[
Lists Home |
Date Index |
Thread Index
]
* Daniela Florescu <dflorescu@mac.com> [2004-12-27 16:31]:
> > I've gotten pretty good at assembling a SAX engine quickly by
> > composing one using resuable strategies.
>
> Alan,
>
> you didn't answer my question. Besides the personal interest and
> fun, where there any technical reasons why Xquery/XSLT wouldn't
> have been the right choice ?
There are reasons for programming besides personal intereset
and fun? Explain.
> Instead of writing Java over SAX, isn't it more convenient
> to write:
> for $x in //foobar
> where some:predicate($x)
> return my:function($x)
I don't know XQuery. There might be some genius in that
statement that I'm missing, but...
I can't see where I'm supposed to put logic that invokes Java
reflection in the above statement. Now I have to write a huge
switch statement explictly in my:function, or implicitly in the
XQuery file.
> and leave the smart streaming strategies to implementors to figure
> out ?
My streaming strategies are pretty dumb.
The core strategy is to first choose an XML expression of the
task at hand, so that as the events arrive, the next event is
the next thing that has to be done.
I made the my XML as explicit as possible, so my SAX processor
could be as simple as possible. I did so, under the assumption
that if there was a more concise way to express the same thing,
then I could use XSLT.
I use XSLT to generate my Ant files from a more document like
project description for example.
> Did you try XQuery/XSLT and fail ? If yes, where did they
> fell short ?
Like I said. I didn't want the document. I wanted the parser.
I don't want to generate a document, I want to generate action.
Cheers.
--
Alan Gutierrez - alan@engrm.com
|
