Two people have built separate sets of scripts to produce an ebook based on Fairfield Project contents.
2008 version
Download the Fairfield Project as an ebook on the Amazon Kindle here. Please note this version, by Jürgen Hubert, contains wiki articles from 2008, nothing more recent.
2012 automated export
A work in progress by BookMan.
- Current (2013-02-17) state:
- preview export:
- please don't upload elsewhere, because this is not final!
- History:
- 2012-12-09: Image caching module working, mirroring process optimized
- 2012-10-28: ebook generation finished, now working on all the special cases
- 2012-10-14: refactored whole mirroring process into classes, to enable use for other backends (besides ePub generation)
- 2012-10-08: mirroring (with changes update function) and basic page parsing done
- Todo:
- structure generation (might need tags)
- stylesheet special cases
Brainstorming
I think the easiest way to get a book like structure is to tag 'section' pages (those that are in the new menu bar?) with a special tag book:section, which then serve as root pages for everything below. This gives a more booklike result, than a mere concatenation of pages.
- perhaps an book index page like nav:top would be even easier. And it gives the ability to provide a sequence.
- This ebook-index will also hold configuration information. That way the whole export can be configured via that wiki page, and now fiddeling with unaccessible config files is needed.
- is this bad? I don't think so… the wiki needs a user and misbehaving users can be blocked, right?
Pages that should not go into the book could be tagged _ebook-ignore (already done this for this page). In addition special pages are also ignored.
unformatted pages could also be skipped… I will see how I can include them into the book.
I also think that we need to rethink the timeline. I made a sortable timeline module for the MoinMoin wiki, perhaps something semilar can be found for this thing. But this is something I would start after a basic book export is finished.
Need to detect page redirects! Look for wikirequest in header of page html, example st-jerome-in-japan.
WIKIREQUEST.info.requestPageName = "order-of-the-sword-of-st-jerome";
Our choice to move fixed pages that don't have a live version (mostly this would be fiction) to the archive: category requires me to change the mirroring so this pages are found. Currently all archive: pages are ignored.
Structure
Test for the structure