search

rankmath / seo-by-rank-math

Rank Math is a revolutionary WordPress SEO Plugin that combines the features of many SEO tools and lets you multiply your traffic in the easiest way possible :bulb: :chart_with_upwards_trend: →
https://rankmath.com
107 stars 52 forks source link

`og:image:alt` meta has changed from v1.0.99 to v1.0.100 #128

Closed strategio closed 1 year ago

strategio commented 2 years ago

Describe the bug og:image:alt meta has changed from v1.0.99 to v1.0.100

To Reproduce Steps to reproduce the behavior:

  1. Go to the Media library
  2. Add a new image with the title "My image title"
  3. Edit the image and set "Alternative Text" to "My alternative image content"
  4. Display the image page on the frontend
  5. Inspect the source

Here's the results: => v1.0.99: <meta property="og:image:alt" content="My alternative image content" /> => v1.0.100: <meta property="og:image:alt" content="My image title" />

Additional context We have automated end-to-end tests with WPML and this started to fail as soon as we updated to v1.0.100. If it was a bug that you corrected, we'll have to adjust our tests. Otherwise, if it's a regression on your side, we'll wait for your fix to restore our test.

WPML internal ref: wpmlwpseo-217

strategio commented 1 year ago

Hi @surajv, is this something you want to fix?

It's not impacting us directly but since it's failing on SEO By Rank Math side, we had to disable one assertion in our end-to-end tests. We were waiting for this to be fixed to restore our test assertion.

Now if you tell us it won't be fixed, we'll just close the pending ticket on our side (definitely remove the assertion).

surajv commented 1 year ago

Hi @strategio, sorry for not following up, issue is not happening for us.

In version v1.0.100, we have fixed the related issue, this filter (https://rankmath.com/kb/filters-hooks-api-developer/#change-opengraph-image) was not working. Please check if you are using this filter which is modifying the output.

strategio commented 1 year ago

Thanks @surajv, we've restored the failing assertion in our tests and we confirm it's passing again.