You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Test case bind03 passes by happy coincidence rather than by design (attached for convenience).
Currently, to evaluate
{code}
PREFIX : http://example.org/
SELECT ?z ?s1
{
?s ?p ?o .
BIND(?o+1 AS ?z)
?s1 ?p1 ?z
}
{code}
depends on the order in which the statements are passed to the evaluation engine. But the QueryJoinOptimizer can rearrange the order. In this case, the order is unchanged, but the following fails:
{code}
PREFIX : http://example.org/
SELECT ?z ?s1
{
?s :p ?o .
BIND(?o+1 AS ?z)
?s1 ?p1 ?z
}
{code}
.
Should sparql be working out the dependency ordering or should it respect the ordering given by the user? i.e. should this work or not:
{code}
PREFIX : http://example.org/
(Migrated from https://openrdf.atlassian.net/browse/SES-2275 )
Test case bind03 passes by happy coincidence rather than by design (attached for convenience).
Currently, to evaluate
{code}
PREFIX : http://example.org/
SELECT ?z ?s1
{
?s ?p ?o .
BIND(?o+1 AS ?z)
?s1 ?p1 ?z
}
{code}
depends on the order in which the statements are passed to the evaluation engine. But the QueryJoinOptimizer can rearrange the order. In this case, the order is unchanged, but the following fails:
{code}
PREFIX : http://example.org/
SELECT ?z ?s1
{
?s ?p ?o .
BIND(?o+1 AS ?z)
?s1 :p ?z
}
{code}
whereas this works:
{code}
PREFIX : http://example.org/
SELECT ?z ?s1
{
?s :p ?o .
BIND(?o+1 AS ?z)
?s1 ?p1 ?z
}
{code}
.
Should sparql be working out the dependency ordering or should it respect the ordering given by the user? i.e. should this work or not:
{code}
PREFIX : http://example.org/
SELECT ?z ?s1
{
?s1 ?p1 ?z
BIND(?o+1 AS ?z)
?s :p ?o .
}
{code}
The text was updated successfully, but these errors were encountered: