Composer 6.4 Enhancements
This topic describes the significant enhancements in Composer 6.4.
Raw Data Table Changes
The following changes were made to raw data tables in this release.
- All UI references to "raw data tables" or "raw data table" have been changed to simply "tables" or "table." Pivot tables are still referred to as "pivot tables."
The column headings of a table now include a context menu you can use to perform various functions on the table, as described below. Select next to the column heading name to access the menu.
Menu Option Description Aggregation Shown only for numeric fields or metrics, this option allows you to change the aggregation method for the field. Valid options are AVG, MIN, MAX, SUM, NONE, or LAST VALUE. See Change Metric Aggregation in Tables. Autosize all columns This option causes Composer to autosize all the column widths in the table. See Change Column Widths. Group by <field> This option allows you to group the table by the field represented by the table column. See Group and Ungroup Table Data. Granularity Shown only for time fields used to group the table, this option allows you to change the granularity of the grouped time field. Valid options are Millisecond, Second, Minute, Hour, Day, Week, Month, and Year. See Change Time Field Granularity in Tables. Include Blanks Shown only for time fields used to group the table, this option allows you apply even time intervals for the grouped time field. Select this option once to apply even time intervals. Select it a second time to remove even time intervals. See Even Time Intervals. Ungroup <field> This option removes grouping by the table column. See Group and Ungroup Table Data. You can also select and hide a column from a table using the column heading context menu. See Select Columns Using the Table Context Menu and Hide Columns Using the Table Context Menu.
You can now group tables by one or more fields in the raw data. Summarized totals are provided for numeric fields in each group. The fields used for grouping are automatically moved to the leftmost columns of the table. Grouping can be set on the table itself using the column heading context menus, on the Table Settings sidebar (a new Group area has been added), and in the default settings for a data source. See Group and Ungroup Table Data.
In addition, the ability to group table data is now controlled by the Group interactivity setting. If the Group interactivity setting is switched off, the end user cannot group table data. See Control How Users Interact With a Visual.
You can change the aggregation of metrics in a table using the new table context menu. See Change Metric Aggregation in Tables.
In addition, the ability to change the aggregation of table metrics is now controlled by the Metrics interactivity setting. If the Metrics interactivity setting is switched off, the end user cannot change the aggregation of metrics in tables. See Control How Users Interact With a Visual.
You can now set even time intervals for a time field (Include Blanks setting) dynamically in a table using the column heading context menus if the table is grouped by the time field. See Apply Even Time Intervals on Tables and Even Time Intervals.
You can now alter the granularity of a time field dynamically in a table using the column heading context menus if the table is grouped by the time field. See Change Time Field Granularity in Tables.
Data Source Configuration Changes
The following changes have been made to data source configurations:
A new "Data Unavailable" message now appears when an error occurs obtaining data from a data source for a row security filter. In addition, if a row security filter returns no results, a new "No search results" message appears. See Restrict Access to Data Using Row Security.
The default settings available for tables in a data source configuration have changed in this release. You can now specify default:
Aggregation methods for numbers and metrics in the table data
Columns to display in tables produced for the data source
Grouping for the table data.
Caching Changes
This release introduces a new metadata cache to cache field statistics and improve overall metadata caching performance. In past releases, both visual data and field statistics were stored in the same cache. This new metadata cache is largely controlled by the API at this time. However, the clear cache button () on the Sources page still clears all caches for the data source and the Caching switch at the bottom of the Tables/Indices Tab of the data source configuration wizard still enables or disables all caching for a data source.
New API endpoints have been introduced to support cache clearing.
Endpoint | Method | Description |
---|---|---|
/api/sources/<source_id>/cache/visdata | DELETE | Clears the visual data cache. |
/api/sources/<source_id>/cache/metadata | DELETE | Clears the metadata cache. |
/api/sources/<source_id>/cache | DELETE | Clears both the visual and metadata caches. |
In addition, several new properties have been introduced for source definitions using the API:
A new boolean
cacheableMetadata
property indicates whether metadata caching is enabled for a data source. The existingcacheable
property indicates whether caching, in general, is enabled for the data source.A new
statsCached
property allows you to control metadata (field statistics) caching on a per-field basis. This property can be used to configure the field statistics look-up via materialized views (ignoring internal caches and redirecting the requests from the original source) and to improve the performance of statistics requests when precreated statistics data exists. Please contact your Technical Support representative to get guidance on how to configure this.
Comments
0 comments
Please sign in to leave a comment.