fideloper/porter

Extendable link-parser and meta-data obtainer for various services such as Vimeo and Youtube

dev-develop 2013-07-04 15:29 UTC

Requires

• shuber/curl: dev-master

Requires (Dev)

• mockery/mockery: >=0.7.2

MIT

• Chris Fidao

• parser
• youtube
• vimeo
• meta
• meta data
• link parser

README

Extendable url-parser thingy and meta-data obtainer for various services such as Youtube and Vimeo

What does it do?

Say you have a piece of content, maybe from a user on your site.

I love Sriracha sauce!!! I even put it on my Cinnamon Toast Struddle Crunch Tiger Blood Leprechaun cereal!
Here's a video review of it http://www.youtube.com/watch?v=GYamE6G1EZo,
and a spiffy ad for Sriracha Sauce!!! http://vimeo.com/41852814

You like how Twitter shows an embed and information about linked Youtube videos, so you figure

Hey, Twitter is cool! I'm gonna do exactly what Twitter does so I'm cool too!

Well you're in luck! Using this spicy library, you can grab the URLs from content and get meta information about them.

How do I use it?

$some_weirdos_comment = "I love Sriracha sauce!!! I even put it on my Cinnamon Toast Struddle Crunch Tiger Blood Leprechaun cereal! Here's a video review of it http://www.youtube.com/watch?v=GYamE6G1EZo, and a spiffy ad for Sriracha Sauce!!! http://vimeo.com/41852814";

$parser = new \Porter\Parser();

// I want YouTooooooob!
$parser->addService( new \Porter\Service\Youtube() );

// I want Vimeooooooooo!
$parser->addService( new \Porter\Service\Vimeo() );

$parsed_urls = $parser->parse( $parser->matchUrls($some_weirdos_comment) );

// ID's Parsed from URLs directly
foreach ( $parsed_urls as $parsed )
{
    $item_id = $parsed->getId(); //Youtube: GYamE6G1EZo, Vimeo: 41852814
}

// Now, let's get fancy with some API requests
foreach ( $parsed_urls as $parsed )
{
    $meta = $parsed->getMetadata( new \Porter\Request\Curl() ); // stdClass of data
}

Foreach service (Currently Youtube, Vimeo), you can get:

1. Video ID
2. Meta Data Available via public api
   • ID
   • Image
   • Play Time
   • Title
   • etc

Contribute!

There's room for improvement. For instance:

1. I've abstracted out a Request implementation, when the code using it pretty clearly expects a cURL request to happen (Gives it a URL to grab data from)
2. Only supports Youtube/Vimeo at the moment (Flickr? Twitpic? etc etc)
3. The Parser class loops through each service, and creates a new instance of each service on each iteration. Needed for multiple URLs, but I dislike the implementation.

Licence

MIT