adding partition info when we try to explain on partitioned table

[zhexuany]

Table schema is

| CREATE TABLE `trb7` (
  `id` int(11) DEFAULT NULL,
  `name` varchar(50) DEFAULT NULL,
  `purchased` date DEFAULT NULL
) ENGINE=InnoDB DEFAULT CHARSET=utf8 COLLATE=utf8_bin
PARTITION BY RANGE ( year(`purchased`) - 1 ) (
  PARTITION p0 VALUES LESS THAN (1990),
  PARTITION p1 VALUES LESS THAN (1995),
  PARTITION p2 VALUES LESS THAN (2000),
  PARTITION p3 VALUES LESS THAN (2005)
)

When we want to run explain on sql like select count(*) from trb7 where purchased > date'1992-10-10' and purchased < date'1995-10-10, current output only come with key range. It is better to inlucde partition need to be scaned.

The following is current explain info:

scala> spark.sql("select count(*) from trb7 where purchased > date'1992-10-10' and purchased < date'1995-10-10'").explain
2018-12-17 16:18:48 WARN  ObjectStore:568 - Failed to get database test, returning NoSuchObjectException
2018-12-17 16:18:48 WARN  ObjectStore:568 - Failed to get database test, returning NoSuchObjectException
2018-12-17 16:18:48 WARN  ObjectStore:568 - Failed to get database test, returning NoSuchObjectException
2018-12-17 16:18:48 WARN  ObjectStore:568 - Failed to get database test, returning NoSuchObjectException
== Physical Plan ==
*(2) HashAggregate(keys=[], functions=[specialsum(count(1)#26L, LongType, 0)])
+- Exchange SinglePartition
   +- *(1) HashAggregate(keys=[], functions=[partial_specialsum(count(1)#26L, LongType, 0)])
      +- TiSpark CoprocessorRDD{[table: trb7] , Columns: [purchased], Residual Filter: Not(IsNull([purchased])), [[purchased] LESS_THAN 1995-10-10T00:00:00.000+08:00], [[purchased] GREATER_THAN 1992-10-10T00:00:00.000+08:00], KeyRange: [[116,128,0,0,0,0,0,0,32,95,114,0,0,0,0,0,0,0,0], [116,128,0,0,0,0,0,0,32,95,115,0,0,0,0,0,0,0,0])[[116,128,0,0,0,0,0,0,33,95,114,0,0,0,0,0,0,0,0], [116,128,0,0,0,0,0,0,33,95,115,0,0,0,0,0,0,0,0]), Aggregates: Count(1)}