From 5585ed93b09bc05cdd7a731650eca50d43d7159b Mon Sep 17 00:00:00 2001
From: Takeshi YAMAMURO <linguin.m.s@gmail.com>
Date: Thu, 12 Jan 2017 09:46:53 -0800
Subject: [PATCH] [SPARK-17237][SQL] Remove backticks in a pivot result schema
## What changes were proposed in this pull request?
Pivoting adds backticks (e.g. 3_count(\`c\`)) in column names and, in some cases,
thes causes analysis exceptions like;
```
scala> val df = Seq((2, 3, 4), (3, 4, 5)).toDF("a", "x", "y")
scala> df.groupBy("a").pivot("x").agg(count("y"), avg("y")).na.fill(0)
org.apache.spark.sql.AnalysisException: syntax error in attribute name: `3_count(`y`)`;
at org.apache.spark.sql.catalyst.analysis.UnresolvedAttribute$.e$1(unresolved.scala:134)
at org.apache.spark.sql.catalyst.analysis.UnresolvedAttribute$.parseAttributeName(unresolved.scala:144)
...
```
So, this pr proposes to remove these backticks from column names.
## How was this patch tested?
Added a test in `DataFrameAggregateSuite`.
Author: Takeshi YAMAMURO <linguin.m.s@gmail.com>
Closes #14812 from maropu/SPARK-17237.
---
.../org/apache/spark/sql/catalyst/analysis/Analyzer.scala | 2 +-
.../org/apache/spark/sql/DataFrameAggregateSuite.scala | 8 ++++++++
.../scala/org/apache/spark/sql/DataFramePivotSuite.scala | 2 +-
3 files changed, 10 insertions(+), 2 deletions(-)
diff --git a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/Analyzer.scala b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/Analyzer.scala
index 3c58832d34..1957df89e6 100644
--- a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/Analyzer.scala
+++ b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/Analyzer.scala
@@ -477,7 +477,7 @@ class Analyzer(
} else {
val suffix = aggregate match {
case n: NamedExpression => n.name
- case _ => aggregate.sql
+ case _ => toPrettySQL(aggregate)
}
value + "_" + suffix
}
diff --git a/sql/core/src/test/scala/org/apache/spark/sql/DataFrameAggregateSuite.scala b/sql/core/src/test/scala/org/apache/spark/sql/DataFrameAggregateSuite.scala
index 7853b22fec..e7079120bb 100644
--- a/sql/core/src/test/scala/org/apache/spark/sql/DataFrameAggregateSuite.scala
+++ b/sql/core/src/test/scala/org/apache/spark/sql/DataFrameAggregateSuite.scala
@@ -530,4 +530,12 @@ class DataFrameAggregateSuite extends QueryTest with SharedSQLContext {
limit2Df.groupBy("id").count().select($"id"),
limit2Df.select($"id"))
}
+
+ test("SPARK-17237 remove backticks in a pivot result schema") {
+ val df = Seq((2, 3, 4), (3, 4, 5)).toDF("a", "x", "y")
+ checkAnswer(
+ df.groupBy("a").pivot("x").agg(count("y"), avg("y")).na.fill(0),
+ Seq(Row(3, 0, 0.0, 1, 5.0), Row(2, 1, 4.0, 0, 0.0))
+ )
+ }
}
diff --git a/sql/core/src/test/scala/org/apache/spark/sql/DataFramePivotSuite.scala b/sql/core/src/test/scala/org/apache/spark/sql/DataFramePivotSuite.scala
index a8d854ccbc..51ffe34172 100644
--- a/sql/core/src/test/scala/org/apache/spark/sql/DataFramePivotSuite.scala
+++ b/sql/core/src/test/scala/org/apache/spark/sql/DataFramePivotSuite.scala
@@ -200,7 +200,7 @@ class DataFramePivotSuite extends QueryTest with SharedSQLContext{
test("pivot preserves aliases if given") {
assertResult(
- Array("year", "dotNET_foo", "dotNET_avg(`earnings`)", "Java_foo", "Java_avg(`earnings`)")
+ Array("year", "dotNET_foo", "dotNET_avg(earnings)", "Java_foo", "Java_avg(earnings)")
)(
courseSales.groupBy($"year")
.pivot("course", Seq("dotNET", "Java"))
--
GitLab