Github user srowen commented on a diff in the pull request:
https://github.com/apache/spark/pull/5786#discussion_r29386539
--- Diff: pom.xml ---
@@ -1629,17 +1629,21 @@
</profile>
<profile>
- <id>hadoop-2.2</id>
+ <id>hadoop-1</id>
<properties>
- <hadoop.version>2.2.0</hadoop.version>
- <protobuf.version>2.5.0</protobuf.version>
- <hbase.version>0.98.7-hadoop2</hbase.version>
- <avro.mapred.classifier>hadoop2</avro.mapred.classifier>
- <codehaus.jackson.version>1.9.13</codehaus.jackson.version>
+ <hadoop.version>1.2.0</hadoop.version>
--- End diff --
None of this should actually affect the releases people use since they
build with profiles explicitly enabled as desired. It's a good change of
defaults, for hygiene though.
The story here is that Spark 1.2 was sorta accidentally released as
depending (in Maven) on 2.2, and so that was expressed later, but this is the
logical conclusion of that process. The default build which you'd get if
approaching this as a developer is and should be a consistent Hadoop 2.2 build.
However more needs to change besides this. The release script has to set a
"hadoop-1" profile in some cases now. The docs about building vs various Hadoop
versions has to be updated too.
---
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.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
