CODEWITHSUNDEEP
phpmysqlsql
idris
idris

Reputation: 1049

File get contents returns false

So I'm trying to have users verify that they own the domain. So I generated a file and have them upload it to their site. So I then have to verify it, so what I do is 

file_get_contents($url.'/'.$token.'.html');

All this returns is 

bool(false)

Here's more of the code 

$url = $_POST['url'];

//Get site info
$gin = $con->prepare("SELECT * FROM verify WHERE url = :url");
$gin->bindValue(':url', $url);
$gin->execute();

//Get token
$t = $gin->fetch(PDO::FETCH_ASSOC);
$token = $t['token'];
$url = $t['url'];


//Get content
var_dump(file_get_contents($url.'/'.$token.'.html'));

I have 3 columns in the table token, which is the string in the file. url which is the url obviously, its in example.com format. And a verified column which is either 1 or 0. Any ideas?

Upvotes: 1

Views: 4236

Answers (3)

Hyder B.
Hyder B.

Reputation: 12256

Based on my experience with fetching third-party content from more than 1 million domain names, I would not recommend you to use file_get_contents() because this PHP function cannot handle page redirects, site that requires a valid user-agent etc. The issue you are experiencing might be specific to a certain domain names only. A better approach to your problem is to use curl.

    function download_content($url) {
    $ch = curl_init();
    $timeout = 5;
    curl_setopt($ch, CURLOPT_URL, $url);
    curl_setopt($ch, CURLOPT_RETURNTRANSFER, 1);
    curl_setopt($ch, CURLOPT_CONNECTTIMEOUT, $timeout);
    curl_setopt($ch, CURLOPT_FOLLOWLOCATION, true);
    curl_setopt($ch, CURLOPT_USERAGENT, "Firefox 32.0");
    $data = curl_exec($ch);
    curl_close($ch);
    return $data;
}

Usage:

$returned_content = download_content('http://stackoverflow.com');

Upvotes: 5

Isaac
Isaac

Reputation: 983

Try to add http:// before $url and put a valid url. It will work

var_dump(file_get_contents('http://'.$url.'/'.$token.'.html')); // With a valid URL

Upvotes: 0

Lekensteyn
Lekensteyn

Reputation: 66465

If you enable error_reporting(E_ALL), then you will probably see that the use of HTTP URLs are disallowed due to an ini setting.

Warning: you are possibly opening a hole by allowing arbitrary prefixes in file_get_contents. Try to use parse_url to validate that you actually have a HTTP URL. Then you should probably consider using cURL and disable external redirects (otherwise one could pass a URL such as http://bit.ly/something# and still pass your tests).

Upvotes: 0

Related Questions