From de14086e1f6a2474bb9ba1452ada94e0ce58cf9c Mon Sep 17 00:00:00 2001
From: gatorsmile <gatorsmile@gmail.com>
Date: Wed, 5 Jul 2017 10:40:02 +0800
Subject: [PATCH] [SPARK-21295][SQL] Use qualified names in error message for
 missing references

### What changes were proposed in this pull request?
It is strange to see the following error message. Actually, the column is from another table.
```
cannot resolve '`right.a`' given input columns: [a, c, d];
```

After the PR, the error message looks like
```
cannot resolve '`right.a`' given input columns: [left.a, right.c, right.d];
```

### How was this patch tested?
Added a test case

Author: gatorsmile <gatorsmile@gmail.com>

Closes #18520 from gatorsmile/removeSQLConf.
---
 .../sql/catalyst/analysis/CheckAnalysis.scala  |  2 +-
 .../results/columnresolution-negative.sql.out  | 10 +++++-----
 .../results/columnresolution-views.sql.out     |  2 +-
 .../sql-tests/results/columnresolution.sql.out | 18 +++++++++---------
 .../sql-tests/results/group-by.sql.out         |  2 +-
 .../sql-tests/results/table-aliases.sql.out    |  2 +-
 .../org/apache/spark/sql/SubquerySuite.scala   |  4 ++--
 7 files changed, 20 insertions(+), 20 deletions(-)

diff --git a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/CheckAnalysis.scala b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/CheckAnalysis.scala
index fb81a7006b..85c52792ef 100644
--- a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/CheckAnalysis.scala
+++ b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/CheckAnalysis.scala
@@ -86,7 +86,7 @@ trait CheckAnalysis extends PredicateHelper {
       case operator: LogicalPlan =>
         operator transformExpressionsUp {
           case a: Attribute if !a.resolved =>
-            val from = operator.inputSet.map(_.name).mkString(", ")
+            val from = operator.inputSet.map(_.qualifiedName).mkString(", ")
             a.failAnalysis(s"cannot resolve '${a.sql}' given input columns: [$from]")
 
           case e: Expression if e.checkInputDataTypes().isFailure =>
diff --git a/sql/core/src/test/resources/sql-tests/results/columnresolution-negative.sql.out b/sql/core/src/test/resources/sql-tests/results/columnresolution-negative.sql.out
index 60bd8e9cc9..9e60e592c2 100644
--- a/sql/core/src/test/resources/sql-tests/results/columnresolution-negative.sql.out
+++ b/sql/core/src/test/resources/sql-tests/results/columnresolution-negative.sql.out
@@ -90,7 +90,7 @@ SELECT mydb1.t1.i1 FROM t1, mydb1.t1
 struct<>
 -- !query 10 output
 org.apache.spark.sql.AnalysisException
-cannot resolve '`mydb1.t1.i1`' given input columns: [i1, i1]; line 1 pos 7
+cannot resolve '`mydb1.t1.i1`' given input columns: [t1.i1, t1.i1]; line 1 pos 7
 
 
 -- !query 11
@@ -161,7 +161,7 @@ SELECT db1.t1.i1 FROM t1, mydb2.t1
 struct<>
 -- !query 18 output
 org.apache.spark.sql.AnalysisException
-cannot resolve '`db1.t1.i1`' given input columns: [i1, i1]; line 1 pos 7
+cannot resolve '`db1.t1.i1`' given input columns: [t1.i1, t1.i1]; line 1 pos 7
 
 
 -- !query 19
@@ -186,7 +186,7 @@ SELECT mydb1.t1 FROM t1
 struct<>
 -- !query 21 output
 org.apache.spark.sql.AnalysisException
-cannot resolve '`mydb1.t1`' given input columns: [i1]; line 1 pos 7
+cannot resolve '`mydb1.t1`' given input columns: [t1.i1]; line 1 pos 7
 
 
 -- !query 22
@@ -204,7 +204,7 @@ SELECT t1 FROM mydb1.t1
 struct<>
 -- !query 23 output
 org.apache.spark.sql.AnalysisException
-cannot resolve '`t1`' given input columns: [i1]; line 1 pos 7
+cannot resolve '`t1`' given input columns: [t1.i1]; line 1 pos 7
 
