[Conversion Output] KFX Output

[maximalmobi7]

the x-ray plugin for calibre seems only to work for azw3 and mobi files, it doesn't seem to support kfx format that I can tell anyway.

referring to my edited post under page 5 about this plugin, also I have to learn how to manually put the thumbnail under the device for kfx format still, I am new to understanding it since I just found out in the last day about this new format.
again it seems best to use calibre input format rb format to out format kfx format, without conversion errors from epub since the kfx plugin seems to convert the RB format into epub then outputs it to kfx format, that seems to keep the conversion errors from happening in the few ebooks I tested today.

to not get the error below when converting from epub format to RB format into kfx format, make sure your source from epub format to rb format, has under structure detection the box label insert metadata as page of start of book that feature needs to be uncheck at all times when going from epub format to rb format, also needs to be unchecked when converting from rb format to kfx format, otherwise you will get the error below, I was able to convert my ebook once I made sure that feature was not selected in calibre.

Error(YJConvValidator):E120013: Data table found in the book


https://www.mobileread.com/forums/sho...=272407&page=5

[jhowell]

Originally Posted by maximalmobi7

So I was playing around with kfx format plugin and found It seems to work best if I convert an epub format into the rb format, then selecting the rb format to kfx format, that seems to of gotten me past the converting bugs that happen when you try epub to kfx format...

Converting first to a more limited format and then to KFX can certainly work around some conversion problems, but at the cost of removing some of the formatting from the book. That approach should be used only as a last resort for books that do not convert otherwise. Not converting to KFX at all may yield better results for those books.

Originally Posted by maximalmobi7

... I just have to figure out how to place the thumbnail image their for kfx format...

My post on ExtractCoverThumbs was meant for the the specific user that I replied to. It is possible to combine the various pieces called out in that post to produce a test version of ExtractCoverThumbs that supports KFX, but that change has not been vetted by creator of that program and there may be flaws. Also, ExtractCoverThumbs is a stand-alone program, not a calibre plugin. It requires use of Python.

For now, if you want your KFX book to have a cover thumbnail on an e-ink kindle then you should set the "amazon" identifier of the book in calibre to the 10-character ASIN of an equivalent book sold by Amazon (if such exists.) Here are instructions for doing this:

• Search amazon.com with a web browser for an Amazon equivalent of the book you are converting and bring up its page. The web browser address bar will have something like "https://www.amazon.com/gp/product/B005MZN2B2/...". The ASIN (Amazon Standard Identification Number) for the book is the 10-character string that begins with a "B".
  
• In calibre select your book, choose "Edit Metadata", and in the "Ids:" field add "amazon:B005MZN2B2" (without the quotes and using the ASIN you discovered). If there is something already in the "Ids:" field then separate the new information with a comma.
  
• Convert the book to KFX and sideload it to your kindle device. It should show the cover thumbnail of the book you found previously on Amazon.

Originally Posted by maximalmobi7

...in any case I hope calibre is updated at some point to support thumbnail and kfx format tranfering in the future versions of it.

I agree that it would be nice if calibre supported KFX sideloading and thumbnail creation. I am not interested in working on that at the moment myself.

Originally Posted by maximalmobi7

the x-ray plugin for calibre seems only to work for azw3 and mobi files, it doesn't seem to support kfx format that I can tell anyway.

That is probably true. I have no interest in X-ray and have not investigated how it might work with KFX.

[maximalmobi7]

so I tried the automatic download metadata and covers , with selects the books asin with calibre control D command, under calibre, first got the metadata from amazon.com but when converting the ebook, it fails to load the thumbnail information when you copy the file.kfx to your kindle paperwhite 2, their was no thumbnail loaded with the file, is their something wrong with using the download metadata and covers feature in calibre to make the kfx get the thumbnail image included in the ebook file?


under id tags its formatted like this after downloading the metadata, isbn: number, amazon:number, mobi-asin:number

so I changed it around so that the amazon asin number was listed first under id tags
but the output kfx image still didn't have the thumbnail image included, I may of did something wrong since the information metadata downloaded from amazon source for asin didn't make the output kfx included the thumbnail, I recopy the output file to my kindle paperwhite 2 still no thumbnail image being loaded so something else needs to be done for that I think, or I did something wrong with the metadata tab under calibre.

