Window Analytics
===========



Type
--------- 

transform

Class
--------- 

fire.nodes.etl.NodeWindowAnalytics

Fields
--------- 

.. list-table::
      :widths: 10 5 10
      :header-rows: 1

      * - Name
        - Title
        - Description
      * - partitionByCols
        - PartitionBy
        - partition column names separated by comma(,)
      * - orderByCols
        - OrderBy
        - order by column names separated by comma(,)
      * - windowFunction
        - Window Function
        - Window Function Name
      * - analyticsCol
        - Analytics Column
        - 
      * - window_offset
        - Window Offset
        - It's used in lead and lag functions.


Details
-------
This node Generates a new Dataframe with Analytics Column appended to the incoming Dataframe.


Analytics Column is populated with value based on the Window Function selected.


Examples
-------
Incoming Dataframe has following rows:


::

    EMP_CD    |    EMP_NAME    |    DEPT    |    SALARY    |    AGE    
    ------------------------------------------------------------------------
    E01       |    ANTHONY     |    HR      |    50000     |    40
    E02       |    LISA        |    HR      |    45000     |    35
    E03       |    MARTIN      |    HR      |    20000     |    25
    E04       |    DAVID       |    SALES   |    55000     |    40
    E05       |    MARK        |    SALES   |    60000     |    45
    E06       |    JOE         |    SALES   |    40000     |    25
    E07       |    BELLA       |    HR      |    60000     |    24


If WindowAnalytics node is configured as below:
+++++++++++++++


PARTITIONBY      :     DEPT

ORDERBY          :     AGE

WINDOW FUNCTION  :     first_value

ANALYTICS COLUMN :     SALARY : integer

WINDOW OFFSET    :     1


then outgoing Dataframe would be created as below 

where incoming Dataframe is partitioned by [DEPT] and data is sorted by [AGE] and [FIRST VALUE] of Analytics Column [SALARY] within a partition is appended as new column:


::

    EMP_CD    |    EMP_NAME    |    DEPT    |    SALARY    |    AGE    |    first_value
    ---------------------------------------------------------------------------------------
    E07       |    BELLA       |    HR      |    60000     |    24     |    60000
    E03       |    MARTIN      |    HR      |    20000     |    25     |    60000
    E02       |    LISA        |    HR      |    45000     |    35     |    60000
    E01       |    ANTHONY     |    HR      |    50000     |    40     |    60000
    E06       |    JOE         |    SALES   |    40000     |    25     |    40000
    E04       |    DAVID       |    SALES   |    55000     |    40     |    40000
    E05       |    MARK        |    SALES   |    60000     |    45     |    40000


If WindowAnalytics node is configured as below:
+++++++++++++++


PARTITIONBY      :     DEPT

ORDERBY          :     AGE

WINDOW FUNCTION  :     lead

ANALYTICS COLUMN :     SALARY : integer

WINDOW OFFSET    :     2


then outgoing Dataframe would be created as below 

where incoming Dataframe is partitioned by [DEPT] and data is sorted by [AGE] and 

Analytics Column [SALARY] value of leading 2 or [WINDOW OFFSET] rows within a partition is appended as new column:


::

    EMP_CD    |    EMP_NAME    |    DEPT    |    SALARY    |    AGE    |    first_value
    ---------------------------------------------------------------------------------------
    E07       |    BELLA       |    HR      |    60000     |    24     |    45000
    E03       |    MARTIN      |    HR      |    20000     |    25     |    50000
    E02       |    LISA        |    HR      |    45000     |    35     |    
    E01       |    ANTHONY     |    HR      |    50000     |    40     |    
    E06       |    JOE         |    SALES   |    40000     |    25     |    60000
    E04       |    DAVID       |    SALES   |    55000     |    40     |    
    E05       |    MARK        |    SALES   |    60000     |    45     |