Re: [SPARQL EXISTS] where and how to record engine status of tests

here is a  version of the summary table which includes the results for
BlazeGraph,
MillenniumDB, QLever, Virtuoso, dydra, and rdflib.
this is based on the manifest versions in the repository on 19.05.2025.

the cells with a single "x" depict the failed result in its original form
and canonicalized.
the left column contains the respective query text and expected results.

the double "x" indicates that the result file was not found.
most are because the correlation is by query file name, rather than by test
identification, as it should be.
a few are because the result is absent.

the generation process attempts to account for empty results by replacing
all of the variants with a uniform document.

On Fri, 16 May 2025 at 22:31, Peter F. Patel-Schneider <
pfpschneider@gmail.com> wrote:

> Where should the results of running the SPARQL EXISTS tests on engines go?
>
> I have results for MillenniumDB, QLever, and Virtuoso Open Source.
>
> peter
>
>
>