I've been asked to handle a security issue for a site which was set up by another programmer. As of yet, I haven't seen any of the code, so I'm going off of assumptions at this point and I want to cover my bases. The group hosting the site ran a security check and found that they had code vulnerable to SQL injection.
Example: www.example.com/code.php?pid=2&ID=35 (GET parameter ID is vulnerable to SQL Injection)
Now, because I'm a novice, I've explained that I can likely resolve the issue with the host, but their site would still need to be looked over by someone who has a deeper knowledge of security.
So, to take care of potential SQL Injections (and without seeing the code), I would use mysql_real_escape_string:
$query = sprintf("SELECT * FROM table WHERE pid='%s' AND ID='%s'",
mysql_real_escape_string($pid),
mysql_real_escape_string($id));
Additionally, I would consider mysqli_real_escape_string and prepared statements, but I don't know how they're configured. But would mysql_real_escape_string take care of potential SQL Injection?
Answer
Skip the old mysql_*
stuff if you can and use PDO.
$pdo = new PDO('mysql:host=localhost;dbname=whatever', $username, $password);
$statement = $pdo->prepare('SELECT * FROM table WHERE pid=:pid AND ID=:id');
$statement->bindParam(':pid', $_GET['pid']);
$statement->bindParam(':id', $_GET['id']);
$results = $statement->execute();
var_dump($results->fetchAll());
No comments:
Post a Comment