Skip to content

[SPARK-16904] [SQL] Removal of Hive Built-in Hash Functions and TestHiveFunctionRegistry #14498

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
gatorsmile wants to merge 5 commits into from
Closed

[SPARK-16904] [SQL] Removal of Hive Built-in Hash Functions and TestHiveFunctionRegistry #14498

Show file tree
Hide file tree
Changes from 2 commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
a26122a
remove hash
gatorsmile Aug 4, 2016
3b3f3c8
Merge remote-tracking branch 'upstream/master' into removeHash
gatorsmile Aug 4, 2016
790f9e6
Merge remote-tracking branch 'upstream/master' into removeHash
gatorsmile Aug 12, 2016
abdeadf
Merge remote-tracking branch 'upstream/master' into removeHash
gatorsmile Aug 12, 2016
05390ad
Merge remote-tracking branch 'upstream/master' into removeHash
gatorsmile Nov 7, 2016
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
80 changes: 36 additions & 44 deletions ...atibility/src/test/scala/org/apache/spark/sql/hive/execution/HiveCompatibilitySuite.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -57,8 +57,6 @@ class HiveCompatibilitySuite extends HiveQueryFileTest with BeforeAndAfter {
TestHive.setConf(SQLConf.COLUMN_BATCH_SIZE, 5)
// Enable in-memory partition pruning for testing purposes
TestHive.setConf(SQLConf.IN_MEMORY_PARTITION_PRUNING, true)
// Use Hive hash expression instead of the native one
TestHive.sessionState.functionRegistry.unregisterFunction("hash")
// Ensures that the plans generation use metastore relation and not OrcRelation
// Was done because SqlBuilder does not work with plans having logical relation
TestHive.setConf(HiveUtils.CONVERT_METASTORE_ORC, false)
Expand All @@ -76,7 +74,6 @@ class HiveCompatibilitySuite extends HiveQueryFileTest with BeforeAndAfter {
TestHive.setConf(SQLConf.IN_MEMORY_PARTITION_PRUNING, originalInMemoryPartitionPruning)
TestHive.setConf(HiveUtils.CONVERT_METASTORE_ORC, originalConvertMetastoreOrc)
TestHive.setConf(SQLConf.CROSS_JOINS_ENABLED, originalCrossJoinEnabled)
TestHive.sessionState.functionRegistry.restore()

// For debugging dump some statistics about how much time was spent in various optimizer rules
logWarning(RuleExecutor.dumpTimeSpent())
Expand Down Expand Up @@ -553,7 +550,42 @@ class HiveCompatibilitySuite extends HiveQueryFileTest with BeforeAndAfter {
"union31",
"union_date",
"varchar_2",
"varchar_join1"
"varchar_join1",

// These tests are based on the Hive's hash function, which is different from Spark
"auto_join_nulls",
"auto_join0",
"auto_join1",
"auto_join2",
"auto_join3",
"auto_join4",
"auto_join5",
"auto_join6",
"auto_join7",
"auto_join8",
"auto_join9",
"auto_join10",
"auto_join11",
"auto_join12",
"auto_join13",
"auto_join14",
"auto_join14_hadoop20",
"auto_join15",
"auto_join17",
"auto_join18",
"auto_join19",
"auto_join20",
"auto_join22",
"auto_join25",
"auto_join30",
"auto_join31",
"correlationoptimizer1",
"correlationoptimizer2",
"correlationoptimizer3",
"correlationoptimizer4",
"multiMapJoin1",
"orc_dictionary_threshold",
"udf_hash"
)

