Secretly adding (extra) HTTP GET Variables to a swf file through PHP - php

I'm trying to build a workaround for embedding my (downloaded) flash videoplayer. (it's the JWplayer...)
At the moment, when somebody wants to embed the videoplayer they have to include the swf with all the added flashvars (ex: www.site.be/core/swf/player.swf?streamer=url&file=file.mp4&image=file.jpg&plugin=analytics...).
That's messy, and it feels a bit risky... people who know what they are doing can also just remove the plugin and other added data, resolving in me not being able to track pageviews etc.
My workaround is like this:
$data = file_get_contents('URL');
header("content-type: application/x-shockwave-flash");
echo $data;
Turns out that, when I use file_get_contents on a regular test file, info.php, that responds through a $_GET['var'], the above stated code works, but when I use it on the flashplayer, it doesn't...
As in: the flash file does not seem to be accepting (or responding to) the added header variables...
Can somebody tell me why this is? Is this a 'flash' related problem or a 'Php' related problem?
Or are there suggestions on how to handle my "flash-embed-with-to-much-junk"-problem in a different way?
(thanks)

The flash is expecting GET parameters so you can't force them any other way.
What I would do is store the GET variables in a SESSION (called swf_vars in my example) if you want it a secret, then have the <embed> code point to a PHP script that does something like..
<?php
session_start();
// Full URL path to SWF
$url = "http://www.site.be/core/swf/player.swf?";
// These are the GET variables you want
foreach ($_SESSION['swf_vars'] as $key => $value) {
$url .= $key . "=" . urlencode($value) . "&";
}
$url = rtrim("&", $url);
// Fetch the SWF
header("Content-Type: application/x-shockwave-flash");
echo file_get_contents($url);
?>

Related

file_get_contents api link two connections with session

I have two api pages, api1.php and api2.php. On 1 there is set an session and on 2 this session needs to be returned.
Ofcourse there will be additional functions but my goal of this is to link those two api connections to one and eachother by using a session.
api1.php:
session_start();
$api_key = 'dfdsakdsfjdskfjdskfdsjfdfewfifjjsd';
$_SESSION['api_key'] = $api_key;
setcookie('api_key', $api_key);
api2.php:
session_start();
echo $_SESSION['api_key'];
echo $_COOKIE['api_key'];
test.php:
$url = 'http://example.com/api1.php';
$content1 = file_get_contents($url);
$url2 = 'http://example.com/api2.php';
$content2 = file_get_contents($url2);
echo $content2;
As you may have noticed, i'm visiting the page test.php to obtain a result.
But no result is being returned.
Can somebody tell me why this is not working and what may be an additional way of making all of this happen?
(Notice: the example.com are both the same site (mine))
You're code "links" correctly. The problem is actually in test.php! Instead of executing the code contained in both files, it retrieves the entire file. If you view source you will note the PHP tags and your code. A better solution to check if this is working is to go to api1.php and api2.php separately. With some code adjustments you could also just use the include() or require() functions. Which would look like this:
api2.php
echo $_SESSION['api_key'] . "\n<br/>\n";
echo $_COOKIE['api_key'];
test.php
include('api1.php');
include('api2.php');
It's worth noting the using the include and require functions executes the code in api1.php and api2.php as if that code were a part of test.php.

What's wrong with this include statement?

