Bookshelf | Reports | Community | KDP Select

Home » Amazon KDP Support » Ask the Community » Formatting

Thread: Any LibreOffice users also have working toc.ncx Table of Contents ?


This question is not answered. Helpful answers available: 2. Correct answers available: 1.

Reply to this Thread Reply to this Thread Search Forum Search Forum Back to Thread List Back to Thread List

Permlink Replies: 5 - Pages: 1 - Last Post: Jun 17, 2012 4:18 PM Last Post By: cjeasyaspie
jeyw

Posts: 9
Registered: 06/08/12
Any LibreOffice users also have working toc.ncx Table of Contents ?
Posted: Jun 11, 2012 12:26 AM
 
Click to report abuse...   Click to reply to this thread Reply
I'm not sure exactly how to phrase my question as I've gone from being a complete novice to devouring more tech and programming stuff in 48 hours than I've done in many years. So, being that my head is still swimming with a ton of new stuff, maybe someone can help me clarify the following:

I've successfully completed an e-book and cleaned it up using LibreOffice. I saved as a .doc file, which actually came out truer to format than the html file I first tried.

Within the .doc file, I was able to create a working, linked Table of Contents. As such, my book is currently uploaded with everything except the toc.ncx navigation from the Go To menu within Kindle. I also do not have a Cover to navigate to, but am unsure if this is also part of the html (toc.ncx) I'm missing, or if it's because I emailed a .doc file to my Kindle instead of downloading a book from the Kindle Store?

Anyway, apart from the cover, I've done enough reading to know I have to have the toc.ncx to get the navigational TOC from the Go To menu. However, I'm still confused about KindleGen.

What exactly is its purpose? Is this what I [i]must [/i]]have to create the TOC correctly and if so, does this mean I will be forced to use the html file of the text, which did not format as well as the .doc?

Also, when creating a toc.ncx file and putting all through KindleGen, does KindleGen then put the e-book document file and the Toc file together to create one, single file to upload as your book?

I've also read a lot about Mobi and Calibre but am on information overload and can't quite sort out if these are all the same thing or if KindleGen is different or what the purpose of each one is.

Thanks
mrlasers

Posts: 1,633
Registered: 07/06/11
Re: Any LibreOffice users also have working toc.ncx Table of Contents ?
Posted: Jun 11, 2012 11:18 AM   in response to: jeyw in response to: jeyw
 
Click to report abuse...   Click to reply to this thread Reply
KindleGen is an eBook compiler for Kindle's "mobi" format. It processes the input source file(s) and produces a single binary file containing the final eBook that contains both mobi7 and KF8 formatted versions of the book, along with any auxiliary features like NCX.

The mobi file that it produces is what you would upload to KDP as your book contents. Many users who are less familiar with commandline interfaces prefer to use Amazon's Kindle Previewer, which will use an integrated copy of KindleGen to convert an HTML, OPF, or ePub file.

The NCX table of contents will not be automatically created when converting from HTML. However, it's fairly straight forward to include this manually by providing an OPF and NCX file for your HTML document. I've attached examples of both of these files, which you can edit for your specific needs. The navPoint structure in the NCX example can be copied and edited for each point that you want referenced within the book. You should be able to copy the title and src values directly from the HTML version of your TOC and paste them into the appropriate section of the NCX navPoint.

A basic overview of how these files work is included in the Kindle Publishing Guidelines PDF, or you may wish to read up on their implementation for ePub 2.0, which will be source-compatible with the requirements for Kindle.

--
MrLasers
eBook Formatting & Production
http://www.mrlasers.com
cjeasyaspie

Posts: 2,378
Registered: 05/28/09
Re: Any LibreOffice users also have working toc.ncx Table of Contents ?
Posted: Jun 12, 2012 8:55 AM   in response to: mrlasers in response to: mrlasers
 
Click to report abuse...   Click to reply to this thread Reply
The NCX table of contents will not be automatically
created when converting from HTML.

MRLasers,

Do you have any idea as to why Amazon does not include the NCX when converting from HTML?

It's entirely possible for them to do it... as you know, if you, in the past, uploaded an MS Word DOC directly to KDP.

As recently as 3 weeks ago, this method produced a very nice MOBI, complete with NCX pointers, and without over-riding block paragraphs with first line indented paragraphs.

With that feature in place, there was no reason for anyone to resort to 3rd-party programs such as Calibre.

Unfortunately, Amazon has recently changed the software, and gone back to the original... with forced first line indents and no NCX.

Very disappointing!

CJ, at CJ's Easy as Pie Kindle Tutorials
http://www.cjs-easy-as-pie.com/
cj-01@cjs-easy-as-pie.com
jeyw

Posts: 9
Registered: 06/08/12
Re: Any LibreOffice users also have working toc.ncx Table of Contents ?
Posted: Jun 15, 2012 8:15 PM   in response to: mrlasers in response to: mrlasers
 
Click to report abuse...   Click to reply to this thread Reply
Many users who are less familiar with commandline interfaces prefer
to use Amazon's Kindle Previewer, which will use an
integrated copy of KindleGen to convert an HTML, OPF,
or ePub file.

