Add a custom site

Add a site manually

To add a site, you'll need to pass as a second argument to add() an array that contains at least 3 elements:

host is the domain name of the URLs you want to match, e.g. example.com (including subdomains such as www.example.com)
At least one of the following:
- extract is a regexp used to extract values from the URL.
- scrape is an array with at least one extract value and other optional values:
  - extract is a regexp used to extract values from the scraped page.
  - match is a regexp used to determine whether to scrape the content of the URL. If it's not specified, every URL is scraped.
  - url is an optional URL used for scraping. It can contain values extracted from the original URL, e.g. http://example.org/scrape/{@id}.
  - header is a HTTP header, e.g. User-agent: not Mozilla
Exactly one of the following:
- A choose array that contains a when array and an otherwise array.
  - when must contain the following two elements:
    - A test string that contains an XPath condition.
    - An iframe or flash element as specified below.
  - otherwise must contain one iframe or flash element as specified below.
- An iframe or flash element that contains an array of attributes:
  - src contains the URL of the iframe or Flash object.
  - width and height are optional and default to 640 × 360.
  - Other attributes such as allowfullscreen or scrolling are automatically added where necessary.

You can specify multiple host, scheme, scrape, header, extract, match or when values using arrays.

How to configure multiple `host` and `extract` values

$configurator = new s9e\TextFormatter\Configurator;

$configurator->MediaEmbed->add(
    'youtube',
    [
        'host'    => ['youtube.com', 'youtu.be'],
        'extract' => [
            "!youtube\\.com/watch\\?v=(?'id'[-0-9A-Z_a-z]+)!",
            "!youtu\\.be/(?'id'[-0-9A-Z_a-z]+)!"
        ],
        'iframe'  => [
            'width'  => 560,
            'height' => 315,
            'src'    => 'http://www.youtube.com/embed/{@id}'
        ]
    ]
);

// Get an instance of the parser and the renderer
extract($configurator->finalize());

$text = 'http://www.youtube.com/watch?v=-cEzsCAzTak';
$xml  = $parser->parse($text);
$html = $renderer->render($xml);

echo $html;

<span data-s9e-mediaembed="youtube" style="display:inline-block;width:100%;max-width:560px"><span style="display:block;overflow:hidden;position:relative;padding-bottom:56.25%"><iframe allowfullscreen="" loading="lazy" scrolling="no" src="http://www.youtube.com/embed/-cEzsCAzTak" style="border:0;height:100%;left:0;position:absolute;width:100%"></iframe></span></span>

How to configure the `iframe` renderer

$configurator = new s9e\TextFormatter\Configurator;

$configurator->MediaEmbed->add(
    'youtube',
    [
        'host'    => 'youtube.com',
        'extract' => "!youtube\\.com/watch\\?v=(?'id'[-0-9A-Z_a-z]+)!",
        'iframe'  => [
            'width'  => 560,
            'height' => 315,
            'src'    => 'http://www.youtube.com/embed/{@id}'
        ]
    ]
);

// Get an instance of the parser and the renderer
extract($configurator->finalize());

$text = '[media]http://www.youtube.com/watch?v=-cEzsCAzTak[/media]';
$xml  = $parser->parse($text);
$html = $renderer->render($xml);

echo $html;

<span data-s9e-mediaembed="youtube" style="display:inline-block;width:100%;max-width:560px"><span style="display:block;overflow:hidden;position:relative;padding-bottom:56.25%"><iframe allowfullscreen="" loading="lazy" scrolling="no" src="http://www.youtube.com/embed/-cEzsCAzTak" style="border:0;height:100%;left:0;position:absolute;width:100%"></iframe></span></span>

How to configure the `flash` renderer

$configurator = new s9e\TextFormatter\Configurator;

$configurator->MediaEmbed->add(
    'dailymotion',
    [
        'host'    => 'dailymotion.com',
        'extract' => "!dailymotion\.com/(?:video/|user/[^#]+#video=)(?'id'[A-Za-z0-9]+)!",
        'flash'   => [
            'width'  => 560,
            'height' => 315,
            'src'    => 'http://www.dailymotion.com/swf/{@id}'
        ]
    ]
);

// Get an instance of the parser and the renderer
extract($configurator->finalize());

$text = '[media]http://www.dailymotion.com/video/x222z1[/media]';
$xml  = $parser->parse($text);
$html = $renderer->render($xml);

echo $html;

<span data-s9e-mediaembed="dailymotion" style="display:inline-block;width:100%;max-width:560px"><span style="display:block;overflow:hidden;position:relative;padding-bottom:56.25%"><object data="http://www.dailymotion.com/swf/x222z1" style="height:100%;left:0;position:absolute;width:100%" type="application/x-shockwave-flash" typemustmatch=""><param name="allowfullscreen" value="true"></object></span></span>

How to scrape content

Some media sites don't put all of the necessary data (e.g. the ID of a video) in the URL. In that case, you may have to retrieve it from the page itself.