This function is returning the content of the file rather the result of fetch_link_settings_overide() within it.
The issue is not with the overide function as after the initial error I commented out my modification just to be sure it wasn't something I had done there.
function fetch_link_settings(){
include( plugins_url()."/plugin-child/plugin_overrides.php");
return fetch_link_settings_override();
}
Adding the content of the derived function plugin-child/plugin_overrides.php as we are not getting anywhere currently.
function fetch_link_settings_override(){
global $post;
// If the destination url is set by the user, use that. Otherwise, use the permalink
$destination_url = get_post_meta($post->ID, '_promo_slider_url', true);
// ASAdd additional place to look in the case of the post being via the PODS advert track
if( ! $destination_url )
$destination_url = get_post_meta($post->ID, 'okd_advert_link', true);
if( ! $destination_url )
$destination_url = get_permalink($post->ID);
// If the target attribute is set by the user, use that. Otherwise, set it to _self
$target = get_post_meta($post->ID, '_promo_slider_target', true);
if( ! $target ) $target = '_self';
// Setup the disable links variable
$disable_links = get_post_meta($post->ID, '_promo_slider_disable_links', true);
return compact('destination_url', 'target', 'disable_links');
}
You write this:
include( plugins_url()."/plugin-child/plugin_overides.php");
Why is plugins_url() there? The include function is strictly based on the file system:
The `include` statement includes and evaluates the specified file.
As explained in the WordPress docs, the plugins_url() would give you the full web URL which is 100% different than the file system WordPress is installed on:
Retrieves the absolute URL to the plugins directory (without the
trailing slash) or, when using the $path argument, to a specific file
under that directory.
So perhaps it should be like this:
include("/plugin-child/plugin_overides.php");
Or perhaps you need the plugin_dir_path()?
include(plugin_dir_path( __FILE__ ) . "/plugin-child/plugin_overides.php");
But that seems wrong. Where would /plugin-child/plugin_overides.php? Try doing this:
include("/full/path/to/wordpress/and/this/plugin-child/plugin_overides.php");
Just replace /full/path/to/wordpress/and/this/ with the actual file system path to /plugin-child/plugin_overides.php.
EDIT: Since the original poster is persistent in using plugins_url() despite all of the suggestions otherwise, here is my detailed response:
…you said “you cannot load raw functions via a URL with include” well
this is not relevant because even if I add $some_var = 'smith'; as the
first statement in the included file, it is not visible in the
function using the include.
Apologies. Functions, classes, strings, constants… Just about anything that you want to be raw, unprocessed PHP will simply not be passed via an http:// or https:// URL because Apache will parse the PHP instructions & simply return the output of that file and not the raw, unprocessed contents of the PHP in that file.
Additionally the original poster contents the following:
You can’t help me because what you are saying does not make sense or
you are not explaining yourself adequately. Look at these examples:
include realpath(dirname(FILE) . "/" . "relative_path");
include("data://text/plain;base64,".base64_encode($content));
include("data://text/plain,".urlencode($content));
All taken from the official PHP documentation. They all use
functions returning components that are concatenated with the rest of
the url. I also tried this typing the filepath explicitly and the
result is the same.
The examples cited are as follows:
include realpath(dirname(FILE) . "/" . "relative_path");
This is a filesystem level include which is the most common way PHP files are included into other files.
include("data://text/plain;base64,".base64_encode($content));
include("data://text/plain,".urlencode($content));
These are both data URLs. Not http or https. So again when you use plugins_url() what you are getting is a full http:// or https:// URL in which Apache parses the PHP instructions & simply return the output of that file and not the raw, unprocessed contents of the PHP in that file. Or as very clearly explained in the PHP documentation you are linking to; emphasis mine:
If "URL include wrappers" are enabled in PHP, you can specify the file
to be included using a URL (via HTTP or other supported wrapper - see
Supported Protocols and Wrappers for a list of protocols) instead of a
local pathname. If the target server interprets the target file as PHP
code, variables may be passed to the included file using a URL request
string as used with HTTP GET. This is not strictly speaking the same
thing as including the file and having it inherit the parent file's
variable scope; the script is actually being run on the remote server
and the result is then being included into the local script.
Going back to your example, you say now the contents of plugin_overides.php is $some_var = 'smith';. How exactly? If it is a PHP file like this:
<?php
$some_var = 'smith';
?>
When you call that file via a URL generated by the following code:
include(plugins_url() . "/plugin-child/plugin_overrides.php");
Assuming your website is http://some.cool.website/ the you are basically making a call like this:
http://some.cool.website/plugin-child/plugin_overides.php
So the output of plugin_overides.php would be 100% blank. If you wanted to get output of that file, you could do the following:
<?php
$some_var = 'smith';
echo $some_var;
?>
And that would return smith. Meaning the absolute ONLY output you would get from that call is pure text. Nothing else.
Now I see you actually have posted the contents of plugin_overides.php. My example explanation above is still apt, but still a basic question. This is your function; just the interface & return for example:
function fetch_link_settings_override(){
// Other code removed. Just a structural illustration for now.
return compact('destination_url', 'target', 'disable_links');
}
Do you actually call fetch_link_settings_override() in plugin_overides.php when it runs? Well, if that function does not run, then there is 100% no way you will ever get any output. But assuming good faith, look at your return statement here:
return compact('destination_url', 'target', 'disable_links');
If you are returning compact, then you are returning an array. You cannot simply return a bare array as a URL call like this http://some.cool.website/plugin-child/plugin_overides.php. The output at most would be simply the word Array.
If the goal is to take that array & do something, then you should use json_encode in fetch_link_settings_override and then use json_decode on the receiving side of that. So the return statement would be something like this:
return json_encode(compact('destination_url', 'target', 'disable_links'));

