Page Numbers for Kindle

[HBIT]

I am currently creating an eBook in Sigil on a Mac and will be converting it to a mobi file for Kindle. How do I create page numbers for the Kindle? I read the post thread by goaspy on the Calibre forum. I see that I need an apnx file. I have a print book and ISBN number to reference to, but how can I generate an apnx file with a Mac. and how is it incorporated in the mobi file?

[Doitsu]

Originally Posted by HBIT

I am currently creating an eBook in Sigil on a Mac and will be converting it to a mobi file for Kindle. How do I create page numbers for the Kindle?

You could theoretically use page number anchors in the book and a pagelist in the .ncx file or a pagemap file to define page numbers. (There are example files in this post.)
However, AFAIK, only large publishing houses can publish books with "real page numbers." I.e., if you plan to publish your book with KDP, you can save yourself the hassle.

BTW, KindleGen won't generate an .apnx file for you, but you can generate an .apnx file for testing purposes with the latest KindleUnpack version. However, IIRC, you'll also need to add ASINs with dualmetafix_mmap.py, before you generate the .apnx file with KindleUnpack.

[RbnJrg]

Originally Posted by Doitsu

You could theoretically use page number anchors in the book and a pagelist in the .ncx file or a pagemap file to define page numbers. (There are example files in this post.)
However, AFAIK, only large publishing houses can publish books with "real page numbers." I.e., if you plan to publish your book with KDP, you can save yourself the hassle.

BTW, KindleGen won't generate an .apnx file for you, but you can generate an .apnx file for testing purposes with the latest KindleUnpack version. However, IIRC, you'll also need to add ASINs with dualmetafix_mmap.py, before you generate the .apnx file with KindleUnpack.

Please, could you explain a bit more the procedure?
That is, if I have an epub with a pagemap file, and with that epub I build a .mobi with Kindlegen and after that I use the last version of KindleUnpack to unpack the generated .mobi, will I have an .apnx file?
And regarding ASIN, can it be any number? (I ask you this because the generated .mobi could be a public domain book whose ASIN doesn't exist).

Many thanks in advance.
Rubén

[thewitt]

Originally Posted by RbnJrg

And regarding ASIN, can it be any number? (I ask you this because the generated .mobi could be a public domain book whose ASIN doesn't exist).

The ASIN number is what Amazon assigns you when you upload your book through their KDP program.

I can't answer the rest of your question, sorry.

[Notjohn]

As Thewit suggests, ASIN is merely an Amazon stock number. I believe every digital e-tailer (including Apple) now has a similar inventory system, making ISBNs an unnecessary and rather pointless expense for those of us in countries where they aren't free.

As for Real Page Numbers, if it were possible through the KDP (I suspect that it isn't), I really don't think the effort is worth it. Academics are quite accustomed to citing e-books with location numbers or (my preference) chapter title.

[Doitsu]

Hi Ruben,

To be honest, I've only limited experience with this topic. I only toyed it with it when KevinH added the APNX generation feature to KindleUnpack. If you have super-technical questions he's the guy to ask.

Adding a page number basically consists of the following:

1. Adding unique page number targets to the .html files. For example.
<a id="page1" /> (which Sigil will save as <a id="page1"></a>). You can, of course, add ids to all tags that can have one.

2. Adding a pagelist to the NCX file or a separate pagemap file. Since Kindlegen supports both methods and pagemap is an ADE extension, I'd recommend pagelists, unless you're going for cross-plattform compatibility.

Originally Posted by RbnJrg

And regarding ASIN, can it be any number? (I ask you this because the generated .mobi could be a public domain book whose ASIN doesn't exist).

AFAIK, yes. Calibre uses UUIDs and my simple KindleGen wrappper (which includes KevinH's dualmetafix_mmap.py script) uses only the last 12 characters of the UUID to generate a fake ASIN and it seems to be working. However, the downside is that if you sideload books with a fake ASIN and an EBOK book type, Paperwhites will not generate a cover, even if the book contains one. (For more background information on this see the dualmetafix_mmap.py thread.)
However, my plugin can optionally also generate a thumbnail image, if you set the corresponding kindlegen.ini parameter. For more details, see the kindlegen wrapper thread. Note that in order to generate thumbnails, you'll need to open a command line prompt as an administrator and enter the following command to download and install the Python Imaging Library.

pip install Pillow

Here's practical workflow example that uses my plugin, you can of course also use Kindlegen and dualmetafix_mmap.py manually.

1. Install my simple Kindlegen Sigil plugin.
2. Change the following kindlegen.ini parameters before running the plugin:

  azw3_only = True
  donotaddsource = True
  add_asin = True

3. Download PW2_pagelist_2.zip and extract pagelist.epub.

4. Open pagelist.epub with Sigil and use the Kindlegen plugin to generate an .azw3 file with it. If all goes well, you should end up with a Page number test [paglist]_e9b1fce6366b.azw3 file and a thumbnail_e9b1fce6366b_EBOK_portrait.jpg file. (Your random ASIN will of course be different.)

