opt join

src/Columns/ColumnConst.h

@@ -121,6 +121,11 @@ class ColumnConst final : public COWHelper<IColumn, ColumnConst>
         return data->isNullAt(0);
     }
 
+    void insertIndicesFrom(const IColumn &, const Selector & selector) override
+    {
+        s += selector.size();
+    }
+
     void insertRangeFrom(const IColumn &, size_t /*start*/, size_t length) override
     {
         s += length;

src/Columns/ColumnDecimal.cpp

@@ -342,6 +342,19 @@ void ColumnDecimal<T>::insertData(const char * src, size_t /*length*/)
     data.emplace_back(tmp);
 }
 
+template <is_decimal T>
+void ColumnDecimal<T>::insertIndicesFrom(const IColumn & src, const IColumn::Selector & selector)
+{
+    size_t old_size = data.size();
+    data.resize(old_size + selector.size());
+    const ColumnDecimal & src_dec = assert_cast<const ColumnDecimal &>(src);
+    const Container & src_data = src_dec.getData();
+    for (size_t i = 0; i < selector.size(); i++)
+    {
+        data[old_size + i] = src_data[selector[i]];
+    }
+}
+
 template <is_decimal T>
 void ColumnDecimal<T>::insertRangeFrom(const IColumn & src, size_t start, size_t length)
 {

src/Columns/ColumnDecimal.h

@@ -62,6 +62,7 @@ class ColumnDecimal final : public COWHelper<ColumnVectorHelper, ColumnDecimal<T
     void insertDefault() override { data.push_back(T()); }
     void insertManyDefaults(size_t length) override { data.resize_fill(data.size() + length); }
     void insert(const Field & x) override { data.push_back(x.get<T>()); }
+    void insertIndicesFrom(const IColumn & src, const IColumn::Selector & selector) override;
     void insertRangeFrom(const IColumn & src, size_t start, size_t length) override;
     void insertRangeSelective(const IColumn & src, const IColumn::Selector & selector, size_t selector_start, size_t length) override;
 

src/Columns/ColumnNullable.cpp

@@ -239,6 +239,15 @@ const char * ColumnNullable::skipSerializedInArena(const char * pos) const
     return pos;
 }
 
+void ColumnNullable::insertIndicesFrom(const IColumn & src, const Selector & selector)
+{
+    const ColumnNullable & nullable_col = assert_cast<const ColumnNullable &>(src);
+    const IColumn & src_null_map = *nullable_col.null_map;
+    const IColumn & src_nested_column = *nullable_col.nested_column;
+    getNullMapColumn().insertIndicesFrom(src_null_map, selector);
+    getNestedColumn().insertIndicesFrom(src_nested_column, selector);
+}
+
 void ColumnNullable::insertRangeFrom(const IColumn & src, size_t start, size_t length)
 {
     const ColumnNullable & nullable_col = assert_cast<const ColumnNullable &>(src);
@@ -280,6 +289,12 @@ void ColumnNullable::insertFromNotNullable(const IColumn & src, size_t n)
     getNullMapData().push_back(0);
 }
 
+void ColumnNullable::insertIndicesFromNotNullable(const IColumn & src, const IColumn::Selector & selector)
+{
+    getNestedColumn().insertIndicesFrom(src, selector);
+    getNullMapColumn().insertMany(0, selector.size());
+}
+
 void ColumnNullable::insertRangeFromNotNullable(const IColumn & src, size_t start, size_t length)
 {
     getNestedColumn().insertRangeFrom(src, start, length);

src/Columns/ColumnNullable.h

@@ -66,12 +66,14 @@ class ColumnNullable final : public COWHelper<IColumn, ColumnNullable>
     StringRef serializeValueIntoArena(size_t n, Arena & arena, char const *& begin, const UInt8 * null_bit) const override;
     const char * deserializeAndInsertFromArena(const char * pos) override;
     const char * skipSerializedInArena(const char * pos) const override;
+    void insertIndicesFrom(const IColumn & src, const Selector & selector) override;
     void insertRangeFrom(const IColumn & src, size_t start, size_t length) override;
     void insertRangeSelective(const IColumn & src, const IColumn::Selector & selector, size_t selector_start, size_t length) override;
     void insert(const Field & x) override;
     void insertFrom(const IColumn & src, size_t n) override;
 
     void insertFromNotNullable(const IColumn & src, size_t n);
+    void insertIndicesFromNotNullable(const IColumn & src, const IColumn::Selector & selector);
     void insertRangeFromNotNullable(const IColumn & src, size_t start, size_t length);
     void insertManyFromNotNullable(const IColumn & src, size_t position, size_t length);
 

src/Columns/ColumnString.cpp

@@ -107,6 +107,11 @@ void ColumnString::updateWeakHash32(WeakHash32 & hash) const
     }
 }
 
