Fix Range when starting at zero (#1073)

The existing behavior results in the `0` value being added twice. Since `lo` is always added to `dst`, we never want to explicitly add `0` if `lo` is equal to `0`.

Fix Range when starting at zero (#1073)
17a6b21e · Scott K Logan · GitHub · 7fa6f1f9 · 17a6b21e · 17a6b21e
Unverified Commit 17a6b21e authored Nov 26, 2020 by Scott K Logan Committed by GitHub Nov 26, 2020
Hide whitespace changes
Inline Side-by-side

Showing with 7 additions and 1 deletion

benchmark_register.h src/benchmark_register.h +1 -1

benchmark_gtest.cc test/benchmark_gtest.cc +6 -0

No files found.
--- a/src/benchmark_register.h
+++ b/src/benchmark_register.h
@@ -87,7 +87,7 @@ void AddRange(std::vector<T>* dst, T lo, T hi, int mult) {
  }

  // Treat 0 as a special case (see discussion on #762).
-  if (lo <= 0 && hi >= 0) {
+  if (lo < 0 && hi >= 0) {
    dst->push_back(0);
  }


--- a/test/benchmark_gtest.cc
+++ b/test/benchmark_gtest.cc
@@ -90,6 +90,12 @@ TEST(AddRangeTest, ZeroOnlyRange) {
  EXPECT_THAT(dst, testing::ElementsAre(0));
 }

+TEST(AddRangeTest, ZeroStartingRange) {
+  std::vector<int> dst;
+  AddRange(&dst, 0, 2, 2);
+  EXPECT_THAT(dst, testing::ElementsAre(0, 1, 2));
+}
+
 TEST(AddRangeTest, NegativeRange64) {
  std::vector<int64_t> dst;
  AddRange<int64_t>(&dst, -4, 4, 2);