0.5.2

• Fixes issues where intellisense is unavailable when used alongside discord.js

---

• fix(TypeScript): bundle types at prepublishOnly (#149)

  • fix(typescript): bundle types at prepublishOnly

  This is required to support proper module augmentation with
  declare module '@discordjs/voice' { } because TypeScript doesn't
  propegate module augmentations when only
  targeting the root index.d.ts, even when that
  file is re-exporting other types.

  It's an issue we faced with Sapphire Framework & Pieces and easily fixed
  by bundling up all the types using Rollup.

  • chore: switch to cli package

  • chore: add rtb config file

  • chore: remove config and update cli tool

• chore(Release): v0.5.2

0.5.1

Changes

• Better type inference when using event listeners for specific transitions, e.g.

  voiceConnection.on(VoiceConnectionStatus.Ready, (oldState, newState) => {
    // newState is now inferred to be the Ready state
  });

• Fixes to the rejoin and disconnect methods of VoiceConnection

---

Commit Log

• docs(Examples): add a JavaScript example (#127)

  • docs(Examples): create radio bot example

  • chore(Examples): create overview README

  • chore(Examples): linting and consistency

• feat: strengthen new state type for status events (#132)

  • fix(VoiceConnection): strengthen new state type for status events

  • fix(AudioPlayerEvents): strengthen new state type for status events

  • fix(Events): type error when emitting status

• fix(VoiceConnection): stuck on signalling state when rejoining the same channel on ready state (#139)

  • fix(VoiceConnection): stuck on signalling state

  when rejoining the same channel on ready state

  • test(VoiceConnection): rejoins in a ready state

  • fix: remove checking channelId

• fix(VoiceConnection): disconnect method now actually disconnects (#140)

  • test(VoiceConnection): expose broken disconnect

  • fix(VoiceConnection): disconnect method

  • refactor(VoiceConnection): remove Object.assign

• chore(Release): v0.5.1

0.5.0

VoiceConnection changes

• new disconnect() method if you want to disconnect the connection without destroying it
• reconnect() -> rejoin(joinConfig?)
  • This and the below change were to improve clarity. What this method does is actually try to re-join a voice channel (rather than try to fix a broken voice connection)
• reconnectAttempts -> rejoinAttempts
• calling joinVoiceChannel when an existing connection exists that is disconnected will now call rejoin() on that connection

Other

• Nicer errors are thrown when you don't have a valid encryption package installed
• Some improvements to the music bot example

---

Commit log

• fix(Examples): youtube-dl-exec args

  Passing q: true worked on Ubuntu, but on some other systems, this was parsed as -q true rather than -q.

• refactor(Examples): remove unnecessary guard in music bot example

  Resolves #125

• feat!(VoiceConnection): add disconnect and rename reconnect to rejoin (#129)

  • fix(VoiceConnection): reconnect now sets proper disconnect reason

  • feat(VoiceConnection): disconnect method

  • refactor(VoiceConnection): reconnect -> rejoin

  • feat(VoiceConnection): allow updating joinConfig

  • refactor(VoiceConnection): allow rejoin in non-destroyed states

  • refactor(VoiceConnection): reconnectAttempts -> rejoinAttempts

  • test(VoiceConnection): new methods

  • refactor(VoiceConnection): remove negated if statement

• feat(Secretbox): throw fallback error when no encryption package installed (#130)

  • feat(Secretbox): display fallback error when no valid installation

  • test(Secretbox): no errors thrown with package installed

  • test(Secretbox): fix tweetnacl test

• refactor(Secretbox): type methods object (#131) - Antonio Román [email protected]

• chore(Release): v0.5.0

0.4.0

New features

Opus audio probing utility

The new probing utility lets you figure out whether it's possible to demux your ReadableStream with an Ogg/WebM demuxer, rather than passing it to FFmpeg straight away. This enables you to make some performance improvements:

import { createReadStream } from 'fs';
import { StreamType, demuxProbe, createAudioResource } from '@discordjs/voice';

async function start() {
  // Example verifications
  assert.equal((await demuxProbe(createReadStream('file.mp3'))).type, StreamType.Arbitrary);
  assert.equal((await demuxProbe(createReadStream('file.opus'))).type, StreamType.OggOpus);
  assert.equal((await demuxProbe(createReadStream('file.webm'))).type, StreamType.WebmOpus);

