* reclaimed.
   *
   * If any of this fails along the way, we fall back to loading Finalizer directly in the
   * application class loader.
   *
   * NOTE: The tests for this behavior (FinalizableReferenceQueueClassLoaderUnloadingTest) fail
   * strangely when run in JDK 9. We are considering this a known issue. Please see
   * https://github.com/google/guava/issues/3086 for more information.

   * reclaimed.
   *
   * If any of this fails along the way, we fall back to loading Finalizer directly in the
   * application class loader.
   *
   * NOTE: The tests for this behavior (FinalizableReferenceQueueClassLoaderUnloadingTest) fail
   * strangely when run in JDK 9. We are considering this a known issue. Please see
   * https://github.com/google/guava/issues/3086 for more information.

					throughputHighestPut = totalPut
					// let the client see lower value as well
					throughputHighestGet = totalGet
				}
				sendResult()
				break
			}

			// We break if we did not see 2.5% growth rate in total GET
			// requests, we have reached our peak at this point.
			doBreak := float64(totalGet-throughputHighestGet)/float64(totalGet) < 0.025

			throughputHighestGet = totalGet

	m.mu.Lock()
	defer m.mu.Unlock()
	// See if someone else fetched it while we waited for the lock.
	b, ok = m.buckets[bucket]
	if ok {
		if b.bucket != bucket {
			logger.Info("getBucket: newly cached bucket %s does not match this bucket %s", b.bucket, bucket)
			debug.PrintStack()
		}
		return b
	}

	// New bucket. If we fail return the transient bucket.
	b = newBucketMetacache(bucket, true)

## Decision

We do not use Java serialization.
Instead, we use custom serialization where we explicitly describe how data objects should be serialized and deserialized.

For internal purposes, we use binary formats for their brevity.
We use the `Serializer` abstraction to separate the actual implementation of serialization from its uses.

When sharing data with external tools, we use JSON.

## Status

ACCEPTED

      }
      updateLastKnownIndex();
      /*
       * If the entry has been removed from the map, we return null, even though that might not be a
       * valid value. That's the best we can do, short of holding a reference to the most recently
       * seen value. And while we *could* do that, we aren't required to: Map.Entry explicitly says

**Experimental!**

We'd like to capture our architectural decisions about the build tool as [Architectural Decision Records (ADRs)](https://adr.github.io/).
For now we just have this global repository of ADRs.
If we see fit, we can break these out to per-platform ones, or keep a hybrid approach to having global and platform-specific ADSs.

Our aim is to keep the process lightweight and approachable.

 * of project dependencies. This optimizes module loading during runtime, as we will only load external
 * modules that are not loaded transitively by other project modules.
 *
 * We perform this filtering, since if we simply include all external dependencies, regardless of whether
 * they are already loaded transitively, there is a measurable performance impact during module-loading.

    // cheats and does not attempt every possible permutation. Instead, it only considers wildcards
    // in the leftmost position. We assert this fact when we generate the public suffix file. If
    // this assertion ever fails we'll need to refactor this implementation.
    var wildcardMatch: String? = null
    if (domainLabelsUtf8Bytes.size > 1) {
      val labelsWithWildcard = domainLabelsUtf8Bytes.clone()

        List<String> parentCommits = parentCommitsOf(commit)
        if (parentCommits.size() != 2) {
            println("$commit is not a merge commit we're looking for. Parents: $parentCommits")
            return
        }

        // The correct state we are looking for is:
        // 1. It's a merge commit.
        // 2. One of its parent commits is from master only.