Object Orientation in arrays

[David Bovill]

On 22 Nov 2005, at 12:40, xavier.bury at clearstream.com wrote:

> David,
>
> The thing with XUL or the basis for this question is to just  
> translate XML
> type data or streams of data
> to memory for faster management....
>
> I also tried to work with xml in a recent project and as soon as i  
> had 500
> objects in this text variable
> things started slowing down drastically... Which is what i would  
> like to
> avoid.
>
> Since puting arrays into arrays is out of the question in rev  
> (though it
> would be idea), and XML does
> require a bit more parsing than arrays, there must be an easier, or  
> more
> efficient method...
>
> BTW, i did try xml into an array, and the problem is still parsing  
> xml...
> Parsing Arrays is much more
> efficient...

Yes - would be a problem for large documents - but XUL and web  
services are not that large? For anything larger I guess the only  
choice is a database - Valentina comes to mind due to its clean  
object-relational mapping? Maybe Ruslan can give some pointers?

However in general, as I am not strong in database design, I do not  
know how to structure such dbs with variable nesting - for these  
reasons I like to shift anything that complex to the domain of open  
source projects that have spent years tackling it and call them with  
web services?

NB - XUL (XML user interface language) is used by Mozilla to define  
in an open way how rich applications lay out their interfaces. This  
is another example of how Revolution can be used as a key component  
of an open source strategy.