RE: XSL controversy

["Didier PH Martin" <martind@xxxxxxxxxxxxx>]

Hi Sebastian,

Didier PH Martin writes:
 > Didier says:
 > Sebastian I would interested to know how you can do that with Both XSL
and
 > Oracle tools (actuals not planed). Be specific on which XSL software
version
 > and which Oracle software version and name. Thanks.
 >

Sebastian said:
I didn't say you could do it today; I was assuming that some future
(production) release of Oracle XSL tools would support it. After all, no-one
says that XSL has to apply only to to the result of parsing a text
file. If all Sean's data was in Oracle, then id("Cuchulain") could be
used to return a record, no?

Since we have _no_ production-ready XSL tools today (necessarily,
since we  have no XSL), I feel entitled to assume everything.....

Didier says:
Optimism is a good thing but according to laws of probabilities you have now
50% chances that this scenario will happen and 50% this scenario won't. (or
if you take different premises this could be different) I may also say that
because my future is not yet happened that I will be as rich as Gate is
:-))). Or that before the end of our life we'll all take a week end on the
moon :-) who knows?

Talking about a would be future do not help us, determine what we can do and
what we can't with the current implementations does.

Now about the premises of XSL and Oracle. You scenario would implied that we
do a query to the Oracle data base, apply on the returned data set (i.e. a
XML document) a XSL script. The XSL script will have to pull data from more
than one document and not only from the returned document. The basis to all
this is to be able to add to the actual query language the capability to
include a XSL script (or nay other kind of script like for instance
Omnimark, Balise or DSSSL) to the returned XML document. So that the
returned document will have the following processing instruction included
<?xml-stylesheet type="text/xsl" href="myscript"?>

Now the question is: what is the probability that Oracle does that? If they
do like for the Sedona project the probability is quite low :-). If they do
like for internet files, it will - but not necessarily on time :-)

Also about your comment on id("Cuchulain"), we have to suppose that the
processing is to be done on a single table and that all elements are
identified by their primarily key. This latter has to be unique. Now we have
to check that Oracle 8 returns a data set with the primarily key embedded as
an element's attribute (no so sure of that, they have a particular XML
document model, not necesarily this one).

Jee, that's a lot of premises. Sebastian, what is the next lottery winning
number? just kidding :-) What we can objectively say now, is that this is
not existing yet. We don't know if that will exist one day in the XSL world.
It is probable that a data base vendor will do it, but not necessarily
Oracle. Oracle may do it but we don't know when. Too much maybes and perhaps
for me.

regards
Didier PH Martin
mailto:martind@xxxxxxxxxxxxx
http://www.netfolder.com

regards
Didier PH Martin
mailto:martind@xxxxxxxxxxxxx
http://www.netfolder.com


 XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list