  // Example of creating an audio resource using the probe
  const input = createReadStream('./song.unknownextension');
  const { stream, type } = await demuxProbe(input);
  createAudioResource(stream, { inputType: type });
}

Silence padding for audio resources

By default, all audio resources now end with 5 packets of silence to avoid interpolation noises. See #120 on examples of configurability.

Other

• You can now use event handlers that return void | Promise<void>, since the latter is handled exactly the same as a void return in Node.js v16
• The music bot example has been refactored to use youtube-dl instead of ytdl-core/ytdl-core-discord for streaming videos due to its reliability.

---

Commits

• feat(Util): add Opus audio probing utility (#118)

  • chore(Deps): update prism-media

  • feat(Util): opus demux probe tool

  • feat: update music bot example

  • refactor(Util): remove unnecessary log

  • chore: update typescript dep in example

  • chore(Util): add docs to demuxProbe

  • chore: update discord.js dep and set download limit cap in example

  • refactor: use youtube-dl-exec instead of child_process

  • test(Util): initial demuxProbe test

  • test(Util): remaining demuxProbe tests

  • refactor: combine event listener bindings in docs

  • refactor: use promises properly in example

  • docs(Util): write docs for ProbeInfo

  • refactor: use probe.stream in music bot example

• feat!: add silence padding frames to AudioResources (#120)

  • feat(AudioResource): add ended getter

  • feat(AudioResource): silence padding frames

  • test(AudioResource): silence padding frames

  • feat(AudioPlayer): stop() respects silent padding by default

  • test(AudioPlayer): stop() will play silence frames

  • feat(AudioResource): use silenceRemaining in ended

  • test(AudioResource): read() after silence ends

• feat: allow awaitable void event listeners, fix VoiceConnection disconnected states (#121)

  • feat: use Awaited for event emitters

  Credit to Kyra for the Awaited implementation in discord.js

  Co-authored-by: Antonio Román [email protected]

  • fix(Example): improve reconnection logic

  Co-authored-by: Antonio Román [email protected]

• chore(Examples): re-enable eslint rule

• chore(Release): v0.4.0

0.3.1

• chore(Legal): added UNLICENSE in examples (#111)

• chore: update package description

• chore(README): improve wording

• fix(Util): increase search depth for dependencies

• fix(AudioResource): improve accuracy of function signatures (#113)

  • fix(AudioResource): invalid typings

  • refactor(AudioResource): improve typing

  • fix(AudioResource): improve typings

  • fix(AudioResource): function signature types

  • fix(AudioResource): function signature type

• chore(Deps): update discord.js dep in music bot example (#114)

• chore(Release): v0.3.1

0.3.0

• chore(README): use new blurple for discord.js server badge (#108)

• chore(Deps): update dependencies (#109)

  • chore(Deps): update dependencies

  • chore: disable eslint(no-redeclare)

  • chore(Deps): use tiny-typed-emitter instead of typed-emitter

  • test(VoiceUDPSocket): fix failing tests after updating jest

  • chore(Deps): move tiny-type-emitter

  • chore(Deps): update lockfile

  • chore(Deps): make discord-api-types regular dep

  • chore(Deps): add @types/ws as normal dep

• docs: music bot example (#107)

  • docs: create music bot example

  • chore: add README to the example

  • chore: fix typo in auth example file

  • refactor: remove TrackMetadata, use Track

  • docs: add explanations to music bot example

  • fix: queueLock not being released

  • docs: document music bot discord.js code

  • refactor: use commands.set in example

  • chore: create eslint file for example

  • fix: defer interaction on play command

  • docs: respond with errors to invalid command usage

  • docs: fix linter error

  • fix: use followUp for deferred interactions

  • chore: update discord-api-types dep in example

  • feat: apply suggestions from code review

  Co-authored-by: Antonio Román [email protected]

