* If you know Python types you know how to use Pydantic.
* Plays nicely with your **<abbr title="Integrated Development Environment: similar to a code editor">IDE</abbr>/<dfn title="A program that checks for code errors">linter</dfn>/brain**:
    * Because pydantic data structures are just instances of classes you define; auto-completion, linting, mypy and your intuition should all work properly with your validated data.
* Validate **complex structures**:

# Tests

tf_cuda_library(
    name = "c_test_util",
    testonly = 1,
    srcs = ["c_test_util.cc"],
    hdrs = ["c_test_util.h"],
    visibility = [
        "//learning/brain:__subpackages__",
        "//tensorflow:__subpackages__",
    ],
    deps = [
        ":c_api",
        ":c_api_experimental",
        "//tensorflow/core:lib",
        "//tensorflow/core:protos_all_cc",

* **不烧脑**：
    * 没有新的模式定义 micro-language 需要学习。
    * 如果你知道 Python types，你就知道如何使用 Pydantic。
* 和你 **<abbr title="Integrated Development Environment - 集成开发环境: 类似于代码编辑器">IDE</abbr>/<dfn title="一个检查代码错误的程序">linter</dfn>/brain** 适配:
    * 因为 pydantic 数据结构仅仅是你定义的类的实例；自动补全，linting，mypy 以及你的直觉应该可以和你验证的数据一起正常工作。
* 验证**复杂结构**:
    * 使用分层的 Pydantic 模型，Python `typing` 的 `List` 和 `Dict` 等等。
    * 验证器使我们能够简单清楚地将复杂的数据模式定义、检查并记录为 JSON Schema。

* **更簡單**：
    * 不需要學習新的 micro-language 來定義結構。
    * 如果你知道 Python 型別，你就知道如何使用 Pydantic。
* 和你的 **<abbr title="Integrated Development Environment - 整合開發環境: 類似於程式碼編輯器">IDE</abbr>/<dfn title="檢查程式碼錯誤的程式">linter</dfn>/brain** 都能好好配合：
    * 因為 Pydantic 的資料結構其實就是你自己定義的類別實例，所以自動補齊、linting、mypy 以及你的直覺都能很好地在經過驗證的資料上發揮作用。
* 驗證**複雜結構**：
    * 使用 Pydantic 模型時，你可以把資料結構分層設計，並且用 Python 的 `List` 和 `Dict` 等型別來定義。

	// for this instance of unlocker, we should let the
	// caller know that lock is not successfully released
	// yet.
	if len(locks)-tolerance == tolerance {
		// In case of split brain scenarios where
		// tolerance is exactly half of the len(*locks)
		// then we need to make sure we have unlocked
		// upto tolerance+1 - especially for RUnlock
		// to ensure that we don't end up with active

    ],
)

tf_cuda_library(
    name = "c_api_test_util",
    testonly = 1,
    srcs = ["c_api_test_util.cc"],
    hdrs = ["c_api_test_util.h"],
    visibility = [
        "//learning/brain:__subpackages__",
        "//tensorflow:__subpackages__",
    ],
    deps = [
        ":c_api",
        ":c_api_experimental",
        ":c_api_internal",
        "//tensorflow/c:c_test_util",

#include "tensorflow/core/platform/types.h"
#include "tensorflow/core/public/release_version.h"
#include "tensorflow/core/public/session.h"

// The implementation below is at the top level instead of the
// brain namespace because we are defining 'extern "C"' functions.
using tensorflow::AttrValueMap;
using tensorflow::DataType;
using tensorflow::ExtendSessionGraphHelper;
using tensorflow::FullTypeDef;
using tensorflow::Graph;

  }

  private static <T> int drain(
      BlockingQueue<T> q,
      Collection<? super T> buffer,
      int numElements,
      long timeout,
      TimeUnit unit,
      boolean interruptibly)
      throws InterruptedException {
    return interruptibly
        ? Queues.drain(q, buffer, numElements, timeout, unit)

        if (result.isOverflow()) {
          // Not enough room in output buffer--drain it, creating a bigger buffer if necessary.
          startDraining(true);
          continue DRAINING;
        } else if (result.isUnderflow()) {
          // If encoder underflows, it means either:
          // a) the final flush() succeeded; next drain (then done)
          // b) we encoded all of the input; next flush

            @Nullable Throwable error) {}

    /**
     * If this is an accumulating log, it will "drain" this instance. It returns the accumulated log entries, and
     * also "resets" this instance to empty (initial) state.
     */
    @Nonnull
    default List<Entry> drain() {
        return List.of();
    }