README.md

github.com/lestrrat-go/slack/rtm

Slack RTM Client

Synopsis

Simple RTM Client:

package rtm_test

import (
  "context"
  "fmt"

  "github.com/lestrrat-go/slack/rtm"
  "github.com/lestrrat-go/slack"
)

func processMessageEvent(e *rtm.MessageEvent) {
  // Dummy
}

func ExampleClient() {
  ctx, cancel := context.WithCancel(context.Background())
  defer cancel()

  // Create a new Slack REST client. This is required,
  // as the RTM client needs to hit the rtm.start API
  // endpoint in order to initiate a websocket session
  restcl := slack.New(slackToken)

  // Create a new Slack RTM client. Use `Run()` to start
  // listening to incoming events. If you need to stop the
  // RTM client, use the cancel function for the context.Context
  // object that you passed
  rtmcl := rtm.New(restcl)
  go rtmcl.Run(ctx)

  // Now, consume your events. They come from a channel.
  // If you care about properly bailing out of a processing
  // loop like below, you should be using select {} on the
  // channel returned by `rtmcl.Events()`
  for e := range rtmcl.Events() {
    // Event types are listed in event_types.go file, which
    // is auto-generated from Slack's web page describing
    // the events.
    switch typ := e.Type(); typ {
    case rtm.MessageType:
      // It's a message. Event objects carry a unique payload
      // depending on the event type, so you will have to do
      // a type assertion.
      // TODO: come up with a mapping of Type -> Data's underlying
      // type, so it's easier for the users to see
      processMessageEvent(e.Data().(*rtm.MessageEvent))
    default:
      fmt.Printf("Unhandled event: %s", typ)
    }
  }
}