I Just Found a Simple WordPress Bug That’s Been Open/Unfixed for Eight Years
Eight years is a long time. It covers the period from birth to third grade, or an entire two-term US presidency. But, as I just learned, eight years has not yet been enough time to get a very simple fix to the way WordPress generates permalinks included into WordPress Core.
A few weeks ago, I noticed that if you write a slash character (
/) in a post title, the permalink that gets generated mushes together the words separated by the slash, rather than separating them with a – character as I’d expected. I’ve made this article an example: notice how the URL says
openunfixed rather than
open-unfixed, because of the slash.
In addition to being odd, this has some minor negative SEO consequences, since nobody’s doing a Google search for the non-word “openunfixed.” It’s also very easy to fix: a minor change to one line of code in
wp-includes/formatting.php. Putting all this together, I filed a Trac ticket outlining the problem and the fix earlier this afternoon.
Here’s where it gets weird: my Trac ticket was immediately marked as a duplicate of an eight-year-old ticket on the same topic that has still not been resolved. The fix in that ticket appears ready to go—there have even been unit tests written and run—but it’s not at all clear to the people who wish to move it forward what the next steps are to get it into Core.
I checked on that ticket’s most recent progress, and it turns out to have been two months ago, when a person asked the following in the WordPress #core Slack channel:
I’m trying to get bug #10792 fixed – I wrote a patch (twice, by now), but I don’t know how to get it actually included in a release. I have all the keywords set, but then… nothing. What more can I do?
In response, that person was given the following: a context-free link to the “Contribute” front page on wordpress.org.
I’m wondering what to do about this as a more general problem. The first thing that came to mind is a team of people who hunt down small, old, “ready-to-go” bugfixes and provide the guidance and willpower to shepherd them into Core. Maybe such a team exists—if so, it’s presumably swamped, but an eight-year (or three-year or whatever) wait time doesn’t seem like something the community should just cheerfully accept.
I won’t pretend to be an experienced enough WordPress contributor to have any firm answers. Thoughts?
Image credit: Henry Burrows