OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.

 


Help: OASIS Mailing Lists Help | MarkMail Help

xslt-conformance message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [Elist Home]


Subject: Document for processor developers -- first draft


The following would serve as the introduction to the questionnaire
on discretionary items. This is our one document aimed at the XSLT
processor developers, though we encourage them to read the rest.

===== START =====
The XSLT and XPath Recommendations were written in hopes that XML
data could be transformed as needed by the products of many vendors.
Builders of Web sites should be able to use tools to create XSLT
stylesheets regardless of the XSLT processor that will ultimately
execute the styling. The OASIS XSLT/XPath test suite is the most
comprehensive measure of the necessary interoperability.

As a developer trying to meet the W3C Recommendations, you know how
the verbiage of the Recommendations contains precisely-stated
provisions that interlock in complex and sometimes surprising ways.
There are also cases where you have leeway to choose one of two
allowable behaviors or decide output details. Below is our
questionnaire about the choices. We want you to specify the choices
you made when creating your XSLT processor. You are encouraged to
post the answers on your own Web site as well as sending them to us.
When you prepare to use our test suite, an XML document containing
your answers is used to filter out test cases that assume the other
choice, leaving a rendition that is custom-fit to your processor
insofar as allowed by the Recommendations.

Conformance to the Recommendations is tested by running thousands
of small test cases, each with processor inputs and an expected
output. Comparison of actual and expected output neutralizes all
details that have been defined as irrelevant by the XML, InfoSet,
Namespaces, and XSLT Recommendations. There is no reason to think
that all tests have equal weight, hence we discourage talk of
percentages of conformance and the like. If you have some tests
that you would like to be included in a later edition of our
suite, see [URL] for submission instructions.
===== END =====

Let me know if you have any comments on the above prose.
.................David Marston



[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [Elist Home]


Powered by eList eXpress LLC