View Single Post
Old 01-23-2024, 04:45 AM   #7
shamanNS
Wizard
shamanNS ought to be getting tired of karma fortunes by now.shamanNS ought to be getting tired of karma fortunes by now.shamanNS ought to be getting tired of karma fortunes by now.shamanNS ought to be getting tired of karma fortunes by now.shamanNS ought to be getting tired of karma fortunes by now.shamanNS ought to be getting tired of karma fortunes by now.shamanNS ought to be getting tired of karma fortunes by now.shamanNS ought to be getting tired of karma fortunes by now.shamanNS ought to be getting tired of karma fortunes by now.shamanNS ought to be getting tired of karma fortunes by now.shamanNS ought to be getting tired of karma fortunes by now.
 
Posts: 1,115
Karma: 12345678
Join Date: Feb 2010
Location: Serbia
Device: Kindle PW5, Kobo Libra 2, Kindle PW1
@BionicGecko The script actually still does its job, as in it creates thumbnail files in /system/thumbnail with correct filenames.
Further behavior that is firmware version dependant and book format and cdetype flag (EBOK vs PDOC) dependant is the thing that determines if Kindle uses those files or not.
The thing that doesn't work on newer firmware (prior to 5.15.1 IIRC that fixed that issue?) is the combination of azw3 + PDOC flag = what you used to get when you send files via email or Send to Kindle, before Amazon started converting "personal documents" to KFX format (and probably you still get azw3 PDOCs on Kindle devices with firmware older than some arbitrary cutoff version).

Anything that goes through Amazon's conversion servers ends up with a file that has the cdetype flag set to value of "PDOC".

If you "sideload" your book using USB cable and make sure that book files have cdetype set to EBOK Kindle should use and display cover thumbnails... after you deal with that stupid overwritting of existing thumbnail images (= you would need to use the script 2 times with letting Kindle go online in between).

The only issue with that Python script that I can recall is that it had issues with filenames or title metadata?? when book title or author contained non-ASCI characters (= the script code doesn't support Unicode?).

I remember using modified version (modified to only "parse" KFX files + don't downsize the image + put the files inside a folder where ScreenSaver hack caches screensaver images; and repackaged into a KUAL extension so the script can be used directly on the Kindle itself) of that script to workaround a bug with ScreenSaver hack not extracting correct image from KFX books created by Calibre plugin + Kindle Previewer.
And I definitely used that on Kindle PW3 with firmware 5.9.7 which I've used for years and years before updating it to 5.13.6.

Last edited by shamanNS; 01-23-2024 at 04:58 AM.
shamanNS is offline   Reply With Quote