RE: [oic] Groups - ConformanceAssessmentMethod-wd02.odt uploaded

[Hanssens Bart <Bart.Hanssens@fedict.be>]

Hello Ming Fei,


Thanks for the interesting remarks ! I have to take some time to think about
it and integrate them in a new draft.

Allow me to add some first impressions:

> 3. As to how contributors generate xml file, html file, how contributors
> submit test cases to SVN. Maybe these are too many technical details, and
> not necessary to state in the proposal. But could you write a simple
> guideline in the wiki so that other members can start to work based on
> your previous experimental result?

That's something I'm surely willing to do, but perhaps after we reach
consensus on the procedure and the XML files...

The way I see it, we first have to agree on the way we want to work and
which XML schemas we'll use, then we can think about making it easier to
create them and to generate HTML (or, why not, an ODT) versions.


> 4. Since conformance test suite and interoperability test suite are
> different deliverables of OIC TC, it will be better we create different
> trunks in the SVN so that convenient to manage test suites and create
> corresponding zip package.

Sounds like a good idea, so the outcome of this would be something like:

\svn\TestSuite\trunk\conformance\odf11
\svn\TestSuite\trunk\conformance\odf12
...
\svn\TestSuite\trunk\interop\odf11
\svn\TestSuite\trunk\interop\odf12
...

> 7. First step, we need to start from creating conformance test suite.
> I understand conformance test suite should cover the whole
> specification, and we should try to avoid duplicate or intersect the same
> feature in the specification. So we need to provide a tools or means that
> are indexing the existed test cases so that TC members are easy to
> search if some test case already exist. I do not if SVN can do such things.

Mmm, SVN itself won't provide the tools for this. But if we have the XML
files with test assertions pointing to specific parts of the ODF spec, it is
fairly easy to create a script and/or XSLT that processes all the XML's
and generates an index. Not perfect, but probably good enough...


Best regards,

Bart