Flush Opcache with Varnish: a WordPress Plugin Announcement

Flush Opcache with Varnish is here to rescue you from the constant annoyance of flushing yet another cache!

If, like me, you use PHP’s Opcache to speed up a site, and you have a Varnish cache, and maybe some other server-side caches, you probably want to flush the opcache, Varnish cache, and any other server-side caches at the same time after changing the PHP files on your server, (like, when you update WordPress, plugins, or themes). I don’t want to click a button for each cache type, or worse, have to log in to the command-line for a sudo service restart varnish after doing updates, especially if I’m doing repeated updates. Flush Opcache with Varnish hooks into Mika Epstein’s excellent Varnish HTTP Cache plugin and flushes the PHP Opcache and the WP Super Cache cache, (if you use WP Super Cache), every time you manually flush the Varnish cache, turning the Varnish HTTP Cache “Clear Cache” buttons into a three-for-one deal.

If you also use the plugin WP Opcache to manage your opcache then WP Opcache will be called to do the actual Opcache flushing so you can take advantage of its automatic rebuilding of the opcache.

I’m already using Flush Opcache with Varnish in production and it’s saving me time every time I update a theme or plugin. You can get it in the WordPress.org plugin directory.

Why X Theme and Pro sometimes cause PHP warnings on WPEngine

The Problem

Earlier this fall I was working on an e-commerce site that’s hosted on WPEngine that used Themeco’s Pro as its parent theme most pages hadn’t seen any customization yet, most of the code was stock, and this appeared on the Products page:

PHP Warning:  Invalid argument supplied for foreach() in wp-content/mu-plugins/wpengine-common/plugin.php

Hmm. Who is passing what to a foreach?

It turns out that both X Theme and it’s progeny Pro do this turn off WordPress’s automatic responsive images:

// Disable WordPress 4.4 Responsive Images
// =============================================================================

if ( ! function_exists( 'x_disable_wp_image_srcset' ) ) :
  function x_disable_wp_image_srcset( $source ) {
    return false;
  add_filter( 'wp_calculate_image_srcset', 'x_disable_wp_image_srcset' );

They’re hooking into the wp_calculate_image_srcset filter and returning false, which stops WordPress from adding a srcset attribute to an image, (more on why in a moment). The problem here is that the value being passed to the filter, $sources, is an array, but x_disable_wp_image_srcset() turns that array into a boolean false, so then anything else hooking into wp_calculate_image_srcset, for example, WPEngine’s CDN system, tries to iterate over the $sources array, but it’s not an array, hence the error. By filtering $sources to a boolean false Themeco is breaking the expected, and documented, behaviour of the wp_calculate_image_srcset filter.


Luckily, there are two easy fixes for this problem. The first is to make sure you’re using the latest version of WPEngine’s must-use plugin. They’ve caught on to the problem and check to make sure $sources is actually an array:

if ( is_array( $sources ) ) {
    foreach ( $sources as $source ) {
        // do the stuff

If you can’t do that, at least Themeco made the function pluggable, so you can add your own version of x_disable_wp_image_srcset() to your theme’s functions.php or a plugin that returns an empty array and plays nicely with other plugins:

// This is a pluggable function. If we don't define it here then Pro does in 
// pro/framework/functions/global/admin/thumbnails/setup.php
function x_disable_wp_image_srcset( $sources, $size_array, $image_src, $image_meta, $attachment_id ) {
    return [];

Or, you could use responsive images.

Confusion, (aka, why this happens in the first place)

When I first discovered this problem I E-mailed Themeco to tell them about the problem, however they pointed me at the documentation for the wp_calculate_image_srcset() function, which states the function can return false, and sent me on my way:

Hey John,

Thanks for writing in. In looking over WordPress' official documentation for that function/hook, I believe that boolean false should be the correct value to return:


In the "Return" section you'll see that it is supposed to return a string or a boolean false if there is an "error or only one source exists." It seems as though WPEngine should need to revise their function to account for these instances anyway even if not for X or Pro (and at least work to catch any fatal errors like this).

