Finding duplicate ISBNs & compare size

[realzi]

Hey there,
trying to find an answer to my question i did read a lot on this forum, but to no avail. I understand that automatic deletion of duplicates, which are not binary duplicates, seems to be frowned upon. I want to do it anyway.
I would love to find duplicate ISBN numbers (making sure it is the same edition of the book) and then to compare the size of the books to only keep the biggest. My assumption being that in this case the quality of the cover and the book per se is better.
The finding duplicates helps a lot in this case. But as my library is quite big i would also like to have an automated process, where the file sizes of books with the same isbn get compared.
Is there any easy way to achieve this?
Regards realzi

[theducks]

Your Q prompted me to run an Find Duplicates:ISBN (the plugin) search.
1) DON'T Automate
2) See #1, don't trust a simple compare.

Each and every one was uniquely wrong (and tedious to find a resolution).

[realzi]

Well obviously a prerequesite to this procedure would be to verify the ISBNs are correct!

There is a plugin that extracts the ISBNs from ebooks. The scenario you are describing sounds like you might have a couple of wrong numbers there. As the plugin was able to identify/verify a bit more than 70% of the ebooks in my library i am quite confident that an automated approach would be useful.

Also i tested the find duplicates plugin yesterday and in the first 50 Books the results were on point. But i don't feel like comparing all of the books by hand. That is why i made this post.

So is there a way to compare the results of the plugin?

As far as i understand the plugin iterates through the whole library and makes a list of ISBNs. If the most recent one was already on the list, it gets marked as a duplicate, if it was not, it is put on the list. So far so simple, i just struggle to see a way to compare filesizes in this process.

[chaley]

Originally Posted by realzi

As far as i understand the plugin iterates through the whole library and makes a list of ISBNs. If the most recent one was already on the list, it gets marked as a duplicate, if it was not, it is put on the list. So far so simple, i just struggle to see a way to compare filesizes in this process.

You can't with the find duplicates plugin. It is possible with Action Chains using a template that for each book searches for others with the same ISBN then for each of the results check for identical book sizes, but the template will be very complicated and the process horribly slow. I am not interested in helping write such a template.

If you really want to do this ( ) then I suggest you make a custom column containing the book format size (you are comparing the same formats?). Then use find duplicates to find ISBN duplicate sets. For each set, sort the booklist by the format file size and manually delete any that your Mark 1 eyeball say are the same size.

Or write an Action Chains template that iterates through the selection sets finding books with the same size and deleting one of them. Personally I think this is madness, but I am not you.

The easiest solution is probably to write a python program that uses the calibre API to do the nested scans, but that is "easiest" only if you are a programmer.

[realzi]

Thx @chaley, you have given me an idea. I can Just save the duplicates into folders, delete them from the library and let a python script or another program do the size comparison. Obviously i will need to have all the ISBNs in Order and all the Metadata downloaded first, which will take quite some time.