[sword-devel] Alternate versification ....

[Chris Little]

On Feb 4, 2008, at 3:33 AM, Barry Drake wrote:

> Jonathan Morgan wrote:
>> If you look in the archives of this mailing list you can see some
>> things that people have been persistently asking for (alternate
>> versification sprints to mind...)

I don't want to be discouraging, but given the context of this, I  
think it's important to say this. Tasks have varying levels of  
difficulty, and tasks as deeply internal and integral to Sword as this  
one will require a certain level of experience with the API itself. It  
may be better to work on a more limited task or assist with a frontend  
for a while before jumping right into what could be a fairly difficult  
task.

> The recent thread on flexible versification seems to have died.  I had
> hoped that Chris would point us to a spec. somewhere.  Chris?

What's to spec? We have GenBooks. When you run an OSIS module through  
a xml2gbs, you get a GenBook Bible (like the DRCgb I posted last  
week). We have versekey modules. We just need to make the versekey  
interface talk to Bibles stored as GenBooks.

A first pass should just get KJV-type Bibles working via the GenBook  
driver. A second pass might extend that to work with only 66-book  
canon books, but permit different chapter/verse counts. A third pass  
might extend that to permit additional books. (We have an exhaustive  
list of canonical books and a nearly exhaustive list of extant non- 
canonical books within the genre, so a static, but extendable, list  
could be utilized.)

Troy has some additional implementation ideas, I believe, if he wishes  
to post them.

>
> The suggestion that a flexible canon be supported only in general  
> books
> raises one concern.  It's terribly slow!  For this reason, no atttempt
> has been made to work with general books in SwordReader so far.

I'm not aware of their being slow. What's your basis for saying that?  
Be careful not to confuse access times required by the module drivers  
themselves with particular UI implementations.

They will be slower. Any dynamic system will be slower than a  
corresponding static system. I don't think a user will necessarily  
notice the difference in the course of normal usage.

--Chris