SAP BI 4.2 Beta - What's New in Web Intelligence
Following the enhancements introduced in SAP BI 4.1 SP6, the Web Intelligence and Semantic Layer stack keeps on improving in SAP BI 4.2. This article describes the Web Intelligence new features ~70 beta customers and partners have tested last month during SAP BI 4.2 Beta Program:
- Shared Elements
- Geo maps
- SAP HANA Direct Access
- SAP HANA Online Mode
- Parallel Data Provider
- Big Number
- Custom Elements
- REST Web Service
As this article describes beta material please remember that this presentation and SAP's strategy and possible future developments are subject to change and may be changed by SAP at any time for any reason without notice.
You might also be interested by this other article that covers Semantic Layer new features in SAP BI 4.2 Beta.
In order to ease the report creation, Web Intelligence offers in SAP BI 4.2 the ability to save report elements in the CMS repository. These report elements are named shared elements. Before SAP BI 4.2, creating a new document required several steps to be able to retrieve data from your data source:
- Selecting the data source
- Creating the query
- Optional creation of a few variables
- Defining the display of the data set (tables, charts…) and setting up the properties of this report element
- Optional definition of the report element format
Hence creating a new document may require some time. When you have to create similar documents, instead of recreating a new document from scratch, you can simply copy documents. But when a change is required, it must be done manually in all documents; which is cumbersome from a maintenance standpoint.
With shared elements, expert designers can save some report elements that can be used by casual users. When a shared element is inserted in the document, a link between this shared element and the document is created. Thanks to this link, when a document is opened, you can check whether a new version of the shared element has been published and, if needed, resynchronize the content of the shared element into the document.
Creating and Managing Shared Elements
Any report element can be saved as a shared element. When it is saved, all items required to create this report element are also saved:
- The report element itself and its properties: description, type…
- The variables used by the report element, if any
- The data providers used by the report element or its variables, if any.
- Their data sources, if any. If the report element is based on several data sources, then they are all saved in the shared element.
- The format defined for the report item (labels, colors, chart and their properties), if any
Shared elements can be saved in the CMS repository, in Favorites or Public folder or sub-folder.
To create a shared element, you must be in Design mode, select a report element and open the Export Element to Enterprise dialog box:
- Right-click it and select the Linking > Save as Shared element command in the contextual menu.
- Click the Save as a shared element button in the Report Element > Linking > Shared element toolbar.
Export shared element dialog box.
Select the destination folder for the shared element, then click Save to save it.
To manage the shared panels saved in the CMS repository, a new side panel Shared Elements has been introduced. From this side panel, you can:
- Navigate in the CMS repository to see the shared element
- Search, copy, move or delete shared elements
- Edit shared elements’ properties (name and description).
- Select a shared element and insert it in the document.
Available Shared Elements in Side Panel.
Inserting a Shared Element into a Document
Inserting a shared element can be done from the Shared Elements side panel, from the toolbar or from the contextual menu.
When the shared element is inserted in a document, then all its content (report element, query, data source, variable, format) is copied in the document. If a similar data source, query or variable is already used in the document, it is not recreated and is reused.
A link to the source shared element is created in order to check for shared element updates.
Currently Used Shared Elements.
The shared elements linked to the document are listed in the Currently used pane of the Shared Elements side panel. The version of the copied shared element content is also displayed.
Updating Shared Elements
When a shared element is inserted in a document, its content is explicitly copied into the document. A link between the shared element and the document is created but as this is a copy and not a dynamic reference, some differences may exist between the shared element and its copied content.
By default, when a document linked to shared elements is opened, Web Intelligence checks if the version of the instance copied in the document is still aligned with the one in the CMS repository. If a shared element version in the CMS repository is different from the one saved in the document, a green icon is displayed before the shared element name.
Available Shared Element Update.
To resynchronize the copied content in the document with the shared element, select the checkbox next to the shared element name and click the Update button.
You can remove the link between the shared element and the document by selecting it in the Currently used pane and clicking the Unlink button. This unlinks the document from the shared element, but the shared element content remains in the document.
You can stop checking the shared element version when the document is opened by unselecting the “Check for Shared Element update on open” option in the Document Summary dialog box.
“Check for Shared Element update on open” Parameter.
Geo maps allow you to create charts that take advantage of geographical data your reports contain. The implementation of these geo maps in Web Intelligence is similar to the one in SAP Explorer and SAP Lumira. Note that geo maps are for the moment not supported in shared elements.
Mapping Geographical Data
To create geo-maps, you first need to map geographical data contained in the report with the ones available in the geographical database embedded in Web Intelligence:
1. In Design Mode, in the Available Objects side panel, select a dimension that returns geographical data.
2. Right click it and in the contextual menu, select the Edit as a Geography command.
Edit as a Geography Command.
3. In the Edit as a Geography dialog box, use the Level drop-down list to define the geographical level you want to assign to the dimension. It can be Country, Region, Sub-Region or City and assumes all data returned by the dimension is on the same level.
