Bug #6244
closed
Test case higher-order-functions-034
Category:
XPath conformance
Applies to branch:
12, trunk
Fix Committed on Branch:
12, trunk
Fixed in Maintenance Release:
Description
This test case is failing because it considers that
element(e)? is a subtype of element(e, xs:anyType)*
To quote https://www.w3.org/Bugs/Public/show_bug.cgi?id=21595, this is not the case, because element(e) allows a nilled element, while element(e, xs:anyType) does not.
I think that what is happening is that somewhere along the line, we're assuming that because the stylesheet isn't schema-aware, no element is ever going to have the nilled property, and therefore there's no difference between element(E, T) and element(E, T?). However, although there's no difference in what the two types match, there is a difference when it comes to the subtyping relationship.
The culprit seems to be XPathParser#3683 which ignores the ", xs:anyType" - code which I added on 2023-07-17 in the course of adding support for name test unions in 4.0.
Fixed for 12.x.
However, the code for 11.x is rather different, but the problem still occurs... Wrong: the test passes in 11.x
- Status changed from New to Resolved
- Applies to branch deleted (
11)
- Fix Committed on Branch 12, trunk added
- Platforms .NET, Java added
Note, this bug never made it into an issued release.
- Status changed from Resolved to Closed
- % Done changed from 0 to 100
- Fixed in Maintenance Release 12.4 added
Bug fix applied in the Saxon 12.4 maintenance release
Please register to edit this issue
Also available in: Atom
PDF
.