Kindlegen for mac neophyte needs some help

[petercrowell]

Well, thanks! And thanks for all your help. But I've still got some work ahead of me. Formatting issues (more curly quotes, I suspect.)

Here's what I don't get: I've got a toc.ncx, a book.html, a book.opf, a book.mobi, a couple of .png files and a .jpg. How does all this get into the book?

In the kindle previewer the table of contents is not working and the clickable line at the bottom is not active. Also, my graphics are not showing up.

I think I've got some gaps in my understanding. How does this all hang together?

[1611mac]

just happened across your thread and thought I'd share though I don't understand exactly what you are doing.....

You are converting html? I make nice eBooks by dropping the html into "Pages" (Mac app) , and then using eBook Styles (Apple has a "Best Practices" document to use as a template) to set set main text, headings, captions, then export as an ePub book. (Search for "ePub" in Pages help.) I then place the ePub into Calibre where a click converts it to .mobi for the Kindle app on my iPad.

If this doesn't apply then please ignore, and I'm sorry to butt in.....

[petercrowell]

double post deleted

[ATDrake]

The OPF is the key.

Think of it as both being like a combination shopping list/recipe instruction which says "pick up 1 jar of pickles from aisle 7, 2 tubs of ice cream from the Frozen Foods department, and a 200mL bottle of Pepto-Bismol from the Pharmacy", then "eat one small dill pickle, eat one spoonful each of Maple Walnut Swirl and Wasabi Butternut Crunch, if queasiness results, take 10mL of Pepto-Bismol".

All your html and image files must be declared in the <manifest>. Otherwise the parser won't know they exist and they don't get included in the book. If your images are in a subfolder, that has to be reflected in the href you put in the <item> so it knows where to find them.

Then your <spine> has to list your html files (images not necessary) in the order you want them to appear in, like chapters in a book.

To get the TOC showing, you have to put a line in the <guide> section to let the parser know it exists. I can't remember the exact syntax off the top of my head, but it's pretty simple and the Amazon Kindle Publishing Guidelines has some detailed info on how to do it.

The .mobi is just the "baked" version of the book once you've assembled and fed all the ingredients into the "mixer" which is KindleGen (sometimes "scrambler" might be more appropriate).

The location progress bar in KPreviewer is non-clickable. To simulate the 5-way controller flicking, use the double-arrow buttons on the toolbar.

Hope this helps and wasn't too incomprehensible an analogy.

[petercrowell]

Originally Posted by ATDrake

Hope this helps and wasn't too incomprehensible an analogy.

Helps a LOT. Thanks again!

[petercrowell]

Originally Posted by 1611mac

If this doesn't apply then please ignore, and I'm sorry to butt in.....

No this seems like great advice. But I'm so far down the path I've taken I have to see it through! And I think I'm getting close.

Thanks for your suggestion.

[petercrowell]

Originally Posted by ATDrake

If your images are in a subfolder, that has to be reflected in the href you put in the <item> so it knows where to find them.

So, if all the files are together in one folder, and they are properly referenced in the opf file, they appear as needed?

[1611mac]

Originally Posted by petercrowell

No this seems like great advice. But I'm so far down the path I've taken I have to see it through! And I think I'm getting close.

Thanks for your suggestion.

Actually, it sounds crazy using Pages to make an ePub but it actually works well with simple books. I've done 300-400 page books in 45 minutes but they have just a couple of pics and perhaps 20 chapters. In Pages you simply set all text to "body" style, then go back and set the Chapter Numbers and Heading styles as well as the Title. Export and you have an ePub books with TOC (generated from styles). And Pages is compatible with Microsoft word so you can even generate eBooks from Word docs....

[ATDrake]

Originally Posted by petercrowell

So, if all the files are together in one folder, and they are properly referenced in the opf file, they appear as needed?

Yup. You do have to list each individual image (only once; you can reuse a referenced image as often as you like) and put in their file types.

But as long as that's done and the images are also properly attached within the HTML file (easiest way to check is to drag and drop it into a browser window and see if you have any "broken image" blank spaces), then they should show up.

Also, don't worry about the order that stuff goes into in the <manifest>. You can jumble it up as much as you like. File appearance order is only important in the <spine> section.

[petercrowell]

Originally Posted by ATDrake

Yup. You do have to list each individual image (only once; you can reuse a referenced image as often as you like) and put in their file types.

Right now my images are listed this way at the top of the body in the html file:
<body>
<h1 height="20″><img src="sl_cover" alt="img" /></h1>
<p height="10″>by</p>
<p height="20″><img src="ptc_logo" alt="img" /></p>
<p>&nbsp;</p>
<p height="30″><img src="sl_logo.png" alt="img" /></p>

So, these images need to be listed in the manifest?
Are they listed properly in the html file?

[ATDrake]

Well, technically the alt text is supposed to be for the benefit of the blind and put in a short textual equivalent or description of what your image represents for those who can't see it (e.g. alt="My Book Title").

But aside from that, yes*, and now you need an <item> in the manifest for each image that is expected to appear inside the finished book, with the appropriate media-type="" filled in.

* Assuming your first two images really are named exactly that and are not missing a file type extension which they might have in the actual filenames but not in the version listed in the html.

[petercrowell]

I'll fix those extensions, and I do plan to fill in the alt text. Thanks for pointing that out.

As for designating the correct media type, can you tell me what I'd need to change with the code below to designate an image?

<item id="item1" media-type="text/x-obeb1-document" href="nameofbook.html">

[pdurrant]

Originally Posted by petercrowell

I'll fix those extensions, and I do plan to fill in the alt text. Thanks for pointing that out.

As for designating the correct media type, can you tell me what I'd need to change with the code below to designate an image?

<item id="item1" media-type="text/x-obeb1-document" href="nameofbook.html">

I think you might be better off using Sigil you do your base editing and adding of pictures, etc. Then the opf will be created correctly for you, and you can move on to any manual editing you might need.

Trying to edit the opf, xhtml and css manually with your level of knowledge is likely to lead to lots of problems.

Alternatively, you could use Sigil to generate sample code that you then adapt into your manual code.

[ATDrake]

Originally Posted by petercrowell

As for designating the correct media type, can you tell me what I'd need to change with the code below to designate an image?

The standard mimetypes for png and jpeg are "image/png" and "image/jpeg", respectively. And the id needs to be unique, but you probably knew that already.

[petercrowell]

Originally Posted by pdurrant

Trying to edit the opf, xhtml and css manually with your level of knowledge is likely to lead to lots of problems.

But I'm so close! I can feel it!

Plus, hashing through it like this helps me learn. Subsequent ebooks will benefit from this experience.

I will check out Sigil, however. Thanks for the tip!