Recommendations for tagging cloud-uploaded photos?

Started by jch2103, September 13, 2013, 09:37:15 PM

Previous topic - Next topic

jch2103

I've uploaded quite a few photos to the cloud over the years (in my case, to Smugmug). Many of these are scanned family photos, not in any particular order.

I've belatedly realized that it would have been a good idea to have added a 'Smugmug' keyword (or perhaps Attribute) to my IM database to keep track of which images I've uploaded (and which ones I still need to upload). I can download a list of galleries and download images from Smugmug, but I'm looking for an efficient way to load that information into IMatch.

The brute-force approach would be to make a list of the cloud photos and then update my IM folders and images individually by hand. Tedious at best. I could also download a reduced-size set of the cloud images and load them into my IMatch database, but aside from searching for individual visually similar images I don't see an easier way to tag the originals in IM.

In theory, I could load the information into IM through an import of Categories or Attributes (via manipulating a text list of previously uploaded photos), but I'd somehow need to transform the text list into an xml-based .imcsx or .imas file.

Does anyone have a recommendation on the best way to approach this problem? I suspect I'm not alone...

John

Mario

If your files have a unique identification attribute, it can be done with a small script.
E.g. if you use unique file names, or the metadata of your uploaded files contains an XMP document id that has not been changed, or the uploaded file is unmodified and you can get the 32-Bit CRC checksum...

Else it will most likely not be possible to automate this.
-- Mario
IMatch Developer
Forum Administrator
http://www.photools.com  -  Contact & Support - Follow me on 𝕏 - Like photools.com on Facebook

jch2103

I had in mind being able to some kind of SQL database-style join, using file names in a text file joined with file names in the IM database.

Is this at all feasible?
John

Mario

No. The IMatch database is opaque and not accessible from the outside. IMatch uses a object-oriented storage schema based on a relation standard product. There is no table which contains 'file names'.

You will reach your goal much easier with a script. You can iterate all files in the database easily (there are many sample scripts which show how to do that). Produce a text or XML file from the files you have uploaded to your photo sharing web site, load this list in your script, iterate over all files in the database and when you find a match, bookmark, pin, flag or dot the image. Then you have all uploaded files in a collection, for further processing. Or put them into a category, also easy from a script.
-- Mario
IMatch Developer
Forum Administrator
http://www.photools.com  -  Contact & Support - Follow me on 𝕏 - Like photools.com on Facebook

jch2103

Thanks. That should help, after I've gotten a better handle on scripting.
John

Mario

You can always post questions in the scripting forum. I'm sure you'll get help if you provide details about what you have (input data) and what you need (what to do with matching files etc.)
-- Mario
IMatch Developer
Forum Administrator
http://www.photools.com  -  Contact & Support - Follow me on 𝕏 - Like photools.com on Facebook