- From: C. M. Sperberg-McQueen <cmsmcq@blackmesatech.com>
- Date: Wed, 16 Jun 2010 12:10:26 -0600
- To: ht@inf.ed.ac.uk (Henry S. Thompson)
- Cc: "C. M. Sperberg-McQueen" <cmsmcq@blackmesatech.com>, public-xml-schema-testsuite@w3.org
On 16 Jun 2010, at 10:17 , Henry S. Thompson wrote: > ... > > I think this example (or see the resolution of bug 4078, testGroups > addB156 and addB156a in msMeta/Additional_w3c.xml for a parallel > instanceTest example) illustrates my answer to your larger "Why did > you do it this way" question -- since the resolution of many bugs > seemed to be going to be "label this one 1.0, and make a copy and > label it 1.1, with changes as follows", that putting the version and > edition attr. on the {schema,instance}Test element would support that > straightforwardly. It would be equally straightforward, and somewhat more useful, not to copy the test but to specify that for this particular test different versions give different results. That makes is significantly easier to find tests illustrating areas which have changed between 1.0 and 1.1. Copying the test seems to have no advantage except to make it harder (not impossible, just unnecessarily inconvenient) to find concrete instances of constructs which have changed expected behavior between 1.0 and 1.1, thus helping to obscure changes in the spec, and to make it harder for the WG to check to see whether changes in the spec have been adequately covered in the test suite, thus making it harder to move to PR. Every little bit helps. > It was also then straightforward to update my test harness to run only > e.g. 1.0 non-implDe tests, or to find all implDe 1.1 tests. You'll be pleased to see, when you look at http://www.w3.org/XML/2008/xsdl-exx/ancillary/xsts-schema.sketch.xml that the necessary test harness changes are also straightforward in that design. Michael -- **************************************************************** * C. M. Sperberg-McQueen, Black Mesa Technologies LLC * http://www.blackmesatech.com * http://cmsmcq.com/mib * http://balisage.net ****************************************************************
Received on Wednesday, 16 June 2010 18:10:56 UTC