Row Filter

The Row Filter node is one of the fundamental nodes that every KNIME developer must be familiar with. It allows users to include or exclude records in a variety of ways depending on their specific needs.

1) Include/Exclude rows by attribute value: These two options allow you to include or exclude rows according to either a pattern, a range of values, or by missing value matching. To begin, select the column at the top of the configuration menu by which you will filter the table.

In general, pattern matching is used for string data types and range checking is for numeric data types. Further, the pattern matching option allows for wildcard and regular expression matching. Wildcard matching uses the * character, such as "New*" would match all states that begin with "New".

2.) Include/Exclude rows by row number: These two options allow you to include or exclude rows according to their row position in the table. This is useful if you with to include the first N records in a table, or exclude unneccessary records that are always in the same location. The end of the row filter can either be set manually or dynamically to the end of the table. The beginning of the row filter must be set manually.

3.) Include/Exclude rows by Row ID: These two options allow you to include or exclude rows according to their Row ID values. Every KNIME table has a Row ID field that is separate from the other fields in the table. By default, these have names like "Row0," "Row1," and so on (although these can be modified in various ways.

This option allows you to write a regular expression to include or exclude rows according to these Row ID values.