}

  /**
   * Waits for the {@link ServiceManager} to become {@linkplain #isHealthy() healthy}. The manager
   * will become healthy after all the component services have reached the {@linkplain State#RUNNING
   * running} state.
   *
   * @throws IllegalStateException if the service manager reaches a state from which it cannot
   *     become {@linkplain #isHealthy() healthy}.
   */
  public void awaitHealthy() {

  }

  /**
   * Waits for the {@link ServiceManager} to become {@linkplain #isHealthy() healthy}. The manager
   * will become healthy after all the component services have reached the {@linkplain State#RUNNING
   * running} state.
   *
   * @throws IllegalStateException if the service manager reaches a state from which it cannot
   *     become {@linkplain #isHealthy() healthy}.
   */
  public void awaitHealthy() {

## Status

ACCEPTED

## Consequences

* Classes are forced to have consistent nullability annotations.
* IDE warnings become more accurate, reducing warnings fatigue.
* Public API nullability annotations become closer to reality.
* Java compilation of Gradle code suffers from small performance penalty (up to 10%).
* Time has to be allocated to clean up existing projects.

    }

  /**
   * A callback to be invoked each time the dispatcher becomes idle (when the number of running
   * calls returns to zero).
   *
   * Note: The time at which a [call][Call] is considered idle is different depending on whether it
   * was run [asynchronously][Call.enqueue] or [synchronously][Call.execute]. Asynchronous calls
   * become idle after the [onResponse][Callback.onResponse] or [onFailure][Callback.onFailure]

        // TODO: in a moment Maven stop pushing user properties to system properties (and maybe something more)
        // and allow multiple instances per JVM, this may become a pool? derive key based in invokerRequest?
        MavenContext result = residentContext.computeIfAbsent(
                "resident",
                k -> new MavenContext(invokerRequest, false, (MavenOptions)

We use *Specification* and *Discovery* documents stored in Google Drive, but they present some downsides:

* They are rarely updated after creation and initial review, and then become hard to follow, especially after important decisions are made
* They are not synced with the code to reflect the eventual solution that is committed
* Google Docs is not a "code-oriented" tool, like Markdown can be

compliant, then the rights granted under this License from a particular
Contributor are reinstated (a) provisionally, unless and until such
Contributor explicitly and finally terminates Your grants, and (b) on an
ongoing basis, if such Contributor fails to notify You of the
non-compliance by some reasonable means prior to 60 days after You have

can more easily get a good answer, or even solve the problem yourself even before asking. And in GitHub I can make sure I always answer everything, even if it takes some time. I can't personally do that with the chat systems. 😅

Conversations in the chat systems are also not as easily searchable as in GitHub, so questions and answers might get lost in the conversation. And only the ones in GitHub count to become a [FastAPI Expert](fastapi-people.md#fastapi-experts){.internal-link target=_blank},...

The client does not wait for the remote handler to finish before returning.
Returning any error will also cancel the stream remotely.

CAREFUL: When utilizing two-way communication, it is important to ensure that the remote handler is not blocked on a send.
If the remote handler is blocked on a send, and the client is trying to send without the remote receiving,
the operation would become deadlocked if the channels are full.

### Typed handlers

          if (minDelayNanos < coordinatorWakeUpAt - now) {
            backend.coordinatorNotify(this@TaskRunner)
          }
          return null
        }

        // No other thread is coordinating. Become the coordinator!
        else -> {
          coordinatorWaiting = true
          coordinatorWakeUpAt = now + minDelayNanos
          try {
            backend.coordinatorWait(this@TaskRunner, minDelayNanos)