## Setting up your development environment

In order to make changes to Gradle, you'll need:

* [Adoptium JDK](https://adoptium.net/temurin/archive/?version=17) (Java Development Kit) **version 17**. Fixed version is required to use [remote cache](#remote-build-cache).

            }
        }

        for (String key : versions.keySet()) {
            if (versions.get(key) == null && managedVersions.get(key) == null) {
                InputLocation location = plugins.get(key).getLocation("");
                problems.add(new ModelProblemCollectorRequest(Severity.WARNING, Version.V20)
                        .setMessage("'build.plugins.plugin.version' for " + key + " is missing.")

      Class.forName("java.nio.file.Path");
      return new JavaNioCreator();
    } catch (ClassNotFoundException runningUnderAndroid) {
      // Try another way.
    }

    try {
      int version = (int) Class.forName("android.os.Build$VERSION").getField("SDK_INT").get(null);
      int jellyBean =
          (int) Class.forName("android.os.Build$VERSION_CODES").getField("JELLY_BEAN").get(null);
      /*

/// info | 信息

FastAPI 在 0.95.0 版本添加了对 `Annotated` 的支持（并开始推荐使用它）。

如果你使用的是更旧的版本，尝试使用 `Annotated` 会报错。

请确保在使用 `Annotated` 之前，将 FastAPI 版本[升级](../deployment/versions.md#upgrading-the-fastapi-versions)到至少 0.95.1。

///

## 声明元数据 { #declare-metadata }

你可以声明与 `Query` 相同的所有参数。

例如，要为路径参数 `item_id` 声明 `title` 元数据值，你可以这样写：

A **context manager** in Python is something that you can use in a `with` statement, for example, `open()` can be used as a context manager:

```Python
with open("file.txt") as file:
    file.read()
```

In recent versions of Python, there's also an **async context manager**. You would use it with `async with`:

```Python
async with lifespan(app):
    await do_stuff()
```

        // this generates timestamped versions like maven-resolver-provider:
        // https://github.com/apache/maven/blob/03df5f7c639db744a3597c7175c92c8e2a27767b/maven-resolver-provider/src/main/java/org/apache/maven/repository/internal/RemoteSnapshotMetadata.java#L79
        String version = artifact.getVersion();
        String qualifier = formatDate(timestamp, true) + '-' + buildNumber;

				DeleteMarker:     tc.isDelMarker,
				IsLatest:         !tc.isNoncurrent,
				SuccessorModTime: tc.objectSuccessorModTime,
				VersionID:        tc.versionID,
			}
			opts.NumVersions = 1
			if tc.hasManyVersions {
				opts.NumVersions = 2 // at least one noncurrent version
			}
			if res := lc.Eval(opts); res.Action != tc.expectedAction {
				t.Fatalf("Expected action: `%v`, got: `%v`", tc.expectedAction, res.Action)

- plugin execution error due to something that is not expected to go wrong (the compiler executable missing)
- asking to use a plugin for which you do not have a version defined - tools to easily select versions
- goal not found in a plugin (probably could list the ones that are)

 */

// PluginNotFoundException, PluginResolutionException, PluginDescriptorParsingException,

    export flags="--config=linux -k"
    ```

    If the tests are to be run on the GPU:

    *   For TensorFlow versions starting from v.2.18.0: Add the `cuda` option
        flag.

        ```bash
        export flags="--config=linux --config=cuda -k"
        ```

    *   For TensorFlow versions prior v.2.18.0: Add CUDA paths to
        LD_LIBRARY_PATH and add the `cuda` option flag.

        ```bash

	// Represents Erasure backend - single drive
	formatBackendErasureSingle = "xl-single"

	// formatErasureV1.Erasure.Version - version '1'.
	formatErasureVersionV1 = "1"

	// formatErasureV2.Erasure.Version - version '2'.
	formatErasureVersionV2 = "2"

	// formatErasureV3.Erasure.Version - version '3'.
	formatErasureVersionV3 = "3"

	// Distribution algorithm used, legacy
	formatErasureVersionV2DistributionAlgoV1 = "CRCMOD"