Richard Quirk

It is slightly possible, but it isn't particularly well suited to doing it this way. The problem is that the script by default only has 1 path at a time...

When I get round to it... it isn't a priority for me atm though just another thing to not forget to do :)

That's because it has not been ported to Python 3 yet. The first line has "#!/usr/bin/env python" so should pick up the default python 2 in the path, it's unusual...

It's a good idea, I might have to see if it's possible. Some things do seem hellishly slow on Flickr's end too - creating the sets takes minutes occasionally, and...

Nice. I wasn't expect that. I'll see if I can integrate that change :-)

Originally it did this, but checked every photo in the local db against every photo on flickr. With 10k photos it would take years to finish, as each remote check...

Yeah, look up all photos in a few queries and compare that with the DB. Definitely not 1 by 1. I have a ton of sets, so even that might...

I see that photosets also returns the count - https://secure.flickr.com/services/api/flickr.photosets.getList.html - so maybe get all the sets, check the counts, then check any that don't look good would be a...

It's a python "shelve" file, which is a serialized dictionary. The keys are the flickr photo ids to file names, and the file names to flickr keys. e.g. python >...

Can you try with the latest, which removed the old embedded exif.py and requires the upstream python-exif package?