datalog: magic-sets benefit test on disjoint-cluster graph (205/205)
Some checks failed
Test, Build, and Deploy / test-build-deploy (push) Failing after 24s
Some checks failed
Test, Build, and Deploy / test-build-deploy (push) Failing after 24s
Two disjoint chains, query rooted in cluster 1. Semi-naive derives the full closure over both clusters (6 ancestor tuples). Magic-sets only seeds magic_ancestor^bf for cluster 1, so only 2 query-relevant tuples are returned (a→b, a→c). The test asserts both numbers, demonstrating the actual perf-shape benefit of goal-directed evaluation.
This commit is contained in:
@@ -1,6 +1,6 @@
|
||||
# datalog scoreboard
|
||||
|
||||
**204 / 204 passing** (0 failure(s)).
|
||||
**205 / 205 passing** (0 failure(s)).
|
||||
|
||||
| Suite | Passed | Total | Status |
|
||||
|-------|--------|-------|--------|
|
||||
@@ -13,5 +13,5 @@
|
||||
| negation | 10 | 10 | ok |
|
||||
| aggregates | 18 | 18 | ok |
|
||||
| api | 14 | 14 | ok |
|
||||
| magic | 20 | 20 | ok |
|
||||
| magic | 21 | 21 | ok |
|
||||
| demo | 18 | 18 | ok |
|
||||
|
||||
Reference in New Issue
Block a user