5. Run Kindleunpack and unpack the .azw3 file with it using the default settings. Kindleunpack will unpack the file and automatically generate mobi8-Page number test [paglist]_e9b1fce6366b.apnx. Rename this file so that it has the same name as the .azw3 file, but an .apnx extension. I.e. Page number test [paglist]_e9b1fce6366b.apnx

6. Copy the .azw3 file to a folder in the Kindle /documents folder and the thumbnail .jpg file to the (hidden) /system/thumbnails folder. Create a new folder in the folder that contains the .azw3 using the same name as the book but with the extension .sdr. In my specific example, I'd need to create Page number test [paglist]_e9b1fce6366b.sdr. Copy the .apnx file into this folder.

For example if you put the .azw3 file in the folder /document/test, it should contain the following:

Page number test [paglist]_e9b1fce6366b.azw3
Page number test [paglist]_e9b1fce6366b.sdr

(The Page number test [paglist]_e9b1fce6366b.sdr folder should contain Page number test [paglist]_e9b1fce6366b.apnx.)

To test the page numbers, open the new .azw3 file and select Go To > Page or Location. If the .apnx is valid, the Page button should be no longer greyed out and you should be able to select a page by entering its number.

To display the page number while reading a book, tap the lower left corner to toggle between page number, min left in this chapter/book and location.

[HBIT]

Thanks for all the responses.

It is a lot to digest. As I am just starting the book I will have plenty of time to consider it. BTW if I publish with KDP, does that mean that it will automatically have page numbers?

[RbnJrg]

Originally Posted by Doitsu

Hi Ruben,

To be honest, I've only limited experience with this topic. I only toyed it with it when KevinH added the APNX generation feature to KindleUnpack. If you have super-technical questions he's the guy to ask.

Adding a page number basically consists of the following:

1. Adding unique page number targets to the .html files. For example.
<a id="page1" /> (which Sigil will save as <a id="page1"></a>). You can, of course, add ids to all tags that can have one.

2. Adding a pagelist to the NCX file or a separate pagemap file. Since Kindlegen supports both methods and pagemap is an ADE extension, I'd recommend pagelists, unless you're going for cross-plattform compatibility.


AFAIK, yes. Calibre uses UUIDs and my simple KindleGen wrappper (which includes KevinH's dualmetafix_mmap.py script) uses only the last 12 characters of the UUID to generate a fake ASIN and it seems to be working. However, the downside is that if you sideload books with a fake ASIN and an EBOK book type, Paperwhites will not generate a cover, even if the book contains one. (For more background information on this see the dualmetafix_mmap.py thread.)
However, my plugin can optionally also generate a thumbnail image, if you set the corresponding kindlegen.ini parameter. For more details, see the kindlegen wrapper thread. Note that in order to generate thumbnails, you'll need to open a command line prompt as an administrator and enter the following command to download and install the Python Imaging Library.

pip install Pillow

Here's practical workflow example that uses my plugin, you can of course also use Kindlegen and dualmetafix_mmap.py manually.

1. Install my simple Kindlegen Sigil plugin.
2. Change the following kindlegen.ini parameters before running the plugin:

  azw3_only = True
  donotaddsource = True
  add_asin = True

3. Download PW2_pagelist_2.zip and extract pagelist.epub.

4. Open pagelist.epub with Sigil and use the Kindlegen plugin to generate an .azw3 file with it. If all goes well, you should end up with a Page number test [paglist]_e9b1fce6366b.azw3 file and a thumbnail_e9b1fce6366b_EBOK_portrait.jpg file. (Your random ASIN will of course be different.)

5. Run Kindleunpack and unpack the .azw3 file with it using the default settings. Kindleunpack will unpack the file and automatically generate mobi8-Page number test [paglist]_e9b1fce6366b.apnx. Rename this file so that it has the same name as the .azw3 file, but an .apnx extension. I.e. Page number test [paglist]_e9b1fce6366b.apnx

6. Copy the .azw3 file to a folder in the Kindle /documents folder and the thumbnail .jpg file to the (hidden) /system/thumbnails folder. Create a new folder in the folder that contains the .azw3 using the same name as the book but with the extension .sdr. In my specific example, I'd need to create Page number test [paglist]_e9b1fce6366b.sdr. Copy the .apnx file into this folder.

For example if you put the .azw3 file in the folder /document/test, it should contain the following:

Page number test [paglist]_e9b1fce6366b.azw3
Page number test [paglist]_e9b1fce6366b.sdr

(The Page number test [paglist]_e9b1fce6366b.sdr folder should contain Page number test [paglist]_e9b1fce6366b.apnx.)

To test the page numbers, open the new .azw3 file and select Go To > Page or Location. If the .apnx is valid, the Page button should be no longer greyed out and you should be able to select a page by entering its number.