Hopefully that helps, cheers.

Best Regards,
Your Themeco Team

This confusion occurs because the convention in WordPress is that if a filter and a function have the same name, the filter filters the output of the function. However, in this case the convention is broken – when I raised the issue in the Advanced WordPress Facebook group a very experienced WP developer made the exact same mistake and assumed that allowable return types for the function should be fine to return from the filter. After we cleared up the confusion I filed a bug about the confusing names, go star it so it gets some attention!

A Fluke of PHP

So why does returning false in the wp_calculate_image_srcset filter stop WordPress from using responsive images? Here’s the relevant code from media.php:

$sources = apply_filters( 'wp_calculate_image_srcset', $sources, $size_array, $image_src, $image_meta, $attachment_id );

// Only return a 'srcset' value if there is more than one source.
if ( ! $src_matched || count( $sources ) < 2 ) {
    return false;

To follow along, if $sources is turned into a boolean false at the apply_filters() call, then a count( false ) happens in the condition on the next non-comment line. Instead of an error like I expected, running count() on any non-array value returns 1, so count( false ) == 1, also count( 0 ) == 1


First, be a good citizen of the WordPress ecosystem and don’t change the type of a filtered value to something other than the documented types for that value. Second, if you have any non-core themes or plugins installed you can’t trust that a value coming into a filter will be the type that the docs say, so check! When WPEngine realized there was a problem an is_array() fixed the problem. Third, read support E-mails thoroughly and make sure you understand them, it hurts when I go to the effort to document a problem and come up with a solution, and I’m told it’s not a problem. Finally, go +1 my ticket so maybe others won’t get bitten.

PhpStorm Review

Late last year I kept hearing a lot about PhpStorm 8, especially from WordPress people and wanted to try it out but the price was stopping me, (it turns out the price is probably worth paying – but it’s hard to know if the price is worth paying until you’ve already paid the price). Luckily for me WP e-Commerce has some licenses for core contributors, and apparently I count, (JetBrains, the people who make PhpStorm, provide licenses to open-source projects at no charge – thanks!).

One of the things that JetBrains asked in return for the licenses was that we try to write reviews of our experience, and after many months of using PhpStorm, here are my thoughts.

PhpStorm crashed several times in the 24 hours after I installed it, not a good start. Support suggested installing the EAP version, which is like a beta stream, which fixed the problem. Since then I’ve switched back & forth between the current release and EAP versions of PhpStorm, but I usually use the EAP, because I like shiny things.


I had heard that PhpStorm was fast, but it’s a Java app, (like Netbeans or Eclipse), so I had my doubts. When first installed there were some pretty big performance issues, especially related to scrolling. PhpStorm was exhibiting what I refer to as “that Java scroll lag” – one of the reasons I tend to avoid Java apps. There was also a problem that when scrolling with the mouse the text would disappear, which makes mouse scrolling nearly useless. Eventually I learned that there’s a version of PhpStorm for OS X that comes with a bundled, tweaked, Java Run-time Environment, and using that version of PhpStorm solved these problems.

When comparing PhpStorm’s performance to Sublime Text, (2 or 3), Sublime Text is still way faster. PhpStorm feels like a normal app, but ST3 has that amazing snappiness that Sublime Text users have come to love, and want to see everywhere. While PhpStorm’s performance is satisfactory, it’s not as blazing fast as the chatter led me to believe, but the combination of decent speed, and the other benefits that of using a real IDE, mean it helps me write cleaner code faster.


Like most coding programs, especially Java ones, PhpStorm hurt my eyes out of the box, but the look is tweakable enough to make it look how I want. There are two built-in themes, (“Default” and “Darcula”), and you can load 3rd-party colour schemes for the text editor. I think it might be possible to load 3rd-party themes as well but I haven’t investigated this. Switching to the Darcula theme, and switching to the Predawn colour scheme, (I can’t remember where I found the PhpStorm port, maybe it even came with it), and Inconsolata made PhpStorm feel like home.

