FB2 creation

[Mneme]

Originally Posted by Stephanos

I second the suggestion of using Calbre. Unlike OpenOffice, it can import Epub and PDF, as the original poster specified.

I was excited about the possibility of converting my pdf files to other small-screen-friendly formats, so I gave calibre a shot at converting pdf to mobi or fb2. Unfortunately calibre is very picky with pdf files it can convert - out of 6 I tried (none were DRM), only one was converted successfully, the others failed with some error in one of the pdftohtml.pyo calls. Edit: pdf import success rate is 100% improved as of current calibre release.
I wasn't impressed with the one successful result of the pdf to fb2 conversion, either. For one thing, the resulting fb2 file didn't validate, even after I opened it in another editor and added compulsory elements, like language and genre. It appeared to have invalid tags, which did not conform to the schema. In addition, file structure detection was very poor. The file was one solid block instead of being divided into sections (chapters). Admittedly, I've only tried the GUI interface for conversion, it might be a different story with the command line.

Overall, calibre is a really impressive tool, but its fb2 conversion could be improved.

Edit: two updated versions of calibre later, fb2 conversion is still not up to scratch (IMHO), because converted files fail validation. An invalid file = unpredictable behaviour in readers.

[BobC]

I convert most of my books to FB2 as my experience is that on the BeBook that format provides the best display. In addition the Table of Contents for the book is automatically generated by the reader and using it is much simpler than, say, the Mobipocket, approach..

Most of the time I have used Book Designer 5 to build the books however there are a couple of shortcomings that you need to be aware of when using it. The most frequent problem is that it does not correctly handle the ampersand character (&) which it halfheartedly renders as " amp; " (instead of the HTML entity " & "). No amount of playing around with BD5 seems to overcome this so I now post-process the resultant fb2 output file to correct the problem, changing instances of amp; to & using a text editor. You do have to be careful if doing this that you don't change words like "lamp" followed by a semi-colon to l& so include a leading space before the search and replace string. There is a slight possibility that you could be caught out by a phrase in a document about electrical matters where amp is used as a common shortform for ampere and is immediately followed by a semi-colon but I suspect documents including this would be rare.

The more problematic issue with Book Designer, however, is its inability to generate the correct fb2 code for tables. If you have a book where the display of a table is essential then trying to hack the output with a text editor, while possible, is messy. In fact other fb2 tools seem to also suffer from table-blindness. This is where OpenOffice can come to the rescue. The OpenOfficeOrg add-on OOOFBTools will produce tables that CoolReader (but not FBReader) will render sensibly. OOOFBTools can take a bit of getting used to as all the documentation is in Russian, however a little experimentation and practice will pay dividends. The important thing to remember is that if you want something to appear as a fb2 Title (and thus appear in the automatically generated TOC) then you need to use the style "level 1" (or level 2 etc) not the Title style


BobC

[Kravejs]

Originally Posted by Mneme

If all you need is English documentation for the OOoFBTools-1.7.0 plugin, I could probably probably translate the Help file.

Thank you for you offers!

At this moment I opened English site: http://www.fbtools.org/ where I planed to collect all information about tools to create fiction book files. Moreover I can delegate rights for English part of this site to fix errors and add useful information because English is not my native language. Contact me by email: yegor.chemisov AT fbtools DOT org.

Yegor