*   [Google Objective-C Style Guide](https://google.github.io/styleguide/objcguide.html)

#### Running sanity check

If you have Docker installed on your system, you can perform a sanity check on
your changes by running the command:

```bash
tensorflow/tools/ci_build/ci_build.sh CPU tensorflow/tools/ci_build/ci_sanity.sh
```

    NodeVisitState state = visitedNodes.get(node);
    if (state == NodeVisitState.COMPLETE) {
      return false;
    }
    if (state == NodeVisitState.PENDING) {
      return true;
    }

    visitedNodes.put(node, NodeVisitState.PENDING);
    for (N nextNode : graph.successors(node)) {
      if (canTraverseWithoutReusingEdge(graph, nextNode, previousNode)

   * implemented as "no-effort". No interrupts or other attempts are made to stop threads executing
   * tasks. Second, the returned list will always be empty, as any submitted task is considered to
   * have started execution. This applies also to tasks given to {@code invokeAll} or {@code
   * invokeAny} which are pending serial execution, even the subset of the tasks that have not yet

// to filter out unsuitable nodes. If MaximumVolumeSize is unset, it falls back
// to a comparison against the less precise Capacity. If that is also unset,
// the scheduler assumes that capacity is insufficient and tries some other
// node.
message CSIStorageCapacity {
  // Standard object's metadata. The name has no particular meaning. It must be

And the **container** itself (in contrast to the **container image**) is the actual running instance of the image, comparable to a **process**. In fact, a container is running only when it has a **process running** (and normally it's only a single process). The container stops when there's no process running in it.

## Container Images

  ~DeviceThread();

  // Requests that the worker thread execute the specified operation. Blocks
  // until the previously pending operation (a StartExecute without a Join) has
  // finished, if any.
  //
  // `cancellation_manager` must live until after `Join` finishes and pending
  // `is_async` operations finish. In addition to allowing the caller to cancel
  // the operation, its `StartCancel` method will be called if op execution

		select {
		case <-m.ctx.Done():
			if debugPrint {
				fmt.Println("Client sending disconnect to mux", m.MuxID)
			}
			m.addErrorNonBlockingClose(internalResp, context.Cause(m.ctx))
			errState = true
			continue
		case req, ok := <-requests:
			if !ok {
				// Done send EOF
				if debugPrint {
					fmt.Println("Client done, sending EOF to mux", m.MuxID)
				}
				msg := message{
					Op:    OpMuxClientMsg,

// to filter out unsuitable nodes. If MaximumVolumeSize is unset, it falls back
// to a comparison against the less precise Capacity. If that is also unset,
// the scheduler assumes that capacity is insufficient and tries some other
// node.
message CSIStorageCapacity {
  // Standard object's metadata.

            "CombinedFuture@\\w+\\[status=PENDING,"
                + " info=\\[futures=\\[SettableFuture@\\w+\\[status=PENDING],"
                + " SettableFuture@\\w+\\[status=PENDING]]]]");
    Integer integerPartial = 1;
    futureInteger.set(integerPartial);
    assertThat(futureResult.toString())
        .matches(
            "CombinedFuture@\\w+\\[status=PENDING,"

 * href="http://docs.oracle.com/javase/tutorial/essential/exceptions/tryResourceClose.html"
 * >try-with-resources</a> statement in JDK6-compatible code. Running on Java 7, code using this
 * should be approximately equivalent in behavior to the same code written with try-with-resources.
 * Running on Java 6, exceptions that cannot be thrown must be logged rather than being added to the
 * thrown exception as a suppressed exception.
 *