To display the page number while reading a book, tap the lower left corner to toggle between page number, min left in this chapter/book and location.

Many, many thanks Doitsu!! Very very clear your answer; for the rest, you know a lot about this stuff The best for you.

[Doitsu]

Originally Posted by HBIT

Thanks for all the responses.

It is a lot to digest. As I am just starting the book I will have plenty of time to consider it. BTW if I publish with KDP, does that mean that it will automatically have page numbers?

Amazon will display an estimated page number count on the product page, but the Kindle book will only have so-called "locations."

[Hitch]

Originally Posted by Doitsu

Amazon will display an estimated page number count on the product page, but the Kindle book will only have so-called "locations."

To be clear, NO: the published-at-Amazon Kindle book will NOT HAVE the results of your efforts, which is why, in Doitsu's first reply to you, he told you to save yourself the hassle. The APNX plug-in at Calibre only works for those books that will NOT be published at Amazon; that you plan to distribute privately, or off your own website, or use yourself on your own device.

Hitch

[HBIT]

Originally Posted by Hitch

To be clear, NO: the published-at-Amazon Kindle book will NOT HAVE the results of your efforts, which is why, in Doitsu's first reply to you, he told you to save yourself the hassle. The APNX plug-in at Calibre only works for those books that will NOT be published at Amazon; that you plan to distribute privately, or off your own website, or use yourself on your own device.

Hitch

Thanks, I did not understand that. So, there is no way to have page numbers on a book that is published by Amazon?

[cybmole]

Originally Posted by HBIT

Thanks, I did not understand that. So, there is no way to have page numbers on a book that is published by Amazon?

well clearly there is because amazon sell some kindle books which do show page numbers.
so the question is what tools do the publishers of those books use to put them there?

IIRC there was some reverse engineering done when A introduced page numbers, & those numbers were found to live in a separate file, but you'd need to search the boards - maybe the calibre or kindle ones - for more info.

[Hitch]

Originally Posted by cybmole

well clearly there is because amazon sell some kindle books which do show page numbers.
so the question is what tools do the publishers of those books use to put them there?

IIRC there was some reverse engineering done when A introduced page numbers, & those numbers were found to live in a separate file, but you'd need to search the boards - maybe the calibre or kindle ones - for more info.

The difference isn't the tools to put them there; it's the way the books are uploaded to Amazon that makes the difference. BPH's do not use the KDP to upload books. Just as--if you tried to upload an audio/video-enhanced MOBI, the audio and video portions would be stripped via the KDP upload, whereas BPH's can patently have that in their books--you can't have an APNX that will survive uploading at the KDP. BPH's use FTP direct-to-server for their uploads, with zero pre-processing; KDP books do not; they are ALL reprocessed in the PW (Publishing Workflow) after the publisher hits "save and publish."

After Amazon had started really moving books on the KDP, and the original, more-tecchie publisher had petered out, so to speak, they discovered how onerous the unsupported tech support burden was, in supporting the DIY publishers with everything from why their indented paragraphs didn't work to why their covers showed up twice, etc., etc., etc. They were surprised at the overhead. So they've made very deliberate decisions to keep the DIY-KDP process as simple as possible--so as to not add further to that already massive support overhead--for support, remember, that they NEVER intended to provide.

Thus, I wouldn't expect to see Amazon allow more-advanced features to self-pubs any time soon, because for every poster here who can figure out how to do a pagelist correctly, there will be 1,000 who can't, and who (wrongly) expect Amazon to fix it for them when it doesn't work. I don't say that to sound discouraging or disparaging--just factual.

HTH,
Hitch

[Tex2002ans]

Originally Posted by HBIT

Thanks, I did not understand that. So, there is no way to have page numbers on a book that is published by Amazon?

Most likely what occurs, is similar to Audio/Video support.

Self-publishers, and all of the "small publishers", are forced to go through KDP. Using this method, you are not able to create books with page numbers.

Only the "big publishers" are able to add those into their books. They are able to upload their books directly to Amazon via FTP, and are able to go through a completely different system.

Most likely, what you have spotted, is one of these books with "page numbers" in them, which was created by a large publisher (Random House, Macmillan, etc. etc.).

Edit: I see Hitch already answered above while I was typing.

[Hitch]

Originally Posted by Tex2002ans

Most likely what occurs, is similar to Audio/Video support.

Self-publishers, and all of the "small publishers", are forced to go through KDP. Using this method, you are not able to create books with page numbers.

Only the "big publishers" are able to add those into their books. They are able to upload their books directly to Amazon via FTP, and are able to go through a completely different system.

Most likely, what you have spotted, is one of these books with "page numbers" in them, which was created by a large publisher (Random House, Macmillan, etc. etc.).

Edit: I see Hitch already answered above while I was typing.

Well, THAT is a first! I have never been able to outtype you, Tex! I doubt very many can. ;-)

Hitch