It's a slightly weird idea/request for sure but the idea came about this way. I've created a script that mirrors all my metadata files into folders named after the archive in question (some managed by mylar, some custom created, lots scene based so no real consistency in file name itself) using the library's folder structure mirrored as well to keep things somewhat organized.

I did this as a personal backup, but decided well I know CV API is rough and I've got lots of data from sources that can't be automatically grepped from (think creator pages, publisher sites, previewsworld, errr fetlife yes there's a story there, etc) and for books the 3 DBs don't even recognize because it's apparently me and their mom that bought them.

However my own lack of organization is making things rough for folks and they asked about matching by hash, so I had to explain my current hashes include the very files they're looking at embedded and thus will not match.

Then I had the dim bulb turn on, "what if?" Jeffrey Wright said in my brain and boom the idea to record filename and original + new file hash for each archive written when the option is turned on into either a spreadsheet or DB table in the comictagger sqlite DB.