Implemented javaHashUTF16LE()

dbms/src/Functions/FunctionsHashing.cpp

@@ -23,6 +23,7 @@ void registerFunctionsHashing(FunctionFactory & factory)
     factory.registerFunction<FunctionIntHash64>();
     factory.registerFunction<FunctionURLHash>();
     factory.registerFunction<FunctionJavaHash>();
+    factory.registerFunction<FunctionJavaHashUTF16LE>();
     factory.registerFunction<FunctionHiveHash>();
     factory.registerFunction<FunctionMurmurHash2_32>();
     factory.registerFunction<FunctionMurmurHash2_64>();

dbms/src/Functions/FunctionsHashing.h

@@ -353,6 +353,41 @@ struct JavaHashImpl
     static constexpr bool use_int_hash_for_pods = false;
 };
 
+struct JavaHashUTF16LEImpl
+{
+    static constexpr auto name = "javaHashUTF16LE";
+    using ReturnType = Int32;
+
+    static Int32 apply(const char * raw_data, const size_t raw_size)
+    {
+        char * data = const_cast<char *>(raw_data);
+        size_t size = raw_size;
+
+        // Remove Byte-order-mark(0xFFFE) for UTF-16LE
+        if (size >= 2 && data[0] == '\xFF' && data[1] == '\xFE')
+        {
+            data += 2;
+            size -= 2;
+        }
+
+        if (size % 2 != 0)
+            throw Exception("Arguments for javaHashUTF16LE must be in the form of UTF-16", ErrorCodes::LOGICAL_ERROR);
+
+        UInt32 h = 0;
+        for (size_t i = 0; i < size; i += 2)
+            h = 31 * h + static_cast<UInt16>(static_cast<UInt8>(data[i]) | static_cast<UInt8>(data[i + 1]) << 8);
+
+        return static_cast<Int32>(h);
+    }
+
+    static Int32 combineHashes(Int32, Int32)
+    {
+        throw Exception("Java hash is not combineable for multiple arguments", ErrorCodes::NOT_IMPLEMENTED);
+    }
+
+    static constexpr bool use_int_hash_for_pods = false;
+};
+
 /// This is just JavaHash with zeroed out sign bit.
 /// This function is used in Hive for versions before 3.0,
 ///  after 3.0, Hive uses murmur-hash3.
@@ -1102,6 +1137,7 @@ using FunctionMurmurHash3_32 = FunctionAnyHash<MurmurHash3Impl32>;
 using FunctionMurmurHash3_64 = FunctionAnyHash<MurmurHash3Impl64>;
 using FunctionMurmurHash3_128 = FunctionStringHashFixedString<MurmurHash3Impl128>;
 using FunctionJavaHash = FunctionAnyHash<JavaHashImpl>;
+using FunctionJavaHashUTF16LE = FunctionAnyHash<JavaHashUTF16LEImpl>;
 using FunctionHiveHash = FunctionAnyHash<HiveHashImpl>;
 
 #if USE_XXHASH

dbms/tests/queries/0_stateless/00800_function_java_hash.reference

@@ -1,4 +1,7 @@
 96354
 -676697544
+138768
+-2143570108
+2145564783
 96354
 1470786104

dbms/tests/queries/0_stateless/00800_function_java_hash.sql

@@ -1,4 +1,7 @@
 select javaHash('abc');
 select javaHash('874293087');
+select javaHashUTF16LE(convertCharset('a1가', 'utf-8', 'utf-16le'));
+select javaHashUTF16LE(convertCharset('가나다라마바사아자차카타파하', 'utf-8', 'utf-16le'));
+select javaHashUTF16LE(convertCharset('FJKLDSJFIOLD_389159837589429', 'utf-8', 'utf-16le'));
 select hiveHash('abc');
 select hiveHash('874293087');