Thanks. I think I understand KindleGen now - basically a program that combines files to make one uploadable file, correct?

However, you lost me at the line that starts "many users who are less familiar with commandline interfaces..."

Did you mean "prefer to use" as in "in place of KindleGen" or am I reading that the wrong way? I have used the Kindle Previewer, though not with a TOC attached and am unaware of any way to combine several files into one with the Previewer.

Reading further into your reply, I understand tweaking the provided html samples to use my own reference points, but did you still mean to create these separate files and then run everything through KindleGen to make one file?

Since I have not yet looked at KindleGen myself, I'm assuming that there is a place to attach several files and then a process by which it all becomes one file.

I guess what I am trying to glean from your post is whether the bottom line is, if I want a TOC, I have to use KindleGen to generate a single binary file.

BTW, thanks for taking the time to attach the samples. Appreciate it!
mrlasers

Posts: 1,633
Registered: 07/06/11
Re: Any LibreOffice users also have working toc.ncx Table of Contents ?
Posted: Jun 16, 2012 9:52 AM   in response to: cjeasyaspie in response to: cjeasyaspie
 
Click to report abuse...   Click to reply to this thread Reply
@jeyw:
Kindle Previewer integrates a copy of KindleGen, so those who are not familiar or comfortable with a commandline interface (as KindleGen has) can use it to convert their content in the same manner. If you want to combine multiple files (NCX, additional HTML, etc) into a Kindle book you will need to prepare an OPF file for your publication, which is then run through KindleGen (or Kindle Previewer) in place of the HTML.

While you could also zip these files together and upload to KDP, I don't really see the point. The only thing you're gaining there is the small bit of uncertainty as to whether Amazon may or may not in the future alter the conversion process in a way that will cause your content to appear differently than you intended. With a compiled mobi file you don't need to worry about the content getting re-converted.

@cjeasyaspie:

Do you have any idea as to why Amazon does not
include the NCX when converting from HTML?

Because there is no NCX in HTML? How would they guess what should and should not be in the NCX? I find your question a bit baffling as a premise.

It's entirely possible for them to do it... as you
know, if you, in the past, uploaded an MS Word DOC
directly to KDP.

It should be entirely possible to mirror the automatic TOC when converting a Word document. But the TOC created in that manner is part of the Word DOM (you can walk it in VBA if you want a quick and dirty way to see this), which should make such work trivial.

I do not see how this applies to HTML.

As recently as 3 weeks ago, this method produced a
very nice MOBI, complete with NCX pointers, and
without over-riding block paragraphs with first line
indented paragraphs.

They had a similar automatic conversion for NCX about a year ago, when they first silently enabled DOCX support. I would guess that they never take this further than short tests because it does not apply uniformly enough to the content that they are ingesting. I for one would not want my TOC automatically mirrored into NCX, but I can see how many would find that a huge upgrade from what they're currently able to do themselves.

Suppressing the first line indent (which is actually a feature of the Mobi7 reader, [i]not[/i] the converted content itself) requires making assumptions about the content, how it was formatted, and how it was intended to display. While I should think that implying [b]p { text-indent:0; }[/b] across all ingested content should allow for it to appear as intended in nearly all cases, I suppose it's possible that I'm ignoring a large number of possibly common use cases that I simply haven't seen.

Since my process does not involve Word's HTML exporter at all, you will hopefully understand how my research into this area has been far from robust in recent months.

With that feature in place, there was no reason for
anyone to resort to 3rd-party programs such as
Calibre.

There is [i]never[/i] a good reason to use programs like Calibre in production.

Unfortunately, Amazon has recently changed the
software, and gone back to the original... with
forced first line indents and no NCX.

This is why I would [i]always[/i] recommend, wherever possible, producing content yourself, in the proper way, rather than relying on automated processes to make assumptions about how your formatting should appear. In most cases these automatic assumptions will be themselves based on the initial assumptions of the producer about how Word formatting is handled internally and how that should/will convert to Kindle. Something I've found in most cases is that these initial assumptions are generally wrong. Even among those who have been doing it for a long time.

--
MrLasers
eBook Formatting & Production
http://www.mrlasers.com
cjeasyaspie

Posts: 2,378
Registered: 05/28/09
Re: Any LibreOffice users also have working toc.ncx Table of Contents ?
Posted: Jun 17, 2012 4:18 PM   in response to: mrlasers in response to: mrlasers
 
Click to report abuse...   Click to reply to this thread Reply
Because there is no NCX in HTML? How would they guess what should and should not be in the NCX? I find your question a bit baffling as a premise

There is no need to “guess”… the needed information is in the linked Table of Contents within the MS Word file.

