Add tests for encode/decode utf-32

apache · May 8, 2024 · 9e7debe · 9e7debe
1 parent a785897
commit 9e7debe
Show file tree

Hide file tree

Showing 6 changed files with 33 additions and 0 deletions.
diff --git a/...yst/src/test/scala/org/apache/spark/sql/catalyst/expressions/StringExpressionsSuite.scala b/...yst/src/test/scala/org/apache/spark/sql/catalyst/expressions/StringExpressionsSuite.scala
@@ -489,6 +489,8 @@ class StringExpressionsSuite extends SparkFunSuite with ExpressionEvalHelper {
     // non ascii characters are not allowed in the code, so we disable the scalastyle here.
     checkEvaluation(
       StringDecode(Encode(Literal("大千世界"), Literal("UTF-16LE")), Literal("UTF-16LE")), "大千世界")
+    checkEvaluation(
+      StringDecode(Encode(Literal("大千世界"), Literal("UTF-32")), Literal("UTF-32")), "大千世界")
     checkEvaluation(
       StringDecode(Encode(a, Literal("utf-8")), Literal("utf-8")), "大千世界", create_row("大千世界"))
     checkEvaluation(

diff --git a/sql/core/src/test/resources/sql-tests/analyzer-results/ansi/string-functions.sql.out b/sql/core/src/test/resources/sql-tests/analyzer-results/ansi/string-functions.sql.out
@@ -750,6 +750,13 @@ Project [decode(encode(abc, utf-8, false), utf-8) AS decode(encode(abc, utf-8),
 +- OneRowRelation
 
 
+-- !query
+select decode(encode('大千世界', 'utf-32'), 'utf-32')
+-- !query analysis
+Project [decode(encode(大千世界, utf-32, false), utf-32) AS decode(encode(大千世界, utf-32), utf-32)#x]
++- OneRowRelation
+
+
 -- !query
 select decode(1, 1, 'Southlake')
 -- !query analysis

diff --git a/sql/core/src/test/resources/sql-tests/analyzer-results/string-functions.sql.out b/sql/core/src/test/resources/sql-tests/analyzer-results/string-functions.sql.out
@@ -750,6 +750,13 @@ Project [decode(encode(abc, utf-8, false), utf-8) AS decode(encode(abc, utf-8),
 +- OneRowRelation
 
 
+-- !query
+select decode(encode('大千世界', 'utf-32'), 'utf-32')
+-- !query analysis
+Project [decode(encode(大千世界, utf-32, false), utf-32) AS decode(encode(大千世界, utf-32), utf-32)#x]
++- OneRowRelation
+
+
 -- !query
 select decode(1, 1, 'Southlake')
 -- !query analysis

diff --git a/sql/core/src/test/resources/sql-tests/inputs/string-functions.sql b/sql/core/src/test/resources/sql-tests/inputs/string-functions.sql
@@ -131,6 +131,7 @@ select encode(scol, ecol) from values('hello', 'Windows-xxx') as t(scol, ecol);
 select decode();
 select decode(encode('abc', 'utf-8'));
 select decode(encode('abc', 'utf-8'), 'utf-8');
+select decode(encode('大千世界', 'utf-32'), 'utf-32');
 select decode(1, 1, 'Southlake');
 select decode(2, 1, 'Southlake');
 select decode(2, 1, 'Southlake', 2, 'San Francisco', 3, 'New Jersey', 4, 'Seattle', 'Non domestic');

diff --git a/sql/core/src/test/resources/sql-tests/results/ansi/string-functions.sql.out b/sql/core/src/test/resources/sql-tests/results/ansi/string-functions.sql.out
@@ -961,6 +961,14 @@ struct<decode(encode(abc, utf-8), utf-8):string>
 abc
 
 
+-- !query
+select decode(encode('大千世界', 'utf-32'), 'utf-32')
+-- !query schema
+struct<decode(encode(大千世界, utf-32), utf-32):string>
+-- !query output
+大千世界
+
+
 -- !query
 select decode(1, 1, 'Southlake')
 -- !query schema

diff --git a/sql/core/src/test/resources/sql-tests/results/string-functions.sql.out b/sql/core/src/test/resources/sql-tests/results/string-functions.sql.out
@@ -893,6 +893,14 @@ struct<decode(encode(abc, utf-8), utf-8):string>
 abc
 
 
+-- !query
+select decode(encode('大千世界', 'utf-32'), 'utf-32')
+-- !query schema
+struct<decode(encode(大千世界, utf-32), utf-32):string>
+-- !query output
+大千世界
+
+
 -- !query
 select decode(1, 1, 'Southlake')
 -- !query schema