[SPARK-14021][SQL] custom context support for SparkSQLEnv

[adrian-wang]

What changes were proposed in this pull request?

This is to create a custom context for command bin/spark-sql and sbin/start-thriftserver. Any context that is derived from HiveContext is acceptable. User need to configure the class name of custom context in a config of spark.sql.context.class, and make sure the class in classpath. This is to provide a more elegant way for custom configurations and changes for infrastructure team.

How was this patch tested?

Added a unit test in CliSuite to track this.

cc @chenghao-intel

[SparkQA]

Test build #53613 has finished for PR 11843 at commit 654397e.

• This patch fails Scala style tests.
• This patch does not merge cleanly.
• This patch adds the following public classes (experimental):
  • log.warn(s\"Configured context class $className is not a subclass ofHiveContext,\" +

[SparkQA]

Test build #53614 has finished for PR 11843 at commit 864e1b6.

• This patch passes all tests.
• This patch merges cleanly.
• This patch adds the following public classes (experimental):
  • log.warn(s\"Configured context class $className is not a subclass ofHiveContext,\" +

[chenghao-intel]

instead of hard code, can we make it as classOf[HiveContext].getName()?

[SparkQA]

Test build #54020 has finished for PR 11843 at commit 35c7a8f.

• This patch passes all tests.
• This patch merges cleanly.
• This patch adds no public classes.

[chenghao-intel]

cc @rxin @liancheng

[chenghao-intel]

cc @yhuai , this is critical for our own customized HiveContext, can you please merge this?

[rxin]

HiveContext is going to be deprecated in Spark 2.0. Please see https://issues.apache.org/jira/browse/SPARK-13485

I don't think it makes sense to merge this at this point. If you need this, you should do something once SparkSession is introduced.

[andrewor14]

Let's close this PR since HiveContext is now removed. If there is interest feel free to re-open it and make the changes based on SessionState instead.