Data Tables Overview


With TIBCO Spotfire is possible to work with more than one data table within a single analysis. Below is a short description of the different concepts used when handling multiple data tables.

A data table is either fetched from a data source, or created within the application. Data loaded from a data source can be handled either in-memory or in-database depending on how it is added to the analysis. In-memory data tables have one or more columns and zero or more rows, whereas in-database data tables technically do not contain any data but simply fetch the requested data directly from the source. See Data Overview for more information.

In-memory data tables can be linked or embedded. Linked data tables can be loaded completely into the application, but if the source is an information link they can also be configured to load data on demand only.

Data tables can be related to each other, using primary and/or foreign keys (key columns), but they can also be unrelated. When data tables are related, any marking or filtering in one data table may be propagated to the other related data tables.

Data tables can have column matches between them if columns are of the same data type. A column match is used to aggregate data correctly when the data from different data tables is combined in a single visualization. Column matches are much looser in nature than the relation mentioned above, because they do not join the data tables together. Marking and filtering is still individual for each data table even if they have columns matching between them. To learn more about column matches, see Matching Columns.

On-Demand Data Tables

On-demand data tables are data tables to which only rows related to a defined input are loaded. The input could be something like the marked rows in another, related, data table, the filtered rows of another data table or a property value selected in a text area. Changing the input means changing the "demand", i.e., that more, fewer or other rows are loaded into the data table. On-demand data tables can be used by Details Visualizations, and only data from information links or data connections can be loaded on demand.

In-Database Data Tables

While data from in-database data tables is retrieved only when needed, the use of an in-database data table as a details visualization may also be seen as a type of on-demand visualization.

Related Data Tables

As a means of helping you keep track of which data tables are related, a stripe of color will be added to the left of the filters in the filters panel when more than one data table is available. Filters from related data tables (which may affect each other when they are manipulated) all have the same color. Also, the visualizations that use related data tables will show the same color in the title bar, if it is displayed.

Note: You can specify whether or not filtering in a data table should affect what is shown in visualizations used by other, related data tables. The default setting is to ignore filtering in related data tables. See Filtering in Related Data Tables for more information.

Column Matches in Data Tables

To see which column matches exist in the visualization, open the Column Matches tab in Data Table Properties. You can also see which column matches are used in a specific visualization by opening the Visualization Properties dialog on the Data page for the visualization of interest.

  1. See How to Add Data Tables.

  1. Select Edit > Data Table Properties.

  2. Click on the data table you wish to remove from the analysis.

  3. Click on Delete.

  1. Select Edit > Data Table Properties.

  2. Click on the data table you wish to rename.

  3. Click on Rename....

  4. Type a new data table name and click OK.

Note: Reload may affect embedded data tables, as well as linked ones. See Embedded or Linked Data? for more information.

Note: Reload of in-database data tables will only reload the data, not the schema. See Data Source Settings if you need a full schema refresh.

  1. Select Edit > Data Table Properties.

  2. Click on the data table you wish to reload.

  3. Click Refresh Data.

    Comment: The Refresh Data button may be unavailable for some of your data tables. For example, this happens if  you have added rows or columns to an embedded data table, or if you have frozen some columns in an embedded data table. In that case the data table cannot be reloaded.

Note: Reloading all data tables may take a long time if one or more data tables are very big.

Note: Reload may affect embedded data tables, as well as linked ones. See Embedded or Linked Data? for more information.

  1. Click on the Reload Data button on the toolbar, data_reload_data_i.png.

  1. Select File > Add Data Tables....

  2. Add the data tables of interest to the dialog.

  3. Click on the information link or data table from a data connection holding the desired data to select it in the list.

  4. Select the Load on demand check box.

  5. Click on Settings....

  6. Define what type of input will control the on-demand loading.

  7. Comment: For more details see Loading Data on Demand and Details on Define Input.

  8. Click OK to close all dialogs.

  1. When the input controlling the on-demand data table is changed, a red refresh button is shown on the title bar of the visualization.

    Comment: If the title bar has been hidden, right-click on the visualization and select Properties. On the General page, select the Show title bar check box.

  2. Click on the refresh button, dr_refresh_calculation__icon.png.

  1. See Replacing Data.

  1. When the filtering behind a calculated data table is changed, a red refresh button is shown on the title bar of the visualization.

    Comment: If the title bar has been hidden, right-click on the visualization and select Properties. On the General page, select the Show title bar check box.

  2. Click on the refresh button, dr_refresh_calculation__icon.png.

All data tables currently in the analysis will be saved in the document when saving an analysis file. See Saving an Analysis File or Saving an Analysis File in the Library for more information.

  1. See Exporting Data.

  1. Select Edit > Data Table Properties.

  2. Click Linked to source.

  3. Select Prompt for new settings before loading.

  4. Click OK.

    Comment: You can also change this setting when you save your analysis. Click the Edit button in the Save dialog or in the third step of the Save as Library Item wizard. This will open the Data Table Properties dialog.

  1. Locate the data table header in the filters panel, or select the data table in the drop-down menu in the data panel.

  2. Use the filters in the filter panel, alternatively the data panel, to modify what is shown in the visualizations using the specified data table (and, optionally, in other related data tables).

 

See also:

Working with Data Tables

Multiple Data Tables in One Visualization

How to Edit Column Properties