The question was rhetorical… as in “why do you suppose they did that?”… not actually expecting you to know the answer.

This is why I would always recommend, wherever possible, producing content >yourself, in the proper way, rather than relying on automated processes to make >assumptions about how your formatting should appear. In most cases these automatic >assumptions will be themselves based on the initial assumptions of the producer about >how Word formatting is handled internally and how that should/will convert to Kindle. >Something I've found in most cases is that these initial assumptions are generally >wrong. Even among those who have been doing it for a long time.

“Producing content yourself, in the proper way”, is -- contrary to your assumption -- producing content as instructed by Amazon -- with MS Word.

As is the case with many others, you are making it much more difficult than it needs to be.

MS Word is the ideal software to create a Kindle MOBI book, and, in fact, is Amazon’s first recommendation, as can be seen in this excerpt from Amazon’s Simplified Formatting Guide:

=================
Building Your book
Word is a great tool to use because it's extremely easy to format. We suggest writing your book in Word or converting an existing source file into Word (.doc or .docx) format before continuing. Remember these important tips below to ensure an excellent eBook presentation.

https://kdp.amazon.com/self-publishing/help?topicId=A17W8UM0MMSQX6
=================

Note that Amazon goes so far as to suggest, not only “writing your book in Word”, but “converting an existing source file into Word (.doc or docx) before continuing.”

Making things difficult is not always making things better.

Happy Kindling,

CJ, at CJ's Easy as Pie Kindle Tutorials
http://www.cjs-easy-as-pie.com/
cj-01@cjs-easy-as-pie.com


@jeyw:
Kindle Previewer integrates a copy of KindleGen, so
those who are not familiar or comfortable with a
commandline interface (as KindleGen has) can use it
to convert their content in the same manner. If you
want to combine multiple files (NCX, additional HTML,
etc) into a Kindle book you will need to prepare an
OPF file for your publication, which is then run
through KindleGen (or Kindle Previewer) in place of
the HTML.

While you could also zip these files together and
upload to KDP, I don't really see the point. The only
thing you're gaining there is the small bit of
uncertainty as to whether Amazon may or may not in
the future alter the conversion process in a way that
will cause your content to appear differently than
you intended. With a compiled mobi file you don't
need to worry about the content getting
re-converted.

@cjeasyaspie:

Do you have any idea as to why Amazon does not
include the NCX when converting from HTML?

Because there is no NCX in HTML? How would they guess
what should and should not be in the NCX? I find your
question a bit baffling as a premise.

It's entirely possible for them to do it... as you
know, if you, in the past, uploaded an MS Word DOC
directly to KDP.

It should be entirely possible to mirror the
automatic TOC when converting a Word document. But
the TOC created in that manner is part of the Word
DOM (you can walk it in VBA if you want a quick and
dirty way to see this), which should make such work
trivial.

I do not see how this applies to HTML.

As recently as 3 weeks ago, this method produced a
very nice MOBI, complete with NCX pointers, and
without over-riding block paragraphs with first
line
indented paragraphs.

They had a similar automatic conversion for NCX about
a year ago, when they first silently enabled DOCX
support. I would guess that they never take this
further than short tests because it does not apply
uniformly enough to the content that they are
ingesting. I for one would not want my TOC
automatically mirrored into NCX, but I can see how
many would find that a huge upgrade from what they're
currently able to do themselves.

Suppressing the first line indent (which is actually
a feature of the Mobi7 reader, [i]not[/i] the
converted content itself) requires making assumptions
about the content, how it was formatted, and how it
was intended to display. While I should think that
implying [b]p { text-indent:0; }[/b] across all
ingested content should allow for it to appear as
intended in nearly all cases, I suppose it's possible
that I'm ignoring a large number of possibly common
use cases that I simply haven't seen.

Since my process does not involve Word's HTML
exporter at all, you will hopefully understand how my
research into this area has been far from robust in
recent months.

With that feature in place, there was no reason for
anyone to resort to 3rd-party programs such as
Calibre.

There is [i]never[/i] a good reason to use programs
like Calibre in production.

Unfortunately, Amazon has recently changed the
software, and gone back to the original... with
forced first line indents and no NCX.

This is why I would [i]always[/i] recommend, wherever
possible, producing content yourself, in the proper
way, rather than relying on automated processes to
make assumptions about how your formatting should
appear. In most cases these automatic assumptions
will be themselves based on the initial assumptions
of the producer about how Word formatting is handled
internally and how that should/will convert to
Kindle. Something I've found in most cases is that
these initial assumptions are generally wrong. Even
among those who have been doing it for a long time.

--
MrLasers
eBook Formatting & Production
http://www.mrlasers.com

Legend
Helpful Answer
Correct Answer

Point your RSS reader here for a feed of the latest messages in all forums