Conversation
| * | ||
| * @param <T> Object type | ||
| */ | ||
| public class KryoSimpleWrapper<T> implements Writable { |
There was a problem hiding this comment.
What's the difference between this and KryoWritableWrapper, when should we use one and when the other?
There was a problem hiding this comment.
I have added more documentation now. The usage of KryoSimpleWrapper is similar to KryoWritabeWrapper. KryoSimpleWrapper can be used if the object being serialized is not recursive or nested, because it uses a version of kryo object that doesn't track references, hence significantly faster.
| * This serialization does not support recursive/nested structures | ||
| * to provide faster serialization. | ||
| */ | ||
| public abstract class KryoSimpleWritable implements KryoIgnoreWritable { |
There was a problem hiding this comment.
It would be great to switch to Java 8, to be able to keep this an interface :-)
| // non-kryo serialization. | ||
| /** Reusable Input object */ | ||
| private final InputChunked input = new InputChunked(4096); | ||
| private InputChunked input; |
There was a problem hiding this comment.
Should we separate two Kryo implementations in two classes since they have a lot of differences?
There was a problem hiding this comment.
Agreed, we can refactor the existing code to have a better code readability. I have added a TODO for that.
| * no recursive/nested objects should be serialized. | ||
| * @return Hadoop kryo which doesn't track objects. | ||
| */ | ||
| public static HadoopKryo getNontrackingKryo() { |
There was a problem hiding this comment.
How will someone using kryo know which implementation to use? Can we add some guidance around that?
There was a problem hiding this comment.
Added more documentation.
majakabiljo
left a comment
majakabiljo
left a comment
There was a problem hiding this comment.
Discussed offline, @yukselakinci will in a separate diff refactor HadoopKryo.
closes apache#69
closes apache#69
2 participants
Second PR for serialization. There is one more.