If by stream, you mean java.io.OutputStream, you can do this trivially
by subclassing com.jclark.xsl.sax.Destination (typically
com.jclark.xsl.sax.GenericDestination) and then calling
com.jclark.xsl.sax.OutputMethodHandlerImpl.setDestination with an
instance of that class.

Diarmuid Mac Carthy wrote:
> 
> Hi,
> 
> We're investigating the possibility of using James Clarke's XT processor as
> a component in a larger XSLT application.  One key requirement we have is
> support for multiple output documents, which is provided for in XT at the
> expense of having to output to files rather than standard output.
> 
> Is there any way of hooking into the XT processor to intercept this output
> as a stream for use in later processing?  We'd rather not have to dump
> things to files only to read them in again immediately.
> 
> Assuming you can hook into the output stream in this fashion, I can still
> foresee problems.  For example, if the output comes through a single writer
> stream which writes one file stream, then the other, I expect problems
> trying to identify the end of one and the start of another document.  Is
> this the case, or is there an output stream created for each file specified?
> 
> I hope this is an easy one :-)  I've searched through the archives but
> couldn't find anything resembling our problem.
> 
> Any help would be greatly appreciated.
> 
> Regards,
> Diarmuid Mac Carthy
> 
>  XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list


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