Re: Specification of a transform.

I'm not sure if I've interpreted you correctly, but Phil's reply seems to
interpret your "no hierarchical simple relationship" as the key phrase here,
interpreting it as meaning that a multi-phase transformation is needed.

In other words, I think Phil is assuming that your "no hierarchical simple
relationship" is equivalent to saying that in terms of Jackson Structured
Programming

http://en.wikipedia.org/wiki/Jackson_structured_programming

there is a "structure clash" between the input and output schemas; and he
follows JSP by recommending:

"If a structure clash is found, it is usually resolved by splitting the
program into two parts, using an intermediate data structure to provide a
common structural framework with which the two program parts can
communicate."

Unfortunately I don't have copies of Michael Jackson's books on JSP, but I
recall that he goes into considerable detail analysing the types of structure
clash that can arise and the techniques for dealing with them. This work was
all done in the 1970s in the context of processing hierarchical files on
magnetic tape, but it could do with a dusting off and re-interpretation in
terms of XML, especially for streamed transformations.

However, there are other parts of your problem statement where I simply don't
know what you are talking about. Where do "literals" come into it, for
example?

Michael Kay
Saxonica

On 9 Sep 2013, at 08:42, davep wrote:

> Given schema A as input XML. Schema B as XML output.
>  Assume no hierarchical simple relationship.
>  Assume mapping of values needed from input values to output values.
>  Assume literals are needed.
> How do (might) you specify the required transform?
> How do (might) you validate that instance A has been correctly transformed
to instance B, assuming input and output are both valid to the schemas.
>
> Not something I've seen mentioned on this list?
>
> regards
>
> --
> Dave Pawson
> XSLT XSL-FO FAQ.
> http://www.dpawson.co.uk