/**
Expand All @@ -573,39 +605,6 @@ class HiveCompatibilitySuite extends HiveQueryFileTest with BeforeAndAfter {
"annotate_stats_part",
"annotate_stats_table",
"annotate_stats_union",
"auto_join0",
"auto_join1",
"auto_join10",
"auto_join11",
"auto_join12",
"auto_join13",
"auto_join14",
"auto_join14_hadoop20",
"auto_join15",
"auto_join17",
"auto_join18",
"auto_join19",
"auto_join2",
"auto_join20",
"auto_join21",
"auto_join22",
"auto_join23",
"auto_join24",
"auto_join25",
"auto_join26",
"auto_join27",
"auto_join28",
"auto_join3",
"auto_join30",
"auto_join31",
"auto_join4",
"auto_join5",
"auto_join6",
"auto_join7",
"auto_join8",
"auto_join9",
"auto_join_nulls",
"auto_join_reordering_values",
"binary_constant",
"binarysortable_1",
"cast1",
Expand All @@ -618,15 +617,11 @@ class HiveCompatibilitySuite extends HiveQueryFileTest with BeforeAndAfter {
"compute_stats_long",
"compute_stats_string",
"convert_enum_to_string",
"correlationoptimizer1",
"correlationoptimizer10",
"correlationoptimizer11",
"correlationoptimizer13",
"correlationoptimizer14",
"correlationoptimizer15",
"correlationoptimizer2",
"correlationoptimizer3",
"correlationoptimizer4",
"correlationoptimizer6",
"correlationoptimizer7",
"correlationoptimizer8",
Expand Down Expand Up @@ -867,7 +862,6 @@ class HiveCompatibilitySuite extends HiveQueryFileTest with BeforeAndAfter {
"merge2",
"merge4",
"mergejoins",
"multiMapJoin1",
"multiMapJoin2",
"multi_insert_gby",
"multi_insert_gby3",
Expand All @@ -889,7 +883,6 @@ class HiveCompatibilitySuite extends HiveQueryFileTest with BeforeAndAfter {
"nullinput2",
"nullscript",
"optional_outer",
"orc_dictionary_threshold",
"order",
"order2",
"outer_join_ppr",
Expand Down Expand Up @@ -1022,7 +1015,6 @@ class HiveCompatibilitySuite extends HiveQueryFileTest with BeforeAndAfter {
"udf_from_unixtime",
"udf_greaterthan",
"udf_greaterthanorequal",
"udf_hash",
"udf_hex",
"udf_if",
"udf_index",
Expand Down
1 change: 0 additions & 1 deletion sql/hive/src/main/scala/org/apache/spark/sql/hive/HiveSessionCatalog.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -235,7 +235,6 @@ private[sql] class HiveSessionCatalog(
// parse_url_tuple, posexplode, reflect2,
// str_to_map, windowingtablefunction.
private val hiveFunctions = Seq(
"hash",
"histogram_numeric",
"percentile",
"percentile_approx"
Expand Down
28 changes: 0 additions & 28 deletions sql/hive/src/main/scala/org/apache/spark/sql/hive/test/TestHive.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -487,24 +487,6 @@ private[hive] class TestHiveQueryExecution(
}
}


private[hive] class TestHiveFunctionRegistry extends SimpleFunctionRegistry {

private val removedFunctions =
collection.mutable.ArrayBuffer.empty[(String, (ExpressionInfo, FunctionBuilder))]

def unregisterFunction(name: String): Unit = synchronized {
functionBuilders.remove(name).foreach(f => removedFunctions += name -> f)
}

def restore(): Unit = synchronized {
removedFunctions.foreach {
case (name, (info, builder)) => registerFunction(name, info, builder)
}
}
}


private[hive] class TestHiveSessionState(
sparkSession: TestHiveSparkSession)
Copy link
Contributor

@cloud-fan cloud-fan Nov 5, 2016

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

We can still remove this class if we add back the removed spark builtin hash function manually right?

Copy link
Member Author

@gatorsmile gatorsmile Nov 7, 2016

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yeah, I think so

extends HiveSessionState(sparkSession) { self =>
Expand All @@ -520,16 +502,6 @@ private[hive] class TestHiveSessionState(
}
}

override lazy val functionRegistry: TestHiveFunctionRegistry = {
// We use TestHiveFunctionRegistry at here to track functions that have been explicitly
// unregistered (through TestHiveFunctionRegistry.unregisterFunction method).
val fr = new TestHiveFunctionRegistry
org.apache.spark.sql.catalyst.analysis.FunctionRegistry.expressions.foreach {
case (name, (info, builder)) => fr.registerFunction(name, info, builder)
}
fr
}

override def executePlan(plan: LogicalPlan): TestHiveQueryExecution = {
new TestHiveQueryExecution(sparkSession, plan)
}
Expand Down