I added comments to your attachment document.

We’re definitely aware of the similarity you suggest here. Our current position is an evolution from these earlier ideas intended to simplify and keep us out of trouble, handle output as well as input, and stay away from extensibility issues we don’t understand fully.

…mike

---

From: dfdl-wg-bounces@ogf.org [mailto:dfdl-wg-bounces@ogf.org] On Behalf Of RPost
Sent: Tuesday, June 10, 2008 8:59 PM
To: dfdl-wg@ogf.org
Subject: [DFDL-WG] comments and concerns re choice-discriminator-example

I have some questions and possible concerns re the sample xml file Mike posted.

The attached document ‘SmartlengthStringComments1 discusses them.

I attached the other doc that I found from 2003 written by Martin Westhead.

The current sample and an XSLT sample in the old doc bear an uncanny resemblance and thought it might be worthwhile to look at the two side-by-side since they are for the exact same use case.

It raises the question of whether DFDL syntax could allow actual XSLT snippets to be included in the annotation section to use the power of XSLT to do these things. XSLT also supports extensions that could handle complex or unusual corner cases.

A parser should be able to (but I haven’t confirmed it) create a mini XSL stylesheet from the snippet and use a transform to generate the result value that is needed. If so this would allow an existing standard language to be used.

See my item #5 in my doc since there I ask the question of how I might contribute without being disruptive. It is not my intention to disrupt your efforts but the resemblance between these two approaches is so similar I had to bring it to your attention.

Rick