Maintaining session between SSL and non-SSL pages in CakePHP

It’s funny, in a way, that cms34 has been around for nearly five years now, and it’s only just become a real issue that we were not maintaining sessions between SSL and non-SSL pages. This is somewhat understandable: practically speaking, the only time it really matters to carry over the session between the two is when a user is logged in. (This might not be the case with all web applications, but for us, at least, there’s rarely enough happening in the session when a user is not logged in for it to matter.)

As it happens, not that many cms34 sites use SSL; not that many cms34 sites use the user login capability on the front end. And very few use both. But we’ve had a couple of new sites come online lately that do use both, and it’s become a bit of an issue.

The issue was exacerbated by the fact that I recently modified the Users controller to require SSL on the login page, if the site has an SSL certificate. Consequently there were issues with trying to access login-restricted, but non-SSL pages… redirect loops and other such fun.

What’s the problem?

The problem is simple: for obvious security reasons, sessions and cookies cannot be shared directly between two different domains. It’s possible (although less secure) to share them between both SSL and non-SSL on the same domain, and it’s also relatively easy to set them up to work between different subdomains. But if your SSL pages use a different domain name than the non-SSL pages, even if they’re on the same server, there’s no way to get them to automatically use the same session.

The solution (though still not ideal, as it can introduce the risk of session hijacking), as you’ll find in lots of places, is to pass the session ID as a query string variable. Then you can use that to restore the same session ID, even if it’s on another domain — as long as it’s on the same physical server.

Some improvements

There are two key improvements I made to the basic “pass the session ID in the query string” scenario.

First, when the session is created I am writing the user’s IP address (using $_SERVER['REMOTE_ADDR']) as a session variable. Then, when I am attempting to restore the session with the ID passed as a query string variable, I read the session file on the server first, and make sure the IP address in the file matches still matches the user’s current IP address. Only then do I restore the session.

Second, and this is an aesthetic issue almost as much as a security one, once the session has been re-established, and before any response has been sent, I strip the session ID out of the requested URL and redirect to that new URL. It’s all invisible to the user, and the session ID never actually appears in the browser’s address bar.

A look at the code

There’s a lot going on in the cms34 code, much of which is highly specific to this application. But in short the keys to making this work happen in two places:

UsersController::login()

I have a login() action in UsersController that handles all of the special functionality that needs to happen when a user logs in. The actual login itself happens “automagically” via AuthComponent, but Auth doesn’t know everything I need to have happen when a user logs in, so after Auth does its work, my login() action takes it from there.

Honestly not a lot needs to happen here to make this work. Just two things: you have to write the user’s IP address to the session as I noted above, and you have to pass the session ID in a query string variable on the redirect that happens when login is successful. My code looks a little something like this (note that I have an array in the session called Misc that I use for… miscellaneous stuff like this):

class UsersController extends AppController {

  var $name = 'Users'
  // Other controller variables go here, of course.

  function login() {

    // All of this should only run if AuthComponent has already logged the user in.
    // Your session variable names may vary.
    if ($this->Session->read('Auth.User')) {

      // Various session prep stuff happens here.

      // Write IP address to session (used to verify user when restoring session).
      $this->Session->write('Misc.remote_addr(',$_SERVER['REMOTE_ADDR']);

      // Some conditionals for special redirects come here but we'll skip that.

      // Redirect user to home page, with session ID in query string.
      // Make up a query string variable that makes sense for you.
      $this->redirect('/?cms34sid=' . session_id());

    }
  }
}

So far, so good. The rest of the excitement happens in…

AppController::beforeFilter()

Ah yes, the magical beforeFilter() method. There’s a whole lot of stuff going on in AppController::beforeFilter() in cms34, most of which is highly specific to our application. But this is where you will need to put your code to retrieve the session ID from the query string and restore the session… this function runs at the beginning of every page load on your site.

I’ve put this logic almost at the beginning of beforeFilter(), because we really do want that session restored as soon as possible.

Here’s a look…

class AppController extends Controller {

  function beforeFilter() {

    // Additional code specific to your app will likely come before and after this.

