ePub->fb2 (what am I doing wrong?)

[MichaelV]

Hello. I'm trying to convert some of my ePubs into FB2 (in batch, but I don't think it's important - butch or no butch). Here's the problem:
The resulting file FB2 is readable, it includes all images, so far so good, but... It's completely unstructured, that is, no parts, chapters, and any other titles-subtitles tags. So I can't go to, let's say, Chapter 4, and unable to see the structure of the book at all.
On the other hand, ePub is a compressed collection of HTMLs, so it should be easy enough for Calibre to replace HTML headers wih FB2 tags. How should I force the program do what I need and what would be a more or less nice and usable FB2 file?

(I'm sure there's some answer to this.)

Thank you in advance.

[exaltedwombat]

I had to look up FB2 to find out what it was!

I take it you have experience of other FB2 books which ARE structured - i.e. the format WILL do it?

Open your ePub in Sigil. IS it divided into chapters, each a seperate XHTML file? If not, it's easy to split them.

IS there a title at the beginning of each chapter, in a header style? Has the TOC been generated?

[MichaelV]

Originally Posted by exaltedwombat

I take it you have experience of other FB2 books which ARE structured - i.e. the format WILL do it?

The thing is NOT ONE of the FB2 converted from ePub has ever been structured (with Parts, Chapters, etc.)

Here's a snippet from a structured FB2:
<section>
<image xlink:href="#Any2FbImgLoader0"/>
</section><section><title><p><strong>Burglars Can’t Be Choosers</strong></p></title>
<p id="AutBody_0DocRoot">By Lawrence Block</p>
<p>Book 1 in the Bernie Rhodenbarr series</p>
<p>Copyright © 1977 by Lawrence Block</p>
<p><emphasis>For Steve and Nancy Schwerner</emphasis></p>
</section><section><title><p><strong>Chapter One</strong></p></title>

Originally Posted by exaltedwombat

Open your ePub in Sigil. IS it divided into chapters, each a seperate XHTML file? If not, it's easy to split them.

IS there a title at the beginning of each chapter, in a header style? Has the TOC been generated?

Haali Reader doesn't have the TOC, it has an icon allowing to see the structure. And all Parts, Chapters, etc. are in the text in bold and in a larger font, which is not the case when I convert epubs->FB2 in Calibre.
As for checking one of the ePubs in Sigil, there's everything as you wrote above.
1.t is divided into chapters, each a separate XHTML file.
2, There a title at the beginning of each chapter, in a header style.
3. And yes, the TOC has been generated by Sivil.

Still, while converting this ePub in Calibre I'm getting something close to TXT file (with images, though, but it's not an even semi-healthy FB2).

[exaltedwombat]

There's a "Sectionise" option in Calibre's "Convert" page, under "FB2 Output". Mine defaulted to "nothing". You could try some other settings.

[MichaelV]

Originally Posted by exaltedwombat

There's a "Sectionise" option in Calibre's "Convert" page, under "FB2 Output". Mine defaulted to "nothing". You could try some other settings.

I've tried to play with all settings. No go. Sometimes I got 2-3 headers for chapters or subchapters, sometimes none, sometimes the whole text is one endless header.
I think that there's a necessity to choose "Structure Detection" on the left pane, but what to do with the right pane with all XPath expresssions, beats me. I've tried different stuff according to XPath help - nothing works!

[exaltedwombat]

So, what reads a FB2 file? What are the format's advantages? I can't find much about it.

[user_none]

Originally Posted by MichaelV

It's completely unstructured

Look at the sectionize options for FB2. The default if files which puts a section tag around the files that comprise the epub. You might have better luck with sectionizing via toc.

If the file already is split into sections (should be) then it could be something with your reader not supporting splitting on sections.

That said the FB2 output is super advanced. EPUB doesn't translate into FB2 very wel...

Originally Posted by exaltedwombat

So, what reads a FB2 file? What are the format's advantages? I can't find much about it.

It's popular in Russia. There isn't much out there that reads the format. It's an XML based format and is pretty easy to parse.

The major issue with FB2 conversion is it has lots of very specific tags. Such as a tag to denote poetry. This is really easy to turn into nicely formatted html for epub. However, since html doesn't have a way to truly mark the use/intent of a section of text it's non-trival (nearly impossible) to convert a section of text into the specialized FB2 tags.

Some advantages, it's an open format. I believe it predates epub 2. It's also single file. Truly single file. Not a package of multiple files like EPUB. It's very book centric. A reading system should be able to easily understand and display a book in the format very well. But as with any format it doesn't matter how good it is if no one has interest in work with it. Writing an FB2 reader is still going to be a large undertaking.