • Cache must handle EmojisCachingPolicy
  • Cache must handle RolesCachingPolicy
  • Cache must handle PresencesCachingPolicy
  • Cache must handle UsersCachingPolicy
  • Cache must handle OverwritesCachingPolicy
  • Cache must handle InvitesCachingPolicy
  • Cache must handle StickersCachingPolicy
  • Cache must handle CommandsCachingPolicy
  • All the policies listed above must be able to handle custom caching functions

SID - still in development

Entities (discord structures)

• Must be extendable
• Anti monkey-patch defence
• Implement Guilds (SID)
• Implement Messages
• Implement Channels (SID)
• Implement Members
• Implement Emojis
• Implement Stickers
• Implement Roles
• Implement Presences
• Implement Reactions
• Implement Users
• Implement Interactions
• Implement Overwrites (SID)
• Implement Invites (SID)

User-land APIs

• Collection
• Wrapper
• BitField wrappers
• Entities managers
  • EntitiesManager
  • EntitiesCacheManager
  • ApplicationGuildsManager (SID)
  • ChannelMessagesManager/ApplicationMessagesManager (SID)
  • GuildChannelsManager/ApplicationChannelsManager (SID)
  • GuildEmojisManager/ApplicationEmojisManager (SID)
  • GuildRolesManager/ApplicationRolesManager (SID)
  • GuildPresencesManager/ApplicationPresencesManager
  • GuildMembersManager/ApplicationGuildMembersManager (SID)
  • ThreadMembersManager/ApplicationThreadMembersManager (SID)
  • MessageReactionsManager/ApplicationReactionsManager
  • UsersManager (SID)
  • ApplicationInteractionsManager/ApplicationInteractionsApplicationCommandsManager/GuildApplicationCommandsManager
  • ApplicationInvitesManager/GuildInvitesManager (SID)

Contributing

Feel free to create a PR, but check if there is an existing one. See Contributing Guide.

JetBrains Open Source Support

Many thanks to the JetBrains team for OSS licenses for their IDE! JetBrains make the best IDEs, seriously, you should try!