Coming from ST3 I re-mapped some keyboard shortcuts to match ST3 make the transition easier and make it easier to switch between the two when needed.

I had to download extra drivers to connect to MySQL, which was kind of annoying, but I suspect might be license-related.

And Java. PhpStorm 8 requires Java 6, which is no longer installed on Macs because it’s so old, that’s why there’s a version offered for OS X with a bundled JRE. It’s possible to install Java 6 with a package from Apple, because we all love having several versions of Java installed, or it’s possible to tweak some plist files to let PhpStorm run with Java 8, (the current version). Fortunately, it looks like PhpStorm 9, (which I’m running the EAP of now), will do away with this requirement.

Code Inspection

This is an IDE, so it’s aware of any databases that get used. This is great most of the time, (the code inspector tells you if you mis-type a column name, for example), but the it also fusses when I build big SQL queries by concatenating strings. So far I’ve just ignored these complaints.

If I include a file like this in WP: include( get_stylesheet_directory() . ‘/myfile.php’); PhpStorm thinks that myfile.php doesn’t exist, even if it does. Since this is the “right way” to include files in WP themes, and PhpStorm advertises built-in WP knowledge, PhpStorm should understand it.

When making a commit to your preferred Source-Control system, (Git, right?), PhpStorm inspects the code you’re committing, and if there are errors or warnings it pops up a window saying so, and gives you a chance to fix these problems. This is a great way to slowly clean up the code of older projects with lots of legacy, (read: “written before I knew better”), code. However, it’s also kind of annoying when unfixable errors or warnings are included, like when I’ve concatenated together an SQL query.

I was working on a plugin that works with WooCommerce and PhpStorm complained that I was assigning the return value of a function that returns void to a variable. It turns out that WooCommerce has several PHP DocBlocks that state a function returns void when it actually returns a value. Thanks to a WP e-Commerce copy of PhpStorm for pointing out errors in WooCommerce!


If someone from Jetbrains reads this, consider these bug reports!


Sometimes the fonts go bonkers. When I was first setting PhpStorm to use Inconsolata they did, (that was using the release version), and a few times on the EAP I committed some code, and when the commit finished the display went crazy again.

PHPStorm Power Consumption

Power consumption – I’m not sure how it compares to Sublime Text, but it’s consuming more power than Photoshop today – #2 on the list behind Safari with 11 tabs open.


The first time I used the “Pull Up” refactor tool, which lets you pull methods up from a class into the parent class, I pulled up a bunch of static methods. PhpStorm didn’t actually go and find the places where the methods were called and change the calls, which I thought was the point of the “smart” IDE. Inheritance would mean that these methods would still be available, but there’s no reason to load the child class if everything I need is in the parent.

The auto-complete can be really bizarre. Often doesn’t pick the logical thing, or even the first thing in the suggestion list. For example, if I type “col” then press tab or enter autocomplete completes as “columns” not “color,” even though color is the first suggestion. It even insists on completing as “columns” if I type “colo” and press tab/enter. This drives me mad.

Search for any file, (I’ve got it mapped to Cmd+P to match Sublime Text): If a project has several files with the same name there doesn’t seem to be mapping that makes sense. To me it would make sense for the first option to be the one I used the most recently, or the one “closest” to whatever file I’m already looking at in the directory structure. If the files are sorted at the moment, it looks like it’s alphabetical, which should change.

Auto-change formatting: I try to use WordPress style in most of my PHP & Javascript – especially for WordPress-related work! WP style suggests putting a space around function parameters in both Javascript and PHP. However, if I’m doing chained calls in Javascript PhpStorm removes the space after the last function argument when I press the . key. For example:

$( 'body' )
// becomes
$( 'body').
// as soon as the period is typed

When PhpStorm 8.0.3 came out I started getting SSL error messages on startup. Apparently there’s a problem in some SSL-related part of the bundled JRE. 8.0.3 came out on February 13 and there’s still no fix. The current EAP doesn’t have this problem.