 
 -- !query 24
@@ -221,7 +221,7 @@ SELECT mydb1.t1.i1 FROM t1
 struct<>
 -- !query 25 output
 org.apache.spark.sql.AnalysisException
-cannot resolve '`mydb1.t1.i1`' given input columns: [i1]; line 1 pos 7
+cannot resolve '`mydb1.t1.i1`' given input columns: [t1.i1]; line 1 pos 7
 
 
 -- !query 26
diff --git a/sql/core/src/test/resources/sql-tests/results/columnresolution-views.sql.out b/sql/core/src/test/resources/sql-tests/results/columnresolution-views.sql.out
index 616421d6f2..7c451c2aa5 100644
--- a/sql/core/src/test/resources/sql-tests/results/columnresolution-views.sql.out
+++ b/sql/core/src/test/resources/sql-tests/results/columnresolution-views.sql.out
@@ -105,7 +105,7 @@ SELECT global_temp.view1.i1 FROM global_temp.view1
 struct<>
 -- !query 12 output
 org.apache.spark.sql.AnalysisException
-cannot resolve '`global_temp.view1.i1`' given input columns: [i1]; line 1 pos 7
+cannot resolve '`global_temp.view1.i1`' given input columns: [view1.i1]; line 1 pos 7
 
 
 -- !query 13
diff --git a/sql/core/src/test/resources/sql-tests/results/columnresolution.sql.out b/sql/core/src/test/resources/sql-tests/results/columnresolution.sql.out
index 764cad0e39..d3ca4443cc 100644
--- a/sql/core/src/test/resources/sql-tests/results/columnresolution.sql.out
+++ b/sql/core/src/test/resources/sql-tests/results/columnresolution.sql.out
@@ -96,7 +96,7 @@ SELECT mydb1.t1.i1 FROM t1
 struct<>
 -- !query 11 output
 org.apache.spark.sql.AnalysisException
-cannot resolve '`mydb1.t1.i1`' given input columns: [i1]; line 1 pos 7
+cannot resolve '`mydb1.t1.i1`' given input columns: [t1.i1]; line 1 pos 7
 
 
 -- !query 12
@@ -105,7 +105,7 @@ SELECT mydb1.t1.i1 FROM mydb1.t1
 struct<>
 -- !query 12 output
 org.apache.spark.sql.AnalysisException
-cannot resolve '`mydb1.t1.i1`' given input columns: [i1]; line 1 pos 7
+cannot resolve '`mydb1.t1.i1`' given input columns: [t1.i1]; line 1 pos 7
 
 
 -- !query 13
@@ -154,7 +154,7 @@ SELECT mydb1.t1.i1 FROM mydb1.t1
 struct<>
 -- !query 18 output
 org.apache.spark.sql.AnalysisException
-cannot resolve '`mydb1.t1.i1`' given input columns: [i1]; line 1 pos 7
+cannot resolve '`mydb1.t1.i1`' given input columns: [t1.i1]; line 1 pos 7
 
 
 -- !query 19
@@ -270,7 +270,7 @@ SELECT * FROM mydb1.t3 WHERE c1 IN
 struct<>
 -- !query 32 output
 org.apache.spark.sql.AnalysisException
-cannot resolve '`mydb1.t4.c3`' given input columns: [c2, c3]; line 2 pos 42
+cannot resolve '`mydb1.t4.c3`' given input columns: [t4.c2, t4.c3]; line 2 pos 42
 
 
 -- !query 33
@@ -287,7 +287,7 @@ SELECT mydb1.t1.i1 FROM t1, mydb2.t1
 struct<>
 -- !query 34 output
 org.apache.spark.sql.AnalysisException
-cannot resolve '`mydb1.t1.i1`' given input columns: [i1, i1]; line 1 pos 7
+cannot resolve '`mydb1.t1.i1`' given input columns: [t1.i1, t1.i1]; line 1 pos 7
 
 
 -- !query 35
@@ -296,7 +296,7 @@ SELECT mydb1.t1.i1 FROM mydb1.t1, mydb2.t1
 struct<>
 -- !query 35 output
 org.apache.spark.sql.AnalysisException
-cannot resolve '`mydb1.t1.i1`' given input columns: [i1, i1]; line 1 pos 7
+cannot resolve '`mydb1.t1.i1`' given input columns: [t1.i1, t1.i1]; line 1 pos 7
 
