4952 2007-08-15 13:11:31 +0000 Sequence:Sequence Restriction (Recurse) 2010-03-26 15:43:40 +0000 1 1 1 Unclassified XML Schema Test Suite Microsoft tests 2006-11-06 PC Windows XP NEW disputedTest P2 normal --- 1 mhpchan ht David_E3 public-xml-schema-testsuite oldest_to_newest 16155 0 mhpchan 2007-08-15 13:11:31 +0000 The schema has base particle of sequence restricted to a derived particle of sequence. Within the restricted sequence, the particles are not valid restrictions with the corresponding particles in the base type. However, the spec says, "2.1 Each particle in the {particles} of R is a ·valid restriction· of the particle in the {particles} of B it maps to as defined by Particle Valid (Restriction) (§3.9.6)." Since the particles are not valid restrictions, the scenario used should be scenario that checks invalidity testcase affected: msData/particles/particlesW006.xsd 16198 1 mike 2007-08-18 21:28:32 +0000 Personal response. In my view this is a case where the bug is in the spec, not in the test. This is an example of a situation where all instances of R are in fact valid instances of B (because of the maxOccurs="0"), but the algorithm in the spec fails to recognize the fact. In Schema 1.1 this will be a valid restriction. I don't think there is really much to be gained at this stage in testing conformance of products to rules that are known to be deficient. Nor is there much to be gained by patching up the Schema 1.0 rules. 31445 2 ht 2010-02-04 14:03:09 +0000 Even if Mike's analysis is right, this should be changed to invalid, and some of us (e.g. XSV) will get it wrong 33945 3 David_E3 2010-03-26 15:43:40 +0000 The WG believes this test is wrong for 1.0, but that behavior is less than desirable. A new test, properly defined, is required for 1.1.