4. The table displays all values returned by the dimension and a match proposed by Web Intelligence regarding the selected level. You can manually modify these proposals by using the drop-down lists.
Matching Geographical Data.
5. When you have finalized your mapping, close the Edit as a Geography dialog box. In the Available Objects side panel, the dimension appears as a geographical dimension.
Dimension Defined as a Geography.
Using Geographic Charts
To create a geo map, create and run a query containing a geographic dimension as well as the objects to add in the chart. Right-click the report element containing the returned data set and select the Turn to command in the contextual menu.
In the Turn into dialog box, you can select the Geographic section that proposes three geographic chart types:
- Choropleth chart
- Bubble chart
- Pie chart
Select the chart to use as well as the objects to feed it with data. In the Map Item Key, enter the geographic dimension.
Geographic Charts Types.
Validate your chart selection. In the report, the chart is displayed as the selected geo map. You can then edit the chart properties to fine tune its parameters (like the color of the seas) and finalize how your data is displayed in it.
Geographic Pie Chart.
If the document uses a universe as a data source and a drill path containing geographic dimensions has been defined, then the drill path is also supported in geographic charts.
In SAP BI 4.2, BI Platform proposes a common service for comments that Web Intelligence uses. This comments framework relies on an external database, where the comments are stored. By default the comment database is stored within the audit database, installed and configured at install time.
As these comments rely on the BI Platform. They are not available in Web Intelligence standalone or offline modes.
In SAP BI 4.2, only global report comments are supported. Block (table & chart) comments and table cell comments are planned in a future release.
Adding a Comment
A document comment is materialized by a new cell type. To add a new comment in a document, click the Insert Comment in the Report Elements > Comment toolbar. Then position the cell in the report and type the comment to display into the Insert Comment dialog box.
Insert Comment Dialog Box.
As any cell, you can modify its properties (size, position, color…)
A new side panel Comments is available to manage the document’s comments. This side panel is available in Reading and Design modes. When you select a comment in the report, it also appears in this side panel.
Comments Side Panel.
From this side panel, you can edit the comment or copy its content.
You can also reply to the comment created by other users if you have appropriate security rights, the comment is seen as a thread where several users can exchange.
Replying to a Comment.
A new property in the Web intelligence document can be used to set which comments of the thread must be displayed in the report comment:
- The last comment in the thread, by default
- The first comment in the thread
Advanced Display: Support of Workflows
You have the possibility to display comment pending on a value that you will position in an optional column of the Comment Database.
In order to do so you need to write a specific value of your choice in the database in the comment you want to look for.
Then in Design mode, select the Comment cell hosting your comment thread. Click it and check that the formula display by default is =Comment().
This formula supports two arguments:
- “ColumnName”: a column where the search is done
- “Text”: the string to look for
The comment returns the first or the last comment (depending on the value of the option you have selected), if the string is found in the column.
For example, Comment(“OptionKey1”;”Validated”) displays the comment only if the text “Validated” is found in the column.
SAP HANA Direct Access
In order to ease reporting on SAP HANA, the same Direct Access available to connect to SAP BEx query is also available in Web Intelligence for SAP HANA SPS 09 and SAP HANA SPS 10. With this direct access you do not need to create a universe in the information design tool. You can directly select a SAP HANA view (analytic or calculation view) as a data source.
SAP HANA Direct Access Data Source (DHTML).
In this access mode, after connecting to the SAP HANA view, the query panel opens and directly generates the metadata from the SAP HANA view. You can then select your objects to create your query and run it. You can then finalize your report with the data retrieved in the Web Intelligence cube.
Depending on the connection you select as a data source, the model generated in the query panel can be multidimensional or relational.
- Selecting a SAP HANA OLAP connection as a data source allows you to browse HANA models as multidimensional cubes (with dimensions, hierarchies, levels, measures, etc.), and to run multi-dimensional query scripts using the MDX script language.
SAP HANA View Selection – OLAP Connection.
Multidimensional Query Panel in SAP HANA Direct Access.
- Selecting a SAP HANA relational connection allows you to query SAP HANA views through the Connection Server and HANA SQL Access. The system generates a HANA relational transient universe. The transient universe generation strategies are similar to those of relational authored universes on HANA views in the information design tool.
SAP HANA View Selection - Relational Connection.
Relational Query Panel in SAP HANA Direct Access.
To improve performance, before refreshing a report, Web Intelligence checks if the SAP HANA view has been modified since the last refresh. If this is not the case, then it does not regenerate the metadata layer and can reuse the one saved in the document, leading to better performance.
HANA Variables and Input Parameters are surfaced as Web Intelligence prompts that appear at refresh time. You are then invited to provide answers before continuing the data refresh.
Web Intelligence displays these HANA variables or Input Parameters as a list of prompts in the Web Intelligence prompt user interface in the order defined in HANA Studio.
You can select the “Merge Prompts (BEx/HANA Variables)” parameter in the document options to factorize or not the SAP HANA Variable/Input Parameter based on their captions.