  • feat: fallback for unknown commands

  Co-authored-by: Antonio Román [email protected]

• feat(AudioResource): add encoder property, fix type for metadata (#110)

  • fix(AudioResource): inaccuracy in types

  • refactor(AudioResource): include encoder stream

  • refactor(AudioResource): simplify constructor

  • test(AudioResource): check values of encoder property

  • refactor(AudioResource): rename pipeline to edges

  • refactor(AudioResource): mark edges array as readonly

  • refactor(AudioResource): use null instead of undefined

• chore(README): add features list, useful links

• chore(Release): v0.3.0

0.2.1

• fix(AudioPlayer): enter Buffering state for resources that are not yet readable (#98)

  • fix(AudioPlayer): enter Buffering state

  • test(AudioPlayer): fix some existing tests

  • test(AudioPlayer): test Buffering state

  • test(AudioPlayer): remove unnecessary assertion

• fix(VoiceConnection): handle nullable endpoint in voice server update payload (#100)

  • fix(VoiceConnection): handle null endpoint

  • test(VoiceConnection): null endpoint handling

  • chore(Deps): bump discord-api-types to 0.15.1

• chore: update package author email address

• refactor(Adapter): sendPayload now returns boolean (#105)

  • refactor(Adapter): sendPayload now returns boolean

  • test(VoiceConnection): behaviours with failed sendPayload

  • chore: fix linter errors

  • chore(Deps): downgrade TypeScript for TypeDoc support

  • chore: update adapter example

• feat: reason in VoiceConnectionDisconnectedState

• ci: add Node.js v16 runs to workflows

• fix(AudioPlayer): strengthen error event type

• fix(VoiceUDPSocket): catch socket close failure

• chore(Release): v0.2.1

0.2.0

• feat!: keep alive mechanism to detect connectivity to Discord (#89)

  • refactor: simplify IP and port udp socket code

  • test: make sure to destroy udp sockets after test runs

  • feat: udp keep alive

  • chore: remove debug statement

  • feat(VoiceConnection): expose udp socket keep alive ping

  • test(VoiceUDPSocket): keep alive mechanism

  • fix: emit close instead of error

  • feat(VoiceUDPSocket): emit debug message

• feat: implement voice receive (#87)

  • feat: allow specifying selfDeaf/selfMute when joining a voice channel

  • feat: emit messages from UDP

  • feat: create VoiceReceiver class

  • feat: export receiver

  • feat: ssrc discovery

  • feat: packet decryption

  • feat: use fix from sudo-carson

  See discordjs/discord.js@c84a157

  • style: remove unnecessary object spread

  • feat: ssrc map logic

  • feat: add audio receive stream

  • refactor(VoiceReceiver): rename some things

  • fix(VoiceReceiver): listen to stream close instead of end

  • feat(VoiceReceiver): destroy stream on error

  • docs(VoiceReceiver): add documentation

  • docs(AudioReceiveStream): add documentation

  • docs: add documentation for SSRCMap

  • fix(VoiceReceiver): use new ssrc map methods

  • refactor: used typed emitter for SSRCMap

  • test(VoiceUDPSocket): fix tests

  • test: add tests for SSRCMap

  • fix(VoiceReceiver): ignore short messages

  • test(VoiceReceiver): onUdpMessage

  • test(VoiceReceiver): susbcribe

  • test(VoiceReceiver): onWsPacket

  • test(VoiceReceiver): tracking state changes

  • test(VoiceReceiver): immediate binding

  • chore: fix coverage rule

  • test(VoiceReceiver): use RTP packets from different clients

  • feat(VoiceReceiver): play silence frame if nothing has been played yet

  • refactor(VoiceReceiver): reduce duplicated code

• feat(VoiceUDPSocket): use new IP discovery packet format (#90)

