[Conversion Plugin] KFX Input

[Ma'am-I-Am]

Originally Posted by bkennelly

I see that regularly. I have been assuming that it happens for books without DRM.

This book came from the Big river, so definitely came with DRM.

[Ma'am-I-Am]

@JSWolf - I'll try to remember to do it that way. The B/W covers make it a third choice, though.

@jhowell - Did what you asked per DeDRM FAQ, but the book imported properly this time. I have the log if you still want it, but I see little benefit. And I'm not sure a readable KFX-ZIP is exactly a problem, but I also didn't try to convert it.

[jhowell]

Originally Posted by Ma'am-I-Am

@jhowell - Did what you asked per DeDRM FAQ, but the book imported properly this time. I have the log if you still want it, but I see little benefit. And I'm not sure a readable KFX-ZIP is exactly a problem, but I also didn't try to convert it.

No need for a log if a book imports properly. It is only of interest to me if a book imports as KFX-ZIP but despite that is able to be read or converted in calibre.

[jhowell]

Version 2.25.0 - 19 May 2025

Handle unexpected image formatting found in some comics. Fixes "Unexpected image style for horizontal fxl: ...".

Remove invalid URL references present in some print replica books.

[mallomar]

Before putting in a support ticket on KOReader, thought I would ask here if there's something off with the page list in this book. I bought the Kindle edition, which says it contains real page numbers based on the physical book, imported it as KFX to Calibre, converted it to EPUB, yet when I add that EPUB to KOReader it does not have the reference pages. I tried polishing the EPUB to see if that makes a difference but it doesn't. The EPUB appears to have a page list in the nav.xhtml file so I'm not sure what's going wrong. This is the first time I've had this issue in KOReader when Calibre appears to show a functional page list in nav.xhtml or page-map.xml.

[JSWolf]

Originally Posted by mallomar

Before putting in a support ticket on KOReader, thought I would ask here if there's something off with the page list in this book. I bought the Kindle edition, which says it contains real page numbers based on the physical book, imported it as KFX to Calibre, converted it to EPUB, yet when I add that EPUB to KOReader it does not have the reference pages. I tried polishing the EPUB to see if that makes a difference but it doesn't. The EPUB appears to have a page list in the nav.xhtml file so I'm not sure what's going wrong. This is the first time I've had this issue in KOReader when Calibre appears to show a functional page list in nav.xhtml or page-map.xml.

Try reading the eBook as KePub as it's possible the page numbers are not compatible with KOReader. Do not read a ePub as the method used for page number is not compatible with Kobo's old version of RMSDK.

[jhowell]

Originally Posted by mallomar

Before putting in a support ticket on KOReader, thought I would ask here if there's something off with the page list in this book. I bought the Kindle edition, which says it contains real page numbers based on the physical book, imported it as KFX to Calibre, converted it to EPUB, yet when I add that EPUB to KOReader it does not have the reference pages.

That book has page numbers and they should be retained when it is converted from KFX to EPUB. I am unfamiliar with KOReader so I cannot comment on how it handles page numbers.

There are different ways that page numbers can be expressed in EPUB format. The method and options used when converting from KFX to EPUB can affect that.

[mallomar]

Originally Posted by JSWolf

Try reading the eBook as KePub as it's possible the page numbers are not compatible with KOReader. Do not read a ePub as the method used for page number is not compatible with Kobo's old version of RMSDK.

Thanks, but I'm reading in KOReader on Boox and an Android phone, so KePub isn't compatible. FWIW, every other book that I've tried that has a page list or map in the EPUB code has shown it in KOReader.

Originally Posted by jhowell

That book has page numbers and they should be retained when it is converted from KFX to EPUB. I am unfamiliar with KOReader so I cannot comment on how it handles page numbers.

There are different ways that page numbers can be expressed in EPUB format. The method and options used when converting from KFX to EPUB can affect that.

Thanks. I haven't changed any of my Calibre conversion options, so I'm guessing it's a bug in KOReader and will put in a ticket on their Github. It's weird that it's worked on every other book I've tried.

[mallomar]

Originally Posted by jhowell

That book has page numbers and they should be retained when it is converted from KFX to EPUB. I am unfamiliar with KOReader so I cannot comment on how it handles page numbers.

There are different ways that page numbers can be expressed in EPUB format. The method and options used when converting from KFX to EPUB can affect that.

