From 59147981c16b966c4dcbda012f951d7230d7df28 Mon Sep 17 00:00:00 2001
From: taiyang-li <654010905@qq.com>
Date: Wed, 4 Sep 2024 16:00:06 +0800
Subject: [PATCH] add uts

---
 .../tpch/GlutenClickHouseTPCHSaltNullParquetSuite.scala     | 6 ++++++
 .../gluten/utils/clickhouse/ClickHouseTestSettings.scala    | 2 +-
 .../gluten/utils/clickhouse/ClickHouseTestSettings.scala    | 2 +-
 .../gluten/utils/clickhouse/ClickHouseTestSettings.scala    | 4 +++-
 .../gluten/utils/clickhouse/ClickHouseTestSettings.scala    | 2 +-
 5 files changed, 12 insertions(+), 4 deletions(-)

diff --git a/backends-clickhouse/src/test/scala/org/apache/gluten/execution/tpch/GlutenClickHouseTPCHSaltNullParquetSuite.scala b/backends-clickhouse/src/test/scala/org/apache/gluten/execution/tpch/GlutenClickHouseTPCHSaltNullParquetSuite.scala
index 1db37e00f946..bdf285cd5aa6 100644
--- a/backends-clickhouse/src/test/scala/org/apache/gluten/execution/tpch/GlutenClickHouseTPCHSaltNullParquetSuite.scala
+++ b/backends-clickhouse/src/test/scala/org/apache/gluten/execution/tpch/GlutenClickHouseTPCHSaltNullParquetSuite.scala
@@ -2936,5 +2936,11 @@ class GlutenClickHouseTPCHSaltNullParquetSuite extends GlutenClickHouseTPCHAbstr
         checkBHJWithIsNullAwareAntiJoin(df)
       })
   }
+
+  test("soundex") {
+    runQueryAndCompare("select soundex(c_comment) from customer limit 50") {
+      checkGlutenOperatorMatch[ProjectExecTransformer]
+    }
+  }
 }
 // scalastyle:on line.size.limit
diff --git a/gluten-ut/spark32/src/test/scala/org/apache/gluten/utils/clickhouse/ClickHouseTestSettings.scala b/gluten-ut/spark32/src/test/scala/org/apache/gluten/utils/clickhouse/ClickHouseTestSettings.scala
index e99b33f18ae6..8571871ab37c 100644
--- a/gluten-ut/spark32/src/test/scala/org/apache/gluten/utils/clickhouse/ClickHouseTestSettings.scala
+++ b/gluten-ut/spark32/src/test/scala/org/apache/gluten/utils/clickhouse/ClickHouseTestSettings.scala
@@ -911,7 +911,7 @@ class ClickHouseTestSettings extends BackendTestSettings {
     .exclude("ParseUrl")
     .exclude("SPARK-33468: ParseUrl in ANSI mode should fail if input string is not a valid url")
     .exclude("FORMAT") // refer https://github.com/apache/incubator-gluten/issues/6765
-    .exclude("soundex unit test")
+    .exclude("soundex unit test") // CH and spark returns different results when input non-ASCII characters
     .excludeGlutenTest("SPARK-40213: ascii for Latin-1 Supplement characters")
   enableSuite[GlutenTryCastSuite]
     .exclude("null cast")
diff --git a/gluten-ut/spark33/src/test/scala/org/apache/gluten/utils/clickhouse/ClickHouseTestSettings.scala b/gluten-ut/spark33/src/test/scala/org/apache/gluten/utils/clickhouse/ClickHouseTestSettings.scala
index 6b418c4c98a7..a357004af52a 100644
--- a/gluten-ut/spark33/src/test/scala/org/apache/gluten/utils/clickhouse/ClickHouseTestSettings.scala
+++ b/gluten-ut/spark33/src/test/scala/org/apache/gluten/utils/clickhouse/ClickHouseTestSettings.scala
@@ -907,7 +907,7 @@ class ClickHouseTestSettings extends BackendTestSettings {
     .exclude("ParseUrl")
     .exclude("SPARK-33468: ParseUrl in ANSI mode should fail if input string is not a valid url")
     .exclude("FORMAT") // refer https://github.com/apache/incubator-gluten/issues/6765
-    .exclude("soundex unit test")
+    .exclude("soundex unit test") // CH and spark returns different results when input non-ASCII characters
   enableSuite[GlutenTryCastSuite]
     .exclude("null cast")
     .exclude("cast string to date")
diff --git a/gluten-ut/spark34/src/test/scala/org/apache/gluten/utils/clickhouse/ClickHouseTestSettings.scala b/gluten-ut/spark34/src/test/scala/org/apache/gluten/utils/clickhouse/ClickHouseTestSettings.scala
index 8d7797c20574..e8ea718fcb6e 100644
--- a/gluten-ut/spark34/src/test/scala/org/apache/gluten/utils/clickhouse/ClickHouseTestSettings.scala
+++ b/gluten-ut/spark34/src/test/scala/org/apache/gluten/utils/clickhouse/ClickHouseTestSettings.scala
@@ -811,7 +811,9 @@ class ClickHouseTestSettings extends BackendTestSettings {
     .exclude("ParseUrl")
     .exclude("SPARK-33468: ParseUrl in ANSI mode should fail if input string is not a valid url")
     .exclude("FORMAT") // refer https://github.com/apache/incubator-gluten/issues/6765
-    .exclude("soundex unit test")
+    .exclude(
+      "soundex unit test"
+    ) // CH and spark returns different results when input non-ASCII characters
   enableSuite[GlutenDataSourceV2DataFrameSessionCatalogSuite]
   enableSuite[GlutenDataSourceV2SQLSessionCatalogSuite]
   enableSuite[GlutenDataSourceV2SQLSuiteV1Filter]
diff --git a/gluten-ut/spark35/src/test/scala/org/apache/gluten/utils/clickhouse/ClickHouseTestSettings.scala b/gluten-ut/spark35/src/test/scala/org/apache/gluten/utils/clickhouse/ClickHouseTestSettings.scala
index 0e7dca03dcf0..bde904b55b08 100644
--- a/gluten-ut/spark35/src/test/scala/org/apache/gluten/utils/clickhouse/ClickHouseTestSettings.scala
+++ b/gluten-ut/spark35/src/test/scala/org/apache/gluten/utils/clickhouse/ClickHouseTestSettings.scala
@@ -811,7 +811,7 @@ class ClickHouseTestSettings extends BackendTestSettings {
     .exclude("ParseUrl")
     .exclude("SPARK-33468: ParseUrl in ANSI mode should fail if input string is not a valid url")
     .exclude("FORMAT") // refer https://github.com/apache/incubator-gluten/issues/6765
-    .exclude("soundex unit test")
+    .exclude("soundex unit test") // CH and spark returns different results when input non-ASCII characters
   enableSuite[GlutenDataSourceV2DataFrameSessionCatalogSuite]
   enableSuite[GlutenDataSourceV2SQLSessionCatalogSuite]
   enableSuite[GlutenDataSourceV2SQLSuiteV1Filter]