Embedding fonts - strange issue

Originally Posted by slantybard

So I tried doing this with Calibre, but it does not work. I tried using negative margins in the p tag using various px and pt from 10 to 40 with no change. There must be something magical about KindleGen. I haven't learned how to use kindlegen yet but it was disappointing to see that the ebook mothership, Calibre, couldn't handle the job.

I am going to look at the provided samples here and compare css codings with the attempts I made.

I don't use calibre to do conversions because it does its own thing. I prefer to have total control. Calibre tries to dumb things down from the users perspective so they don't have to know what they are doing to convert a book. That is fine, but it means tricks like this one tend to not work with calibre.

[Blossom]

Originally Posted by jswinden

I don't use calibre to do conversions because it does its own thing. I prefer to have total control. Calibre tries to dumb things down from the users perspective so they don't have to know what they are doing to convert a book. That is fine, but it means tricks like this one tend to not work with calibre.

You got me curious, If you don't use Calibre how do you make a Kindle book that is KF8?

[Katsunami]

Kindlegen and Kinde Previewer create a mobi that has both old mobi and new KF8 in it.

[Blossom]

Originally Posted by Katsunami

Kindlegen and Kinde Previewer create a mobi that has both old mobi and new KF8 in it.

How do you do that with Kindle Previewer?

[Katsunami]

Just by opening an epub with it

[Blossom]

Originally Posted by Katsunami

Just by opening an epub with it

Did not know you could that!

[Katsunami]

If you'd like to convert a huge number of books, it would be best to use kindlegen itself. Kindle Previewer uses this in the background.

When I get at my desktop computer, I'll post a zip-file with a mini-script that creates AZW3 files using KindleGen and KindleUnpack, saving the original EPUB and intermittent mobi KF8:Joint file as well.

Warning in advance: there's no error-checking. If the script finishes, you should have the same number of EPUBs, MOBI's and AZW3s. If you don't, you'll have to check the logs and/or compare the output directories to see which book didn't convert. Even though the script saves the original EPUBs, I highly recommend running it on a copy of your files.

[Katsunami]

Conversion using Kindlegen: usage instructions (Windows only)
- Download the attached zip-file.
- Open it in your favorite archiving program, and put the folder "KindleScript" somewhere. C:\ would be a good place.
- If you don't have Python installed, download version 2.7.5

http://www.python.org/download/

Pick the X86-64 version if you have a 64-bit Windows.
- Install Python.
- Now open Kindlegen.bat (Found in the KindleScript folder) in your favorite text editor.
- Edit this line:

"C:\Programs\Python 2.7.5\python.exe" "c:\KindleScript\lib\kindleunpack.py" -s "%%f" ".\Unpacked\%%~nf"

- Replace C:\Programs\Python 2.7.5\ with the location where you have installed Python. (Example: C:\Program Files\Python\)
- Replace C:\KindleScript\ with the location where you have put the KindleScript folder.
- Save the file.

The above is a one-time setup.

Now collect some EPUBs in a folder on your desktop. Copy them there, or save them there from Calibre. Convert as follows:

- Press and hold SHIFT.
- Rightclick the folder with the EPUBs, and click "Open Command Window Here".
- Type the following:

c:\kindlescript\kindlegen.bat

(Replace c:\kindlescript\ with the actual location where you put the KindleScript folder.)

- Wait.
- Pray.

If you doubleclick the folder that contains the EPUBs, you should see the script processing the files. In the end, you should have the following:

1. OriginalEPUB (the original EPUBs you started with)
2. OriginalMobi (the KF8:Joint combo-files that KindleGen generates)
3. AZW3 (the AZW3 files extracted from the KF8:Joint combo-format)
4. Unpacked (the stuff where everything is extracted to get the AZW3.)

If you started with 10 EPUBs, you should end up with 10 files in OriginalEPUB, 10 files in OriginalMobi, 10 files in AZW3, and 10 folders in Unpacked. If stuff is missing, a file did not convert correctly.

As I wrote before: this mini-script is for my personal use. Run it at your own risk, and use one COPIES OF YOUR FILES. If you run it on the folder where you store all your bought EPUBs, and something goes wrong, you could end up with an empty folder. There is NO protection for errors in here.

[Blossom]

Originally Posted by Katsunami

As I wrote before: this mini-script is for my personal use. Run it at your own risk, and use one COPIES OF YOUR FILES. If you run it on the folder where you store all your bought EPUBs, and something goes wrong, you could end up with an empty folder. There is NO protection for errors in here.

K for you! I can't wait till I am better to play with this! I already have python installed so it should work okay.

[Katsunami]

Thanks.

If it's Python 2.7.5, it should. The script depends on KindleUnpack.py, and I have not tested that on any other Python than 2.7.5.

Don't forget to change that line to point to the correct install locations of Python and the script.

After I'm done with my Linux conversion, I may make one for Linux too. It's no more than some loops to run along the files, really.

[Barty]

What kind of file is azw3. It doesn't seem to be a zip file like epub

Originally Posted by Barty

What kind of file is azw3. It doesn't seem to be a zip file like epub

It is a compiled file for Kindle books. You have to unpack it to see the contents.

[JSWolf]

If anyone wants, I can post the CSS to an ePub that I converted to an AZW3 via Kindlegen (then split via KindleUnpack) that has -30px margins for the Kindle Touch.

-30px for the Touch
-40px for the PW1/2

[dordale]

Originally Posted by JSWolf

If anyone wants, I can post the CSS to an ePub that I converted to an AZW3 via Kindlegen (then split via KindleUnpack) that has -30px margins for the Kindle Touch.

I would like to see this--Thanks! I'm trying to learn this stuff--not sure if I'll ever figure it out, but I'm trying.

[Blossom]

Originally Posted by Katsunami

Thanks.

If it's Python 2.7.5, it should. The script depends on KindleUnpack.py, and I have not tested that on any other Python than 2.7.5.

Don't forget to change that line to point to the correct install locations of Python and the script.

After I'm done with my Linux conversion, I may make one for Linux too. It's no more than some loops to run along the files, really.

I plan to work on this on the weekend. I am getting a replacement for the HDX on Monday. Mine is acting up. So I will have time to convert more books.