$_GET parameter not coming through in php

I'm trying to debug a problem with a wordpress plugin.
It creates some html like this:
<audio src="http://mydomain.com/wordpress/wp-content/plugins/the-stiz-audio-for-woocommerce/server/preview.mp3?nonce=e38553d840&audio=4mgvi4h3Jg5qD_Vj5Qj5Kp9V3WxcfxSqhZ71zoEfqGPM3xQL-KqT7fsCoywEk3_VmJSZpMnCa5FNYB_H_50Uwuvy6G8iDgNbMP2ezlV3moj8T0mUkoj7R5tcegYlyRAu3cNTZvTkza4A0P7_g2AhhvQ60FwFAvGZ9jjy9iPJIdU"
class="wcjd-audio-individual" type="audio/mpeg">
</audio>
preview.mp3 is actually a php script.
The 'nonce' param makes it into the script ok, but 'audio' is empty, although it is set. It must be getting stripped off for some reason but I don't know why.
EDIT:
Thanks for the replies so far...
I didn't include the php code originally because it's very simple, but I should have:
/**
* #file Instantiate a WCJDServerAudio instance and allow it handle the request.
*/
define('WCJD_ROOT', realpath(dirname(__FILE__).'/../'));
include_once WCJD_ROOT.'/../../../wp-config.php';
include_once WCJD_ROOT.'/include.php';
$fp = fopen("../../log.txt", "w");
fwrite($fp, "nonce: ".$_GET['nonce']."\n");
fwrite($fp, "audio: ".$_GET['audio']);
fclose($fp);
// The code above results in a file with
// nonce: a90c436753
// audio:
$server = new WCJDServeAudio($_GET);
// Determine whether this is a valid request
if ($server->validRequest()) {
$server->output();
die();
} else {
header('HTTP/1.1 403 Forbidden');
include WCJD_ROOT.'/views/error/403.php';
}
If I simply run this script in a browser it's able to see the 'audio' param, but when it runs as a script from inside the 'audio' tag, the audio param is stripped.
I added some code before the '$server =' part that wrote the contents of $_GET to a log file so I could see in the latter case - and only 'nonce' is set.
The .htaccess is indeed set to make sure the .mp3 extension gets treated as a .php in that directory.
The way it behaves makes me think there is some kind of size limit or magic-quotes issue causing it to drop the audio param. But I have no idea.
Seems like its working as it should. We can't help you unless you post the PHP code.
This works on my localhost pretty well.
<?php
echo $_GET['nonce'];
echo "<br>";
echo $_GET['audio'];
OUTPUT:

Equivalent of PHP's dirname(__FILE__) in JavaScript?

