Will Berkeley has posted comments on this change. ( http://gerrit.cloudera.org:8080/9852 )
Change subject: [examples] Improve the basic Java example ...................................................................... Patch Set 1: (4 comments) http://gerrit.cloudera.org:8080/#/c/9852/1/examples/java/java-example/pom.xml File examples/java/java-example/pom.xml: http://gerrit.cloudera.org:8080/#/c/9852/1/examples/java/java-example/pom.xml@24 PS1, Line 24: <groupId>kudu-examples</groupId> > Hmm, perhaps we should be publishing into org.apache.kudu, just like the re You're right, I think it makes sense to use the same group id. The useful bits in these examples are the source code, which we could consider publishing to Maven in the jar, but also there may be other examples in the future we definitely want to publish, in which case it's good to start using org.apache.kudu for everything. http://gerrit.cloudera.org:8080/#/c/9852/1/examples/java/java-example/src/main/java/org/apache/kudu/examples/Example.java File examples/java/java-example/src/main/java/org/apache/kudu/examples/Example.java: http://gerrit.cloudera.org:8080/#/c/9852/1/examples/java/java-example/src/main/java/org/apache/kudu/examples/Example.java@75 PS1, Line 75: cto.setRangePartitionColumns(rangeKeys); > Now that we support hash partitioning, and we kinda think that hash partiti Done http://gerrit.cloudera.org:8080/#/c/9852/1/examples/java/java-example/src/main/java/org/apache/kudu/examples/Example.java@89 PS1, Line 89: KuduSession session = client.newSession(); > I wonder if it'd be useful to demonstrate or at least draw some attention t Flush mode: AUTO_FLUSH_BACKGROUND is the default and it's what we recommend for most workloads, so I prefer letting the curious find the other flush modes in the JavaDoc. What needs to be covered here is how to discover errors when using a background flush mode. Nullable column: Done. Hash partitioning: I'm splitting the difference between your suggestion to have both types of partitioning and Todd's suggestion to use hash partitioning for the simple example. I've made the table be partitioned by hash on key, and added a comment about range partitioning with a link to the schema design docs. http://gerrit.cloudera.org:8080/#/c/9852/1/examples/java/java-example/src/main/java/org/apache/kudu/examples/Example.java@129 PS1, Line 129: // Scan with a predicate on the 'key' column, returning the 'value' and "added" columns. > maybe we can refactor out some methods from this enormous function? eg 'voi Done -- To view, visit http://gerrit.cloudera.org:8080/9852 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: kudu Gerrit-Branch: master Gerrit-MessageType: comment Gerrit-Change-Id: I79ab5b0b2c30e8fd7c799857a0e3754b7618a580 Gerrit-Change-Number: 9852 Gerrit-PatchSet: 1 Gerrit-Owner: Will Berkeley <[email protected]> Gerrit-Reviewer: Adar Dembo <[email protected]> Gerrit-Reviewer: Andrew Wong <[email protected]> Gerrit-Reviewer: Grant Henke <[email protected]> Gerrit-Reviewer: Kudu Jenkins Gerrit-Reviewer: Todd Lipcon <[email protected]> Gerrit-Reviewer: Will Berkeley <[email protected]> Gerrit-Comment-Date: Mon, 02 Apr 2018 06:42:00 +0000 Gerrit-HasComments: Yes
