FluentIterable<String> filtered =
        FluentIterable.from(asList("foo", "bar")).filter(equalTo("foo"));

    List<String> expected = singletonList("foo");
    List<String> actual = Lists.newArrayList(filtered);
    assertEquals(expected, actual);
    assertCanIterateAgain(filtered);
    assertThat(filtered.toString()).isEqualTo("[foo]");
  }

  private static class TypeA {}

  private interface TypeB {}

* `refactor`: Refactors
    * This is normally for changes to the internal code that don't change the behavior. Normally it improves maintainability, or enables future features, etc.

///

/// info

Of course, the actual `Content-Type` header, status code, etc, will come from the `Response` object you returned.

///

### Document in OpenAPI and override `Response` { #document-in-openapi-and-override-response }

   * table is internally partitioned to try to permit the indicated number of concurrent updates
   * without contention. Because assignment of entries to these partitions is not necessarily
   * uniform, the actual concurrency observed may vary. Ideally, you should choose a value to
   * accommodate as many threads as will ever concurrently modify the table. Using a significantly

    private static final Map<Class<?>, InterpolateObjectAction.CacheItem> CACHED_ENTRIES =
            new ConcurrentHashMap<>(80, 0.75f, 2);
    // Empirical data from 3.x, actual =40

    private interface InnerInterpolator {
        String interpolate(String value);
    }

    @Override
    public Model interpolateModel(

  transient int modCount;

  /**
   * The hashtable. Its values are indexes to the keys, values, and entries arrays.
   *
   * <p>Currently, the UNSET value means "null pointer", and any non negative value x is the actual
   * index.
   *
   * <p>Its size must be a power of two.
   */
  private transient int[] table;

  /**
   * Contains the logical entries, in the range of [0, size()). The high 32 bits of each long is the

 * <p>The crawling process involves the following steps:
 * <ol>
 *   <li>Initialization: Sets up the crawler context and initializes the URL filter.</li>
 *   <li>Thread Creation: Creates a group of crawler threads to perform the actual crawling.</li>
 *   <li>Execution: Starts the crawler threads and waits for them to complete.</li>
 *   <li>Cleanup: Deletes the crawled data and clears the URL filter.</li>
 * </ol>
 *

    }

    // Test ProcessDestroyer inner class
    @Test
    public void test_processDestroyer_creation() throws Exception {
        // We can't directly test ProcessDestroyer without creating actual processes,
        // but we can test that the class exists and basic functionality
        try {
            // Use reflection to verify ProcessDestroyer class exists

In most of the cases, these blocks of code are actual complete applications that can be run as is.

In fact, those blocks of code are not written inside the Markdown, they are Python files in the `./docs_src/` directory.

And those Python files are included/injected in the documentation when generating the site.

### Docs for tests

Most of the tests actually run against the example source files in the documentation.

// but it's easy enough for us to keep it working. https://github.com/square/okhttp/issues/8826
tasks.withType<KotlinCompile> {
  compilerOptions {
    freeCompilerArgs.addAll("-module-name=okhttp", "-Xexpect-actual-classes")
  }