Page Mapping Using toc.ncx

[lorddon]

In my brief research I've found that adobe's method of page mapping is non standard and frowned upon by epub checkers. I discovered the preferred method is to use <pageList> under <navMap> in the toc.ncx, as described here.

<pageList id="page-mapping">

<navLabel><text>Paper Edition Page Mapping</text></navLabel>

<pageTarget id="page-iii" value="3" type="front" playOrder="82">
<navLabel><text>Page iii</text></navLabel>
<content src="frontmatter.html#pageiii"/>
</pageTarget>

<!-- ... -->

<pageTarget id="page-105" value="105" type="normal" playOrder="192">
<navLabel><text>Page 105</text></navLabel>
<content src="chap5.html#page105"/>
</pageTarget>

</pageList>

My difficulty now is I don't understand how or where you add the anchors to mark where the pages begin in the xhtml files themselves.

[frabjous]

I wouln't bother. Unless someone knows something I don't, or something has changed recently, I'm pretty sure the "preferred" method is not supported by any ePub software.

We've discussed this before.

I imagine however, that you're just supposed to use anchor tags. E.g. <a id="page105">first-word-of-page</a> inside chap5.html.

[charleski]

It would be interesting to know if iBooks supports pageList.

[st_albert]

Originally Posted by charleski

It would be interesting to know if iBooks supports pageList.

Apparently they do. This is from page 13 of the iTunesConnect_PublisherUserGuide.pdf :

Page Mapping
To provide a reading experience similar to printed books, the person reading a book on the iPad should be able to refer to and navigate to a particular page. This is especially useful in a classroom setting when the teacher instructs the students to go to a particular page in a textbook. An NCX (Naviagtion Control file for XML) file lets the reader jump directly to any of the major structural elements of the document (for example, chapter or section) or to a particular page. To provide navigation to a particular page, use the NCX defined pageList. Below is an example:

<pageList>
<pageTarget id=”p1” type=”normal” value=”1”>
<navLabel><text>1</text></navLabel>
<content src=”content.html#p1”/>
</pageTarget>
<pageTarget id=”p2” type=”normal” value=”2”>
<navLabel><text>2</text></navLabel>
<content src=”content.html#p2”/>
</pageTarget>
</pageList>


Further documentation on the pageList can be found here:
http://www.daisy.org/epub/issues/cla...m-requirements

[CrystalEYE]

As far as I understand, you don`t have to use anchors, but simply ids.
e.g. <p id="page105">First paragraph on page 105 in the printed version</p>

[soup]

I think I found the answer to this the other day, I copy and pasted a line of code I thought would work and it did.

I don't have the code on me right now, but if you find an ePub with a working table of contents, maybe what I can remember off hand might be enough, however I'll post what I use up when I get the chance, I found it works remarkably well.

in the book I had with the line of code that worked, the code was around the chapter title, it read something like:

"calibre5_calibre3_toc_id1<title_chapter 1>"

now I KNOW that isn't the exact code, but it should help you find the right code if you have a book that has a working table of contents, also this was an ePub converted from a PDF using calibre.

if you do find the line, the '1' in "toc_id1" can be any number, and only have to be in chronological order, 1, 100, 1000, 10000 will wok just as well as 1, 2, 3, 4 etc.

the "chapter 1" part can read as anything you like, it will read on the TOC the same as that line of the page, but the font will be different.

you can place them anywhere in the page, it doesn't have to be at the beginning.

this worked on my kobo, I don't know what your reader is or if this will work on it, I'm new to this.

hope this helps. I should be able to give the exact code in about 10 hours time after this post.

[soup]

it took me a little longer to get my hands on it for you, but fially, it's here.

<h1 class="calibre7 calibre3 sgc-1 sgc-8" id="calibre_toc_3">Chapter 1</h1>

I just paste that code into any line I want to reference in my table of contents, and it comes up.

someone who knows more than me would probably be able to clean in it up for you, and give you a shorter line of code, but it does the job.

as I said before, the "3" in " id="calibre_toc_3" " is changable, and provided that the numbers are increasing order, they don't have to be in next number order.

the "Chapter 1" can be anything you like, but it will come up on both the line of the page you're referencing aswell as the table of contents, the font will be different on the table of contents.

as I said, this is from an epub that was converted from a PDF with calibre, edited with sigil, and worked correctly on the kobo; just a few points of reference for you.