 
 -- !query 36
@@ -313,7 +313,7 @@ SELECT mydb1.t1.i1 FROM t1, mydb1.t1
 struct<>
 -- !query 37 output
 org.apache.spark.sql.AnalysisException
-cannot resolve '`mydb1.t1.i1`' given input columns: [i1, i1]; line 1 pos 7
+cannot resolve '`mydb1.t1.i1`' given input columns: [t1.i1, t1.i1]; line 1 pos 7
 
 
 -- !query 38
@@ -402,7 +402,7 @@ SELECT mydb1.t5.t5.i1 FROM mydb1.t5
 struct<>
 -- !query 48 output
 org.apache.spark.sql.AnalysisException
-cannot resolve '`mydb1.t5.t5.i1`' given input columns: [i1, t5]; line 1 pos 7
+cannot resolve '`mydb1.t5.t5.i1`' given input columns: [t5.i1, t5.t5]; line 1 pos 7
 
 
 -- !query 49
@@ -411,7 +411,7 @@ SELECT mydb1.t5.t5.i2 FROM mydb1.t5
 struct<>
 -- !query 49 output
 org.apache.spark.sql.AnalysisException
-cannot resolve '`mydb1.t5.t5.i2`' given input columns: [i1, t5]; line 1 pos 7
+cannot resolve '`mydb1.t5.t5.i2`' given input columns: [t5.i1, t5.t5]; line 1 pos 7
 
 
 -- !query 50
diff --git a/sql/core/src/test/resources/sql-tests/results/group-by.sql.out b/sql/core/src/test/resources/sql-tests/results/group-by.sql.out
index 14679850c6..e23ebd4e82 100644
--- a/sql/core/src/test/resources/sql-tests/results/group-by.sql.out
+++ b/sql/core/src/test/resources/sql-tests/results/group-by.sql.out
@@ -202,7 +202,7 @@ SELECT a AS k, COUNT(b) FROM testData GROUP BY k
 struct<>
 -- !query 21 output
 org.apache.spark.sql.AnalysisException
-cannot resolve '`k`' given input columns: [a, b]; line 1 pos 47
+cannot resolve '`k`' given input columns: [testdata.a, testdata.b]; line 1 pos 47
 
 
 -- !query 22
diff --git a/sql/core/src/test/resources/sql-tests/results/table-aliases.sql.out b/sql/core/src/test/resources/sql-tests/results/table-aliases.sql.out
index c318018dce..7abbcd834a 100644
--- a/sql/core/src/test/resources/sql-tests/results/table-aliases.sql.out
+++ b/sql/core/src/test/resources/sql-tests/results/table-aliases.sql.out
@@ -60,4 +60,4 @@ SELECT a AS col1, b AS col2 FROM testData AS t(c, d)
 struct<>
 -- !query 6 output
 org.apache.spark.sql.AnalysisException
-cannot resolve '`a`' given input columns: [c, d]; line 1 pos 7
+cannot resolve '`a`' given input columns: [t.c, t.d]; line 1 pos 7
diff --git a/sql/core/src/test/scala/org/apache/spark/sql/SubquerySuite.scala b/sql/core/src/test/scala/org/apache/spark/sql/SubquerySuite.scala
index 820cff655c..c0a3b5add3 100644
--- a/sql/core/src/test/scala/org/apache/spark/sql/SubquerySuite.scala
+++ b/sql/core/src/test/scala/org/apache/spark/sql/SubquerySuite.scala
@@ -870,9 +870,9 @@ class SubquerySuite extends QueryTest with SharedSQLContext {
 
   test("SPARK-20688: correctly check analysis for scalar sub-queries") {
     withTempView("t") {
-      Seq(1 -> "a").toDF("i", "j").createTempView("t")
+      Seq(1 -> "a").toDF("i", "j").createOrReplaceTempView("t")
       val e = intercept[AnalysisException](sql("SELECT (SELECT count(*) FROM t WHERE a = 1)"))
-      assert(e.message.contains("cannot resolve '`a`' given input columns: [i, j]"))
+      assert(e.message.contains("cannot resolve '`a`' given input columns: [t.i, t.j]"))
     }
   }
 }
-- 
GitLab