thevar1able/ClickHouse
1
0
Fork 0
mirror of https://github.com/ClickHouse/ClickHouse.git synced 2024-11-21 15:12:02 +00:00
Code Issues Packages Projects Releases Wiki Activity

Merge 2d32356ad7 into 44b4bd38b9

Browse Source
This commit is contained in:
ricky 2024-11-20 16:25:57 -08:00 committed by GitHub
commit 7405c7e527
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
3 changed files with 38 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
contrib/CMakeLists.txt vendored
View File

@ -195,10 +195,10 @@ if (ARCH_S390X)
add_contrib(crc32-s390x-cmake crc32-s390x)
endif()
add_contrib (SimSIMD-cmake SimSIMD)
option(ENABLE_USEARCH "Enable USearch" ${ENABLE_LIBRARIES})
if (ENABLE_USEARCH)
add_contrib (FP16-cmake FP16)
add_contrib (SimSIMD-cmake SimSIMD)
add_contrib (usearch-cmake usearch) # requires: FP16, SimdSIMD
else ()
message(STATUS "Not using USearch")

4
src/Functions/CMakeLists.txt
View File

@ -134,6 +134,10 @@ list (APPEND OBJECT_LIBS $<TARGET_OBJECTS:clickhouse_functions_array>)
add_subdirectory(Kusto)
list (APPEND OBJECT_LIBS $<TARGET_OBJECTS:clickhouse_functions_kusto>)
if (TARGET ch_contrib::simsimd)
list (APPEND PRIVATE_LIBS ch_contrib::simsimd)
endif()
if (TARGET ch_contrib::datasketches)
add_subdirectory(UniqTheta)
list (APPEND OBJECT_LIBS $<TARGET_OBJECTS:clickhouse_functions_uniqtheta>)

41
src/Functions/bitHammingDistance.cpp
View File

@ -1,6 +1,8 @@
#include <bit>
#include <Functions/FunctionBinaryArithmetic.h>
#include <Functions/FunctionFactory.h>
#include <bit>
#pragma clang diagnostic ignored "-Wused-but-marked-unused"
#include <simsimd/simsimd.h>
namespace DB
@ -26,17 +28,40 @@ struct BitHammingDistanceImpl
if constexpr (sizeof(A) <= sizeof(UInt64) && sizeof(B) <= sizeof(UInt64))
{
UInt64 res = static_cast<UInt64>(a) ^ static_cast<UInt64>(b);
return std::popcount(res);
if constexpr (std::is_same_v<A, B>)
{
simsimd_distance_t distance;
simsimd_hamming_b8(
reinterpret_cast<const simsimd_b8_t *>(&a), reinterpret_cast<const simsimd_b8_t *>(&b), sizeof(A), &distance);
return static_cast<UInt64>(distance);
}
else
{
UInt64 res = static_cast<UInt64>(a) ^ static_cast<UInt64>(b);
return std::popcount(res);
}
}
else if constexpr (is_big_int_v<A> && is_big_int_v<B>)
{
auto xored = a ^ b;
if constexpr (std::is_same_v<A, B>)
{
simsimd_distance_t distance;
simsimd_hamming_b8(
reinterpret_cast<const simsimd_b8_t *>(&a.items),
reinterpret_cast<const simsimd_b8_t *>(&b.items),
sizeof(a.items),
&distance);
return static_cast<ResultType>(distance);
}
else
{
auto xored = a ^ b;
ResultType res = 0;
for (auto item : xored.items)
res += std::popcount(item);
return res;
ResultType res = 0;
for (auto item : xored.items)
res += std::popcount(item);
return res;
}
}
else
throw Exception(ErrorCodes::ILLEGAL_COLUMN, "Unsupported data type combination in function 'bitHammingDistance'");