The following node is available in the Open Source KNIME predictive analytics and data mining platform version 2.7.1. Discover over 1000 other nodes, as well as enterprise functionality at http://knime.com.

Variable Based File Reader

This node can be used to read data from an ASCII file or URL location. The location is read from a scope variable, provided by the input port. It can be configured to read in various formats.
When you open the node's configuration dialog and provide a variable name that contains a valid file location, it tries to guess the reader's settings by analyzing the content of the file. Check the results of these settings in the preview table. If the data shown is not correct or an error is reported, you can adjust the settings manually (see below).
In order for the available variables to have values assigned to them, the predecessor node should be executed before the settings are adjusted.
In most cases a node will be connected to the port that assigns different values to variables in a loop.

If a big file is selected (or the analysis of the file takes longer) the dialog shows a "Start Analysis" button. The analysis is then not triggered every time you change one setting, but only when you click this button. You can also cut the analysis short by clicking the "Stop Analysis" button. Then the analyzer doesn't examine the entire file to check the column types but stops after determining the types. Execution of the node could then fail, if an incompatible data item occurs in the data file.
It is recommended to trigger file analysis before applying the dialog settings to the node. If you click OK (or Apply) and didn't analyze the data file before, an analysis is launched that can't be canceled and that will examine the entire file. Make sure a preview table is visible before you apply the settings.

Dialog Options

Location variable name
Select the name of the variable that holds the location of the file. When you press ENTER, the file is analyzed and the settings pre-set.
Preserve user settings
If checked, the checkmarks and column names/types you explictly entered are preserved even if you select a new file. By default, the analyzer starts with fresh default settings for each new file location.
Read row IDs
If checked, the first column in the file is used as row IDs. If not checked, default row headers are created.
Read column headers
If checked, the items in the first line of the file are used as column names. Otherwise default column names are created.
Column delimiter
Enter the character(s) that separate the data tokens in the file, or select a delimiter from the list.
Ignore spaces and tabs
If checked, spaces and the TAB characters are ignored (not in quoted strings though).
Java style comment
Everything between '/*' and '*/' is ignored. Also everything after '//' until the end of the line.
Single line comment
Enter one or more characters that will indicate the start of a comment (ended by a new line).
Advanced...
Opens a new dialog with advanced settings. There is support for quotes, different decimal separators in floating point numbers, and character encoding. Also, for ignoring whitespaces, for allowing rows with too few data items, for making row IDs unique (not recommended for huge files), for missing value pattern, and for limiting the number of rows read in.
Click on the table header
If the column header in the preview table is clicked, a new dialog opens where column properties can be set: name and type can be changed (and will be fixed then). A pattern can be entered that will cause a "missing cell" to be created when it's read for this column. Additionally, possible values of the column domain can be updated by selecting "Domain". And, you can choose to skip this column entirely, i.e. it will not be included in the output table then.

Ports

Output Ports
0 Datatable just read from the file
This node is contained in KNIME Base Nodes provided by KNIME GmbH, Konstanz, Germany.