does the amazon id tag need to come first in the order of information or does that matter for getting the thumbnail image to show up when converting from rb format to kfx format?

also I like to keep my kindle paperwrite 2 offline to not be getting the downloaded kindle ads in the device, does that effect the thumbnail image not loading for normal transfer of azw3 and mobi files from calibre it doesn't seem to hurt it being offline, for their thumbnails images to load? that was one more question I had if the thumbnail not loading since I am keeping the device offline.

notice that the rb ebook format is black and white, no color images, I don't no if another format better to use first before converting it to kfx if you want to keep color images in your ebook file. I only reason may be if your using the kindle app on android tablet and want to keep the ebook in color if it has color images in the first place. since I do have an android tablet, I may want to try other formats for the ebooks with color images just not sure which formats to use first that all.

also I am not sure if the kfx format unpacks the mobi and azw3 files from from the included kfx output since the size of the output kfx file is bigger than the source epub file also its bigger than the rb format ebook file too that size was also a lot smaller before Kindle Previewer 3, I have to see if it can be treated as an zip file if I can extract the other outputs of mobi and azw3 if their included in the structure if 7 zip works on the output, I have to reinstall 7 zip to see if that can be used. -- nope 7 zip can't extract it, I just tried that so I have to wait for kindleunpack to update to an version that can extract the mobi and azw3 formats from the kfx format if their included in the output since the output file seems too big in its current format.


also note that renameing it to mobi from kfx format does not work for kindleunpack plugin under calibre , I did test that it doesn't read it as an format that can be unpack if you changed the end extension of the file name. again that's if the converted file by kfx output format plugin doesn't remove the source mobi and azw3 formats from the output, that would explain why the output size is bigger than the source input formats.

or if theirs an beta kindleunpack plugin for calibre yet that can do this feature.

[maximalmobi7]

****
The ASIN (Amazon Standard Identification Number) for the book is the 10-character string that begins with a "B".
****

****
In calibre select your book, choose "Edit Metadata", and in the "Ids:" field add "amazon:B005MZN2B2" (without the quotes and using the ASIN you discovered). If there is something already in the "Ids:" field then separate the new information with a comma.

*****

Information above from other users source post, included here since I wanted to comment on that.

when you download the metadata information from amazon source under calibre, the amazon source number that was included with control-d command feature under calibre doesn't not start with an B in the string it pulls up for the ebook, I may have to manually find the correct information from amazon page for the ebook since again the download feature under calibre doesn't not have asin numbers that start with B in them. I just wanted to point this out if may be why the output kfx format is not including the thumbnail when automatic downloading the metadata from amazon under calibre with that download metadata and covers option under calibre. again the asin automatic download of metadata included of amazon: number under ids: is only numbers and no B text before the number from amazon source from calibre feature for metadata automatic download from calibre.

I think It the ids: amazon source it download the amazon 10 numbers without the B before the number that's not included in the ids: tag for the ebook. again the ebooks string doesn't start with an B in it.
I don't no if its an simple matter of putting a B before the 10 number if that not included in the source ebook.

so I deleted the ids : isbn: number , , mobi-asin: number
I have yet to reconvert the rb book to kindle book, I edit that if that works or if putting an B in the ids tag works either I don't no yet since I have yet to try it in the calibre program, still I hope at some point I can use kindleunpack since again the kfx format output size is too big from the source, its still better to use azw3 format if the source size matter to you too since kindleunpack is not useable yet for kfx format.

I don't no if the above ids: tags interfere with the thumbnails feature for kfx ebook or not, I tried putting amazon: first, then isbn and mobi-asin, but after reopening the metadata page in always put the amazon tag in the middle between the isbn 10 number and mobi-asin 10 number the ids: tag box under calibre.

amazon: keep this one, again it doesn't start with an B under the amazon page, don't no if needed to put an B before the number since it doesn't have that on the link amazon page brings up. also it is the 10 number only that metadata and covers downloads from the amazon source.



so I tried putting an string B before the number and it breaks the link to the ebook on amazon.com website, if you put in the B since the ebook doesn't have that only the number, when its just the number it does bring up the correct page, but with the b string inserted before the number it breaks the online link. I don't no if that matters for the kfx format or not.