+void ColumnString::insertIndicesFrom(const IColumn & src, const Selector & selector)
+{
+    for (auto sel : selector)
+        ColumnString::insertFrom(src, sel);
+}
 
 void ColumnString::insertRangeFrom(const IColumn & src, size_t start, size_t length)
 {

src/Columns/ColumnString.h

@@ -192,6 +192,7 @@ class ColumnString final : public COWHelper<IColumn, ColumnString>
         hash.update(reinterpret_cast<const char *>(offsets.data()), offsets.size() * sizeof(offsets[0]));
         hash.update(reinterpret_cast<const char *>(chars.data()), chars.size() * sizeof(chars[0]));
     }
+    void insertIndicesFrom(const IColumn & src, const Selector & selector) override;
 
     void insertRangeFrom(const IColumn & src, size_t start, size_t length) override;
 

src/Columns/ColumnVector.cpp

@@ -464,6 +464,19 @@ Float32 ColumnVector<T>::getFloat32(size_t n [[maybe_unused]]) const
         throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Cannot get the value of {} as Float32", TypeName<T>);
 }
 
+template <typename T>
+void ColumnVector<T>::insertIndicesFrom(const IColumn & src, const IColumn::Selector & selector)
+{
+    size_t old_size = data.size();
+    data.resize(old_size + selector.size());
+    const ColumnVector & src_vec = assert_cast<const ColumnVector &>(src);
+    const Container & src_data = src_vec.getData();
+    for (size_t i = 0; i < selector.size(); i++)
+    {
+        data[old_size + i] = src_data[selector[i]];
+    }
+}
+
 template <typename T>
 void ColumnVector<T>::insertRangeFrom(const IColumn & src, size_t start, size_t length)
 {

src/Columns/ColumnVector.h

@@ -240,6 +240,8 @@ class ColumnVector final : public COWHelper<ColumnVectorHelper, ColumnVector<T>>
         data.push_back(static_cast<T>(x.get<T>()));
     }
 
+    void insertIndicesFrom(const IColumn & src, const IColumn::Selector & selector) override;
+
     void insertRangeFrom(const IColumn & src, size_t start, size_t length) override;
 
     ColumnPtr filter(const IColumn::Filter & filt, ssize_t result_size_hint) const override;

src/Columns/IColumn.cpp

@@ -35,6 +35,12 @@ void IColumn::insertFrom(const IColumn & src, size_t n)
     insert(src[n]);
 }
 
+void IColumn::insertIndicesFrom(const IColumn & src, const Selector & selector)
+{
+    for (auto sel : selector)
+        insertFrom(src, sel);
+}
+
 ColumnPtr IColumn::createWithOffsets(const Offsets & offsets, const ColumnConst & column_with_default_value, size_t total_rows, size_t shift) const
 {
     if (offsets.size() + shift != size())

src/Columns/IColumn.h

@@ -379,6 +379,7 @@ class IColumn : public COW<IColumn>
     using ColumnIndex = UInt64;
     using Selector = PaddedPODArray<ColumnIndex>;
     [[nodiscard]] virtual std::vector<MutablePtr> scatter(ColumnIndex num_columns, const Selector & selector) const = 0;
+    virtual void insertIndicesFrom(const IColumn & src, const Selector & selector);
 
     /// This function will get row index from selector and append the data to this column.
     /// This function will handle indexes start from input 'selector_start' and will append 'size' times

src/Disks/ObjectStorages/HDFS/HDFSObjectStorage.h

@@ -120,6 +120,7 @@ class HDFSObjectStorage : public IObjectStorage
     const Poco::Util::AbstractConfiguration & config;
 
     HDFSBuilderWrapper hdfs_builder;
+protected:
     HDFSFSPtr hdfs_fs;
     SettingsPtr settings;
     const std::string hdfs_root_path;