    // Only run if session ID query string variable is passed and different from the existing ID.
    // Be sure to replace cms34sid with your actual query string variable name.
    if (!empty($this->params['url']['cms34sid']) && $this->params['url']['cms34sid'] != session_id()) {

      // Verify session file exists.
      // I am using CakeSession; your session files may be elsewhere.
      $session_file = TMP.DS.'sessions'.DS.'sess_'.$this->params['url']['cms34sid'];

      if (file_exists($session_file)) {
        $session_contents = file_get_contents($session_file);

        // Find user's IP address in session file data (to verify identity).
        // The CakePHP session file stores a set of serialized arrays; we're reading raw serialized data.
        // If you used a different session variable name than remote_addr, change it AND the 11 to its string length.
        $session_match = 's:11:"remote_addr";s:'.strlen($_SERVER['REMOTE_ADDR']).':"'.$_SERVER['REMOTE_ADDR'] .'";';

        // User's IP address is in session file; so we can continue.
        if (strpos($session_contents,$session_match) !== false) {

          // Set session ID to restore session
          $this->Session->id($this->params['url']['cms34sid']);

          // Redirect to this same URL without session ID in query string
          $current_url = rtrim(preg_replace('/cms34sid=[^&]+[&]?/','',current_url()),'?&');
          $this->redirect($current_url);
        }
      }
    }
  }
}

A few final thoughts

I didn’t really discuss cookies at all here, but suffice to say there’s a cookie containing the session ID that gets written. If you’re only using cookies for the session ID (which is probably a good idea), then you don’t really need to do anything else with them. But if you’re writing certain cookies when a user logs in (like I do), you’ll need to write additional logic to restore them in AppController::beforeFilter(). In my case, the relevant cookies are all duplicates of session data, but are intended for a few edge cases where I need to access that information through JavaScript or in .htaccess files that are protecting login-restricted downloadable files — in other words, places where I can’t use PHP to look at session data.

You may also notice near the end of the code in the AppController::beforeFilter() example above that I am calling a function called current_url(). This is not a built-in part of PHP or CakePHP; it’s a simple little function I have in my config/functions.php file. Here it is:

function current_url() {
  return (!empty($_SERVER['HTTPS']) ? 'https://' : 'http://') . $_SERVER['SERVER_NAME'] . $_SERVER['REQUEST_URI'];
}

How to keep Internet Explorer from displaying your site in “Compatibility View”

Fortunately this is an edge case for me, but… there’s a time for everything. And that time was today.

If you build websites using modern, open standards-based techniques, the last thing you want is to have your Internet Explorer-using visitors see your site in the dreaded “compatibility view”. This monstrosity was created by Microsoft with the advent of IE8, to allow sites (mostly corporate intranets) that require the quirks of IE6 to still look like they’re supposed to when viewed in IE8.

The problem is, IE6 sucks, and it makes modern sites look terrible. So if your users accidentally view your site in compatibility view, it’s going to suck too.

The good news is, it’s easy to force IE8 and later to display your site the way you intended. You can disable compatibility view with this simple meta tag:

<meta http-equiv="X-UA-Compatible" content="IE=9; IE=8; IE=7; IE=EDGE" />

(Source: http://stackoverflow.com/a/10365780)

How not to update a WordPress plugin

The meaning of this post’s title is twofold: 1) how, as a WordPress user, to avoid having a plugin show up in the built-in updater; and 2) how, as a developer, you probably should not approach releasing a major update to your plugin that is incompatible with earlier versions.

The scapegoat here is Elliot Condon’s excellent Advanced Custom Fields, which has become one of my essential go-to plugins for building highly customized WordPress websites for my clients. I don’t mean to pick on Elliot Condon. He’s clearly a tremendously talented developer and I have a ton of respect for his work and what it has allowed me to do in my work.

But I do feel that he handled the 4.0 release of Advanced Custom Fields poorly. I’m not sure the fault is really his, however, as it is just as much or more the fault of how the built-in updater in WordPress works. (Especially since at least some of the changes he made in version 4 were done solely to conform to changes in the official WordPress plugin requirements.)

Here’s the problem: WordPress has a central Plugin Directory that makes it easy to install and update approved plugins. Perhaps too easy. Because if an update of a plugin is made available, and it’s newer than the version you have installed, it appears as an option in the updater. Yes, there are links to information about the update, which you really should read before doing anything else, but it’s all too easy with a few clicks to just run the update and move on.

Most of the time, it just works. Which can be dangerous. Because users — even “experts” like me — come to assume it will always “just work.”

As it happens, the version 4.0 release of Advanced Custom Fields was made available while I was on a weeklong vacation out west with my family, with only my iPad (and SLP’s MacBook Air) along for the ride. And, based on my own recommendations, a client with a relatively newly-launched website went ahead and ran the update. This particular site is heavily dependent on ACF, and the update broke it.

