A Unified Modeling Language (UML) data set is a database
in table format whose rows and columns are mapped from a UML data
source. The data set is formatted so that you can use Business Intelligence
and Reporting Tools (BIRT) to generate reports from it.
Before you begin
The data source that you use must exist before you start
this activity. If you want to create a new data source, see the related
task "Creating UML data sources."
Procedure
- Open the Report Explorer view. Click .
- In the Report Explorer view, right-click a report design;
then click Open.
- Open the Data Explorer view. Click .
- In the Data Explorer view, right-click Data
Sets; then click New Data Set.
If you want to join data from multiple tables, you can also
click New Joint Data Set. For additional information,
see the topic "How to join data sets" in the Field Guide to BIRT.
- Type a name for the data set, select UML Data Set as the type, select a data source, and click Next.
- On the Query Parameters page, list
all the query parameters that you need to define queries in the data
set; then click Next. Use this page
to add, remove, reorder, and rename parameters. Parameters on this
page are identified by the internal name of the parameter. The internal
name is the name that you use in XPath expressions, so it must conform
to XPath naming conventions for variables; it must start with a letter
or underscore (_), and can contain any sequence of letters, digits,
or underscore characters. Changes that you make on this page also
appear on the Parameters page; see step 10.
Parameters are a flexible and powerful tool that allow you
to include many special types of information in reports. For information
about how you can use parameters, see the related concept named "Parameters
in EMF and UML data sets."
Note: After you define a parameter,
you can refer to it in both row-mapping and column-mapping queries
by using its internal name preceded by a dollar sign ($).
- Define the row mapping for the data set. You
define a collection of elements that form the rows of a table and
that specify the element types. To define the row mapping, complete
the following steps:
- Browse to select one of the models specified in the
data source that you are using.
- To define the row mapping, specify an XPath expression. For example, to map all classes in the model, the expression
is //Class and the type is Class.
Note: When you write row-mapping XPath expressions, remember that
root is the resource instance. For more information about XPath syntax,
see the related links at the end of this topic.
You can type
the expression directly, or build it from elements in the EMF structure,
which you add to the Expression field by selecting
a node in the EMF structure and then clicking the right arrow (>)
button.
- If the expression starts from a parameter as a context,
and not from the root of the resource, you can deselect the check
box for Evaluate for every data source instance model. This can improve performance, because the expression
does not need to run for every resource.
- Select an element in the EMF structure that has the
type required for the XPath expression; then click the right arrow
(>) button for the Type field. For
example, for //Class, the result type is Class.
- Specify any stereotypes to include in the data set.
- Click Next.
- Define the table columns for your data set. For column-defining XPath expressions, root is an object of the type
that you specified for the return type in the previous step. You can
define columns either by typing the values for them directly in the
matrix or by completing the following steps:
- Click Browse to display a type
to browse.
- Select an entry in the tree to use as a table column,
and then click the right arrow (>) button.
- Specify all the columns that you need. You can delete
columns, or change their order, by clicking the buttons to the right
of the Column Mapping matrix.
- Adjust the values for Query and Type for each column
as necessary.
- After you configure the table, click Finish. The Edit Data Set window
opens.
- If the data set includes parameters, use the Parameters page of the Edit Data Set window to complete the definition of the parameters. Specify the
display name, which is initially set to the internal name; the type,
which is initially set to String; the direction,
which is initially set to Input; and the name of
the report parameter, if any, to bind to the data set parameter. When you bind values to a data set parameter on the Property Binding page, you use the parameter display name.
- You can continue to edit, click Preview
Results to check your work, or click OK to close the window.
Note: To modify your data set at
any time, in the Data Explorer view, right-click the data set; then
click Edit.
What to do next
You can now use the BIRT feature to complete the report design,
incorporating elements from the data set. For information, see the
Field Guide to BIRT.
Note: Some information, such as links to Eclipse
documentation or to developer guides, is available only from the help
topics that are installed with the product.
