Trying to pair filename and sha1 hash in collections.json

[Adam23]

Hello.

I was pointed to this amazing forum to find answers to my problem

I need to identify which sha1 hash in collections.json file in Kindle belongs to which ebook file.
I understand the hash is generated from the path to the file in Linux file system. However, there is some step in the process that Kindle uses, when handling spaces and dashes? I am unable to obtain the proper hashes. I have tried many online tools, and kindle collections, but none of them generated the right hashes.

For example for the path to: /mnt/us/documents/Orwell, George/1984 - George Orwell.mobi

the corresponding hash should be: f20904d7-b190-40f0-96a5-ea79b73a9997

Is there any online tool or a simple script/program that really works? Or am I doing it wrong?

Thanks for any useful advice/help!

[NiLuJe]

@Adam23: In your example, that's not the sha1 hash of the absolute path, that's the ASIN (in this case, a fake ASIN created by Calibre from an UUID). The sha1 thing is now a fallback when the book doesn't have an ASIN set in its metadata, and used to be preceded by a star in the old format. Can't remember how the SQL-ized version on FW 5.x handles it.

Check the Kindle Collections plugin for the gory details.
There's surely some smaller and simpler, but older, pieces of code laying around that handle the basics, but the Kindle Collections Calibre plugin is the only one that's still maintained, and AFAIK, probably the most complete & accurate one.

[Adam23]

Thanks for your response, but I still couldn't figure out why some books were identified by ASIN or by the hash numbers. Finally, I decided not to follow this path. I instead manually adjusted each file.

[NiLuJe]

FWIW: If the ASIN is set in the metadata, it's used, if it's not, the fallback to the sha1 sum is used instead.