// Replace YouTube references by embedded videos, ?si= attribute is a must
        // E.g. @youtube(Summary,UN0AFCLASZA?si=9aG5tDzj6nL1_IKT&start=371)@ => https://www.youtube.com/embed/UN0AFCLASZA?si=9aG5tDzj6nL1_IKT&start=371"
        // "&rel=0" is also force-injected to prevent video recommendations from other channels

including tensorflow-io should be sandboxed if used to process untrusted data.

For example, if an attacker were to upload a malicious video file, they could
potentially exploit a vulnerability in the TensorFlow code that handles videos,
which could allow them to execute arbitrary code on the system running
TensorFlow.

It is important to keep TensorFlow up to date with the latest security patches

application/x-bzip=bz\n\
application/x-bzip2=bz2\n\
application/x-tar=tar\n\
application/x-rar-compressed=rar\n\
video/3gp=3gp\n\
video/3g2=3g2\n\
video/x-msvideo=avi\n\
video/x-flv=flv\n\
video/mpeg=mpeg\n\
video/mp4=mp4\n\
video/ogv=ogv\n\
video/quicktime=qt\n\
video/x-m4v=m4v\n\
audio/x-aif=aif\n\
audio/midi=midi\n\
audio/mpga=mpga\n\
audio/mp4=mp4a\n\
audio/ogg=oga\n\

  private static final String AUDIO_TYPE = "audio";
  private static final String IMAGE_TYPE = "image";
  private static final String TEXT_TYPE = "text";
  private static final String VIDEO_TYPE = "video";
  private static final String FONT_TYPE = "font";

  private static final String WILDCARD = "*";

  private static final Map<MediaType, MediaType> KNOWN_TYPES = Maps.newHashMap();

     * application/x-tar=tar<br>
     * application/x-rar-compressed=rar<br>
     * video/3gp=3gp<br>
     * video/3g2=3g2<br>
     * video/x-msvideo=avi<br>
     * video/x-flv=flv<br>
     * video/mpeg=mpeg<br>
     * video/mp4=mp4<br>
     * video/ogv=ogv<br>
     * video/quicktime=qt<br>
     * video/x-m4v=m4v<br>
     * audio/x-aif=aif<br>
     * audio/midi=midi<br>
     * audio/mpga=mpga<br>

	etags = make([]string, len(partsMetadata))
	vidMap := map[string]int{}
	for index, metadata := range partsMetadata {
		if errs[index] != nil {
			continue
		}
		vid := metadata.VersionID
		if metadata.VersionID == "" {
			vid = nullVersionID
		}
		vidMap[vid]++
		etags[index] = metadata.Metadata["etag"]
	}

	for _, count := range vidMap {
		// do we have enough common versions

* Ele utiliza um arquivo "spooled":
    * Um arquivo armazenado na memória até um limite máximo de tamanho, e após passar esse limite, ele será armazenado no disco.
* Isso significa que funcionará bem para arquivos grandes como imagens, vídeos, binários grandes, etc., sem consumir toda a memória.
* Você pode receber metadados do arquivo enviado.

* It uses a "spooled" file:
    * A file stored in memory up to a maximum size limit, and after passing this limit it will be stored in disk.
* This means that it will work well for large files like images, videos, large binaries, etc. without consuming all the memory.
* You can get metadata from the uploaded file.

É possível [contribuir](contributing.md){.internal-link target=_blank} no código fonte fazendo Pull Requests, por exemplo:

* Para corrigir um erro de digitação que você encontrou na documentação.
* Para compartilhar um artigo, video, ou podcast criados por você sobre o FastAPI <a href="https://github.com/fastapi/fastapi/edit/master/docs/en/data/external_links.yml" class="external-link" target="_blank">editando este arquivo</a>.

    * Eine Datei, die bis zu einem bestimmten Größen-Limit im Arbeitsspeicher behalten wird, und wenn das Limit überschritten wird, auf der Festplatte gespeichert wird.
* Das bedeutet, es wird für große Dateien wie Bilder, Videos, große Binärdateien, usw. gut funktionieren, ohne den ganzen Arbeitsspeicher aufzubrauchen.
* Sie können Metadaten aus der hochgeladenen Datei auslesen.