Re: Thought on future of XMLC

On Thursday 21 November 2002 15:26, David Li wrote:
> On Thursday, Nov 21, 2002, at 21:58 Asia/Shanghai, Richard Kunze wrote:
> > Sure it does, and coolness definitely *is* a reason :-)
>
> coolness is THE reason. ;)
>
> >> I think the approach is the same. However, eliminating the need for
> >> the
> >> filesystem could be a good thing, especially used in a products. Less
> >> supporting harsh.
> >
> > Agreed. On the other hand, if I'm not mistaken then you need to build
> > these dynamic classes "by hand" (similar to writing assembler code). Right
> > now, I'm thinking of moving the whole XMLC code generation to a more
> > template driven approach (makes it easier to customize/experiment with new
> > features) - and this pretty much clashes with BCEL and similar lowlevel
> > frameworks, because these would force you to write your templates in
> > "assembler" instead of Java (or something very similar to Java). That's
> > why I'm a bit reluctant to take that route.
>
> That's true. BCEL or other similar tool are harder to use. Maybe
> someone should come up with tool to generated "pseudo assembly" in
> these tools from a compiled class files. That could be useful.

Or even better an on-the-fly java compiler that takes source code from an 
input stream and compiles it in-memory, without ever writing the sources or 
the generated byte code to a file.


-- 
Richard Kunze 

[ t]ivano Software, Bahnhofstr. 18, 63263 Neu-Isenburg
Tel.: +49 6102 80 99 07 - 0, Fax.: +49 6102 80 99 07 - 1
http://www.tivano.de, kunze@xxxxxxxxx