Dealing with Kindle ebooks in calibre

[gbm]

From the KFX input plugin page:
https://www.mobileread.com/forums/sh...76&postcount=1

Importing from E-ink Kindle devices

KFX can be imported from Kindle devices running firmware versions 5.6.5 or later. Older versions do not support KFX. Note that obtaining KFX books from an e-ink Kindle usually results in poor quality images without color (except from a ColorSoft device).

KFX books can be added directly from e-ink Kindles connected via USB using the calibre "Device" view. Calibre version 7.25 or later must be used for newer Kindles that connect using MTP (Scribe and 2024 models).

Connect the Kindle to your computer via USB, wait for the automatically started jobs to complete, click "Device" on toolbar to show available books, and select one or more books. Then either click "Add books to library" in the toolbar, drag them to the library icon, or right click and select "Add books to library".

bernie

Originally Posted by darthyoda6

Yes, I went to the add books, browsed to the documents directory on the Kindle and selected the file and ended up with a kfx-zip. I gave up doing batch Kindle files and just select one at a time. I also removed the book then re-added it. Still didn't work, but read fine on the Kindle. Adding the book from an older Kindle gave me a proper kfx.

[Sirtel]

Originally Posted by darthyoda6

Yes, I went to the add books, browsed to the documents directory on the Kindle and selected the file and ended up with a kfx-zip. I gave up doing batch Kindle files and just select one at a time. I also removed the book then re-added it. Still didn't work, but read fine on the Kindle. Adding the book from an older Kindle gave me a proper kfx.

You do have the serial number in the deDRM settings?

[odamizu]

Originally Posted by darthyoda6

Yes, I went to the add books, browsed to the documents directory on the Kindle and selected the file and ended up with a kfx-zip. I gave up doing batch Kindle files and just select one at a time. I also removed the book then re-added it. Still didn't work, but read fine on the Kindle. Adding the book from an older Kindle gave me a proper kfx.

You removed the book from your Kindle (not Calibre) and re-downloaded it to your Kindle (not Calibre), then imported the new Kindle file to Calibre?

(If you just removed the book from Calibre, then re-added it to Calibre, try removing it from your Kindle and downloading a fresh copy to your Kindle. Wait for each book to complete downloading before moving on to the next one. If that doesn't work, post in the KFX Input plugin thread as jhowell will likely be interested in this.)

ETA: The above applies to "regular" ebooks, not textbooks or other special cases where DeDRM doesn't work.

[technomom]

Originally Posted by Sirtel

Tbh, I don't remember in which thread, but someone recently complained that they could no longer access their books on their 2024 Kindle. I assumed there had been an update, but maybe I was wrong and there was some other reason.

Still, it seems more than likely that Amazon will do something about those Kindles still receiving updates. And maybe older models too, although that's somewhat less likely.

That might have been me. I was asked what version of Calibre I'm using, and that's version 8.4.

Sorry for the delay - I had to go out of town, so I can't try to re-do anything until I get back.

[DNSB]

Originally Posted by Seirious

Method 2a has also been blocked. Kindle for PC 2.4.0 no longer allows book downloads, and although downloads still work with Kindle for PC 1.17, DRM removal is no longer possible.

I just brought up a copy of K4PC 2.40 (70904) in a new VM and it is still able to download books published before 2025-April-22 and calibre is able to import and remove DRM from them (the KFXInput plugin and correct DeDRM plugin are installed).

[lkmiller]

Etextbooks and other books with the tougher DRM have been blocked from downloading to Kindle for PC. This is likely because it turns out they did work with the new method. So now those books can only download to apps and devices where DRM cannot be removed -- mobile apps, Kindle for Mac, and Fires.

[odamizu]

I suppose this was just a matter of time (and it may not even be the first): A Chrome extension that captures content directly from Amazon's web reader and converts it to EPUB format — GitHub and Reddit

(I have not tried it, nor do I intend to, but thought I'd throw it out there in case anyone is interested.)

[mitra1]

Originally Posted by odamizu

I suppose this was just a matter of time (and it may not even be the first): A Chrome extension that captures content directly from Amazon's web reader and converts it to EPUB format — GitHub and Reddit

(I have not tried it, nor do I intend to, but thought I'd throw it out there in case anyone is interested.)

This antidrm is very hard to block because what Amazon can do it? Slow the change of page, the extension will be slower.
A cease and desist? This isn't a drm in a technical sense so it's totally legal.

[lkmiller]

Well, that's interesting. I imagine not all books are available for the webreader, but most probably are.

The creator of the new Kindle for PC method now has a method for Kindle for Android. Should work on anything you can download to it including the tougher DRM. (If you want decrypt those, don't wait.)

[jhowell]

Originally Posted by odamizu

I suppose this was just a matter of time (and it may not even be the first): A Chrome extension that captures content directly from Amazon's web reader and converts it to EPUB format — GitHub and Reddit

Kindle Cloud Reader delivers pages as images instead of text. This method uses uses OpenAI Vision as a fancy OCR to convert those images to Markdown text which is then converted to an EPUB. That process will sometimes produce inaccurate text and formatting.

I would consider this approach to be a last resort.

[tomsem]

Originally Posted by jhowell

Kindle Cloud Reader delivers pages as images instead of text. This method uses uses OpenAI Vision as a fancy OCR to convert those images to Markdown text which is then converted to an EPUB. That process will sometimes produce inaccurate text and formatting.

I would consider this approach to be a last resort.

Well, not exactly: you can select and highlight text and add notes, just as with the Kindle app. But this text is not accessible by web inspector, so I assume that is why the extension has to OCR page images.

For graphic novels it would be fine (no OCR needed), but I don't know if the extension has special handling for that.

[JSWolf]

The day I can no longer easily enough download and remove the DRM from Amazon eBooks, I'm done with eBooks from Amazon.

[xyzzy_foo]

Originally Posted by JSWolf

The day I can no longer easily enough download and remove the DRM from Amazon eBooks, I'm done with eBooks from Amazon.

For me, it's already over because KF8 version of Kindle eBooks after publishing April 22, 2025, are not easily obtainable.
I considered buying a PW1, but the image quality of KF8 files downloaded directly via Wi-Fi is unclear. At least with KFX on the KOA3, the images are terrible—black and white with low resolution.

[pdurrant]

I switched to buying from Kobo in mid-February. It's a little more inconvenient, as I still use Kindles, but given the price advantage of price-match and the gift card offers, I probably should have switched ages ago.

[xyzzy_foo]

Originally Posted by pdurrant

I switched to buying from Kobo in mid-February. It's a little more inconvenient, as I still use Kindles, but given the price advantage of price-match and the gift card offers, I probably should have switched ages ago.

The fact that Rakuten Kobo doesn't offer the Price Match Guarantee in Japan, along with reports from Japanese users who switched to Rakuten Kobo about structural issues in EPUB files for certain publisher's manga titles—problems not seen on Kindle eBooks—makes me hesitant.
I plan to buy a Kobo Sage soon, but it's also disappointing that Kobo eReaders saw a price increase in Japan two months ago.

As for honto and Kinoppy, while Epubor's solution can remove DRM, my tests with the trial version showed that at least the former lacks Japan's industry-standard CSS files, leading me to conclude that the original CSS was lost.
I strongly suspect an issue with Epubor's eCore and plan to contact them soon to request preserving the original CSS. Having multiple purchasing options is always a good thing.