Following up on this – you mentioned that your KFX Input plugin should be preserving that page list information, and I wanted to let you know I finally got to the bottom of it, and it confirms your plugin is doing an excellent job!

It was quite a journey. My initial KFX-to-EPUB conversion (using Calibre's standard converter) resulted in an EPUB that as I said didn't show pages in KOReader. I ran epubcheck and found many pathing errors. I fixed all those errors using Calibre's editor until the EPUB passed epubcheck with zero errors or warnings. However, even this "perfectly valid" EPUB still wouldn't show reference pages in KOReader. This led me back to the conversion process itself.

On a hunch, I went back to the original KFX file and converted it again, but this time, I specifically used your KFX Input plugin's own "Convert to EPUB" feature. The resulting EPUB also passed epubcheck with zero errors, but the crucial difference is that this version works perfectly in KOReader – the reference page numbers show up exactly as they should! This strongly suggests that your plugin produces a 'cleaner' or more robust EPUB structure (at least concerning the page-list and its paths) than Calibre's default KFX conversion, successfully avoiding whatever subtle issue KOReader was sensitive to. Thanks for creating a plugin that handles this conversion so well!

[jhowell]

Originally Posted by mallomar

Following up on this – you mentioned that your KFX Input plugin should be preserving that page list information, and I wanted to let you know I finally got to the bottom of it, and it confirms your plugin is doing an excellent job!

I am glad that you got it working.

Since your results point to calibre's conversion pipeline being the most likely cause of the loss of page numbers I do not intend to pursue this further.

[mallomar]

Originally Posted by jhowell

I am glad that you got it working.

Since your results point to calibre's conversion pipeline being the most likely cause of the loss of page numbers I do not intend to pursue this further.

Totally fair. I think I’ll default to KFX Input’s conversion engine from now and then polish the EPUB to get updated metadata and such.

[odamizu]

Originally Posted by jhowell

... It is only of interest to me if a book imports as KFX-ZIP but despite that is able to be read or converted in calibre.

@jhowell: Not sure if this is what you mean, but I was downloading about 50 ebooks to my Kindle and 11 of them imported as KFX-ZIP. And yet, the KFX-ZIPs could be read in calibre, and when I clicked the "From KFX" button, the KFX-ZIPs converted to epub, albeit with errors (see spoiler below). On quick look, the epubs appeared okay and could be read in calibre viewer despite the error report.

Spoiler:

ERROR: Book is incomplete. All of the KFX container files that make up the book must be combined into a KFX-ZIP file for successful conversion. (Missing containers CR!T68962CCA14BQ4968EJZ2TSTHEK5)

Note: After the above, I went back and deleted the books from my Kindle, re-downloaded to Kindle, and when imported the new files into calibre, they imported as KFX and converted without errors. But I report the above since you expressed an interest in KFX-ZIP that could be read and converted in calibre.

[jhowell]

Originally Posted by odamizu

@jhowell: Not sure if this is what you mean, but I was downloading about 50 ebooks to my Kindle and 11 of them imported as KFX-ZIP. And yet, the KFX-ZIPs could be read in calibre, and when I clicked the "From KFX" button, the KFX-ZIPs converted to epub, albeit with errors (see spoiler below).

That can happen if one or more of the files that make up a book are not downloaded to the Kindle before importing it into calibre. That is most likely to occur if you quickly download multiple books without waiting for each to complete before starting the next one.

Originally Posted by odamizu

On quick look, the epubs appeared okay and could be read in calibre viewer despite the error report.

The missing files will likely contain images that will now be missing from the converted book. That could be important depending on the book.

Originally Posted by odamizu

Note: After the above, I went back and deleted the books from my Kindle, re-downloaded to Kindle, and when imported the new files into calibre, they imported as KFX and converted without errors. But I report the above since you expressed an interest in KFX-ZIP that could be read and converted in calibre.

Thanks for the report. I will add a note about this possibility in the plugin documentation.

[Seirious]

Do you have any plans to support converting multiple KFX files to EPUB?

[jhowell]

Originally Posted by Seirious

Do you have any plans to support converting multiple KFX files to EPUB?

If you mean combining the individual KFX files that make up a single book then that is already done by this plugin as explained in the first post of this thread.

If you mean something else, such as merging separate books into an omnibus, then I have no plans for any such thing.