Christian,
> hm... sounds good as architecture, but doesn't this make it more complex 
> than necessary?What I envision is that the xsl templates could be 
> shipped with the sdk framework and sit there to be invoked only when the 
> developer puts xml files into a  given folder....
>   
Yeah, interesting shift in perspective: You see XSLT transformation as
another potential step in the generator chain, where the generator  -
among the many other things it does - looks for potential XML files that
satisfy a defined schema, applies a standard XSLT transformation to
them, and then goes on to build the project, right?! Then everybody
could pour in valid XML into the build environment, probably tweak a few
build options, and have their project built from XML. (I call it
"qooxdoo-embeds-XSLT").

My perspective was: There are people building an application, which just
happens to be written down in XML (there could be other ways to nail
it); this gets transformed into JavaScript (and there could be a lot of
transformations leading to this result, depending on input data); and
then, at the end of the pipeline, qooxdoo's build process kicks in,
ignorant of all that happened before. (Let's call that
"application-embeds-qooxdoo").

I think both are viable perspectives, and I agree it doesn't look like a
lot of effort to implement your approach. But besides cost of
implementation I'm more concerned about interfaces, as I said. I'm very
much in favour of a tool doing one thing right, rather than doing
multiple things. I think, in any environment you should be able to state
the purpose of a component in a single sentence. And the generator is
already far into the realm of doing multiple things. It's not only about
a quick fix, you also have to think about consequences, side-effects
when refactoring, isolation of concerns, and so on. What happens when we
want to re-organize the qooxdoo build process?!

Of course, if the XSLT integration would solve a lot of people's
problems, and would be general enough in the sense of numerous, because
so many people want to create Web interfaces through XML, and have all
the necessary tools sitting right on their desktop, and are happy using
them, that would change things. But in the absence of such evidence, I
would be really hesitant to add another feature to the generator.

Maybe we could start off with the "application-embeds-qooxdoo" approach,
and think about a tighter intergration when demands are evident. How
does that sound?

=Thomas


-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
_______________________________________________
qooxdoo-devel mailing list
qooxdoo-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/qooxdoo-devel

Reply via email to