Open sanastas opened 3 years ago
how would a FloatSerializer
be different from an IntSerializer
?
wouldn't this suffice:
public class OakFloatSerializer implements OakSerializer<Float> {
private final int size;
public OakFloatSerializer() {
this(Float.BYTES);
}
public OakFloatSerializer(int size) {
this.size = size;
}
@Override
public void serialize(Float value, OakScopedWriteBuffer targetBuffer) {
targetBuffer.putFloat(0, value);
}
Yes. That is exactly what we expect. It would also help if a simple test would be added that validates that this serializer works correctly with OakMap.
Oak requests its users to provide serializers for their objects representing keys and values.
Oak has internal common package including sub-packages for common serializers for common objects: IntBuffer, Integer and String. It would be good to have more of such. What kind of objects? It is up to implementer decision.
@liran-funaro can you please add more details to this issue?
For more explanations feel free to add questions in this issue.