PhpStorm’s licensing scheme for open-source projects has changed this year, so I need a Jetbrains account, the licence is issued from WP e-Commerce to that account, then I’m supposed to sign in to the account with PhpStorm and it’ll activate. Unfortunately the SSL bug prevents PhpStorm from talking to the Jetbrains servers, so I can’t activate my PhpStorm license. Luckily the EAP comes with its own license, otherwise I would be out of luck. A bug that prevents customers from activating software is a very big bug.


PhpStorm is good, very good at some things, especially the IDE-type functionality I was looking for. Integrated code inspections & linting were a welcome surprise that is really helping my code look better and encouraging me to write better in-line documentation. There have been problems, though, that might have caused me to abandon PhpStorm I hadn’t pre-decided to give it a real chance. I nearly always use the EAP both because I like shiny things, and because the bugs in the release version of the software are annoying, or serious, enough to drive me into the arms of the EAP.

So, is PhpStorm worth the purchase price? Probably. It handles code completion, data-source awareness, most refactoring, and all its IDE-related functionality quickly and as promised, (or very nearly so). The problems tend to be in other areas, and from the state of the current PhpStorm 9 EAP, most of these problems will be solved soon.

Xdebug is not friends with WordPress + create_function()

If you use Xdebug for debugging WordPress-based sites there’s something you should be aware of. If function created with PHP’s create_function() are hooked to WordPress actions or filters, any time $wp_filters is in the scope Xdebug sends invalid XML to the Xdebug client, (like your IDE). If the Xdebug client doesn’t deal with the invalid characters before attempting to parse the XML it will fail. IDEs deal with the parsing failure in different ways, SublimeTextXdebug doesn’t show the call stack or list of current variables but doesn’t crash entirely so debugging can continue. I don’t know how other IDEs handle the failure.

Why the XML is Invalid

create_function() makes a new function, gives it a random name that starts with a null character, and returns that function name as a string. When that string is passed to WordPress’s add_filter() or add_action() functions the name that create_function() returned is used as an array key in deep within the $wp_filters global. When Xdebug sends the list of variables back to an Xdebug client it is sent as XML and the null character encoded as &#0; which is illegal in XML.

How to solve the Problem

Ideally Xdebug would only send valid XML to the Xdebug client, but the bug report has been open for a year and doesn’t seem to be high priority so developers should solve the problem by avoiding create_function()

Avoiding create_function() is Good Coding

create_function() eval()‘s the code that’s passed as the second argument, and eval() is something to be avoided, so create_function() is too. This Xdebug bug makes us avoid create_function() for actions and filters in order to keep Xdebug useful, with the side effect of making our code a bit more secure.

Adding to a WordPress page’s URL without changing the URL of its Child Pages.

I was recently asked to add an extra path component to the URL of a WordPress page for SEO reasons. It took some diving into the internals of WordPress. Here’s the setup:

  • WordPress is installed in the site root, at example.com
  • We are using a static homepage
  • The company blog is at example.com/blog/
  • I was asked to make the blog URL be example.com/blog/some-extra-keywords/ without changing the URL of the current and future blog posts, which are currently example.com/blog/post-name/

There are 3 things required to make this work:

  1. A filter to tell WordPress to process requests for /blog/some-extra-keywords/ as if they are for /blog/
  2. A filter to change permalinks from /blog/ to /blog/some-extra-keywords/ (Optional, but recommended).
  3. A redirect from /blog/ to /blog/some-extra-keywords/ (Also optional, but also recommended).