• feat: use typed event emitters internally (#91)

  • refactor(VoiceUDPSocket): use typed event emitter

  • refactor(VoiceWebSocket): use typed event emitter

  • refactor(Networking): use typed event emitter

• chore(Deps): update dependencies

• chore(Release): v0.2.0

0.1.3

• test(VoiceUDPSocket): unit tests (#82)

  • test(VoiceUDPSocket): unit tests

  • fix(VoiceUDPSocket): fix issues found in unit tests

  • refactor(VoiceUDPSocket): code review

  • refactor(VoiceUDPSocket): use .off instead of .removeListener

• feat: add ping stats (#77)

  • feat(VoiceWebSocket): ping recording

  • feat(VoiceConnection): expose ping property

  • chore(Deps): install mock-socket and jest-websocket-mock

  • refactor(VoiceWebSocket): use browser WS API

  • chore: set up ws mock

  • test(VoiceWebSocket): packet parsing

  • test(VoiceWebSocket): event propagation

  • fix(VoiceWebSocket): heartbeat ack logic

  • test(VoiceWebSocket): heartbeat flow

  • test(VoiceWebSocket): ping

• docs(README): fix branch in contributing link (#88)

• feat: createAudioResource now infers stream type (#85)

  • feat: createAudioResource now infers stream type

  • feat: infer from OggOpusDemuxer streams

  • test: first createAudioResource test

  • fix: type in AudioResource test

  • feat(AudioResource): infer stream type for WebmDemuxer

  • test(AudioResource): inferring stream types

  • test(AudioResource): add remaining inference test

• chore: publish documentation and use typed events (#84)

  • chore: install typedoc

  • chore: fix generateDependencyReport export

  • docs: clean up AudioPlayer docs

  • chore: export AudioPlayerError

  • docs: clean up AudioResource docs

  • docs: clean up adapter docs

  • docs: clean up entersState docs

  • docs: clean up joinVoiceChannel docs

  • docs: clean up VoiceConnection docs

  • docs: add typed events for VoiceConnection

  • docs: typed events for AudioPlayer

  • chore(GitHub): add docs CI workflow

  • chore(GitHub): only run docs workflow on push

  • chore: fix docs formatting

• chore(Release): v0.1.3

0.1.2

• feat: stream time statistics (#75)

  • feat(AudioResource): add playbackDuration

  • feat(AudioPlayer): add playbackDuration

  • test(AudioPlayer): playbackDuration

• refactor: split states into a union of interfaces (#76)

  • refactor(AudioPlayer): split states into separate interfaces

  • refactor(AudioPlayer): pass state to preparePacket

  • refactor(VoiceConnection): split states into separate interfaces

  • refactor(Networking): split states into separate interfaces

• test(VoiceConnection): add unit tests (#80)

  • test(VoiceConnection): createVoiceConnection

  • test(VoiceConnection): addServerPacket

  • test(VoiceConnection): addStatePacket

  • refactor: reduce code duplication in tests

  • test(VoiceConnection): configureNetworking

  • test(VoiceConnection): onNetworkingClose

  • fix(VoiceConnection): inc reconnectAttempts when done automatically

  • test(VoiceConnection): onNetworkingStateChange

  • test(VoiceConnection): destroy

  • test(VoiceConnection): reconnect

  • test(VoiceConnection): subscribe

  • test(VoiceConnection): onSubscriptionRemoved

  • refactor: dummy instead of fake

  • test(VoiceConnection): strengthen onSubscriptionRemoved tests

  • refactor(VoiceConnection): make adapter bindings testable

  • test(VoiceConnection): adapter bindings

• feat: allow adapter creator to call a destroy method (#79)

  • feat: add required destroy methods to both implementers

  • feat(VoiceConnection): implement new adapter destroy method

  • feat: update the adapter example

  • refactor: remove non-null assertion from adapter example

  • fix: logical bug in adapter

  • test(VoiceConnection): adapter destroy

• chore(Release): v0.1.2