*   XLA reduction emitter is deterministic when the environment variable
        `TF_DETERMINISTIC_OPS` is set to "true" or "1". This extends
        deterministic `tf.nn.bias_add` back-prop functionality (and therefore
        also deterministic back-prop of bias-addition in Keras layers) to
        include when XLA JIT compilation is enabled.
    *   Fix problem, when running on a CUDA GPU and when either environment

        }

        return when (this) {
            is FirResolvable, is FirVariableAssignment -> {
                when (val calleeReference = toReference(analysisSession.useSiteSession)) {
                    is FirResolvedErrorReference -> transformErrorReference(this, calleeReference)
                    is FirResolvedNamedReference -> when (calleeReference.resolvedSymbol) {

        logger.fine(format(">> CONNECTION ${CONNECTION_PREFACE.hex()}"))
      }
      sink.write(CONNECTION_PREFACE)
      sink.flush()
    }
  }

  /** Applies `peerSettings` and then sends a settings ACK. */
  @Throws(IOException::class)
  fun applyAndAckSettings(peerSettings: Settings) {
    this.withLock {
      if (closed) throw IOException("closed")

     of this section 3.

3.2 When the Program is Distributed as Source Code:

  a) it must be made available under this Agreement, or if the
  Program (i) is combined with other material in a separate file or
  files made available under a Secondary License, and (ii) the initial
  Contributor attached to the Source Code the notice described in
  Exhibit A of this Agreement, then the Program may be made available

### Build session state

A "build session" represents a single invocation of Gradle, for example when you run `gradlew build`.
A session runs the build one or more times.
For example, when continuous build is enabled, the session may run the build many times, but when it is disabled, the session will run the build once only.

The build session state is managed by the `BuildSessionState` class.

                // getSymbol(ClassId) returns a symbol from the original file, so here we avoid using it
                return getContainingDeclarationByPsi(symbol)
            }
        }

        when (symbol) {
            is KtLocalVariableSymbol,
            is KtAnonymousFunctionSymbol,
            is KtAnonymousObjectSymbol,
            is KtDestructuringDeclarationSymbol -> {

            multipleReadyTasks = true
            break@eachQueue
          }

          // We have a task to execute when we complete the loop.
          else -> {
            readyTask = candidate
          }
        }
      }

      // Implement the decision.
      when {
        // We have a task ready to go. Get ready.
        readyTask != null -> {
          beforeRun(readyTask)

The process that happens when your API app calls the *external API* is named a "callback". Because the software that the external developer wrote sends a request to your API and then your API *calls back*, sending a request to an *external API* (that was probably created by the same developer).

* `fastapi-cli` - to provide the `fastapi` command.

When you install `fastapi` it comes these standard dependencies.

## `fastapi-slim`

If you don't want the extra standard optional dependencies, install `fastapi-slim` instead.

When you install with:

```bash
pip install fastapi
```

...it includes the same code and dependencies as:

      },
      logger = logger,
    )

  /** Runs all tasks that are ready. Used by the test thread only. */
  fun runTasks() {
    advanceUntil(nanoTime)
  }

  /** Advance the simulated clock, then runs tasks that are ready. Used by the test thread only. */
  fun advanceUntil(newTime: Long) {
    taskRunner.assertThreadDoesntHoldLock()

    taskRunner.lock.withLock {
      check(currentTask == TestThreadSerialTask)