#include #include #include #include #include namespace DB { namespace ErrorCodes { extern const int LOGICAL_ERROR; } template void SetVariantsTemplate::init(Type type_) { type = type_; switch (type) { case Type::EMPTY: break; #define M(NAME) \ case Type::NAME: (NAME) = std::make_unique(); break; APPLY_FOR_SET_VARIANTS(M) #undef M } } template size_t SetVariantsTemplate::getTotalRowCount() const { switch (type) { case Type::EMPTY: return 0; #define M(NAME) \ case Type::NAME: return (NAME)->data.size(); APPLY_FOR_SET_VARIANTS(M) #undef M } UNREACHABLE(); } template size_t SetVariantsTemplate::getTotalByteCount() const { switch (type) { case Type::EMPTY: return 0; #define M(NAME) \ case Type::NAME: return (NAME)->data.getBufferSizeInBytes(); APPLY_FOR_SET_VARIANTS(M) #undef M } UNREACHABLE(); } template typename SetVariantsTemplate::Type SetVariantsTemplate::chooseMethod(const ColumnRawPtrs & key_columns, Sizes & key_sizes) { /// Check if at least one of the specified keys is nullable. /// Create a set of nested key columns from the corresponding key columns. /// Here "nested" means that, if a key column is nullable, we take its nested /// column; otherwise we take the key column as is. ColumnRawPtrs nested_key_columns; nested_key_columns.reserve(key_columns.size()); bool has_nullable_key = false; for (const auto & col : key_columns) { if (const auto * nullable = checkAndGetColumn(*col)) { nested_key_columns.push_back(&nullable->getNestedColumn()); has_nullable_key = true; } else nested_key_columns.push_back(col); } size_t keys_size = nested_key_columns.size(); bool all_fixed = true; size_t keys_bytes = 0; key_sizes.resize(keys_size); for (size_t j = 0; j < keys_size; ++j) { if (!nested_key_columns[j]->isFixedAndContiguous()) { all_fixed = false; break; } key_sizes[j] = nested_key_columns[j]->sizeOfValueIfFixed(); keys_bytes += key_sizes[j]; } if (has_nullable_key) { /// At least one key is nullable. Therefore we choose a method /// that takes into account this fact. if ((keys_size == 1) && (nested_key_columns[0]->isNumeric())) { /// We have exactly one key and it is nullable. We shall add it a tag /// which specifies whether its value is null or not. size_t size_of_field = nested_key_columns[0]->sizeOfValueIfFixed(); if ((size_of_field == 1) || (size_of_field == 2) || (size_of_field == 4) || (size_of_field == 8)) return Type::nullable_keys128; /// Pass to more generic method } if (all_fixed) { /// Pack if possible all the keys along with information about which key values are nulls /// into a fixed 16- or 32-byte blob. if (keys_bytes > (std::numeric_limits::max() - std::tuple_size>::value)) throw Exception(ErrorCodes::LOGICAL_ERROR, "Aggregator: keys sizes overflow"); if ((std::tuple_size>::value + keys_bytes) <= 16) return Type::nullable_keys128; if ((std::tuple_size>::value + keys_bytes) <= 32) return Type::nullable_keys256; } /// Fallback case. return Type::hashed; } /// If there is one numeric key that fits into 64 bits if (keys_size == 1 && nested_key_columns[0]->isNumeric() && !nested_key_columns[0]->lowCardinality()) { size_t size_of_field = nested_key_columns[0]->sizeOfValueIfFixed(); if (size_of_field == 1) return Type::key8; if (size_of_field == 2) return Type::key16; if (size_of_field == 4) return Type::key32; if (size_of_field == 8) return Type::key64; if (size_of_field == 16) return Type::keys128; if (size_of_field == 32) return Type::keys256; throw Exception(ErrorCodes::LOGICAL_ERROR, "Numeric column has sizeOfField not in 1, 2, 4, 8, 16, 32."); } /// If the keys fit in N bits, we will use a hash table for N-bit-packed keys if (all_fixed && keys_bytes <= 16) return Type::keys128; if (all_fixed && keys_bytes <= 32) return Type::keys256; /// If there is single string key, use hash table of it's values. if (keys_size == 1 && (typeid_cast(nested_key_columns[0]) || (isColumnConst(*nested_key_columns[0]) && typeid_cast(&assert_cast(nested_key_columns[0])->getDataColumn())))) return Type::key_string; if (keys_size == 1 && typeid_cast(nested_key_columns[0])) return Type::key_fixed_string; /// Otherwise, will use set of cryptographic hashes of unambiguously serialized values. return Type::hashed; } template struct SetVariantsTemplate; template struct SetVariantsTemplate; }