Note that scraping content is a pretty expensive operation that can take several seconds to complete, in part due to network latency and the responsiveness of the target site.

$configurator = new s9e\TextFormatter\Configurator;

$configurator->MediaEmbed->add(
    'slideshare',
    [
        'host'   => 'slideshare.net',
        'scrape' => [
            // Here we ensure that we don't scrape just every link to http://slideshare.net
            'match'   => '!slideshare\\.net/[^/]+/\\w!',
            // Retrieve the presentationId from the embedded JSON
            'extract' => '!"presentationId":(?<id>[0-9]+)!'
        ],
        'iframe' => [
            'width'  => 427,
            'height' => 356,
            'src'    => 'http://www.slideshare.net/slideshow/embed_code/{@id}'
        ]
    ]
);

// Get an instance of the parser and the renderer
extract($configurator->finalize());

$text = 'http://www.slideshare.net/Slideshare/10-million-uploads-our-favorites';
$xml  = $parser->parse($text);
$html = $renderer->render($xml);

echo $html;

http://www.slideshare.net/Slideshare/10-million-uploads-our-favorites

Specify a different URL for scraping

If the URL used for scraping is different from the media's URL, you can specify it in the url element of the scrape array. You can also use variables in the URL using the familiar syntax {@id}. Values for those variables come from named captures in previous extract regexp and from the tag's attributes if applicable.

For example: the dimensions of a Gfycat video are mentionned in the metadata of their page. However, if someone posted a direct link to a Gfycat .gif image such as http://giant.gfycat.com/SereneIllfatedCapybara.gif, the dimensions would not be available. In the following example, we configure scrape with a custom URL that is known to include the original image's dimensions.

In addition, we specify that both the width and height attributes should be filtered as unsigned integer.

$configurator = new s9e\TextFormatter\Configurator;

$configurator->MediaEmbed->add(
    'gfycat',
    [
        'host'       => 'gfycat.com',
        'attributes' => [
            'height' => ['filterChain' => '#uint'],
            'width'  => ['filterChain' => '#uint']
        ],
        'extract'    => "!gfycat\\.com/(?'id'\\w+)!",
        'scrape'     => [
            'url'     => 'http://gfycat.com/{@id}',
            'extract' => [
                '!property="og:image:height"\s*content="(?<height>\d+)!',
                '!property="og:image:width"\s*content="(?<width>\d+)!'
            ]
        ],
        'iframe'     => [
            'width'     => '{@width}',
            'height'    => '{@height}',
            'src'       => '//gfycat.com/iframe/{@id}'
        ]
    ]
);

// Get an instance of the parser and the renderer
extract($configurator->finalize());

$text = 'http://giant.gfycat.com/SereneIllfatedCapybara.gif';
$xml  = $parser->parse($text);
$html = $renderer->render($xml);

echo $html;

<span data-s9e-mediaembed="gfycat" style="display:inline-block;width:100%;max-width:600px"><span style="display:block;overflow:hidden;position:relative;padding-bottom:56.3333333333333%"><iframe allowfullscreen="" loading="lazy" scrolling="no" src="//gfycat.com/iframe/SereneIllfatedCapybara" style="border:0;height:100%;left:0;position:absolute;width:100%"></iframe></span></span>

Add custom HTTP headers when scraping

Custom HTTP headers can be specified in the scrape configuration.

$configurator->MediaEmbed->add(
    'sitename',
    [
        'host'   => 'example.org',
        'scrape' => [
            'extract' => '#example\\.org/video/(?<id>\\d+)#',
            'header'  => [
                'Cookie: mycookie=1',
                'User-agent: foo-agent'
            ]
        ],
        'iframe' => ['src' => '//example.org/embed/{@id}']
    ]
);