[GitHub] [spark] srielau commented on a diff in pull request #42755: [SPARK-45034][SQL] Support deterministic mode function

["srielau (via GitHub)" <gi...@apache.org>]

srielau commented on code in PR #42755:
URL: https://github.com/apache/spark/pull/42755#discussion_r1322715522


##########
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/aggregate/Mode.scala:
##########
@@ -18,15 +18,22 @@
 package org.apache.spark.sql.catalyst.expressions.aggregate
 
 import org.apache.spark.sql.catalyst.InternalRow
-import org.apache.spark.sql.catalyst.expressions.{Expression, ExpressionDescription, ImplicitCastInputTypes}
-import org.apache.spark.sql.catalyst.trees.UnaryLike
+import org.apache.spark.sql.catalyst.analysis.TypeCheckResult
+import org.apache.spark.sql.catalyst.analysis.TypeCheckResult.{DataTypeMismatch, TypeCheckSuccess}
+import org.apache.spark.sql.catalyst.expressions.{Expression, ExpressionDescription, ImplicitCastInputTypes, Literal}
+import org.apache.spark.sql.catalyst.trees.{BinaryLike, UnaryLike}
+import org.apache.spark.sql.catalyst.types.PhysicalDataType
 import org.apache.spark.sql.catalyst.util.GenericArrayData
-import org.apache.spark.sql.types.{AbstractDataType, AnyDataType, ArrayType, DataType}
+import org.apache.spark.sql.catalyst.util.TypeUtils.toSQLExpr
+import org.apache.spark.sql.errors.DataTypeErrors.toSQLType
+import org.apache.spark.sql.types.{AbstractDataType, AnyDataType, ArrayType, BooleanType, DataType}
 import org.apache.spark.util.collection.OpenHashMap
 
 // scalastyle:off line.size.limit
 @ExpressionDescription(
-  usage = "_FUNC_(col) - Returns the most frequent value for the values within `col`. NULL values are ignored. If all the values are NULL, or there are 0 rows, returns NULL.",
+  usage = """
+    _FUNC_(col[, deterministicResult]) - Returns the most frequent value for the values within `col`. NULL values are ignored. If all the values are NULL, or there are 0 rows, returns NULL.

Review Comment:
   +1 deterministic



##########
sql/core/src/test/resources/sql-tests/results/group-by.sql.out:
##########
@@ -1121,3 +1121,108 @@ struct<d:int>
 -- !query output
 0
 2
+
+
+-- !query
+SELECT mode(col) FROM VALUES (-10), (0), (10) AS tab(col)
+-- !query schema
+struct<mode(col, false):int>
+-- !query output
+0
+
+
+-- !query
+SELECT mode(col, false) FROM VALUES (-10), (0), (10) AS tab(col)
+-- !query schema
+struct<mode(col, false):int>
+-- !query output
+0
+
+
+-- !query
+SELECT mode(col, true) FROM VALUES (-10), (0), (10) AS tab(col)
+-- !query schema
+struct<mode(col, true):int>
+-- !query output
+-10
+
+
+-- !query
+SELECT mode(col, 'true') FROM VALUES (-10), (0), (10) AS tab(col)
+-- !query schema
+struct<>
+-- !query output
+org.apache.spark.sql.catalyst.ExtendedAnalysisException
+{
+  "errorClass" : "DATATYPE_MISMATCH.UNEXPECTED_INPUT_TYPE",
+  "sqlState" : "42K09",
+  "messageParameters" : {
+    "inputSql" : "\"true\"",
+    "inputType" : "\"STRING\"",
+    "paramIndex" : "2",
+    "requiredType" : "\"BOOLEAN\"",
+    "sqlExpr" : "\"mode(col, true)\""
+  },
+  "queryContext" : [ {
+    "objectType" : "",
+    "objectName" : "",
+    "startIndex" : 8,
+    "stopIndex" : 24,
+    "fragment" : "mode(col, 'true')"
+  } ]
+}
+
+
+-- !query
+SELECT mode(col, b) FROM VALUES (-10, false), (0, false), (10, false) AS tab(col, b)
+-- !query schema
+struct<>
+-- !query output
+org.apache.spark.sql.catalyst.ExtendedAnalysisException
+{
+  "errorClass" : "DATATYPE_MISMATCH.NON_FOLDABLE_INPUT",
+  "sqlState" : "42K09",
+  "messageParameters" : {
+    "inputExpr" : "\"b\"",
+    "inputName" : "deterministicResult",
+    "inputType" : "\"BOOLEAN\"",
+    "sqlExpr" : "\"mode(col, b)\""
+  },
+  "queryContext" : [ {
+    "objectType" : "",
+    "objectName" : "",
+    "startIndex" : 8,
+    "stopIndex" : 19,
+    "fragment" : "mode(col, b)"
+  } ]
+}
+
+
+-- !query
+SELECT mode(col) FROM VALUES (map(1, 'a')) AS tab(col)
+-- !query schema
+struct<mode(col, false):map<int,string>>
+-- !query output
+{1:"a"}
+
+
+-- !query
+SELECT mode(col, false) FROM VALUES (map(1, 'a')) AS tab(col)
+-- !query schema
+struct<mode(col, false):map<int,string>>
+-- !query output
+{1:"a"}
+
+
+-- !query
+SELECT mode(col, true) FROM VALUES (map(1, 'a')) AS tab(col)
+-- !query schema
+struct<>
+-- !query output
+org.apache.spark.SparkIllegalArgumentException
+{
+  "errorClass" : "_LEGACY_ERROR_TEMP_2005",
+  "messageParameters" : {
+    "dataType" : "PhysicalMapType"
+  }

Review Comment:
   Would it be possible to fix that legacy error while we're in the neighborhood?



-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org