Appboard/2.6/builder/data collections


Data Collections are the datasets consumed by AppBoard widgets to produce visualizations and also subjected to manipulation in the form of configured (default) filtering, filtering actions, sorting, and limiting row counts. Additionally if it is desired to have widgets update automatically without the Viewer being reloaded then data collections can be configured to have a polling interval.

The data for a data collection originates from a data source entity, and in the simplest case without any filtering or sorting etc... should represent the set of data as returned by the external data source.

As an administrator configuring AppBoard you can think of data collections as a step in the pipeline of getting data from an external data source to drive a widget. It is data collections that widgets are configured to use, and it's the data collections that get filtered by filtering actions. It's important to understand that in practice the AppBoard server keeps track of data collections on a per session basis. So if one user clicks a widget that performs a server side filter on a data collection, that will happen for that user but not other clients.

Managing Data Collections

The main page for managing data collections in the Builder is by selecting the Data Collections builder mode. By default a summary view of all data collections is shown with the following fields as shown below:

Data Collections administration page

Field Description
Name The name of this data collection. By default data collections are created for each unique data source entity discovered in the format data_source_name.entity. These can be renamed and new data sources created however.
Rows The number of rows in the data collection. This may show zero if the data collection has never been previewed, or not currently being used (i.e. configured on a visible widget).
Data Source The name of the data source entity associated to this data collection.
Server Polling Shows whether server polling is enabled or disabled. Refer to the Caching & Polling Overview documentation for more information.
Polling Period When a data collection is active, i.e. configured on a visible widget, the polling period determines how frequently the client requests for updates. By default the polling period is set to zero (Disabled). See the Caching and Polling page for more information. Note that polling will be temporarily disabled while editing Widgets or Widget Actions and while in the System Administration page.
Last Poll This is the last time the client requested the data collection. It could be from previewing the data collection, or because of a visible widget configured to use this data collection. For active data collections with polling enabled this should count up to the polling period and reset indicating the client requested new data.
Server Response While the client can request data as frequently as it wants the server may respond with new data or cached data, or sometimes an error condition:
  • (blank): no status available
  • fresh: server responded with new data (less than 5 seconds old)
  • cached (n seconds old): The server returned cached data which is n seconds old. If cached data is returned this indicates the data source is configured to cache results - see the data source configuration for the actual cacheTimeout (maximum cache age). Also see the Caching and Polling page for more information.
  • expired (n seconds ago): The server returned stale data on purpose, most likely because there was cached data but there was an issue with the data source when fetching new data.
Status This indicates the overall status of the data collection based on the current configuration and server reponse. The following states are possible:
  • Gray Check: configuration is valid, but the data collection is inactive, i.e. most likely the data collection has not been accessed.
  • Green Check: Valid data collection with valid response from the server.
  • Yellow !: The server reported an error with the data source or response from the data source.
  • Red X: invalid configuration - this data collection is associated to a data source that no longer exists.
Preview View the data in this data collection by clicking on the blue icon.

Filtering Data Collections

For general information about filters and filtering data in AppBoard, see the Filtering Overview page.

Adding a Data Collection

Use these instructions to create a new Data Collection:

  1. Click Data Collections link in the left side tool palette.
  2. Click Add button at the bottom of the page to launch the Data Collections Wizard.
  3. On the Configure panel, enter the following:
    • Data Collection Name - Provide a unique name for the Data Collection. Only underscores, hyphens, periods, and alphanumeric characters are allowed (no spaces).
    • Data Source - Select the Data Source that will be used for the Data Collection.
    • Options - Check one or more of the following options:
      1. Filter - Provides an interface for adding server-side filters to the Data Collection.
      2. Server Side Sort - Allows you to specify the default sort order in which the Data Collection is displayed using these options:
        • Server Side Sort - Select the field from the Data Source by which the Data Collection will be sorted. You can select multiple columns to use for sorting.
        • Order (the box to the right) - Select either Ascending or Descending.
      3. Limit Row Count - Allows you to limit the number of rows returned by the server for this Data Collection, using the Server Side Row Limit box.
      4. Client Polling - Provides the option of setting a Polling Interval. This is the number of seconds to wait before the client will fetch the latest data from the server cache. For example, if this is set to 30, every 30 seconds AppBoard will refresh the Data Collection to have the latest data from the server. The minimum polling interval value is 5 seconds and the maximum value is 3600 seconds (one hour). Note: To maximize performance, Data Collections are only polled when one or more of the Widgets that uses that Data Collection is in the foreground display.
      5. Server Polling - Provides the option of scheduling jobs on the server that will prepare results for the client and continually update those results. This is a good option when queries take more time to process causing longer wait times on the client. The server scans data collections at start-up and when data collections and/or data sources are saved, creating jobs that will run while the server is running. If those queries are dynamic (session variables or query shim expressions are used in the adapter or entity settings) then temporary jobs will be created when unique queries are made to the server. When a client runs a client query, the server will look up the prepared results and return it. If it is not finished with the initial job; it will wait up to the client's remote timeout for the results (or the maximum timeout specified by the server configuration).
    • Click the Next button.
  4. On the Client Side Filter panel, add one or more Client Side Filter Rules.
  5. Click "Finish" to create the Data Collection.

Editing a Data Collection

Use these instructions to modify an existing Data Collection:

  1. Click Data Collections in the left side tool palette.
  2. Select the Data Collection from the data grid.
  3. Click Edit at the bottom to enter the Data Collections Wizard for that Data Collection.
  4. Make any necessary modifications to options in the Configure section.
  5. Click Next
  6. Make any necessary modifications to the Client Side Filter section.
  7. Click Finish to save the changes to the Data Collection.

Deleting a Data Collection

Use these instructions to delete an existing Data Collection:

  1. Click Data Collections in the left side tool palette.
  2. Click on the Data Collection you would like to delete.
  3. Click the Delete button at the bottom of the page.
  4. Click Yes to confirm the removal of the Data Collection.
It is possible to delete data collections that are currently configured on widgets and in actions. Ensure this is the desired action.