[GitHub] [spark] dtenedor commented on a diff in pull request #38663: [SPARK-41143][SQL] Add named argument function syntax support

2022-11-15 Thread GitBox


dtenedor commented on code in PR #38663:
URL: https://github.com/apache/spark/pull/38663#discussion_r1023072531


##
sql/catalyst/src/main/antlr4/org/apache/spark/sql/catalyst/parser/SqlBaseParser.g4:
##
@@ -769,7 +769,7 @@ inlineTable
 ;
 
 functionTable
-: funcName=functionName LEFT_PAREN (expression (COMMA expression)*)? 
RIGHT_PAREN tableAlias
+: funcName=functionName LEFT_PAREN (functionArgument (COMMA 
functionArgument)*)? RIGHT_PAREN tableAlias

Review Comment:
   here in the `.g4` file, we only change the syntax for table function calls, 
but the PR title and description mention general named argument support. Can we 
either:
   
   1. update the PR description, or
   2. update the `functionCall` rule to replace `argument+=expression` with 
`argument+=functionArgument`?
   
   Up to you, the latter would be nice since we can make all the parser changes 
in the same PR.



-- 
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-unsubscr...@spark.apache.org

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


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



[GitHub] [spark] dtenedor commented on a diff in pull request #38663: [SPARK-41143][SQL] Add named argument function syntax support

2022-11-15 Thread GitBox


dtenedor commented on code in PR #38663:
URL: https://github.com/apache/spark/pull/38663#discussion_r1023072531


##
sql/catalyst/src/main/antlr4/org/apache/spark/sql/catalyst/parser/SqlBaseParser.g4:
##
@@ -769,7 +769,7 @@ inlineTable
 ;
 
 functionTable
-: funcName=functionName LEFT_PAREN (expression (COMMA expression)*)? 
RIGHT_PAREN tableAlias
+: funcName=functionName LEFT_PAREN (functionArgument (COMMA 
functionArgument)*)? RIGHT_PAREN tableAlias

Review Comment:
   here in the `.g4` file, we only change the syntax for table function calls, 
but the PR title and description mention general named argument support. Can we 
either update the PR description or else update the `functionCall` rule to 
replace `argument+=expression` with `argument+=functionArgument`? The latter 
would be nice since we can make all the parser changes in the same PR.



##
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/literals.scala:
##
@@ -324,6 +324,21 @@ object LiteralTreeBits {
   val nullLiteralBits: BitSet = new ImmutableBitSet(TreePattern.maxId, 
LITERAL.id, NULL_LITERAL.id)
 }
 
+case class NamedArgumentExpression(key: String, value: Expression) extends 
LeafExpression {

Review Comment:
   this is in `literals.scala`, but this new expression is not really a literal 
:) can we maybe put this into its own file with a class comment?



##
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/plans/logical/NamedArgumentFunction.scala:
##
@@ -0,0 +1,92 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements.  See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License.  You may obtain a copy of the License at
+ *
+ *http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.spark.sql.catalyst.plans.logical
+
+import java.util.Locale
+
+import scala.collection.mutable
+
+import org.apache.spark.sql.catalyst.expressions.{Expression, 
NamedArgumentExpression}
+import 
org.apache.spark.sql.errors.QueryCompilationErrors.{tableFunctionDuplicateNamedArguments,
 tableFunctionUnexpectedArgument}
+import org.apache.spark.sql.types._
+
+/**
+ * A trait to define a named argument function:
+ * Usage: _FUNC_(arg0, arg1, arg2, arg5 => value5, arg8 => value8)
+ *
+ * - Arguments can be passed positionally or by name
+ * - Positional arguments cannot come after a named argument

Review Comment:
   from the example, named arguments can come after positional arguments. write 
that in the comment too? also maybe we can add other constraints:
   
   * no function call may include two arguments with the same name
   * case sensitivity follows the SQLConf.CASE_SENSITIVE boolean configuration
   * the function signature must specify the argument names, and the provided 
argument names must match the names in the function signature



##
sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/parser/PlanParserSuite.scala:
##
@@ -852,6 +852,31 @@ class PlanParserSuite extends AnalysisTest {
 stop = 43))
   }
 
+  test("table valued function with named arguments") {
+// All named arguments
+assertEqual(
+  "select * from my_tvf(arg1 => 'value1', arg2 => true)",
+  UnresolvedTableValuedFunction("my_tvf",
+NamedArgumentExpression("arg1", Literal("value1")) ::
+  NamedArgumentExpression("arg2", Literal(true)) :: Nil, 
Seq.empty).select(star()))
+
+// Unnamed and named arguments
+assertEqual(
+  "select * from my_tvf(2, arg1 => 'value1', arg2 => true)",
+  UnresolvedTableValuedFunction("my_tvf",
+Literal(2) ::
+  NamedArgumentExpression("arg1", Literal("value1")) ::
+  NamedArgumentExpression("arg2", Literal(true)) :: Nil, 
Seq.empty).select(star()))
+
+// Mixed arguments
+assertEqual(
+  "select * from my_tvf(arg1 => 'value1', 2, arg2 => true)",

Review Comment:
   thanks for adding these parser tests! can we also add some query tests in 
e.g. `SQLQuerySuite` that show what happens if we try to analyze such function 
calls with named arguments? do we get an error message, or does the whole 
feature work end-to-end?



##
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/plans/logical/NamedArgumentFunction.scala:
##
@@ -0,0 +1,92 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor licens