Variáveis de ambiente podem ser úteis para lidar com **configurações** do aplicativo, como parte da **instalação** do Python, etc.

## Criar e Usar Variáveis de Ambiente { #create-and-use-env-vars }

Você pode **criar** e usar variáveis de ambiente no **shell (terminal)**, sem precisar do Python:

//// tab | Linux, macOS, Windows Bash

<div class="termy">

```console

    * `description`: ein `str` mit einer kurzen Beschreibung für die externe Dokumentation.
    * `url` (**erforderlich**): ein `str` mit der URL für die externe Dokumentation.

### Metadaten für Tags erstellen { #create-metadata-for-tags }

Versuchen wir es mit einem Beispiel mit Tags für `users` und `items`.

Erstellen Sie Metadaten für Ihre Tags und übergeben Sie diese an den Parameter `openapi_tags`:

    * `description`: un `str` con una breve descripción para la documentación externa.
    * `url` (**requerido**): un `str` con la URL para la documentación externa.

### Crear metadata para etiquetas { #create-metadata-for-tags }

Probemos eso en un ejemplo con etiquetas para `users` y `items`.

Crea metadata para tus etiquetas y pásala al parámetro `openapi_tags`:

{* ../../docs_src/metadata/tutorial004_py39.py hl[3:16,18] *}

    private Proxy proxy;

    private List<ArtifactRepository> mirroredRepositories = Collections.emptyList();

    private boolean blocked;

    public MavenArtifactRepository() {}

    /**
     * Create a remote download repository.
     *
     * @param id        the unique identifier of the repository
     * @param url       the URL of the repository
     * @param layout    the layout of the repository

Las variables de entorno pueden ser útiles para manejar **configuraciones** de aplicaciones, como parte de la **instalación** de Python, etc.

## Crear y Usar Variables de Entorno { #create-and-use-env-vars }

Puedes **crear** y usar variables de entorno en la **shell (terminal)**, sin necesidad de Python:

//// tab | Linux, macOS, Windows Bash

<div class="termy">

```console

		i--
	}

	j := i
	for j > 0 && !os.IsPathSeparator(dirPath[j-1]) { // Scan backward over element.
		j--
	}

	if j > 1 {
		// Create parent.
		if err := osMkdirAll(dirPath[:j-1], perm, baseDir); err != nil {
			return err
		}
	}

	// Parent now exists; invoke Mkdir and use its result.
	if err := Mkdir(dirPath, perm); err != nil {

export TFCI=py311,linux_x86,public_cache,disk_cache

# Recommended: Configure Docker. (Linux only)
#
#   TF uses hub.docker.com/r/tensorflow/build containers for CI,
#   and scripts on Linux create a persistent container called "tf"
#   which mounts your TensorFlow directory into the container.
#
#   Important: because the container is persistent, you cannot change TFCI

						}
					}
					association.Error = association.DB.Session(&Session{
						NewDB: true,
					}).Model(child.Interface()).Create(mapv.Interface()).Error
				case schema.Many2Many:
					association.Error = association.DB.Session(&Session{
						NewDB: true,
					}).Model(child.Interface()).Create(mapv.Interface()).Error

					for _, key := range mapv.MapKeys() {
						k := strings.ToLower(key.String())

#### Pydantic's `.model_dump()` { #pydantics-model-dump }

`user_in` is a Pydantic model of class `UserIn`.

Pydantic models have a `.model_dump()` method that returns a `dict` with the model's data.

So, if we create a Pydantic object `user_in` like:

```Python
user_in = UserIn(username="john", password="secret", email="******@****.***")
```

and then we call:

```Python
user_dict = user_in.model_dump()
```

When you install FastAPI, it comes with a production server, Uvicorn, and you can start it with the `fastapi run` command.

But you can also install an ASGI server manually.

Make sure you create a [virtual environment](../virtual-environments.md){.internal-link target=_blank}, activate it, and then you can install the server application.

For example, to install Uvicorn:

<div class="termy">

```console