Quote:
Originally Posted by Kest
Calibre is using the number in its "ASIN" metadata, but definitely is not a valid one (has way more than 10 characters). I don't know about that calibre cover metadata but I guess is a random string associated with both the book and the thumbnail's name (that I tried to emulate without success).
|
Yes -- that is the problem in the first place. calibre uses UUIDs, since there is really no way to know what the ASIN should be (although if you manually add an amazon:XXXXXXXX identifier, the Quality Check plugin can embed that).
"Since it depends on Amazon's records and no such book may exist"
If the ASIN isn't valid, there is no cover to download.
Quote:
I just noticed that indeed the cover is being extracted from the personal azw3 and saved as "thumbnail_<random_six_chars>.jpg" (really random chars. They re-roll if I cut off then restore the same book from the documents folder), and another one with four random chars if I insert the mobi counterpart of the book.
The problem is that the azw3's cover is not showing (the mobi one showed perfectly from the beginning, even with the personal tag), so maybe really the problem is the personal thing.
Isn't there a tool to modify the properties or whatever it is, of an azw3 to get rid of that Personal Document property?
|
Yes, calibre can do it quite easily. The proper settings in MOBI Output, or alternatively the Quality Check plugin I mentioned above.
There is also Java Mobi Metadata Editor, which should be floating around somewhere here.
But I cannot think why the Kindle would be extracting the conver from a PDOC *.azw3 and then not using it.