Note that Change Source is not supported for SAP HANA direct access mode.
SAP HANA Online Mode
Another access to SAP HANA SPS 09 and SAP HANA SPS 10supported by Web Intelligence in SAP BI 4.2 is the SAP HANA Online mode. As for SAP Direct Access, this mode does not require you to create a universe in the information design tool to query the server; Web Intelligence is directly connected to the SAP HANA view.
This SAP HANA Online mode is supported only through a relational model, hence it does not benefit from the SAP HANA hierarchies that are flattened as dimensions in the exposed model.
When you create a document in SAP HANA Online mode, you first select a SAP HANA relational connection as data source. In order to ease the usability, the SAP HANA view content is directly displayed in the Available Objects side panel. You do not need to use the query panel, since the document is created as you drag and drop objects from the Available Objects side panel into the page. Each time a new object is added to the document, a new query is sent to the SAP HANA server that returns the dataset; you can interactively explore the data and create your document.
When possible, Web Intelligence calculations are delegated to SAP HANA and are no more performed in the Web Intelligence cube. This is typically the case for calculations on measures. This method takes advantage of the SAP HANA in-memory architecture and enables very fast response times.
Selecting SAP HANA Metadata in SAP HANA Online Mode.
When you create a new document in SAP HANA online mode, SAP HANA variables and input parameters (if any) are prompted before displaying the metadata in the Available Objects pane. To avoid reprompting them each time the corresponding objects are used, the entered values are kept until you explicitly reset them through the Reset Prompts button in the Data Access > Data Providers toolbar.
When you save a document created in SAP HANA Online mode, it is saved without any data. The data set is retrieved from the SAP HANA server when you open the document the next time.
As for SAP HANA Direct Access, you can select to merge them or not using the SAP Merge Prompts (BEx/HANA Variables) parameter in the document options.
This new SAP HANA Online mode has some limitations:
- Merged objects, custom groups and shared elements are not supported.
- As for SAP HANA Direct Access, it is not possible to change the data source.
- You can add a new data source but it must be a SAP HANA view also accessed in SAP HANA Online mode.
It is possible to switch from this mode to the SAP HANA Direct Access mode by clicking the Disconnect from SAP HANA button in the Data Access > SAP HANA toolbar. The objects selected in the document are then used to create a query in the query panel.
Parallel Data Provider
If a Web Intelligence report contains several queries, it is possible to improve the refresh performance by running them in parallel instead of running them one after the other. This capability is available for all Web Intelligence clients i.e. Web Intelligence Rich Client, applet, DHTML, scheduling and REST Web Service.
Supported Data Providers
The parallel data provider refresh is supported for the following data providers:
- Universes created by the information design tool (UNX) and universe design tool (UNV) on both relational or OLAP connections
- Free-hand SQL based on secured relational connections
- Text files (TXT/CSV)
- SAP HANA Direct Access based on HANA views
The following data providers are not in scope of 4.2:
- Microsoft Excel data providers
- SAP BW BEx queries
- Web Services
- Analysis Views
Settings for Web Intelligence
Administrators can control the use of this parallel data provider refresh in order to prevent users from decreasing the database performance by running too many parallel queries.
For Web Intelligence, the administrator can set the maximal number of concurrent data refresh jobs per document. By default, the maximum number of refresh jobs that can run in parallel is set to 64, but it can be decreased if the underlying hardware cannot support the workload.
In SAP BI 4.0 and BI 4.1, the decimal number is limited to 15 digits. To make calculations more precise and follow the IEEE 754-2008 standard for the floating point, Web Intelligence introduces a new attribute named Decimal for measure.
- Web Intelligence’ calculation engine supports these decimal values
- Decimal values have a default precision of 40 digits, with a maximum exponent value of 300 to correctly convert double into decimal.
- Decimal values uses the default “ROUND_HALF_EVEN” rounding method
- Variable measures consuming a decimal, return also as decimals
- Web Intelligence supports a new function:
decimal ToDecimal (value)
where value can be a number or a string. This function transforms its input into a decimal. The string input type is useful to extract high precision values from a data source, without any loss in precision.
To set a measure as a decimal, in Design mode, go to the Available Objects side panel. Right-click a measure and select the Change Type > Decimal contextual menu.
Change Type Command.
In addition to its extension framework, Web Intelligence proposes in SAP BI 4.2 a new framework to allow you to write custom visualizations. SAP partners can typically use this framework to propose predefined components they have written.
A custom element charts are available in the list of available charts.
Available Custom Elements.
The following sample shows how data can be displayed as gauges through a custom element that relies on Google charts.
Google Gauge Chart Used As a Custom Element.
REST Web Services
The Web Intelligence REST Web Services have been enhanced to support:
- Linked universes as a data source
- BOTH and EXCEPT operators in filters
- Dotted and dashed lines in charts
- Big numbers
- Microsoft Excel files in the CMS repository update
- Microsoft Excel file data provider paths change