Stack Node

About:  This article introduces the Stack node, an Input node within Construct.
Location: Node panel

Table of Contents

• Feature Overview
• Using the Stack Node
• Defining the Schema in the Stack Node
• Video Tutorial
• Related Articles

Feature Overview

A Stack node is a special data connection that appends the records of several co-located data files together and makes them available as if they were a single source. Once configured, it is integrated and used in jobs like any other Input node.

Using the Stack Node

Select the files to be combined by entering some common portion of their names. Special Parameter characters can be used here. Wildcard characters (*) may be needed to achieve this.

Example:  "College Enrollment Data FA*.csv"

“Regex:” may also be used to specify groups of files to match when wildcards alone won’t suffice.

Example: “Regex:CI-..\.xls.*\.*$” 

The above example will match all Excel worksheets in workbooks that start with “CI-“, are followed by two characters, and have a file extension that starts with “xls”.

Checking the Create Source ID Column option box and entering a name will add a new column to the incoming data. The name of the source file for each record will appear in the field.

Checking either the Create File Created Column and/or Create File Modified Column option boxes will add new columns to the incoming data. They will be filled with the relevant file information for that data row.

Defining the Schema in the Stack Node

The schema of the first matching table will be used to define the data layout the Stack node presents to a job. In addition to opting in or out of certain fields using the individual column check boxes, users may select a schema mode to further control how this node behaves.

• FIXED will not adapt to changes to the list of columns. Discrepancies will be reported as errors.
• EXTENDABLE will automatically allow new data columns to appear in the output. Missing columns will be reported as errors.
• DYNAMIC will automatically alter the columns output to match the first table schema.
• LOCKED will always output the original columns list. Missing columns will populate with nulls.

Video Tutorial

Related Articles

• Input Node
• Input Nodes