summaryrefslogtreecommitdiff
path: root/benchmark
diff options
context:
space:
mode:
authorKouhei Yanagita <yanagi@shakenbu.org>2023-09-16 12:10:09 +0900
committerGitHub <noreply@github.com>2023-09-16 12:10:09 +0900
commit7d08dbd015a16c27f2d9c77751e80fa6efba2d7a (patch)
treef40369e5c50acc88026c5221695443342fb5c49c /benchmark
parent67dedf8cf634843488a477e53b9995b63e9aa291 (diff)
Optimize Range#bsearch for beginless/endless ranges
On Range#bsearch for endless ranges, we try positions at `begin + 2**i` (i = 0, 1, 2, ...) to find a point that satisfies a given condition. Subsequently, we perform binary searching with the interval `[begin, begin + 2**n]`. However, the interval `[begin + 2**(n-1), begin + 2**n]` is sufficient for binary search because `begin + 2**(n-1)` does not satisfy the condition. The same applies to beginless ranges.
Diffstat (limited to 'benchmark')
-rw-r--r--benchmark/range_bsearch.yml9
1 files changed, 9 insertions, 0 deletions
diff --git a/benchmark/range_bsearch.yml b/benchmark/range_bsearch.yml
new file mode 100644
index 0000000000..9a7388ab04
--- /dev/null
+++ b/benchmark/range_bsearch.yml
@@ -0,0 +1,9 @@
+prelude: |
+ r = (1..)
+
+benchmark:
+ '10**1': r.bsearch { |x| x >= 10 }
+ '10**2': r.bsearch { |x| x >= 100 }
+ '10**3': r.bsearch { |x| x >= 1000 }
+ '10**4': r.bsearch { |x| x >= 10000 }
+ '10**5': r.bsearch { |x| x >= 100000 }