### Tags with Enums { #tags-with-enums }

If you have a big application, you might end up accumulating **several tags**, and you would want to make sure you always use the **same tag** for related *path operations*.

In these cases, it could make sense to store the tags in an `Enum`.

**FastAPI** supports that the same way as with plain strings:

{* ../../docs_src/path_operation_configuration/tutorial002b_py39.py hl[1,8:10,13,18] *}

        @Test
        @DisplayName("should compare versions correctly")
        void shouldCompareVersionsCorrectly() {
            // Based on the actual implementation, it only handles specific cases
            assertTrue(ModelVersionUtils.isVersionGreaterOrEqual("4.1.0", "4.1.0"));
            assertFalse(ModelVersionUtils.isVersionGreaterOrEqual("4.0.0", "4.1.0"));

        assertEscaping(xmlEscaper, "\uFFFD", ch);
      }
    }

    // Test _all_ allowed characters (including surrogate values).
    for (char ch = 0x20; ch <= 0xFFFD; ch++) {
      // There are a small number of cases to consider, so just do it manually.
      if (ch == '&') {
        assertEscaping(xmlEscaper, "&amp;", ch);
      } else if (ch == '<') {
        assertEscaping(xmlEscaper, "&lt;", ch);
      } else if (ch == '>') {

one:
	MOVL $0, R15
	JMP two

// Ensure 3-arg instructions get GOT-rewritten without errors.
// See issue 58735.
TEXT ·a13(SB), 0, $0-0
	MULXQ runtime·writeBarrier(SB), AX, CX
	RET

// Various special cases in the use-R15-after-global-access-when-dynlinking check.
// See issue 58632.
TEXT ·a14(SB), 0, $0-0
	CMPL runtime·writeBarrier(SB), $0

interfaces into a single place. The interfaces themselves are located in
sub-modules, as described below.

Note that the file `__init__.py` in the TensorFlow source code tree is actually
only a placeholder to enable test cases to run. The TensorFlow build replaces
this file with a file generated from [`api_template.__init__.py`](https://www.github.com/tensorflow/tensorflow/blob/master/tensorflow/api_template.__init__.py)
"""

    SettableFuture<String> f = SettableFuture.create();
    MockCallback callback = new MockCallback("foo");
    Futures.addCallback(f, callback, ex);
    f.set("foo");
    assertEquals(1, ex.runCount);
  }

  // Error cases
  public void testSameThreadExecutionException() {
    SettableFuture<String> f = SettableFuture.create();
    Exception e = new IllegalArgumentException("foo not found");
    MockCallback callback = new MockCallback(e);

  private static boolean isProxyOfSameInterfaces(Object arg, Class<?> proxyClass) {
    return proxyClass.isInstance(arg)
        // Equal proxy instances should mostly be instance of proxyClass
        // Under some edge cases (such as the proxy of JDK types serialized and then deserialized)
        // the proxy type may not be the same.
        // We first check isProxyClass() so that the common case of comparing with non-proxy objects

## Dataclasses in Nested Data Structures { #dataclasses-in-nested-data-structures }

You can also combine `dataclasses` with other type annotations to make nested data structures.

In some cases, you might still have to use Pydantic's version of `dataclasses`. For example, if you have errors with the automatically generated API documentation.

`app.add_middleware()` receives a middleware class as the first argument and any additional arguments to be passed to the middleware.

## Integrated middlewares { #integrated-middlewares }

**FastAPI** includes several middlewares for common use cases, we'll see next how to use them.

/// note | Technical Details

For the next examples, you could also use `from starlette.middleware.something import SomethingMiddleware`.

    // may change and break assumptions in this test [*]. This is not a bug in the implementation of
    // Ascii.equalsIgnoreCase(), but it is a signal that its documentation may need updating as
    // regards edge cases.

    // The Unicode point {@code 00df} is the lowercase form of sharp-S (ß), whose uppercase is "SS".
    assertEquals("PASSWORD", "pa\u00dfword".toUpperCase()); // [*]