PHP Regex to get youtube video ID?

Question

Can someone show me how to get the youtube id out of a url regardless of what other GET variables are in the URL.

Use this video for example: http://www.youtube.com/watch?v=C4kxS1ksqtw&feature=related
So between v= and before the next &

Answer 1

This can be very easily accomplished using parse_str and parse_url and is more reliable in my opinion.

My function supports the following urls:

• http://youtube.com/v/dQw4w9WgXcQ?feature=youtube_gdata_player
• http://youtube.com/vi/dQw4w9WgXcQ?feature=youtube_gdata_player
• http://youtube.com/?v=dQw4w9WgXcQ&feature=youtube_gdata_player
• http://www.youtube.com/watch?v=dQw4w9WgXcQ&feature=youtube_gdata_player
• http://youtube.com/?vi=dQw4w9WgXcQ&feature=youtube_gdata_player
• http://youtube.com/watch?v=dQw4w9WgXcQ&feature=youtube_gdata_player
• http://youtube.com/watch?vi=dQw4w9WgXcQ&feature=youtube_gdata_player
• http://youtu.be/dQw4w9WgXcQ?feature=youtube_gdata_player

Also includes the test below the function.

/**
 * Get Youtube video ID from URL
 *
 * @param string $url
 * @return mixed Youtube video ID or FALSE if not found
 */
function getYoutubeIdFromUrl($url) {
    $parts = parse_url($url);
    if(isset($parts['query'])){
        parse_str($parts['query'], $qs);
        if(isset($qs['v'])){
            return $qs['v'];
        }else if(isset($qs['vi'])){
            return $qs['vi'];
        }
    }
    if(isset($parts['path'])){
        $path = explode('/', trim($parts['path'], '/'));
        return $path[count($path)-1];
    }
    return false;
}
// Test
$urls = array(
    'http://youtube.com/v/dQw4w9WgXcQ?feature=youtube_gdata_player',
    'http://youtube.com/vi/dQw4w9WgXcQ?feature=youtube_gdata_player',
    'http://youtube.com/?v=dQw4w9WgXcQ&feature=youtube_gdata_player',
    'http://www.youtube.com/watch?v=dQw4w9WgXcQ&feature=youtube_gdata_player',
    'http://youtube.com/?vi=dQw4w9WgXcQ&feature=youtube_gdata_player',
    'http://youtube.com/watch?v=dQw4w9WgXcQ&feature=youtube_gdata_player',
    'http://youtube.com/watch?vi=dQw4w9WgXcQ&feature=youtube_gdata_player',
    'http://youtu.be/dQw4w9WgXcQ?feature=youtube_gdata_player'
);
foreach($urls as $url){
    echo $url . ' : ' . getYoutubeIdFromUrl($url) . "\n";
}

Answer 2

We know that video ID is 11 chars length and can be preceded by v= or vi= or v/ or vi/ or youtu.be/. So the simplest way to do this:

<?php
$youtube = 'http://youtube.com/v/dQw4w9WgXcQ?feature=youtube_gdata_player
http://youtube.com/vi/dQw4w9WgXcQ?feature=youtube_gdata_player
http://youtube.com/?v=dQw4w9WgXcQ&feature=youtube_gdata_player
http://www.youtube.com/watch?v=dQw4w9WgXcQ&feature=youtube_gdata_player
http://youtube.com/?vi=dQw4w9WgXcQ&feature=youtube_gdata_player
http://youtube.com/watch?v=dQw4w9WgXcQ&feature=youtube_gdata_player
http://youtube.com/watch?vi=dQw4w9WgXcQ&feature=youtube_gdata_player
http://youtu.be/dQw4w9WgXcQ?feature=youtube_gdata_player';

preg_match_all("#(?<=v=|v\/|vi=|vi\/|youtu.be\/)[a-zA-Z0-9_-]{11}#", $youtube, $matches);

var_dump($matches[0]);

And output:

array(8) {
  [0]=>
  string(11) "dQw4w9WgXcQ"
  [1]=>
  string(11) "dQw4w9WgXcQ"
  [2]=>
  string(11) "dQw4w9WgXcQ"
  [3]=>
  string(11) "dQw4w9WgXcQ"
  [4]=>
  string(11) "dQw4w9WgXcQ"
  [5]=>
  string(11) "dQw4w9WgXcQ"
  [6]=>
  string(11) "dQw4w9WgXcQ"
  [7]=>
  string(11) "dQw4w9WgXcQ"
}

Answer 3

fixed based on How to validate youtube video ids?

