Comparison of Two Tags - Category Formula?

[newsweatpantswhodis]

Hi all - I have read what i think are the relevant parts of the help guide and still quite figure out the last step of what I'm trying to achieve (described below).  Can anyone point me in the right direction?

I just want a category (or a set of search results, or whatever the case may be) that includes photos in which gps timestamp is not equal to: {File.MD.XMP::exif\DateTimeOriginal\DateTimeOriginal\0|substr:0,10;format:YYYY:MM:DD hh:mm:ss}{File.MD.XMP::iptcCore\Scene\Scene\0} [i have time zone info in the scene field]. 

i've dropped the variable in as a filter in metadata mechanic, and it seems to work and properly filter down to the appropriate set of photos, but i'm trying to actually view that set of photos rather than process them like metadata mechanic does. 

any idea how to achieve it?  i recognize that this would be pretty slow, so maybe it's not possible via categories - but is there at least a way to build this set on an as-needed basis from within imatch?  as it stands, i've been running the appropriate -if command in exiftool, creating a txt file list, and then finding that same set of results in imatch.  hoping there's a way that cuts out the exiftool step.

[PandDLong]

I have a number of categories setup to do pretty much exactly what you want to do.

In brief, I use a data-driven category.  I set the 'Based On' indicator to 'Variable'.  Then in the Variable field I use {VariableA|is:(VariableB},,Mismatch, TRUE}.   I also set the 'Use other' to 'No' so I don't create a category of all those that are the same.

I always test my variable expression in VarToy and select a photo where these variables match and one where they don't to make sure I get the expected result  (I often find my first attempt has an error).  Then I cut and paste it into the data driven category variable field, run the Preview as a last check, and then I am done. 

I leave it in my category tree under a hierarchy of rarely used categories and collapse at the top-level to spare any performance hits unless I am actively looking at my "mismatches".

Good luck

Michael

[newsweatpantswhodis]

thanks @pandlong! i'm still getting my head aroun data-driven categories, but i think i get the gist - i'll give it a go this weekend.  appreciate it!