The official ACF website offers a migration guide that makes it (somewhat) easy to convert your existing version 3 implementation to work with version 4. But it’s not a “click-it-and-you’re-done” kind of process. It takes time, and you need to know what you’re doing. Which is nothing like the general experience of using the built-in WordPress updater.

I was able to temporarily solve the problem for this website by rolling back to an earlier version (which required some hunting to locate, given the limited Internet access I had with hotel WiFi). Which leads to my first tip:

How to prevent a WordPress plugin from updating

How does WordPress know when a plugin has an update available? Well, it’s easy for it to check the Plugin Directory to see what the latest version of any given plugin is. But how does it know which version you have installed? Also easy. Each plugin’s main PHP file includes a comment block at the top, with information WordPress parses both to display in the Plugins area of the admin interface and for added functionality (e.g. knowing when there’s an update available). For Advanced Custom Fields, the comment block looks like this:

/*
Plugin Name: Advanced Custom Fields
Plugin URI: http://www.advancedcustomfields.com/
Description: Fully customise WordPress edit screens with powerful fields. Boasting a professional interface and a powerfull API, it’s a must have for any web developer working with WordPress. Field types include: Wysiwyg, text, textarea, image, file, select, checkbox, page link, post object, date picker, color picker, repeater, flexible content, gallery and more!
Version: 3.5.8.2
Author: Elliot Condon
Author URI: http://www.elliotcondon.com/
License: GPL
Copyright: Elliot Condon
*/

This is essentially a set of key/value pairs. Version is what WordPress reads to see which version you’re running, and compares it against the master version in the Plugin Directory to see if updates are available.

So how do you keep it from running the update? Simple: Change the version number to something higher. I like to prepend it with “999” so I still know what the “real” version number is that I’m running, like this: 999.3.5.8.2. This way the WordPress updater thinks the major version is “999”, which is almost certainly higher than whatever the real current version is.

Simply change this version number, save the PHP file on your server, and you’re done. The updater will never trigger for this plugin (as long as its real version number is less than 999). You may also want to update Description with an explanation of what you’ve done.

Bear in mind this is a temporary solution. You really should do whatever you need to do to get your site compatible with the latest version, then restore the original version number and let the updater do its magic.

How, as a developer, not to create this mess for your users

I have submitted a few, very simple, plugins and themes to the official WordPress repositories, so I have a bit of experience with this, but I’m no expert on the process. However, what is clear to me is that if you submit changes to a plugin as an update, the built-in updater will pick it up and make it available to any users who have an older version installed. This is dangerous. If your new version includes such radical changes as to make it incompatible with earlier versions, you have to assume that most users will not read your notes, and will believe they can just run the updater with no problems… especially if you’ve made a habit of releasing frequent incremental updates to the plugin in the past that “just worked.”

The only real solution I see to this is to submit the new, incompatible version to the Plugin Directory as an altogether new plugin, instead of an update to the existing plugin. The risk here is that you lose visibility. Your download count and ratings/reviews reset to zero, and anyone who’s using an older version may never know about the new version. So, it’s bad for marketing.

But an incompatible update breaking sites for unsuspecting users is bad for marketing too. It’s going to cause your ratings to take a hit, cause a lot of bad publicity, and turn off your loyal users. The migration is going to be work for them anyway; making them do it after they clean up a mess created when they unsuspectingly ran the update is even worse.

A few specifics as pertains to Advanced Custom Fields

Again, I don’t mean to pick on Elliot Condon for his work with Advanced Custom Fields. I will continue to be a loyal (and, yes, paid) user of this plugin. It’s brilliant. Nonetheless, it’s created a hassle for me this week. After returning from my trip, the first work-related thing I did was go through every client site that’s running ACF and apply my “version 999” trick so those clients won’t run the updater until I have a chance to migrate their sites to version 4.

The biggest challenge I had in rolling back was simply getting my hands on the old version. Sure, I had it on some of my other sites, but I access almost every client site via SSH/SFTP, and the hotel I was staying at had port 22 blocked. Luckily the site I needed to fix was one of the few I access via regular FTP, and port 21 was open. But I still needed to get a copy of version 3.x to reinstall on that client’s site.

As I found along the way, the WordPress Plugin Directory hides old versions under the “Developers” tab, where every previous version can be downloaded as a ZIP file or checked out with Subversion. Previous versions of ACF are available here.

Where’s the option to change the “uploads” path in WordPress 3.5?

Thanks for nothing, WordPress.

It’s not often that I complain about anything in WordPress, much less get genuinely angry about something I think is downright stupid, but today it happened.