Originally Posted by CrystalEYE

As far as I understand, you don`t have to use anchors, but simply ids.
e.g. <p id="page105">First paragraph on page 105 in the printed version</p>

I'm trying this, but Sigil doesn't generate the pageList from it so I probably have to add that part. I'm using something like this:

<a id="page014">first word of new page</a>

So far that doesn't cause an error. Then I would have to add something like the following in the ncx file:

<pageList>
...
<pageTarget id=”page014” type=”normal” value=”14”>
<navLabel><text>14</text></navLabel>
<content src=”Chapter01.xhtml#page014”/>
...
</pageList>

Will this work?

Actually, I'm thinking that I don't need the "<navLabel><text>14</text></navLabel>" part. I'm just using the ncx to guide the creation of the APNX file in Calibre. I don't want the page numbers to show up anywhere in the actual text.

[Doitsu]

Originally Posted by freewheeling

I'm trying this, but Sigil doesn't generate the pageList from it so I probably have to add that part.

Since you mentioned that you're using Sigil, you might find my Pagelist plugin helpful, which'll automatically generate a pageList from page marker ids.

BTW, a typical pageList should look like this:

  <pageList>
    <navLabel>
      <text>Pages</text>
    </navLabel>
    <pageTarget id="page7" type="normal" value="7">
      <navLabel>
        <text>7</text>
      </navLabel>
      <content src="Text/book.xhtml#page7"/>
    </pageTarget>
    <pageTarget id="page8" type="normal" value="8">
      <navLabel>
        <text>8</text>
      </navLabel>
      <content src="Text/book.xhtml#page8"/>
    </pageTarget>
    <pageTarget id="page9" type="normal" value="9">
      <navLabel>
        <text>9</text>
      </navLabel>
      <content src="Text/book2.xhtml#page9"/>
    </pageTarget>
  </pageList>

Doitsu:

I actually tried to use that plugin, and have it installed, but I get an error message. I assumed that if I created the ePub with Sigil 9.9 that it would already be in ePub3 but perhaps I'm wrong and I need to run a conversion program on it.

Ahh, never mind. I forgot to change the epub declaration. DUH!

The plugin appears to process the first 19 page numbers in the first chapter and then spits out this:

IndexError: list index out of range
Error: list index out of range

I suspect that what's happening is that I'm identifying the first word in the new page, which isn't a number. So I'm back to my previous method.

Update: I changed the targets to the empty first example and the plugin ran perfectly. However, using the APNX output in Calibre to send the book to my Kindle still doesn't produce accurate page numbers. They're clearly not using the PageList for anything. Maybe KindleCreate does. (I can't find a KindleGen, so KindleCreate is probably the latest version of that.)

[Doitsu]

Originally Posted by freewheeling

The plugin appears to process the first 19 page numbers in the first chapter and then spits out this:

Can you please paste the whole error message verbatim? It'd help me to check for this error, which should only be displayed, if the pagelist and/or the page target markers are incorrectly formatted.

[Hitch]

Originally Posted by freewheeling

I suspect that what's happening is that I'm identifying the first word in the new page, which isn't a number. So I'm back to my previous method.

Update: I changed the targets to the empty first example and the plugin ran perfectly. However, using the APNX output in Calibre to send the book to my Kindle still doesn't produce accurate page numbers. They're clearly not using the PageList for anything. Maybe KindleCreate does. (I can't find a KindleGen, so KindleCreate is probably the latest version of that.)

Are you thinking that Kindlegen is going to give you "real page numbers" by using the Calibre plugin? That won't work. As far as I know, the APNX plugin, for Calibre does nothing but create them for those books being read VIA Calibre, for self-made/self-read books. Not books intended to be processed via the KDP or Kindlegen, etc.

I've never heard that the APNX plugin, for Calibre, can be used for anything in commercial production. Now, I could be utterly wrong, as we don't use Calibre for anything like this (we use it for its intended purpose, as a catalog, silly buggers that we are), so I suppose it COULD work, but I'd be very, very surprised, especially as the KDP very notoriously tends to reject most Calibre-made files as "unsupported." Moreover, you cannot upload an AZW to the KDP, so...???

Hitch

[jhowell]

EPUBs with real page numbers are rare so few tools exist that work with them.

APNX files produced directly by calibre or by using the APNX plugin do not include pages defined in the NCX pageList, they only create approximately positioned pages based on a specified (or estimated) page count for the book.

Kindlegen will process a pageList and save it in the MOBI file produced, but I do not know of any software that can split out this information to produce a separate APNX file as needed for Kindle devices.

The KFX Output plugin for calibre will make use of an NCX pageList to produce KFX files containing real page numbers that can be sideloaded to Kindle devices. (KFX format does not rely on APNX files.)