{File.Persons.Tag|count:true;cast:int} doesn't return full range of values

[cg]

I set up a "Number of People" data-driven category using {File.Persons.Tag|count:true;cast:int} referenced in this post: https://www.photools.com/community/index.php?msg=76961

It worked great in iMatch 2021. However in iMatch 2023, the category only returns values of 0 and 1, with 1 including all files with 1+ faces.

Has something changed with variables or syntax?

Thanks!

[axel.hennig]

I was not able to reproduce this.

I did set-up the following data-driven category (I set "Name for 'Other'" to "0"):


And got the following result:


So I've got 1, 2 and also 3 (and also Other = 0) which is all correct based on the FaceRecognition result for this test-DB.

[cg]

I had the Data type set to "Integer" (which I take it means it doesn't need the 0 specified for Other). I changed it to Automatic/0 and got the same result of just 0 and 1. This happened for my primary database, as well as for a smaller test database (both Automatic/0 and Integer settings).

Another thing I have noticed (which may be unrelated) is that the shift-F5 shortcut did nothing in the category window. There was no response at all from iMatch to that shortcut. I was only able to update this data-driven category with the contextual menu. And even then, it would only update with the Viewer window closed, or sometimes only upon restarting iMatch completely.

[Mario]

Check your number format settings in Windows.
Does your system use the same separator for lists and 1000 groups, e.g. a comma?
Then see this thread for an answer https://www.photools.com/community/index.php/topic,13272

[cg]

Hi, Mario.

Indeed the default list separator was a , and after changing it to ; the data-driven category appears to return the expected values. I'm pretty sure I had never changed that setting in Windows. I'm wondering now what other applications, if any, use that setting and might behave differently with that setting change, but I'm certainly glad that it's working in iMatch again.

Thank you!