</mxCell>
                <mxCell id="32" value="" style="pointerEvents=1;shadow=0;dashed=0;html=1;fillColor=#505050;labelPosition=center;verticalLabelPosition=bottom;verticalAlign=top;outlineConnect=0;align=center;shape=mxgraph.office.devices.laptop;strokeColor=none;" parent="33" vertex="1">
                    <mxGeometry width="500" height="350" as="geometry"/>
                </mxCell>

This is very useful when you need to:

* Have shared logic (the same code logic again and again).
* Share database connections.
* Enforce security, authentication, role requirements, etc.
* And many other things...

All these, while minimizing code repetition.

## First Steps

    ],
)

cc_library(
    name = "tf_shape",
    srcs = ["tf_shape.cc"],
    hdrs = ["tf_shape.h"],
    copts = tf_copts(),
    visibility = ["//visibility:public"],
    deps = [
        ":c_api_macros",
        ":tf_shape_internal",
        "//tensorflow/core:framework",
    ],
)

cc_library(
    name = "tf_shape_internal",
    hdrs = ["tf_shape_internal.h"],
    copts = tf_copts(),

 *
 * The task runner is also responsible for releasing held threads when the library is unloaded.
 * This is for the benefit of container environments that implement code unloading.
 *
 * Most applications should share a process-wide [TaskRunner] and use queues for per-client work.
 */
class TaskRunner(
  val backend: Backend,
  internal val logger: Logger = TaskRunner.logger,
) {
  val lock: ReentrantLock = ReentrantLock()

                </mxCell>
                <mxCell id="32" value="" style="pointerEvents=1;shadow=0;dashed=0;html=1;fillColor=#505050;labelPosition=center;verticalLabelPosition=bottom;verticalAlign=top;outlineConnect=0;align=center;shape=mxgraph.office.devices.laptop;strokeColor=none;" parent="33" vertex="1">
                    <mxGeometry width="500" height="350" as="geometry"/>
                </mxCell>

        }
      }
    }
  }

  /**
   * Object class to adapter type. This approach limits us to one adapter per Kotlin class, which
   * might be too few for values like UTF_STRING and OBJECT_IDENTIFIER that share a Kotlin class but
   * have very different ASN.1 interpretations.
   */
  private val defaultAnyChoices =
    listOf(
      Boolean::class to BOOLEAN,
      BigInteger::class to INTEGER_AS_BIG_INTEGER,

    assertThat(request.tag(String::class.java)).isSameAs(stringTag)
    assertThat(request.tag(Long::class.javaObjectType)).isSameAs(longTag)
  }

  /** Confirm that we don't accidentally share the backing map between objects. */
  @Test
  fun tagsAreImmutable() {
    val builder =
      Request.Builder()
        .url("https://square.com")
    val requestA = builder.tag(String::class.java, "a").build()

        network stack is offline.

  [[3]] "Remote" repositories referenced using the file:// protocol may
        be available. However, if that file:// url references a
        file-share, as in the case of an NFS or SMB mount, that will
        be unavailable.

  So, offline mode has several implications, some of which may not be
  altogether obvious:

    !!! tip
        Prefer to use the `Annotated` version if possible.

    ```Python hl_lines="105  107-115"
    {!> ../../../docs_src/security/tutorial005.py!}
    ```

## Verify the `username` and data shape

We verify that we get a `username`, and extract the scopes.

    These Pydantic models define more or less a "schema" (a valid data shape).

    So this will help us avoiding confusion while using both.

### Create the Pydantic *models* / schemas

Create all the same Pydantic models as in the SQLAlchemy tutorial: