Merge pull request #28965 from nicelulu/issues_28774

128bit hash-functions accepting arbitrary list of arguments

src/Common/SipHash.h

@@ -18,6 +18,7 @@
 #include <string>
 #include <type_traits>
 #include <Core/Defines.h>
+#include <base/extended_types.h>
 
 
 #define ROTL(x, b) static_cast<UInt64>(((x) << (b)) | ((x) >> (64 - (b))))
@@ -191,6 +192,15 @@ inline void sipHash128(const char * data, const size_t size, char * out)
     hash.get128(out);
 }
 
+inline UInt128 sipHash128(const char * data, const size_t size)
+{
+    SipHash hash;
+    hash.update(data, size);
+    UInt128 res;
+    hash.get128(res);
+    return res;
+}
+
 inline UInt64 sipHash64(const char * data, const size_t size)
 {
     SipHash hash;

src/Functions/FunctionsHashing.h

@@ -103,6 +103,14 @@ struct IntHash64Impl
     }
 };
 
+template<typename T, typename HashFunction>
+T combineHashesFunc(T t1, T t2)
+{
+    T hashes[] = {t1, t2};
+    return HashFunction::apply(reinterpret_cast<const char *>(hashes), 2 * sizeof(T));
+}
+
+
 #if USE_SSL
 struct HalfMD5Impl
 {
@@ -248,8 +256,7 @@ struct SipHash64Impl
 
     static UInt64 combineHashes(UInt64 h1, UInt64 h2)
     {
-        UInt64 hashes[] = {h1, h2};
-        return apply(reinterpret_cast<const char *>(hashes), 16);
+        return combineHashesFunc<UInt64, SipHash64Impl>(h1, h2);
     }
 
     static constexpr bool use_int_hash_for_pods = false;
@@ -258,12 +265,20 @@ struct SipHash64Impl
 struct SipHash128Impl
 {
     static constexpr auto name = "sipHash128";
-    enum { length = 16 };
 
-    static void apply(const char * begin, const size_t size, unsigned char * out_char_data)
+    using ReturnType = UInt128;
+
+    static UInt128 combineHashes(UInt128 h1, UInt128 h2)
     {
-        sipHash128(begin, size, reinterpret_cast<char*>(out_char_data));
+        return combineHashesFunc<UInt128, SipHash128Impl>(h1, h2);
     }
+
+    static UInt128 apply(const char * data, const size_t size)
+    {
+        return sipHash128(data, size);
+    }
+
+    static constexpr bool use_int_hash_for_pods = false;
 };
 
 /** Why we need MurmurHash2?
@@ -380,12 +395,22 @@ struct MurmurHash3Impl64
 struct MurmurHash3Impl128
 {
     static constexpr auto name = "murmurHash3_128";
-    enum { length = 16 };
 
-    static void apply(const char * begin, const size_t size, unsigned char * out_char_data)
+    using ReturnType = UInt128;
+
+    static UInt128 apply(const char * data, const size_t size)
     {
-        MurmurHash3_x64_128(begin, size, 0, out_char_data);
+        char bytes[16];
+        MurmurHash3_x64_128(data, size, 0, bytes);
+        return *reinterpret_cast<UInt128 *>(bytes);
     }
+
+    static UInt128 combineHashes(UInt128 h1, UInt128 h2)
+    {
+        return combineHashesFunc<UInt128, MurmurHash3Impl128>(h1, h2);
+    }
+
+    static constexpr bool use_int_hash_for_pods = false;
 };
 
 /// http://hg.openjdk.java.net/jdk8u/jdk8u/jdk/file/478a4add975b/src/share/classes/java/lang/String.java#l1452
@@ -1093,7 +1118,12 @@ public:
 
     DataTypePtr getReturnTypeImpl(const DataTypes & /*arguments*/) const override
     {
-        return std::make_shared<DataTypeNumber<ToType>>();
+        if constexpr (std::is_same_v<ToType, UInt128>) /// backward-compatible
+        {
+            return std::make_shared<DataTypeFixedString>(sizeof(UInt128));
+        }
+        else
+            return std::make_shared<DataTypeNumber<ToType>>();
     }
 
     ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t input_rows_count) const override
@@ -1115,6 +1145,13 @@ public:
         for (const auto & col : arguments)
             executeForArgument(col.type.get(), col.column.get(), vec_to, is_first_argument);
 
+        if constexpr (std::is_same_v<ToType, UInt128>) /// backward-compatible
+        {
+            auto col_to_fixed_string = ColumnFixedString::create(sizeof(UInt128));
+            col_to_fixed_string->getChars() = std::move(*reinterpret_cast<ColumnFixedString::Chars *>(&col_to->getData()));
+            return col_to_fixed_string;
+        }
+
         return col_to;
     }
 };
@@ -1359,7 +1396,7 @@ using FunctionSHA256 = FunctionStringHashFixedString<SHA256Impl>;
 using FunctionSHA384 = FunctionStringHashFixedString<SHA384Impl>;
 using FunctionSHA512 = FunctionStringHashFixedString<SHA512Impl>;
 #endif
-using FunctionSipHash128 = FunctionStringHashFixedString<SipHash128Impl>;
+using FunctionSipHash128 = FunctionAnyHash<SipHash128Impl>;
 using FunctionCityHash64 = FunctionAnyHash<ImplCityHash64>;
 using FunctionFarmFingerprint64 = FunctionAnyHash<ImplFarmFingerprint64>;
 using FunctionFarmHash64 = FunctionAnyHash<ImplFarmHash64>;
@@ -1370,7 +1407,7 @@ using FunctionMurmurHash2_64 = FunctionAnyHash<MurmurHash2Impl64>;
 using FunctionGccMurmurHash = FunctionAnyHash<GccMurmurHashImpl>;
 using FunctionMurmurHash3_32 = FunctionAnyHash<MurmurHash3Impl32>;
 using FunctionMurmurHash3_64 = FunctionAnyHash<MurmurHash3Impl64>;
-using FunctionMurmurHash3_128 = FunctionStringHashFixedString<MurmurHash3Impl128>;
+using FunctionMurmurHash3_128 = FunctionAnyHash<MurmurHash3Impl128>;
 
 using FunctionJavaHash = FunctionAnyHash<JavaHashImpl>;
 using FunctionJavaHashUTF16LE = FunctionAnyHash<JavaHashUTF16LEImpl>;

tests/queries/0_stateless/00678_murmurhash.sql

@@ -34,4 +34,3 @@ SELECT gccMurmurHash(1);
 
 SELECT hex(murmurHash3_128('foo'));
 SELECT hex(murmurHash3_128('\x01'));
-

tests/queries/0_stateless/00746_hashing_tuples.reference

@@ -1,6 +1,11 @@
 12940785793559895259
 17926972817233444501
 7456555839952096623
+CC45107CC4B79F62D831BEF2103C7CBF
+DF2EC2F0669B000EDFF6ADEE264E7D68
+4CD1C30C38AB935D418B5269EF197B9E
+9D78134EE48654D753CCA1B76185CF8E
+389D16428D2AADEC9713905572F42864
 955237314186186656
 8175794665478042155
 9325786087413524176
@@ -13,6 +18,8 @@
 8163029322371165472
 8788309436660676487
 236561483980029756
+8DD5527CC43D76F4760D26BE0F641F7E
+F8F7AD9B6CD4CF117A71E277E2EC2931
 12384823029245979431
 4507350192761038840
 1188926775431157506

tests/queries/0_stateless/00746_hashing_tuples.sql

@@ -4,6 +4,12 @@ SELECT sipHash64(1, 2, 3);
 SELECT sipHash64(1, 3, 2);
 SELECT sipHash64(('a', [1, 2, 3], 4, (4, ['foo', 'bar'], 1, (1, 2))));
 
+SELECT hex(sipHash128('foo'));
+SELECT hex(sipHash128('\x01'));
+SELECT hex(sipHash128('foo', 'foo'));
+SELECT hex(sipHash128('foo', 'foo', 'foo'));
+SELECT hex(sipHash128(1, 2, 3));
+
 SELECT halfMD5(1, 2, 3);
 SELECT halfMD5(1, 3, 2);
 SELECT halfMD5(('a', [1, 2, 3], 4, (4, ['foo', 'bar'], 1, (1, 2))));
@@ -20,6 +26,9 @@ SELECT murmurHash3_64(1, 2, 3);
 SELECT murmurHash3_64(1, 3, 2);
 SELECT murmurHash3_64(('a', [1, 2, 3], 4, (4, ['foo', 'bar'], 1, (1, 2))));
 
+SELECT hex(murmurHash3_128('foo', 'foo'));
+SELECT hex(murmurHash3_128('foo', 'foo', 'foo'));
+
 SELECT gccMurmurHash(1, 2, 3);
 SELECT gccMurmurHash(1, 3, 2);
 SELECT gccMurmurHash(('a', [1, 2, 3], 4, (4, ['foo', 'bar'], 1, (1, 2))));