[
https://issues.apache.org/jira/browse/OPENNLP-776?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15428454#comment-15428454
]
Tristan Nixon commented on OPENNLP-776:
---------------------------------------
Good point. I thought the only way to provide custom serialization was to use
Externalizable, which does require a no-arg constructor, but now I see one can
put the readObject and writeObject methods into a Serializable and get the same
effect (leaving me wondering what the point of Externalizable is...).
One slight complication with this is that if we rely on Object's no-arg
constructor, the implicit initialization of fields like artifactMap and
artifactSerializers does not happen, so I need to do this explicitly in the
readObject method, meaning they cannot be final anymore (nor can
isLoadedFromSerialized).
Otherwise, it seems to be working fine! See the attached patch.
> Model Objects should be Serializable
> ------------------------------------
>
> Key: OPENNLP-776
> URL: https://issues.apache.org/jira/browse/OPENNLP-776
> Project: OpenNLP
> Issue Type: Improvement
> Affects Versions: tools-1.5.3
> Reporter: Tristan Nixon
> Assignee: Joern Kottmann
> Priority: Minor
> Labels: features, patch
> Fix For: 1.6.1
>
> Attachments: externalizable.patch, serialization_proxy.patch
>
>
> Marking model objects (ParserModel, SentenceModel, etc.) as Serializable can
> enable a number of features offered by other Java frameworks (my own use case
> is described below). You've already got a good mechanism for
> (de-)serialization, but it cannot be leveraged by other frameworks without
> implementing the Serializable interface. I'm attaching a patch to BaseModel
> that implements the methods in the java.io.Externalizable interface as
> wrappers to the existing (de-)serialization methods. This simple change can
> open up a number of useful opportunities for integrating OpenNLP with other
> frameworks.
> My use case is that I am incorporating OpenNLP into a Spark application. This
> requires that components of the system be distributed between the driver and
> worker nodes within the cluster. In order to do this, Spark uses Java
> serialization API to transmit objects between nodes. This is far more
> efficient than instantiating models on each node independently.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
