Package org.apache.druid.query.rowsandcols

Class ArrayListRowsAndColumns<RowType>

java.lang.Object
org.apache.druid.query.rowsandcols.ArrayListRowsAndColumns<RowType>
Type Parameters:
RowType -
All Implemented Interfaces:
RowsAndColumns, AppendableRowsAndColumns
public class ArrayListRowsAndColumns<RowType> extends Object implements AppendableRowsAndColumns
ArrayListRowsAndColumns is a RowsAndColumns implementation that believes it has all of its data on-heap.

It is an AppendableRowsAndColumns and, as with all RowsAndColumns, it is not thread-safe for multiple writes. Under certain circumstances, concurrent reads from multiple threads can be correct, but the code has to be follow a very strict ordering of code that ensures that reads only happen after writes and once a value is read, it will never be overwritten.

Additionally, this object implements various of the semantic interfaces directly to provide some degree of processing and memory optimization.

  • Constructor Details

  • Method Details

    • getColumnNames

      public Collection<String> getColumnNames()
      Description copied from interface: RowsAndColumns
      The set of column names available from the RowsAndColumns
      Specified by:
      getColumnNames in interface RowsAndColumns
      Returns:
      The set of column names available from the RowsAndColumns

    • numRows

      public int numRows()
      Description copied from interface: RowsAndColumns
      The number of rows in the RowsAndColumns object
      Specified by:
      numRows in interface RowsAndColumns
      Returns:
      the integer number of rows

    • findColumn

      @Nullable public Column findColumn(String name)
      Description copied from interface: RowsAndColumns
      Finds a column by name. null is returned if the column is not found. The RowsAndColumns object should not attempt to default not-found columns to pretend as if they exist, instead the user of the RowsAndColumns object should decide the correct semantic interpretation of a column that does not exist. It is expected that most locations will choose to believe that the column does exist and is always null, but there are often optimizations that can effect this same assumption without doing a lot of extra work if the calling code knows that it does not exist.
      Specified by:
      findColumn in interface RowsAndColumns
      Parameters:
      name - the name of the column to find
      Returns:
      the Column, if found. null if not found.

    • as

      @Nullable public <T> T as(Class<T> clazz)
      Description copied from interface: RowsAndColumns
      Asks the RowsAndColumns to return itself as a concrete implementation of a specific interface. The interface asked for will tend to be a semantically-meaningful interface. This method allows the calling code to interrogate the RowsAndColumns object about whether it can offer a meaningful optimization of the semantic interface. If a RowsAndColumns cannot do anything specifically optimal for the interface requested, it should return null instead of trying to come up with its own default implementation.
      Specified by:
      as in interface RowsAndColumns
      Type Parameters:
      T - The interface that the calling code wants a concrete implementation of
      Parameters:
      clazz - A class object representing the interface that the calling code wants a concrete implementation of
      Returns:
      A concrete implementation of the interface, or null if there is no meaningful optimization to be had through a local implementation of the interface.

    • addColumn

      public void addColumn(String name, Column column)
      Description copied from interface: AppendableRowsAndColumns
      Mutates the RowsAndColumns by appending the requested Column.
      Specified by:
      addColumn in interface AppendableRowsAndColumns
      Parameters:
      name - the name of the new column
      column - the Column object representing the new column

    • sort

      public void sort(ArrayList<ColumnWithDirection> ordering)

    • toClusteredGroupPartitioner

      public ClusteredGroupPartitioner toClusteredGroupPartitioner()

    • toNaiveSortMaker

      public NaiveSortMaker toNaiveSortMaker()

    • toCursorFactory

      public CursorFactory toCursorFactory()