#pragma once #include #include #include #include #include #include namespace DB { /** A column of array values. * In memory, it is represented as one column of a nested type, whose size is equal to the sum of the sizes of all arrays, * and as an array of offsets in it, which allows you to get each element. */ class ColumnArray final : public COWHelper { private: friend class COWHelper; /** Create an array column with specified values and offsets. */ ColumnArray(MutableColumnPtr && nested_column, MutableColumnPtr && offsets_column); /** Create an empty column of arrays with the type of values as in the column `nested_column` */ explicit ColumnArray(MutableColumnPtr && nested_column); ColumnArray(const ColumnArray &) = default; public: /** Create immutable column using immutable arguments. This arguments may be shared with other columns. * Use IColumn::mutate in order to make mutable column and mutate shared nested columns. */ using Base = COWHelper; static Ptr create(const ColumnPtr & nested_column, const ColumnPtr & offsets_column) { return ColumnArray::create(nested_column->assumeMutable(), offsets_column->assumeMutable()); } static Ptr create(const ColumnPtr & nested_column) { return ColumnArray::create(nested_column->assumeMutable()); } template ::value>::type> static MutablePtr create(Args &&... args) { return Base::create(std::forward(args)...); } /** On the index i there is an offset to the beginning of the i + 1 -th element. */ using ColumnOffsets = ColumnVector; std::string getName() const override; const char * getFamilyName() const override { return "Array"; } TypeIndex getDataType() const override { return TypeIndex::Array; } MutableColumnPtr cloneResized(size_t size) const override; size_t size() const override; Field operator[](size_t n) const override; void get(size_t n, Field & res) const override; StringRef getDataAt(size_t n) const override; void insertData(const char * pos, size_t length) override; StringRef serializeValueIntoArena(size_t n, Arena & arena, char const *& begin) const override; const char * deserializeAndInsertFromArena(const char * pos) override; void updateHashWithValue(size_t n, SipHash & hash) const override; void updateWeakHash32(WeakHash32 & hash) const override; void updateHashFast(SipHash & hash) const override; void insertRangeFrom(const IColumn & src, size_t start, size_t length) override; void insert(const Field & x) override; void insertFrom(const IColumn & src_, size_t n) override; void insertDefault() override; void popBack(size_t n) override; ColumnPtr filter(const Filter & filt, ssize_t result_size_hint) const override; ColumnPtr permute(const Permutation & perm, size_t limit) const override; ColumnPtr index(const IColumn & indexes, size_t limit) const override; template ColumnPtr indexImpl(const PaddedPODArray & indexes, size_t limit) const; int compareAt(size_t n, size_t m, const IColumn & rhs_, int nan_direction_hint) const override; void compareColumn(const IColumn & rhs, size_t rhs_row_num, PaddedPODArray * row_indexes, PaddedPODArray & compare_results, int direction, int nan_direction_hint) const override; int compareAtWithCollation(size_t n, size_t m, const IColumn & rhs_, int nan_direction_hint, const Collator & collator) const override; void getPermutation(bool reverse, size_t limit, int nan_direction_hint, Permutation & res) const override; void updatePermutation(bool reverse, size_t limit, int nan_direction_hint, Permutation & res, EqualRanges & equal_range) const override; void getPermutationWithCollation(const Collator & collator, bool reverse, size_t limit, int nan_direction_hint, Permutation & res) const override; void updatePermutationWithCollation(const Collator & collator, bool reverse, size_t limit, int nan_direction_hint, Permutation & res, EqualRanges& equal_range) const override; void reserve(size_t n) override; size_t byteSize() const override; size_t byteSizeAt(size_t n) const override; size_t allocatedBytes() const override; void protect() override; ColumnPtr replicate(const Offsets & replicate_offsets) const override; ColumnPtr convertToFullColumnIfConst() const override; void getExtremes(Field & min, Field & max) const override; bool hasEqualOffsets(const ColumnArray & other) const; /** More efficient methods of manipulation */ IColumn & getData() { return *data; } const IColumn & getData() const { return *data; } IColumn & getOffsetsColumn() { return *offsets; } const IColumn & getOffsetsColumn() const { return *offsets; } Offsets & ALWAYS_INLINE getOffsets() { return assert_cast(*offsets).getData(); } const Offsets & ALWAYS_INLINE getOffsets() const { return assert_cast(*offsets).getData(); } const ColumnPtr & getDataPtr() const { return data; } ColumnPtr & getDataPtr() { return data; } const ColumnPtr & getOffsetsPtr() const { return offsets; } ColumnPtr & getOffsetsPtr() { return offsets; } MutableColumns scatter(ColumnIndex num_columns, const Selector & selector) const override { return scatterImpl(num_columns, selector); } void gather(ColumnGathererStream & gatherer_stream) override; ColumnPtr compress() const override; void forEachSubcolumn(ColumnCallback callback) override { callback(offsets); callback(data); } bool structureEquals(const IColumn & rhs) const override { if (auto rhs_concrete = typeid_cast(&rhs)) return data->structureEquals(*rhs_concrete->data); return false; } bool isCollationSupported() const override { return getData().isCollationSupported(); } private: WrappedPtr data; WrappedPtr offsets; size_t ALWAYS_INLINE offsetAt(ssize_t i) const { return getOffsets()[i - 1]; } size_t ALWAYS_INLINE sizeAt(ssize_t i) const { return getOffsets()[i] - getOffsets()[i - 1]; } /// Multiply values if the nested column is ColumnVector. template ColumnPtr replicateNumber(const Offsets & replicate_offsets) const; /// Multiply the values if the nested column is ColumnString. The code is too complicated. ColumnPtr replicateString(const Offsets & replicate_offsets) const; /** Non-constant arrays of constant values are quite rare. * Most functions can not work with them, and does not create such columns as a result. * An exception is the function `replicate` (see FunctionsMiscellaneous.h), which has service meaning for the implementation of lambda functions. * Only for its sake is the implementation of the `replicate` method for ColumnArray(ColumnConst). */ ColumnPtr replicateConst(const Offsets & replicate_offsets) const; /** The following is done by simply replicating of nested columns. */ ColumnPtr replicateTuple(const Offsets & replicate_offsets) const; ColumnPtr replicateNullable(const Offsets & replicate_offsets) const; ColumnPtr replicateGeneric(const Offsets & replicate_offsets) const; /// Specializations for the filter function. template ColumnPtr filterNumber(const Filter & filt, ssize_t result_size_hint) const; ColumnPtr filterString(const Filter & filt, ssize_t result_size_hint) const; ColumnPtr filterTuple(const Filter & filt, ssize_t result_size_hint) const; ColumnPtr filterNullable(const Filter & filt, ssize_t result_size_hint) const; ColumnPtr filterGeneric(const Filter & filt, ssize_t result_size_hint) const; int compareAtImpl(size_t n, size_t m, const IColumn & rhs_, int nan_direction_hint, const Collator * collator=nullptr) const; template void getPermutationImpl(size_t limit, Permutation & res, Comparator cmp) const; template void updatePermutationImpl(size_t limit, Permutation & res, EqualRanges & equal_range, Comparator cmp) const; }; }