[
Lists Home |
Date Index |
Thread Index
]
> So I take it one doesn't have a set of conformance tests?
There's a set of published tests, but they aren't described as conformance
tests. Saxon runs all the tests, but in some cases there is room for debate
as to whether the result is correct, because of uncertainties in
interpreting the spec. The test suite itself, of course, is incomplete (I
don't think it covers any of the edge cases where Saxon is known to be
non-conformant).
> The 98% number
> is entirely subjective, or rather, expert opinion?
Let's say it's an educated guess at the proportion of testable assertions
made in the schema specification that are satisfied by the Saxon
implementation.
>
> If so, short of soliciting your expert opinion on the several other
> schema processors out there
I wouldn't dream of commenting on other schema processors (except to say
that some of them aren't).
Michael Kay
http://www.saxonica.com/
|
