[osis-editors] More Comments

[Jim Schaad]

Gentlemen,

Two issues, one request for  a clarification and one just a please help
me.

1.  I think it would be worthwhile to add a section on "conical"
ordering of elements in an OSIS document.  I was looking at a pargraph
that was an entire verse and an entire quote.  According to my
calculations the arrangement of these three elements alone (without the
problems of adding notes, titles and other such things) is that there
are 24 different encodings to start the paragraph with.  This is just
too much different ways for my poor style sheet to handle, although not
a huge issue for the programmic methods of doing display that can build
and keep state while crossing the different items.

2.  I am looking at trying to get my quote marks correct.  I believe
that the way to determine this is as follows:

A) obtain the correct root for this node - this would either be the
osisText/div or div[@type="book"] ancestor of this node.

B) obtain all q and speech nodes that are predecessors of this node,
including this node.

C) remove all q[@type="block"] nodes from the list

D) Level = count of nodes with sID - count of nodes with eID

E) Add to Level the number of speech and q nodes where @type!="block" in
the ancestor tree.

F) Emit no quote if Level== 0, " if Level==odd, ' if Level==even


IS THERE AN EASIER WAY TO COMPUTE THIS?

3. Possible addition for consideration:  To facilitate the extraction of
chapters from the text, it would be nice to be able to add <q> and
<speech> elements at the beginning of divisions (such as chapter) in
order to have history of these elements moved forward.  I would suggest
tagging them with an attribute of cID that matches the correct sID and
eID.

jim