Process Reports can expose process model, process, or task data to be queried so you can report on them in your interfaces. On this page, we will walk through how to create & configure a process report, call that process report using a!queryProcessAnalytics(), and show the results of the process report in an interface.
There are several out-of-the-box system default reports available to use. If you decide to go with one of these process reports, skip to the Calling the Process Report section.
Let's start by creating a process report.
Once created, process reports start with default data. This depends on the initial report type selected. Additionally, some reports require context prior to showing default data. In our case, I need to pass in a process model to compare. For a full listing of these data options including descriptions, see the Process and Report Data page.
While the default data that's being used is useful, let's add some additional columns to our report. To do this, we'll:
Adding data requires completing the Display Name, Definition, and Formatting fields.
Link to More Information lets us link to other useful data points. However, for the purpose of this report, we don't need to specifically link to other process data.
These will return in the d(n) part of the datasubset when we us
a!queryProcessAnalytics(), which can be used to create links to other things. Specifically, when creating task reports, linking to
tp!id can be useful to create a link to a task in someones task queue.
The last thing we'll do is sort the columns. This step is completely optional since we can choose to sort this data later. We're left with a process report with the data we want.
Once we've added the data columns that we want, we can begin to add a filter.
There are two types of filter options available: default and quick. Only default filters are necessary when querying process reports. Quick filters only provide a filter in this designer interface.
To configure our new default filter click on the Edit link to open the Report Options dialog and then click on the filter tab.
For our filter, we want to get today's orders, so we are going to select one of the out of the box filter options, Started within the last (n) days option. We can now save our process report an use it in an expression to get data to our interface.
`a!queryProcessAnalytics() lets us execute our process report and get the resulting data. Because we're dealing with expressions, we'll need to first create constants that point to our (1) process report, and (2) contextual process model.
Remember that process reports are stored as documents in Appian. For process reports, create a constant with a data type of document.
With these constants, we could return our data simply by calling an expression like this:
1 2 3 4 a!queryProcessAnalytics( report: cons!OM_TODAYS_ORDERS, contextProcessModels: cons!OM_NEW_ORDER_MODEL )
However, because we want to be able to page through our results in a paging grid, we're going to use the
query parameter in
a!queryProcessAnalytics() to add paging information.
1 2 3 4 5 6 7 a!queryProcessAnalytics( report: cons!OM_TODAYS_ORDERS, contextProcessModels: cons!OM_NEW_ORDER_MODEL, query: a!query( pagingInfo: ri!pagingInfo ) )
Now we'll get an expression that returns something like this:
Now that we can query this report and get data from it, we can display that data however we want in interfaces.
We can create an interface either in Design Mode or Expression Mode. What's important is knowing how to use the process report data subset. A process report data subset is almost identical to a typical datasubset with two exceptions:
columnConfigsfields that can be used. Of these the
columnConfigfields help identify which data is which.
datafield will always have
c(n)refer to the data columns we started with and added to our process report.
d(n)refer to linked data on any columns where designers selected Link to more information option.
The takeaways from these considerations means that when we are trying to use that data in an interface, our grid column configurations would look something like this.
1 2 3 4 a!gridTextColumn( label: "Name", data: index(local!dataSubset.data, "c0", null) )
Our finished grid will look something like this:
In the above example, we are displaying the data for every order. If we wanted to make a high-level chart of the amount of money that is currently held in each order status type, we could perform an aggregation on the process report itself before querying it. To do so we can create a new process report point to the same process model, but only include two columns into the report, one for order status and on for order total. We can then:
Set our grouping field as the first column by selecting checkbox next to the name and clicking Move Up until it is the first column. Next select the Group By check box.
Once the grouping is configured, we can edit the Aggregation column (in or example, Order Total.
We can call this process report using
a!queryProcessAnalytics() in the same expression as before and see that our data is now aggregated. That expression can be used to populate other charts or grids.