return failedTests;
    }

    /**
     * Class for identifying test output sources. We use this rather than Gradle's {@link TestDescriptor} as we want
     * to avoid any nasty memory leak issues that come from keeping Gradle implementation types in memory. Since we
     * use this a the key for our HashMap, it's best to control the implementation as there's no guarantee that Gradle's

 - [Confluence/Jira](https://github.com/codelibs/fess-ds-atlassian)
 - [Box](https://github.com/codelibs/fess-ds-box)
 - [CSV](https://github.com/codelibs/fess-ds-csv)
 - [Database](https://github.com/codelibs/fess-ds-db)
 - [Dropbox](https://github.com/codelibs/fess-ds-dropbox)
 - [Elasticsearch](https://github.com/codelibs/fess-ds-elasticsearch)
 - [Git](https://github.com/codelibs/fess-ds-git)

You could also use `from starlette.responses import HTMLResponse`.

**FastAPI** provides the same `starlette.responses` as `fastapi.responses` just as a convenience for you, the developer. But most of the available responses come directly from Starlette.

///

### Multiple File Uploads with Additional Metadata { #multiple-file-uploads-with-additional-metadata }

- Make sure there is a space between an ellipsis and a word following or preceding the ellipsis.

Examples:

Source (English):

```
...as we intended.
...this would work:
...etc.
others...
More to come...
```

Result (German):

```
... wie wir es beabsichtigt hatten.
... das würde funktionieren:
... usw.
Andere ...
Später mehr ...
```

Because we are using a relative URL, if your API was located at `https://example.com/`, then it would refer to `https://example.com/token`. But if your API was located at `https://example.com/api/v1/`, then it would refer to `https://example.com/api/v1/token`.

## TensorFlow models are programs

TensorFlow
[**models**](https://developers.google.com/machine-learning/glossary/#model) (to
use a term commonly used by machine learning practitioners) are expressed as
programs that TensorFlow executes. TensorFlow programs are encoded as
computation
[**graphs**](https://developers.google.com/machine-learning/glossary/#graph).
Since models are practically programs that TensorFlow executes, using untrusted

 * limitations under the License.
 */

package com.google.common.collect;

import static com.google.common.base.Preconditions.checkNotNull;
import static java.util.Collections.singletonMap;
import static java.util.stream.Collectors.collectingAndThen;
import static java.util.stream.Collectors.toMap;

import com.google.common.annotations.GwtCompatible;
import com.google.common.annotations.GwtIncompatible;

 * limitations under the License.
 */

package com.google.common.collect;

import static com.google.common.base.Preconditions.checkNotNull;
import static java.util.Collections.singletonMap;
import static java.util.stream.Collectors.collectingAndThen;
import static java.util.stream.Collectors.toMap;

import com.google.common.annotations.GwtCompatible;
import com.google.common.annotations.GwtIncompatible;

# Middleware Avançado { #advanced-middleware }

No tutorial principal você leu como adicionar [Middleware Personalizado](../tutorial/middleware.md) à sua aplicação.

E então você também leu como lidar com [CORS com o `CORSMiddleware`](../tutorial/cors.md).

Nesta seção, veremos como usar outros middlewares.

## Adicionando middlewares ASGI { #adding-asgi-middlewares }

* Verificar se existe um parâmetro query opcional nomeado como `q` (como em `http://127.0.0.1:8000/items/foo?q=somequery`) para requisições `GET`.
    * Como o parâmetro `q` é declarado com `= None`, ele é opcional.
    * Sem o `None` ele seria obrigatório (como o corpo no caso de `PUT`).
* Para requisições `PUT` para `/items/{item_id}`, lerá o corpo como JSON:
    * Verifica que tem um atributo obrigatório `name` que deve ser `str`.