[SPARK-18714][SQL] Add a simple time function to SparkSession

[rxin]

What changes were proposed in this pull request?

Many Spark developers often want to test the runtime of some function in interactive debugging and testing. This patch adds a simple time function to SparkSession:

scala> spark.time { spark.range(1000).count() }
Time taken: 77 ms                                                               
res1: Long = 1000

How was this patch tested?

I tested this interactively in spark-shell.

[rxin]

cc @marmbrus

[SparkQA]

Test build #69662 has started for PR 16140 at commit 76d2f53.

[dongjoon-hyun]

Retest this please

[rxin]

It's ok no need to retest. The change is fine.

[dongjoon-hyun]

I see. BTW, @rxin . Do we need spark to measure time? Is it possible to add that into SparkSession companion object?

[rxin]

That's much more difficult to type though.

[dongjoon-hyun]

True. Never mind. I just thought we were able to use like the following.

    SparkSession.time { SparkSession.builder.appName("Spark Pi").getOrCreate() }

[SparkQA]

Test build #69664 has finished for PR 16140 at commit 76d2f53.

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

[srowen]

shrug It seems pretty unrelated to SparkSession. I get that it's a convenience method but something Spark really needs to expose as an API method?

[rxin]

The cost to maintain this seems very small though, and I'd definitely use it all the time in the repl. In Databricks this is not an issue since the environment always appends the time, but I really miss this in the repl.

[hvanhovell]

LGTM - merging to master/2.1. Thanks!