I tried the ids: "amazon:number" only in ids: under calibre and no luck still no thumbnail image on kindle paperwrite 2, when coverting rb book into kfx ebook,

their was no thumbnail created since it doesn't load an image for the ebook at all, and the amazon link works in calibre.

[jhowell]

I left out some details in the instructions in my previous post.

Originally Posted by maximalmobi7

so I tried the automatic download metadata and covers , with selects the books asin with calibre control D command, under calibre, first got the metadata from amazon.com but when converting the ebook, it fails to load the thumbnail information when you copy the file.

The ASIN that you use must be for a Kindle edition of the book, not a print edition.

Calibre's automatic metadata download will often set the "amazon" identifier to the ASIN of a print edition. You will need to manually correct it in that case.

Originally Posted by maximalmobi7

under id tabs its formatted like this after downloading the metadata, isbn: number, amazon:number, mobi-asin:number

does the amazon id tag need to come first in the order of information or does that matter for getting the thumbnail image to show up when converting from rb format to kfx format?

The order of identifiers does not matter.

A proper ASIN will be 10 characters (upper case letters and digits). One for a Kindle-edition will begin with a "B", though ASINs for other things can also start with "B". An all numeric value refers to a print edition and will not work.

You can verify that the ASIN is set correctly by selecting the book in calibre and looking at the book details panel (usually on the far right). There should be an "Amazon.com" entry in the "Ids:" section. Click on "Amazon.com" there and your browser will open to the associated web page. Verify that the browser shows the expected book and that it is a Kindle edition, not a print edition. Use "Edit Metadata" to fix the ASIN if not.

Originally Posted by maximalmobi7

also I like to keep my kindle paperwrite 2 offline to not be getting the downloaded kindle ads in the device, does that effect the thumbnail image not loading...

The cover thumbnail will be downloaded directly from Amazon. The Kindle needs to be online.

[maximalmobi7]

**
The cover thumbnail will be downloaded directly from Amazon. The Kindle needs to be online.
**
thanks for the above information, so I fixed the asin number to match an ebook version on amazon.com website, but since it was linking it to an real book asin number instead of the ebook version, I don't no if it would download that thumbnail image since again I am not taking my device online for thumbnails, that's why I use calibre again to transfer both ebooks and their thumbnails to the device, so it may or may not work,

-- but I want to keep my device offline that why I use calibre in the first place for transferring ebooks, otherwise my device would be spamed with amazon web ads, which I don't want on my device, so I just keep it offline, calibre has no problems transferring the thumbnails of mobi and azw3 to the device, so I may wait for the kfx ebook format to come more smaller and once calibre has an native kfx converter if that ever happens without requiring an plugin option. also may wait for it to be more stable with less converting errors from the epub source, since rb format doubles the output size I think. it may not be worth the kfx format if you have to have the output size bigger than the epub file size was to start with.

also the size of the files are as follows , the kfx file format was converted from rb book

epub file 6,913
azw3 file 7,126
mobi file 6,982
rb file 12,861

the azw3 and mobi was kindleunpack with that under calibre, so your seeing its split file sizes

kfx file 17, 388

kfx file 17, 387 is the size when the kindle ids: tag matches an kindle ebook instead of real asin amazon book from web link under calibre's ids: metadata

I don't no if the kfx file is just bigger since I used the rb book source, or if it could be smaller and their no way yet to split out the mobi and azw3 formats from the output that I can tell.

[jhowell]

Originally Posted by maximalmobi7

I don't no if the kfx file is just bigger since I used the rb book source, or if it could be smaller...

KFX files for books purchased from Amazon include data compression in conjunction with their DRM. I don't know of any way to produce a compressed KFX file without DRM. The KFX files produced by the plugin are uncompressed and so are significantly larger than formats that use data compression.

