Aklakan commented on code in PR #3517:
URL: https://github.com/apache/jena/pull/3517#discussion_r2442490931
##########
jena-arq/src/main/java/org/apache/jena/sparql/engine/join/MultiHashProbeTable.java:
##########
@@ -180,9 +176,13 @@ public Iterator<Binding> getCandidates(Binding row) {
* This method is package private so that it can be called from tests.
*/
void doFinalize() {
- // Note: We need to stick with the variable order provided in the
initial index -> don't sort!
+ // If an initial join key is given then its variables are prepended to
seen vars
+ // so that those correspond to the first bits of the bit keys!
+ // This means we need to stick with the variable order provided in the
initial index
+ // -> don't sort!
// Arrays.sort(seenVars, (a, b) -> a.getName().compareTo(b.getName()));
- seenVarsJoinKey = JoinKey.create(seenVarSet);
+ seenVarsJoinKey = JoinKey.newBuilder()
+ .addAll(initialIndex.getSuperJoinKey()).addAll(seenVarSet).build();
Review Comment:
Separated the set of initial variables from the set of variables encountered
in the build phase - so that clear() can just clear the latter.
To recap: The initial variables are obtained from peeking the first binding
of each of lhs and rhs. All bindings in the build phase are initially only
indexed by those variables. Later indexes are only created on-demand in the
streaming phase based on which variables are unbound in the streamed bindings.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