as I already mentioned in the title, I'm looking for a JS-function for getting the same result like I get with this PHP code:
dirname(dirname(__FILE__))
Thanks in advance!
I don't think it is possible because php dirname operates on apache server on local machine. It has access to the filesystem. But javascript operates on browser layer which can't operate with filesystem. I think so you should use ajax and proccess result how you need it. I think so its best solution for you.
I needed a solution to write code like this:
$("#div").load(ROOT_URL + "my/path/to/script.php");
Solution: a PHP script generates one JS-file of all needed JS-files and adds the ROOT_URL to the top of the generated file:
$js = 'ROOT_URL = "' . ROOT_URL . '"; ' . $js;
file_put_contents("file.js", $js);
Now I'm able to use the ROOT_URL (set in a PHP config-file) in JS-code as well. I hope I could help.
You can have PHP output the script. Yes, that's right, you probably can't make php process js files (unless you are in full control of the server). But it doesn't matter. Just make sure that the MIME type is correct, both in the headers PHP returns and the script tag. That way, you can have PHP insert the any values you want in the script, including it's own path.
In script.php:
header("Content-type: text/javascript");
echo 'var myvar = '.$something;
//where $something can be $_SERVER['REQUEST_URI'], __FILE__ or whatever you need.
//You could even use information from session variables, or query the database.
//In fact, this way you can have GET parameters in your javascript.
//Make sure you are not creating a vulnerability with the exposed information.
//Then put the rest of the script as usual. You could even include it*.
*: include
In HTML:
<script type="text/javascript" src="script.php"></script>
Yes, I know I'm repeating the MIME type, do it this way to maximize browser compatibility.
There's no analogue of __FILE__ in browser Javascript; the code does not have direct access to the URL from which it was loaded. But with certain assumptions you can figure it out, as in the answer here.
Once you have the URL of the script (I assume in a variable called scriptURL below) you can set about finding the grandparent URL. This can get tricky with URLs, so it's probably safest to let the URL-savvy bits of Javascript parse the URL for you and get just the pathname component before you start with the string-munging:
var a = document.createElement('a')
a.href = scriptURL
var scriptPath = a.pathname
Then it's unfortunately down to string manipulation; here's one somewhat clunky solution:
var components = scriptPath.split(/\//)
while (components.length > 0 && !components[components.length-1])
components.length -= 1;
var twoDirsUp = components.slice(0,components.length-2).join('/')
And then you can convert the result back into a full URL using the anchor element trick in reverse:
a.pathname = twoDirsUp;
var grandParentUrl = a.href
Why not load what you want from absolute URL?
If you have inse your block of codes: /my/script/to/load.js browser will load the correct file if you are in yoursite.com or whatever like yoursite.com/a/b/c/d/e/f
A little off topic, but if you just want to get the similar of dirname($_SERVER['REQUEST_URI']) for javascript, you can do
window.location.href.substr(0, window.location.href.length - window.location.href.split('/').pop().length)
I use something like that to free from the paths in javascript
var __DIR__ = window.location.pathname.match('(.*\/).*')[1] + 'NameOfThisFolder';
first
window.location.pathname.match('(.*\/).*')[1]
return the current path without the file name or other stuff.
rootFolder/folder1/folder2/
then I add the name of this folder ('NameOfThisFolder').
In this way, I can make for instance ajax request in current page from a page that was called in turn from an ajax request without worry about the path

How to Capture PHP Output into a Variable

I've been wanting to do this because my site does about 3 HTTP requests per page load, because each PHP's output is retrieved with cURL. This is too slow, and I want to avoid using cURL. I found this question on Stack Overflow, and it basically does what I want. The accepted answer's suggestion is to use ob_start(); to start getting output then use ob_get_clean(); to put the output into a variable. My issue now is that the PHP scripts I'm trying to capture output from need variables passed to them using HTTP Get. The access those variables like this:
$term = $_GET['term'];
And I don't want to change that, because although I'm going to access these PHP scripts' outputs from another PHP script, I'm also planning on accessing them from elsewhere. So, is there a way to fool these PHP scripts into accepting some arguments through Get, then capturing their outputs with the method suggested above?
You can $_GET variables from any php script if its set (use isset to check that). Then just cURL to such url's will work.
If you have changed the method to POST earlier, you can use CURLOPT_HTTPGET. See the curl_setopt functions page (http://www.php.net/manual/en/function.curl-setopt.php) for more details.
For a non-cURL method, use jQuery ajax. It is quite simple to use, just read the documentation here.
EDIT: This is what you wanted (haven't checked the code though)
<?php
function get_include_contents($filename, $get) {
if (is_file($filename)) {
ob_start();
$_GET = array();
while (list($key, $val) = each($get)) {
$_GET[$key]=$val;
}
include $filename;
return ob_get_clean();
}
return false;
}
$string = get_include_contents('somefile.php', array('param1'=>'x', 'param2'=>'y'));
?>
And I don't want to change that, because although I'm going to access these PHP scripts' outputs from another PHP script, I'm also planning on accessing them from elsewhere. So, is there a way to fool these PHP scripts into accepting some arguments through Get, then capturing their outputs with the method suggested above?
Your question is a bit unclear as to why you're using cURL in the first place. If your scripts are on the same server, you can simply set the correct $_GET variables and use:
<?php
ob_start( );
// include the file.
$_GET['foo'] = 'bar';
include 'the_file.php';
$output = ob_get_clean( );
If your scripts are located on another server, where include is not viable, you will always have to do a HTTP request to get their contents, regardless of whether your this with cURL or Ajax, or sockets for all I care.
well you can access a $_GET from any script loaded as long as its in the URI, the variable $term can be used in any script. You can also include the script.
When you include a script you can access some of its content after the include.

Categories