<?php

$links = [
    "youtube.com/v/tFad5gHoBjY",
    "youtube.com/vi/tFad5gHoBjY",
    "youtube.com/?v=tFad5gHoBjY",
    "youtube.com/?vi=tFad5gHoBjY",
    "youtube.com/watch?v=tFad5gHoBjY",
    "youtube.com/watch?vi=tFad5gHoBjY",
    "youtu.be/tFad5gHoBjY",
    "http://youtu.be/qokEYBNWA_0?t=30m26s",
    "youtube.com/v/vidid",
    "youtube.com/vi/vidid",
    "youtube.com/?v=vidid",
    "youtube.com/?vi=vidid",
    "youtube.com/watch?v=vidid",
    "youtube.com/watch?vi=vidid",
    "youtu.be/vidid",
    "youtube.com/embed/vidid",
    "http://youtube.com/v/vidid",
    "http://www.youtube.com/v/vidid",
    "https://www.youtube.com/v/vidid",
    "youtube.com/watch?v=vidid&wtv=wtv",
    "http://www.youtube.com/watch?dev=inprogress&v=vidid&feature=related",
    "youtube.com/watch?v=7HCZvhRAk-M"
];

foreach($links as $link){
    preg_match("#([\/|\?|&]vi?[\/|=]|youtu\.be\/|embed\/)([a-zA-Z0-9_-]+)#", $link, $matches);
    var_dump(end($matches));
}

Answer 4

I had some post content I had to cipher throughout to get the Youtube ID out of. It happened to be in the form of the <iframe> embed code Youtube provides.

 <iframe src="http://www.youtube.com/embed/Zpk8pMz_Kgw?rel=0" frameborder="0" width="620" height="360"></iframe>

The following pattern I got from @rob above. The snippet does a foreach loop once the matches are found, and for a added bonus I linked it to the preview image found on Youtube. It could potentially match more types of Youtube embed types and urls:

$pattern = '#(?<=(?:v|i)=)[a-zA-Z0-9-]+(?=&)|(?<=(?:v|i)\/)[^&\n]+|(?<=embed\/)[^"&\n]+|(?<=‌​(?:v|i)=)[^&\n]+|(?<=youtu.be\/)[^&\n]+#';

preg_match_all($pattern, $post_content, $matches);

foreach ($matches as $match) {
    $img = "<img src='http://img.youtube.com/vi/".str_replace('?rel=0','', $match[0])."/0.jpg' />";
    break;
}

Rob's profile: https://stackoverflow.com/users/149615/rob

Answer 5

(?<=\?v=)([a-zA-Z0-9_-]){11}

This should do it as well.

Answer 6

I know that the title of the thread refers to the use of a regex, but just as the Zawinski quote says, I really think that avoiding regexes is best here. I'd recommend this function instead:

function get_youtube_id($url)
{
    if (strpos( $url,"v=") !== false)
    {
        return substr($url, strpos($url, "v=") + 2, 11);
    }
    elseif(strpos( $url,"embed/") !== false)
    {
        return substr($url, strpos($url, "embed/") + 6, 11);
    }

}

I recommend this because the ID of YouTube videos is always the same, independent from the style of the URL, e.g.

• http://www.youtube.com/watch?v=t_uW44Bsezg
• http://www.youtube.com/watch?feature=endscreen&v=Id3xG4xnOfA&NR=1
• `And Other Ulr Form In Which The Word "embed/" Is Placed Before The Id ... !!

and that might be the case for embedded and iframe-ed stuff.