Please find attached the latest version of Part 8, DSRL. This version
contains a new clause in Annex B that shows how free-standing DSRL maps can
be processed using XSLT 2.0. Before releasing this for ballot, however, I
need your advice. The present clause consists of:
a) A XSLT 2.0 transform that generates a stylesheet that can be used to
convert files based on the source schema into files that conform to the
target schema
b) an example file, with validating DTD, in French
c) a target schema, coded using the W3C Schema language (OK, I know ..., but
I plan to use RELAX NG in the following clause and want to show that we are
schema agnostic)
d) a DSRL map that is designed to show off all of the DSRL functions (not a
good idea, but I needed to be able to test everything)
e) the XSLT 2.0 file created when d) is tranformed by a)
f) the result of applying e) to b)
The questions I would very much appreciate it if you could find time to
answer are:
1) Should we have examples of how the schema works, or would it suffice to
publish a) on its own, with or without some explanatory text?
2) At present I simply include the XML files, without explaining their
contents in any way. Is this sufficient, or should each file have a
discursive walkthrough?
3) Should the example try to demonstrate all the features of DSRL, or only a
rational subset thereof? (There is a problem of demonstrating things like
default content for missing elements, in that it is in general impossible to
get them in the right place in the schema using automatically generated
files. You really need programmatic access to the schema to be able to get
their position exactly right. If you look carefully at the added
<locality>Downtown</locality> tag you will find that it is in the wrong
place according to the target schema. Should I, therefore, drop anything
that is problematic in its use, or should I document why it is problematic?)
4) Should the example include elements that appear at different levels in
different parts of the model? If so, how can this be simply demonstrated
(and tested)?
5) So far I have only had time to test a single map and file (though with a
number of variants during testing). Does anyone have time to create and test
some more maps?
Any ideas on how to simplify the XSLT would also be appreciated.
Martin
-- DSDL members discussion list To unsubscribe, please send a message with the command "unsubscribe" to dsdl-discuss-request@dsdl.org (mailto:dsdl-discuss-request@dsdl.org?Subject=unsubscribe)
This archive was generated by hypermail 2.1.8 : Wed Apr 12 2006 - 14:48:02 UTC