Table Indexer

Creates an index from the input table. Each table row is represented as a document in the index. Each document contains at least one index field which contains the row id. The index might contain more fields depending on the selected table columns to index. The generated index can be searched using the Index Query node.

Options

Columns to index

The columns to index. Only the columns that are added to the index can be used in subsequent nodes.

Save original data

Select yes if the original value of the cell should be part of the index. If the original value is stored the data table can be recreated from the index in the Index Query node. However storing the original value might consume considerably more memory and disc space.

Index type

The "Memory" type keeps the index in the memory and is thus faster but consumes more memory. The "Temporary File"" type caches only parts of the index in the memory and stores the rest in a temporary file. It is thus slower than the memory type but can handle much larger data tables.

No of threads

The number of threads to use for indexing. Decrease this number if you experience memory problems.

Views

This node has no views

Workflows

• 01_Advanced_QueriesKNIME Hub
• 01_Advanced_QueriesKNIME Hub
• 01_データテーブルのインデックス作成とクエリKNIME Hub
• 01_データテーブルのインデックス作成とクエリKNIME Hub
• 02_Document_QueriesKNIME Hub
• Show all 18 workflows

Links

• Table Normalization: Mapping (Index) Table for one key and many valuesby mlauber71 on 2019-10-29

  forum.knime.com/p/73489

Developers