I build a lot of WordPress sites for clients, and I frequently post them in a staging area on my server. Rather than having dozens of separate installations of WordPress filling up my server’s disk space, I set up a single installation (but with a separate database for each site), with some simple customizations to the wp-config.php file that tell it which database to use based on the domain name.

Please don’t bother to mention that I should “probably” be running WordPress-MU (or MultiSite or whatever they call it now). Ultimately these sites are going to be hosted elsewhere as standalone WordPress sites so I need to keep their databases separate.

One key to making this arrangement work nicely has always been the simple setting in WordPress that allows you to define a special directory for uploads, instead of dumping them all in the main uploads directory. By setting this to a subdirectory under wp-content/uploads named to match the site’s theme folder name, it was easy to keep everything separate.

It’s been a while since I’ve set up a new site on the staging server, at least since before I updated to WordPress 3.5. But I was doing one today, and I had to do a sanity check as I scoured the settings to try to find where in the hell you set the custom uploads path. I was sure it was under Settings → Media but I couldn’t find it there.

Because they took it out.

Yes, they took it out. Why? Well, apparently there’s a rational explanation, but frankly no explanation would satisfy me because it seems incredibly disruptive to take something like this out, even if it was a bad idea to implement it as it was in the first place.

Fortunately, there’s still a way to do it. You just can’t do it in the admin interface anymore; you need to edit the wp-config.php file directly to add a constant. No problem, I’m already doing that anyway.

Here’s what you need:

define('UPLOADS', '/your/custom/path/under/wordpress/root');

If you’re just putting your uploads into a subdirectory under the main uploads directory, like I am, it would look like this:

define('UPLOADS', '/wp-content/uploads/yourdirectory');

Just remember, if you’re doing this because you’re running multiple sites from a single installation like I am, this constant should be set in the same conditional where you’re setting which database the site should use.

Easy. Not as easy as it was, nor as easy as I think it should be, but…

TinyMCE and the non-breaking space problem

Let’s get right to it then: TinyMCE is great, but I am annoyed by its willingness to take users’ multiple spaces literally! Collapsing multiple spaces is a basic characteristic of HTML, and allowing users to carelessly (or intentionally, which they still shouldn’t do) insert multiple spaces by converting every other one of those spaces into the &nbsp; (non-breaking space) character is BAD!

IMHO.

Anyway… start with a pinch of Stack Overflow, add a dash of the official TinyMCE documentation, along with a heaping tablespoon of reading between the lines, and I have a working solution to the problem. My installation of TinyMCE now automatically converts any &nbsp; characters in the text back into regular ol’ spaces.

It’s a bit draconian; after all there are legitimate uses for non-breaking spaces. But 95% of the times they’re inserted by TinyMCE are user accidents, and another 4.9% of those times are abuses like faked “tabs” that would be solved better by another approach altogether. (There are reasonable CSS-based solutions that work in some cases, but let’s talk HTML’s need for tabs another time.)

Anyway… here’s the gist of the solution. You need to create a callback function. Here’s mine:

function my_cleanup_callback(type,value) {
  switch (type) {
    case 'get_from_editor':
      // Remove &nbsp; characters
      value = value.replace(/&nbsp;/ig, ' ');
      break;
    case 'insert_to_editor':
    case 'submit_content':
    case 'get_from_editor_dom':
    case 'insert_to_editor_dom':
    case 'setup_content_dom':
    case 'submit_content_dom':
    default:
      break;
  }
  return value;
}

It may look like there’s a lot of extra stuff in here you don’t need; I included all possible values for type inside the switch to be prepared for the future. You do want to check for type == 'get_from_editor' though; otherwise your replace() is going to run under way too many conditions and may cause weird behavior like new paragraphs appearing when you just want to insert new text into an existing one, or browser-generated warnings about leaving the page when you try to save. (I ran into both as I was fine-tuning this.)

Now that you have your callback function, you just need to… you know… call it. That’s done inside tinyMCE.init(). You’ll need to include this line somewhere:

cleanup_callback: 'my_cleanup_callback',

Be sure to check if cleanup_callback is already declared somewhere, and also don’t forget the comma at the end, unless you’re inserting this as the last line.

Once you’ve got it all rolled out to your site, you’ll need to clear your cache. I’ve found TinyMCE’s configuration files can be annoyingly persistent in the browser cache.

Yes… you have correctly observed that I had to use non-breaking spaces myself in this post, to get the indents in the code samples to show. Pay no attention to the man behind the curtain. And remember my complaint about the lack of tab characters in HTML. Another day.