Correct return value when substr is not found in INSTR method

miland-db · Mar 22, 2024 · db2453a · db2453a
1 parent 834be70
commit db2453a
Show file tree

Hide file tree

Showing 2 changed files with 7 additions and 3 deletions.
diff --git a/common/unsafe/src/main/java/org/apache/spark/unsafe/types/UTF8String.java b/common/unsafe/src/main/java/org/apache/spark/unsafe/types/UTF8String.java
@@ -900,7 +900,7 @@ private int collatedIndexOf(UTF8String substring, int collationId) {
       }
     }
 
-    return 0;
+    return -1;
   }
 
   /**

diff --git a/sql/core/src/test/scala/org/apache/spark/sql/CollationStringExpressionsSuite.scala b/sql/core/src/test/scala/org/apache/spark/sql/CollationStringExpressionsSuite.scala
@@ -84,16 +84,18 @@ class CollationStringExpressionsSuite extends QueryTest
       Literal.create("Aa", StringType(1))), 1)
     checkEvaluation(StringInstr(Collate(Literal("aaads"), "UTF8_BINARY_LCASE"),
       Collate(Literal("Aa"), "UTF8_BINARY_LCASE")), 1)
+    checkEvaluation(StringInstr(Collate(Literal("aaads"), "UTF8_BINARY_LCASE"),
+      Collate(Literal("de"), "UTF8_BINARY_LCASE")), 0)
     // UNICODE
     checkEvaluation(StringInstr(Literal.create("aaads", StringType(2)),
       Literal.create("Aa", StringType(2))), 0)
     checkEvaluation(StringInstr(Collate(Literal("aaads"), "UNICODE"),
-      Collate(Literal("Aa"), "UNICODE")), 0)
+      Collate(Literal("de"), "UNICODE")), 0)
     // UNICODE_CI
     checkEvaluation(StringInstr(Literal.create("aaads", StringType(3)),
       Literal.create("de", StringType(3))), 0)
     checkEvaluation(StringInstr(Collate(Literal("aaads"), "UNICODE_CI"),
-      Collate(Literal("Aa"), "UNICODE_CI")), 0)
+      Collate(Literal("AD"), "UNICODE_CI")), 3)
   }
 
   test("INSTR fail mismatched collation types") {
@@ -175,6 +177,8 @@ class CollationStringExpressionsSuite extends QueryTest
       Collate(Literal("abc,b,ab,c,dEf"), "UNICODE_CI")), 5)
     checkEvaluation(FindInSet(Collate(Literal("DEFG"), "UNICODE_CI"),
       Collate(Literal("abc,b,ab,c,def"), "UNICODE_CI")), 0)
+    checkEvaluation(FindInSet(Collate(Literal("dsf"), "UNICODE_CI"),
+      Collate(Literal("abc,b,ab,c,def"), "UNICODE_CI")), 0)
   }
 
   test("FIND_IN_SET fail mismatched collation types") {
-Original file line number
+Diff line change
@@ Expand Up @@
           }
         }
-        return 0;
+        return -1;
       }
       /**
@@ Expand Down @@