[SPARK-17208] [SQL] Build Outer Join Test Cases in File-based Testing Framework

[gatorsmile]

What changes were proposed in this pull request?

This PR is part of the original PR: #14625.

Based on file-based SQL end-to-end testing framework in SQLQueryTestSuite, this PR is to create test cases for outer joins.

In addition, this PR also renamed the test data sets to more meaningful names, based on the comments from @cloud-fan

How was this patch tested?

Using the same data and query, the Hive outputs are attached below:
hive.output.txt

After comparison, the Spark-generated results are the same.

[rxin]

do we really need to add these tables here? can we just use temp views on the test cases?

[gatorsmile]

Sure, will move them into the test cases.

[gatorsmile]

Using the same data and query, the Hive outputs are attached below:
hive.output.txt

After comparison, the Spark-generated results are the same.

[SparkQA]

Test build #64334 has finished for PR 14782 at commit dc36ce9.

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

[SparkQA]

Test build #64338 has finished for PR 14782 at commit 02d500b.

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

[rxin]

@gatorsmile maybe you should stop porting more tests for now. I don't want you to waste time. I feel you are just porting the tests over from Hive compatibility suite, but that's not really the point. This is a framework to create test cases that would actually make sense. The best cases are the ones that are very small and simple and can test specific aspects of the query engine.

[gatorsmile]

I see. Thanks! Let me close this one.