[maximalmobi7]

Originally Posted by jhowell

KFX files for books purchased from Amazon include data compression in conjunction with their DRM. I don't know of any way to produce a compressed KFX file without DRM. The KFX files produced by the plugin are uncompressed and so are significantly larger than formats that use data compression.

so is their no way for the plugin to use this command with the output

-c2 -dont_append_source

I use it with kindlegen command when converting epub to mobi format, before using kindleunpack under calibre.

that makes the file size smaller, I am just not sure if the kfx plugin can use these commands when converting or not?

[jhowell]

Originally Posted by maximalmobi7

...that makes the file size smaller, I am just not sure if the kfx plugin can use these commands when converting or not?

Those flags have no effect on the creation of KFX files.

KFX files produced by this plugin will always be larger than other formats due to the lack of compression.

[fardeenah]

can u help me please

******************** Failure Reason ********************

Invalid 'style/tag' combinations: [[STYLE_COMBINATION,IMG,display:none,height:[500.0->1000.0]pt]]


Python function terminated unexpectedly
{"level": "error", "msg": "Morning Star<br><br><b>Kindle Previewer error:</b> Invalid 'style/tag' combinations: [[STYLE_COMBINATION,IMG,display:none,height:[500.0-&gt;1000.0]pt]]", "det_msg": "", "title": "KFX conversion failed"} (Error Code: 1)
Traceback (most recent call last):
File "site.py", line 132, in main
File "site.py", line 109, in run_entry_point
File "site-packages\calibre\utils\ipc\worker.py", line 190, in main
File "site-packages\calibre\gui2\convert\gui_conversion.py", line 31, in gui_convert_override
File "site-packages\calibre\gui2\convert\gui_conversion.py", line 25, in gui_convert
File "site-packages\calibre\ebooks\conversion\plumber.py", line 1222, in run
File "calibre_plugins.kfx_output.__init__", line 69, in convert
File "calibre_plugins.kfx_output.__init__", line 124, in convert_from_epub
File "calibre_plugins.kfx_output.__init__", line 162, in report_failure
calibre.ebooks.conversion.ConversionUserFeedBack: {"level": "error", "msg": "Morning Star<br><br><b>Kindle Previewer error:</b> Invalid 'style/tag' combinations: [[STYLE_COMBINATION,IMG,display:none,height:[500.0-&gt;1000.0]pt]]", "det_msg": "", "title": "KFX conversion failed"}

[jhowell]

Originally Posted by fardeenah

******************** Failure Reason ********************
Invalid 'style/tag' combinations: [[STYLE_COMBINATION,IMG,display:none,height:[500.0->1000.0]pt]]

This is a typical failure of the Kindle Previewer to convert a book due to a combination of element and style that it doesn't like. In this case it is an <img> tag with style "display:none" somewhere in the book. If you want to convert this book you will need to edit it to modify or remove the offending content. (Sorry, but you are on your own for figuring out how to accomplish this.)

[fernando2112]

Thanks for making this plugin!!

So far, most of the books I've converted to kfx are working great.
The only issue I had was yesterday, this one book was not displaying the ToC when using the kindle menu (go to), it was empty. I double checked and the file had the ToC correctly set. Any ideas?

I'm just not using kfx 100% of the time yet because for me, not being able to making my own covers working for sideloaded books is a deal breaker. Do you think it will ever be possible to make covers work for kfx (without using the ASIN trick)?

Another thing: is it possible to make the cover inside the book fill the whole page? In AZW3 I can manage to do that, but when converting to kfx the cover has this white border all over.

[jhowell]

Originally Posted by fernando2112

The only issue I had was yesterday, this one book was not displaying the ToC when using the kindle menu (go to), it was empty. I double checked and the file had the ToC correctly set. Any ideas?

The "GO TO" table of contents is created from the NCX file of the source EPUB. If that is coded correctly, I don't know what could be going wrong. Some things to try:

Check that the TOC of the source file shows in calibre's EPUB viewer when you press the "T" icon.

Run the EPUB though the calibre editor's "run check" function and/or EPUBCheck.

