Github user meiercaleb commented on a diff in the pull request:
https://github.com/apache/incubator-rya/pull/251#discussion_r153529877
--- Diff:
extras/rya.pcj.fluo/pcj.fluo.api/src/main/java/org/apache/rya/indexing/pcj/fluo/api/GetQueryReport.java
---
@@ -132,14 +135,20 @@ private BigInteger countBindingSets(final
SnapshotBase sx, final String nodeId,
checkNotNull(nodeId);
checkNotNull(bindingSetColumn);
+ NodeType type = NodeType.fromNodeId(nodeId).get();
+ Bytes prefixBytes = Bytes.of(type.getNodeTypePrefix());
+
// Limit the scan to the binding set column and node id.
- final RowScanner rows =
sx.scanner().over(Span.prefix(nodeId)).fetch(bindingSetColumn).byRow().build();
--- End diff --
Sharding changes how binding set results can be accessed in the Fluo table.
There is no concern about breaking existing Fluo instances because the
instances can't be updated. Just redeployed with an updated jar.
---
