[Fizzadar]

RTM.SendMessage pushes onto the 20-slot outgoingMessages channel, which is only drained by handleEvents while the websocket is connected. If the connection is down, reconnecting, or ManageConnection was never started, the 21st call blocks forever.

TrySendMessage does the same send in a select with default, returning false instead of blocking when the queue is full. mautrix-slack uses it for typing notifications, which are fine to drop.

Checked with a throwaway program against an unstarted RTM: TrySendMessage returns false on the 21st message, and a following SendMessage is still blocked after 2s.

Tag @indent to continue the conversation here.

[Fizzadar]

@indent go fix

[indent]

Removed the doc comment from TrySendMessage in 71d8679. I also re-pinned mautrix/slack#128 to that commit.

View session

[indent]

Issues

No issues found.

Review agents

Select any unchecked box below to run or rerun that agent.

Passed (1)

• Indent Review Agent · No issues found in TrySendMessage.

Full results

Indent Review Agent

• Summary: No issues found in TrySendMessage.
• Last ran on commit: 71d86795
• Latest result

  {
    "summary": "No issues found in TrySendMessage.",
    "findings": []
  }