Treat /blog/some-extra-keywords/ requests like /blog/

We don’t want to change the URL of the blog page in WordPress’s Edit Page screen because that will change the URL of past and future all blog posts. We need to tell WordPress that a request for /blog/some-extra-keywords/ is a request for /blog/. This is done by adding a filter to WordPress’s request filter hook:

 * Make the URL /blog/some-extra-keywords/ respond as /blog/ 
 * would.
 * @param  array $request The request array that WP generates.
 * @return array          The modified request array.
function jb_filter_blog_url_request( $request ) {

    if ( isset( $request['name'] ) && 'some-extra-keywords' == $request['name'] ) {
                // it's a page we're requesting. If you're doing this to something other
                // than a page, try setting $request['name'], and don't unset it 2 lines
                // later
        $request['pagename'] = 'blog';
        unset( $request['name'] );

    return $request;
add_filter( 'request', 'jb_filter_blog_url_request' );

Since WordPress identifies pieces of content by the last part of the URL, (the “name”), that’s all we have to change. In this case, because the blog is on a page, we had to set $request['pagename'].

Note: This is optional, but recommended. If you don’t do this you must do the next item, (redirect /blog/ to /blog/some-extra-keywords/), but you get the most SEO benefit if you do both.

With #1 it’s time to change links to the blog page to point to the new URL. The old URL still works, and you’ll be redirecting it in Step 3, but modifying the links cuts out a round-trip to the server, making your site faster, and it means that anytime a spider crawls the site it will find the correct URL.

Again, we create a filter function and hook it up. Because I’m doing this all to a page, I’m hooking on to the page_link filter, but the function also works with the post_link filter. In fact, I originally used the post_link filter, which doesn’t work for pages, and spent a while banging my head against a wall trying to figure out why it didn’t work.

By hooking into page_link or post_link you’ll modify the URL everywhere it is generated with get_permalink(), which is almost everywhere. Menus, lists of pages, and even the XML Sitemap that Yoast’s WordPress SEO plugin makes will be have the new URL. If you have typed the URL somewhere this filter will not change it, but that’s why you’ll set up a redirect in Step 3.

The code:

 * Modifies the blog URL when it's requested using get_permalink(). 
 * Note that this filter is set up to work on both posts & pages, and the 
 * post_link and page_link filters don't pass quite the same parameters:
 * - $post is an post object for post_link and a post ID for page_link, so we
 *   get the post object if $post is numeric
 * - $dontuse is completely different between the two filters, but not needed,
 *   so we ignore it.
 * @param  string $url       The URL to be filtered.
 * @param  mixed  $post      The post ID or post object that corresponds to $url
 * @param  mixed  $leavename Differs depending on filter. Ignore.
 * @return string            The possibly-modified URL.
function jb_modify_blog_url( $url, $post, $leavename ) {
    $true = false;
        // don't do it in the admin, I'm afraid the modified URL will get
        // added to the URL slug field on the Edit Page screen, and get
        // permanently added, with another copy of it being added every time
        // the page is saved.
    if( ! is_admin() ) {

        if ( is_numeric( $post ) ) {
            $post = get_post( $post );

        if( 'blog' == $post->post_name ) {
            $url .= 'some-extra-keywords/';
    return $url;
add_filter( 'page_link', 'jb_modify_blog_url', 10, 3 );

The magic is in the third if condition: If the page name is ‘blog’ add the extra keywords to the URL.

Redirect from /blog/ to /blog/some-extra-keywords/

Note: This is optional, as well, but if you don’t do it the blog homepage will be available at both /blog and /blog/some-extra-keywords/ which could lead to a duplicate content penalty from Google.

We’re going to do a 301 Redirect to tell search engines that the blog is now at /blog/some-extra-keywords/. I’m going to do this in a .htaccess file because that way the web server doesn’t have to start PHP or WordPress.

If you’ve already got a rewrite section in your .htaccess file, add the following line to it:

RewriteRule ^blog/?$ /blog/some-extra-keywords/ [R=301,L]

This will redirect both /blog/ and /blog, (the first ? makes the slash right before it optional), to /blog/some-extra-keywords/. If you don’t have a rewrite section, add one:

<IfModule mod_rewrite.c>
RewriteEngine On
RewriteBase /

RewriteRule ^blog/?$ /blog/whole-house-fan-energy-saving-tips/ [R=301,L]


And that’s it.