Blog

Every chart that has been automatically or manually defined will display a title. This title will be composed using the row, column, and value fields and their properties that have been specified for the chart. For example, suppose we create a curved line chart on Orders page of the Northwind sample app that uses OrderDate for rows, EmployeeID for columns, and the sum of Freight for values.

The data for the chart can be seen below.

The title states that the chart shows top three “Employee Last Name”. The value is the sum of “Freight” and the rows are grouped by “Order Date”. While the title may reveal all the necessary information, the title can be better. Let’s manually specify the title by adding the following tag. Note that the tags do not need to be on any particular data field, as long as the correct pivot ID is specified. Open the Project Designer. In the Project Explorer, switch to the Controllers tab and double-click on Orders / Views / grid1 / EmployeeID data field node.

Add the highlighted tag:

Save the data field and regenerate the app. When the page comes up in your browser, navigate to the Orders page. Note that the new title has been applied.

The new title will also be displayed above the chart data.

By default, the axes do not have a title applied. One can use the “haxistitle” or “vaxistitle” properties to manually specify a title for the horizontal or vertical axis, respectively. Let’s specify the following axis titles:

The new chart will display the specified labels on each axis.

The chart data will not be changed.

When using a date as a row in a chart, it is necessary to form groups for each date value. Otherwise, a row will be created for each date. This excess of rows will render the chart useless:

The data behind the chart looks like the following:

Instead of listing each date as a unique value, the dates should be placed into buckets in order to provide a clearer view of the data. For example, let’s group the dates by quarter. Add the “quarter” keyword to calculate the quarter of each date.

The chart is now easier to read.

The data for the chart looks as follows:

However, notice that we have lost information about the year. To solve this problem, we need to group the dates by year first, and then by quarter.

The chart now displays quarters per year.

The data is shown here:

However, what happens when the data is filtered to produce a different subset of data? The chart may become useless again.

The single row is displayed below.

In order to solve this problem, an automatic grouping function can be used. By specifying a group of “date”, the server will make several attempts with various groups. The chart with a number of rows closest to 25 will be selected and displayed to the user.

The function has decided to use year/month/week as the most optimal solution for the currently applied filter.

The data can be seen below.

When the filter is cleared and all data from the Orders table is displayed, the group mode will try to maintain the number of rows. The function has elected on using year/month.

The new data can be seen here:

A full list of available groups is displayed below.