Originally Posted by fernando2112

Do you think it will ever be possible to make covers work for kfx (without using the ASIN trick)?

The ExtractCoverThumbs program can create thumbnails for sideloaded books on a Kindle. I am hoping that it will be updated to support KFX. (I provided some files to the developer to help with that.)

Originally Posted by fernando2112

Another thing: is it possible to make the cover inside the book fill the whole page? In AZW3 I can manage to do that, but when converting to kfx the cover has this white border all over.

I don't know of any way to change how covers are processed.

ETA: I looked into this a bit more. On e-ink Kindles the cover page inside KFX books is rendered with margins and page/loc numbers. This happens even for books purchased from Amazon. It is just the way it is.

[fernando2112]

Originally Posted by jhowell

Run the EPUB though the calibre editor's "run check" function and/or EPUBCheck.

Yes, the file had some erros, my bad. Working fine now.

Originally Posted by jhowell

The ExtractCoverThumbs program can create thumbnails for sideloaded books on a Kindle. I am hoping that it will be updated to support KFX. (I provided some files to the developer to help with that.)

Amazing! Hope to have support for kfx files soon!

Thanks for all the answers

[maximalmobi7]

so I want to point out if you rename the file.kfx format extension to file.azw8 extension it seems to be able to load it the file kfx in Kindle Previewer 3 if it thinks its an azw8 file.

also I was playing with converting files to kepub format then to LRF format removed a few fonts in under styling tab I used a few of the filter style information which can remove fonts , padding etc..
in any case then I converted from LRF back to azw3 and also tried epub then to kfx format, that seems to work too if you play around with it to get it to convert to kfx without errors but you do have to play around with the settings in convert tab in calibre a bit, I hope that helps, since at one point I had fonts in an ebook I didn't want so I removed them, also I think converting to LRF first helps remove some stuff that can crash the kfx plugin , also used kepub since I was trying the other one that has features like kfx ,

again the only way to load kfx format into Kindle Previewer 3 is to rename the file to azw8 format. I guess it seems that azw8 is like the kfx format since once in kfx and rename it seems kindle previewer3 likes that file extension for kfx format to be in an azw8 container or at least it thinks that what it is I guess.

any chance in getting the plugin setup to output to azw8 instead of kfx?
since then you can just click in calibre the view icon it it would open it as azw8 to Kindle Previewer 3 if it converted it to azw8 instead of manually renameing the file, its just an idea anyway.

in the kfx output.zip which files requiring editing if you want it to target the format azw8 instead of kfx like for example the output extension and the output text in calibre? it could be converted to kfx but I want to target just the container output name to azw8 since it seems that kfx format files are azw8 files at least it seems that way since it loads them output problems if you rename kfx to azw8 then open the file in kindle previewer 3.

just wondering what strings need to be edited and what does not need editing in kfx output.zip in order to make it output to azw8 instead for extension, or format? since it seems kindle previewer 3 may be making azw8 files instead of that kfx format is the wrong extension maybe for kindle format or I am just guessing since the program seems to like them if their end with azw8 instead of kfx.

ok so I changed kfx to azw8 in the file __init__.py
rename kfxlib to azw8lib
edited kdf_container.py from kfx to text replacement to azw8
rename and edited kfx_container.py to azw8_container.py with replacing kfx to azw8 in it.
misc.py edit the kfx to (".azw8z:"
keep previewer_kpf.py unedited to kfx
edited yj_book.py kfx to azw8
the rest of the files were unedited, then I used 7 zip to recreate the plugin with the changes that seem to make the output azw8 instead of kfx , I hope that info helps, used wordpad to find the kfx text and update it to azw8 ,

again I keep the keep previewer_kpf.py unedited , so it targets kfx but outputs it as an azw8 container I hope that info helps,

"" I will not provide an edited zip file either the creator of kfx plugin need to do that themselves or someone else needs to do so if you don't want to change it yourself.""

I just want to make some info simple since its was easy to change it again you can't change the previewer_kpf.py since it need to target the libs and stuff like epub to kfx in order to create the files , so it needs to stay the same for now I think.