[sword-devel] Jub oddities

[Chris Little]

Karl Kleinpaste wrote:
> There are a bunch of things that seem peculiar about content at
> upper-level keys.
> 
> First, why is there a single top-level key "/Jub"?  If there is no
> actual division at that level, why does that level exist?

Standard references to the book of Jubilees will appear as "Jub 1:1", 
which should translate to an osisID of Jub.1.1, which will be 
transformed to a TreeKey of /Jub/1/1.

A more natural context for the book of Jubilees would be in a collection 
with other similar material. And we'll probably have such a collection 
some day, so you might expect to call up a key like Pseudepigraph:Jub.1.1.

> Bible references are causing minor difficulty.  /Jub/2 contains:
> <reference osisRef="Gen.1-2">[=Gen.1-2]</reference>
> But GS cannot get a verse list out of this.

That's an encoding error. The osisRef ought to be Gen.1-Gen.2.

> The <title> content at the /Jub/DIGITS level is very inconsistent.  Many
> contain strings such as "[=4Q216]" in varying degrees of complexity
> (e.g. /Jub/23) along with a Gen reference, many contain *only* a Gen
> reference (e.g. /Jub/6), and sometimes apparent verse content bleeds
> into these sections (e.g. /Jub/35, /Jub/38).  Is this intended?

The "[=4Q216]" type titles identify corresponding Qumran documents. They 
should probably have osisRefs, e.g "DSS:4Q.216". The Genesis references 
identify parallel passages in Genesis. Those are intended.

The content of /Jub/35 and /Jub/38 looks like a conversion error since 
these chapters lack content in verses 3 and 6 respectively, precisely 
the verses that are appearing in the chapter content.

I'm not sure whether I have the source file for this, so I don't know 
whether I can fix the problems immediately.

--Chris