David Lingren

Thanks for your update. You wrote “I used an image optimizer to strip everything first.” Once the image is optimized, all of the metadata is lost. You can confirm that by going to the Edit Media screen and looking through the “Attachment File Metadata” text area.

You mentioned the empty image_meta array. Erasing it wouldn’t save you much – it’s better to just ignore it.

If the “Attachment File Metadata” text area shows that the images still contain information like description or copyright, I can help you map that to WordPress variables or Custom Field content. However, with the image optimizer that’s not very likely.

I am marking this topic resolved because I don’t think there’s more I can do, but please post an update if you need more help with this topic. Thanks for your interest in the plugin.

Thanks for your report. You have encountered an MLA defect I introduced in the 3.35 update.

I am curious about your use case. The JPG/PNG thumbnail generation was added to MLA before WP 4.7 introduced “WordPress thumbnails” for PDF documents. I would love to hear how you have found them useful more recently.

I have uploaded a new MLA Development Version dated 20260409 that corrects the problem. You can find step-by-step instructions for using the Development Version in this earlier topic:

How to download & install the current development version of MLA

Once the Development Version is installed you can retest the handling of JPG thumbnail images for PDF documents. The thumbnail fix is the only change from the current version, so you can use the Development Version with confidence.

The fix will be part of my next MLA version, but in the interim it would be great if you could install the Development Version and let me know if it works for you. Thanks for alerting me to this MLA defect and for your interest in the plugin.

I have released MLA v3.35, that contains the fix for the regular expression “named subpatterns” issue uncovered during my work on this topic.

Thanks for your kind words regarding MLA and for your interest in the plugin.

Thanks for your question, and for including the details on your mapping rule; very helpful.

I am happy that I have not been entirely replaced by AI chatbots. Here is a template that will give you better results:

Name: Instrument
Data Source: -- Template (see below) --
Meta/template: [+name_only,extract( '/(.*)_(.*)/' )+]([+matches:2+])
Existing text: Replace
Format: Native
Option: Text
Delete NULL Values: checked
Status: active

You must set the Data Source to -- Template (see below) -- so MLA will process the template. The template provided starts with the name_only data source (the name portion of the file name); more reliable than the WordPress post_title data source. The extract( '/(.*)_(.*)/' ) format/option suffix divides the name into two subpatterns: 1. title and 2. instrument, discarding the underscore that separates them. The extract suffix puts the subpatterns into an array of matches. The ([+matches:2+]) data source returns the second subpattern content, if present.

You can add a second rule to save the title portion of the file name by changing the last part of the template to ([+matches:1+]), the first subpattern.

You can find more information about MLA’s regular expression features in the “Regular Expression Features” section of the Settings/Media Library Assistant Documentation tab. Note, however, that the “named subpatterns” outlined in the documentation do not currently work in mapping rules. That is an MLA defect I will correct in my next update, and I am glad this topic helped me uncover the problem.

I hope that gives you a solution for your application. I am marking this topic resolved, but please update it if you have any problems or further questions regarding the above suggestions. Thanks for your interest in the plugin.

Thanks for taking the time to install the Development Version and confirming that fragments are now handled correctly.

You asked about the pagination parameter persisting when you navigate to a different tab on the page. I believe this is because the Ultimate Blocks tabbed content block is handling the tab switch using local JavaScript code; the page is not being reloaded from the server and so the URL is not changing. I believe both plugins are operating “as intended” and I regret that I don’t see how to make it “preventable”.

Regarding the CSS “active tab” issue, I can see it happening but without knowing more about the specific technique you are using I can’t be of much help. If you have more details I can investigate further.

I haven’t installed Firefox on my new system but I will do that shortly. I will also install Ultimate Blocks and put together a test page with multiple tabs. It would be best if you could contact me at my web site, since email will be more convenient. I can post a summary here when the issue is resolved. Thanks for your understanding and your patience.

Thanks for following up with the example files. You have identified an MLA defect introduced in the 3.31/3.32 updates.

I have corrected the problem and have released v3.34 with the fix. The new version correctly processes the examples you sent.

I am marking this topic resolved, but please update it if you have problems or further questions regarding the new version. Thanks for alerting me to this MLA defect and helping with my investigation and testing!

Thank you both for your reports. The vulnerability in question allows a logged in user to write JavaScript code that modifies taxonomy terms assigned to a Media Library item. It’s listed as “Medium” in Wordfence and “Low priority – No impactful threat” in Patchstack. I’m not sure why Wordfence-Scan shows it as “Critical”.

I will be updating MLA to version 3.34 later today, which should resolve the issue. I will post an update here after the release is completed.