Github user zentol commented on the pull request:
https://github.com/apache/flink/pull/1660#issuecomment-189310673
Overall this PR looks good now. I'll go through the documentation more in
detail when i unify the docs for this PR and #1640 when merging.
Furthermore, I'd like to make a slight change as to how the sink is
created. Although there will be 2 separate sinks providing different guarantees
in terms of data consistency, it would be nice if the sink creation would go
through the same mechanism.
@rmetzger drafted the following when i had a chat with him:
```
CassandraSink sink = CassandraSink.addSink(input)
.setHostname(..) //passing the hostname separately is intended for
simple programs
.setQuery(...)
.setConsistencyLevel(...) //decides which sink is used
.setClusterBuilder(new ClusterBuilder() { //optional, for maximum
configurability
@Override
public Cluster createCluster(Cluster.Builder builder){
builder.addContactPoint(host).build();
}
.otherStuff()
.build();
```
CassandraSink would be a wrapper class, exposing common methods a user
would expect from a sink.
If you have no objections i would inject these changes while merging as
well.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
