ばか！バカ！　馬鹿ー月！

kuroneko wrote:Hello!

Right now Im working on adapting CSR into epub for personal use only since I love reading epub books. Can you add them up on your list when done if it's oky with BT? So far i have done 5 volumes.

kuroneko wrote:The files are stored in my drive account: http://goo.gl/bIKvJ

Though I haven't tested it yet on any tablets I think it should work.

• There is actually no need for a TOC page as the .ncx file is the TOC of the ePUB
• Your .opf file is pointing at things that do not exist, like:

  Code: Select all

  <item href="Fonts/_LDS_Plain_Germanica.ttf" id="_LDS_Plain_Germanica.ttf" media-type="application/x-font-ttf"/>
  <item href="Fonts/_LDS_Alpine.ttf" id="_LDS_Alpine.ttf" media-type="application/x-font-ttf"/>
  <item href="Fonts/CalistoMT-Regular.ttf" id="CalistoMT-Regular.ttf" media-type="application/x-font-ttf"/>

• About fonts, it's not really go to force a font-face onto a text. Everyone has their own favorite font for reading and some have problems with specific fonts. And not every reader[software] can change forced layout. So it's up to you.
• Your .opf file is pointing at things that are there but are not used in the ePUB, like in Vol 8:

  Code: Select all

  <item href="Images/CSR_vol08_000a.jpg" id="CSR_vol08_000a.jpg" media-type="image/jpeg"/>
  <item href="Images/CSR_vol08_000b.jpg" id="CSR_vol08_000b.jpg" media-type="image/jpeg"/>
  <item href="Misc/README" id="README" media-type="text/plain"/>

  If not used, it's garbage and only makes the ePUB bigger.
• I think you should use

  Code: Select all

  preserveAspectRatio="xMidYMid meet"

  on the cover image.
• One more thing, why are you using

  Code: Select all

  class="bg"

  on the div for the cover when you didn't specify any CSS for it ?
• On a side note, Sigil 0.7 is out. But it's a bit of a pain in the ass if you use the visual editor to import text... so go for code.

[size=120][b]Generate ePUB here: [url=http://ln.m-chan.org]BTE-GEN[/url][/b][/size]  :D 

Simon wrote: Well, I took a look at them.

First of all, I like the way the ePUBs are done.

But there are some problems.

• There is actually no need for a TOC page as the .ncx file is the TOC of the ePUB
  

  The TOC page is deliberate. When I read ebook, I tend to browse each page instead of summoning the TOC page generated by the reader. This is the reason why I included the biography, title and toc page. I want the epub to be as close to the professional published books as possible. I don't think it hurts to add one even if you have the ncx which all readers such as Aldiko intercepts to display the TOC.

• Your META-INF/container.xml is incomplete, you forgot to add the ref to the .ncx file. That means that the document structure is invalid and the file is not readable.
  

  Sigil handles META-INF/container.xml and to be honest I don't really know anything about this file all I know is thaty it fetches the opf files. What should I do to complete files one?
  

  Code: Select all

  <?xml version="1.0" encoding="UTF-8"?>
  <container version="1.0" xmlns="urn:oasis:names:tc:opendocument:xmlns:container">
      <rootfiles>
          <rootfile full-path="OEBPS/content.opf" media-type="application/oebps-package+xml"/>
  // What to add here? toc?
     </rootfiles>
  </container>

• Your .opf file is pointing at things that do not exist, like:

  Code: Select all

  <item href="Fonts/_LDS_Plain_Germanica.ttf" id="_LDS_Plain_Germanica.ttf" media-type="application/x-font-ttf"/>
  <item href="Fonts/_LDS_Alpine.ttf" id="_LDS_Alpine.ttf" media-type="application/x-font-ttf"/>
  <item href="Fonts/CalistoMT-Regular.ttf" id="CalistoMT-Regular.ttf" media-type="application/x-font-ttf"/>

  This file ".opf" basically links all files. I will try remove those item list (not files) that points to non-existing files. Will try to update but I think this is not that important compared to broken html tags except for validation purposes.

• About fonts, it's not really go to force a font-face onto a text. Everyone has their own favorite font for reading and some have problems with specific fonts. And not every reader[software] can change forced layout. So it's up to you.
  

  The font (Calisto) is not used. The font family is not really linked to any css classes nor inline classes.
  
  I don't know if every reader supports this code like (font-family:'CalistoMT',Georgia,Arial,etc if calisto doesn't exist will try to use the next font, etc) but I agree with you as much as possible I avoid using custom fonts.

• Your .opf file is pointing at things that are there but are not used in the ePUB, like in Vol 8:

  Code: Select all

  <item href="Images/CSR_vol08_000a.jpg" id="CSR_vol08_000a.jpg" media-type="image/jpeg"/>
  <item href="Images/CSR_vol08_000b.jpg" id="CSR_vol08_000b.jpg" media-type="image/jpeg"/>
  <item href="Misc/README" id="README" media-type="text/plain"/>

  If not used, it's garbage and only makes the ePUB bigger.
  

  I included those files in the content.opf because I want each volume to be as complete as possible. Even if those are not used, some users can extract them (7zip) and view them.
  
  I did not made any changes to the quality/size of the images to make the size smaller. As I said before, I originally made these epub adaptations for myself and if you ask me I don't care about files size.

• I think you should use

  Code: Select all

  preserveAspectRatio="xMidYMid meet"

  on the cover image.
  

  My latest epub framework uses "xMidYMid meet" instead of"none" (none will stretch the image based on the reader's screen resolution which makes the image look crap).
  
  Thanks for pointing this one.

• One more thing, why are you using

  Code: Select all

  class="bg"

  on the div for the cover when you didn't specify any CSS for it ?
  

  If you have that class you can always write code for it, modify in the future. All my files uses the same framework and If I wanted to change a certain divider in the future I can always add a code for it.

• On a side note, Sigil 0.7 is out. But it's a bit of a pain in the ass if you use the visual editor to import text... so go for code.

Yes, I recently downloaded it.
Well I code the html files a little bit first before importing to the visual editor. In my line of work I use notepad++ to code and I berate those who uses dreamweaver or such visual editors but since this is a document not a website I use Sigil to do the dirty job even if I don't agree that it uses <p><br /></p> instead of &nbsp; or how it uses 2 spaces to increase indent to the codes.

If you go to my website and view all my project you will see that I really care on how the source code outputs.

I think that's all. Don't feel discouraged because of me and good luck.

Not at all, in fact I thank you for such thorough perspective.

• To my site, more info updates, etc.

• Google Drive

• Volume 13 and 14 doesn't have inline images.

• Volume 10, 11, 15 and 16 are incomplete but worth reading if you're a fan of the series.

denormative wrote:PDFs for v1-v9, v12-v14 in appropriately named folder here: http://www.mediafire.com/folder/6a6tgxa ... Epimetheus

v15 wasn't done since it isn't complete yet, even if it has a 'Full Text' link.