XML Query Test Suite version 0.8.4

[Michael Kay]

> It still seems to be necessary to normalise white space in many of the
> comparisons (528 tests). In most cases this is due to extra or missing
> white space in the expected result files, although it's possible this
> normalization step is masking white space bugs in xq2xsl.

I've just submitted some bug reports that attempt to analyze these
discrepancies. So far I've identified three categories:

(a) the published results have been produced by stripping all whitespace
text nodes from the source document as it is loaded

(b) the published results are serialized with indent=yes (this applies
mainly to the use cases)

(c) the result of the query is a sequence of elements, and successive
elements are serialized with a newline as a separator.

In all three cases the published results are (I believe) wrong according to
the test spec.

There may be other cases that don't fit into these categories: I'm still
working on it.

Michael Kay
http://www.saxonica.com/