[eschwartz]

Originally Posted by laji

Sorry, I can not agree with you. The principle of strictly dividing datas and temp values (primary, real; and secondary, derived and technical only values, needed only for managing DB) is not my idea, and not accidentaly is so strict, so principle. Cosmetic clearing of a principially erronously structured behavior is not a way to happiness. First, it is not nice, not exact, not real-clear solution, only forced - it is the barrier in the way of developing. Second, if to mix temp values into DB, DB will grew up with extrem speed - without real needs. As I will develop the DB shower interface, so will more and more impracticable the DB's size. By the way: THIS was the cause, why I leaved that second, hierarchical-able book-organizer, that I wrote earlier. The size of the DB was 10× more, then real data's of books. Calibre is going on this, principially erronous way, too.

I'm pretty sure Kovid does whatever increases responsiveness in the library, regardless.
There is no "erroneous", there is merely a values dissonance.

Originally Posted by laji

https://www.mobileread.com/forums/sho...4&postcount=24


Sorry, it does not work, too. I tried:

<?xml version='1.0' encoding='utf-8'?>
<package xmlns="http://www.idpf.org/2007/opf" unique-identifier="uuid_id" version="2.0">
    <metadata xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:opf="http://www.idpf.org/2007/opf">
        <dc:contributor opf:file-as="calibre" opf:role="bkp">calibre (2.37.1) [http://calibre-ebook.com]</dc:contributor>
    </metadata>
    <guide>
        <reference href="cover.jpg" title="Borító" type="cover"/>
    </guide>
</package>

...and tried

<?xml version='1.0' encoding='utf-8'?>
<package xmlns="http://www.idpf.org/2007/opf" unique-identifier="uuid_id" version="2.0">
    <guide>
        <reference href="cover.jpg" title="Borító" type="cover"/>
    </guide>
</package>

The folder were named, as my import naming preferences (author - title). In the folder were an author-title named pdf, a cover.jpg, and the above metadata.opf. Calibre imported the PDF into the existing entry (correctly), but did not import cover. I tried it without PDF (JPG + OPF only): Calibre imported the OPF, as a new book file (like EPUB or any else), with title "metadata", and ignored the cover.

That is because both have been voided of metadata. The minimum required metadata in order for calibre too recognize it as a calibre OPF is:

        <dc:identifier opf:scheme="calibre" id="calibre_id">1</dc:identifier>

This will trigger calibre to read metadata from the OPF instead of the ebook file. One of the pieces of metadata it will read is the cover reference.
At that point, you should have the other DC roles filled out as well, or they will default to unknown.


Alternatively, you could embed the covers in the ebooks -- assuming they are EPUBs, at least.

Run this batch file in the topmost folder.

setlocal enabledelayedexpansion
for /d %%e in (*) do (
    cd %%e
    ebook-meta --cover=cover.jpg some-ebook.epub
)