Re: Quran Topic Indexing

[Riyad Preukschas <rpreukschas at web dot de>]

Assalamu Alaikum,


Ossama Khayat wrote:
> Nice work. Comments:
> 1. In general, why don't we use Verse for Aya (not
> Ayat as it's plural), and Chapter for Sûra.

OK, my mistake. I changed all "ayat"s to "aya" and all "ayats"s to 
"ayat" to make it correct.

If you are referring to the XPath expressions they are dependent on the 
structure of the XML file you are applying them to.

Example:

//sura[ at nr='2']/aya[ at nr='255']

This matches all "aya" elements which have an attribute named "nr" with 
the value "255" and a parent element named "sura" with an attribute 
named "nr" with the value of "2".
If we apply this rule to a quran file there will be exact one match.
because the structure is like this:
<quran>
	<sura nr="">
		<aya nr=""></aya>
	</sura>
</quran>

if we had a source like this:
<book>
	<chapter nr="">
		<verse nr=""></verse>
	</chapter>
</book>

we would have to change the XPath expressions to match the content of 
the file.



> 2. What does version (in index) stand for?

It stands for the version of the document structure.
This gives application programmers the possibility to check whether 
their app supports all features provided with this file. So we have to 
change the value every time we add another feature (after release) that 
is incompatible to prior versions.


Salam,
	Riyad Preukschas