Lab 7: Create a Process API to Create Appointments in Salesforce Health Cloud.

Login to Anypoint Platform and click on the Exchange button.

Click on All Assets→Provided by MuleSoft and search for the Appointment Scheduling API. This API is provided by MuleSoft as a part of the MuleSoft Acclerator for Healthcare. Click on the Appointment Scheduling Process API to view its details.

This takes you to the home page for Appointment Scheduling Process API where you can read the overview and use cases. Take a moment to review the documentation.

So, it looks like this API has all of the functions that you need. In order to implement it you will use Anypoint Studio to download the API and create a new project with all of the functions "scaffolded" (templated) out for you.

Open Anypoint Studio and then click on File→New→Mule Project to start your work.

Name your project <FInitialLastName>-sfhc-accelerator-appointment-papi For example, Max Mule would name his project mmule-sfhc-accelerator-appointment-papi.

A window will appear titled Add Dependencies to Project.

In the Search bar of the Add Dependencies to Project window, type in Appointment Scheduling and press enter. Select the Appointment Scheduling Process API published by MuleSoft, and then click Add and Finish. Once your dependency loads, click Finish on the Project Settings dialog.

In the upper right of Studio locate the Mule Palette tab and click on Search in Exchange…. The Add Dependencies to Project window will reappear.

In the Search bar of the Add Dependencies to Project window, type in FHIR R4 Appointment and press enter. Select the FHIR R4 Appointment API that is published by MuleSoft and then click on the Add button. Under Selected modules → Version select the drop down and choose version 1.0.1. Click on the Finish button to close the Add Dependencies to Project window.

In the main canvas scroll down to locate the post:\Appointment:application\fhir+json:fhir-r4-appointment-scheduling-api-config flow. We will implement this flow to create a new appointment for the indicated patient in Salesforce Health cloud.

When APIs are scaffolded by Anypoint Studio the scaffolding process will add a Logger and, depending on the operation, a Transform Message component to the flows. For this lab you won’t need to use the Logger component for your API. Delete this by right clicking on the Logger component and clicking on the Delete option.

The first thing that you will want to do is save the message that was received into a variable for later reference.

HLS FHIR messages are sent as an array of entries that are referred to as "bundles". A FHIR bundle may contain one or more entries in them. In order to process the entries, it will be necessary to iterate over the bundle. You will use the For Each component to accomplish this.

In the Mule Palette, click on Core and locate the Choice component. Drag and drop the Choice component inside of the For Each component.

Now you need to configure the When component. In the flow, click on When to display its property editor. In the When editor, next to Expression, click on the fx button and then enter payload contains("Patient/") as the Expression.

For every entry that contains "Patient" you will want to transform the payload. In the Mule Palette, click on Core and locate the Transform Message component. Drag and drop this component inside the When component of the diagram.

Rename the Transform Message component to Set Variable patientid by right clicking on the Transform Message component and selecting the Rename option.

Now you need to access the current FHIR entry that is being processed to save the "PatientId" to a variable for later processing. The Transform Message component supports accessing variables as well as a number of other message processing functions.

To save the "PatientId" into a variable for later use click on the Add new target icon.

The Selection dialog window will be displayed. Change the Output to Variable via the drop down. Name the variable patientId. Click on the OK button.

The DataWeave editor will be displayed in the lower right pane. Note that the Output is set to Variable - patientId. This tells you that the Dataweave code you are about to enter will be setting the value of the patientId variable, and not the output of this component. Now, copy and paste the following Dataweave code and replace the code that is currently in the editor:

%dw 2.0
import * from dw::core::Strings
output application/java
---
substringAfter(payload,"/")

You will also want to capture the "Practitioner" value to a variable. You will use a similar set of steps that you completed to capture the "PatientId".

Add another When clause by dragging another Transform Message component from the Mule Palette into the space between the Choice component and the existing When and Default clauses. Drop the component on top of the vertical line that will appear when you are in the proper position. You should see a new When clause with the Transform Message component in it.

Click on the newly created When component. In the When editor, next to Expression, click on the fx button and enter payload contains("Practitioner/")

Rename the Transform Message component to Set Variable practitionerId by right clicking on the Transform Message component and selecting the Rename option.

In the Transform Message editor, click on the Add new target icon, which opens the Selection dialog window.

Change the Output to Variable via the drop down. Name the variable practitionerId and then click on the OK button.

In the DataWeave editor, replace the default DataWeave code with the followiing:

%dw 2.0
import * from dw::core::Strings
output application/java
---
substringAfter(payload,"/")

The next step is to reset the message payload to the value that was captured in the first Set Variable component. In the Mule Palette, click on the Core category and scroll to select the Set Payload component. Drag and drop the component to the right side of the For Each component, but not into the main flow. This may be a bit tricky. Look carefully at the surrounding outline to ensure you place it correctly.

In the Set Payload editor, ensure that the fx button is selected and then enter vars.preservePayload in the Value text box.

You are now ready to begin the process to transform the payload received to the appropriate return value. Another Transform Message component will be used to accomplish this.

In the Mule Palette, drag and drop another Transform Message component into the main flow, to the right of the For Each component.

Sample metadata is typically added to the Message Transform component via files. A sample metadata file for mapping the input message structure has been provided in the C:\Users\workshop\mulesoft\HLSWorkshop\PostAppointment_Input.json file in your virtual desktop.

A sample metadata file for mapping the output message structure has also been provided in the C:\Users\workshop\mulesoft\HLSWorkshop\PostAppointmentSFHC_Output.json file in your virtual desktop.

You are now ready to use your metadata files to define the Input and Output payloads. Start with the input metadata by clicking on the Define metadata hyperlink next to the Payload entry in the Input column.

The Select metadata type window will appear. Click on the Add icon. The Create new type window will appear. Enter PostAppointment_Input as the Type id: and then click on the Create type button.

The Select metadata type window will change to include several new input fields. In the first Type drop down select JSON. In the second drop down select Example and then browse to select the PostAppointment_Input.json file. Next click Select.

Repeat the previous steps for the output, creating a new metadata type called PostAppointmentSFHC_Output using the PostAppointmentSFHC_Output.json file as the example.

Now that the metadata has been added to your project you can enter the appropriate DataWeave statements to create the response message. As mentioned earlier, the use of DataWeave is out of scope for this lab exercise. So, to complete this lab step you can simply replace the existing DataWeave code by copying and pasting the expression below into the DataWeave editor.

%dw 2.0
output application/json
---
{
	resourceType: payload.resourceType,
	id: vars.patientId,
	text: {
		status: payload.text.status,
		div: payload.text.div
	},
	status: payload.status,
	serviceCategory: payload.serviceCategory map ( serviceCategory , indexOfServiceCategory ) -> {
		coding: serviceCategory.coding map ( coding , indexOfCoding ) -> {
			system: coding.system,
			code: coding.code,
			display: coding.display
		}
	},
	serviceType: payload.serviceType map ( serviceType , indexOfServiceType ) -> {
		coding: serviceType.coding map ( coding , indexOfCoding ) -> {
			code: coding.code,
			display: coding.display
		}
	},
	specialty: payload.specialty map ( specialty , indexOfSpecialty ) -> {
		coding: specialty.coding map ( coding , indexOfCoding ) -> {
			system: coding.system,
			code: coding.code,
			display: coding.display
		}
	},
	appointmentType: {
		coding: payload.appointmentType.coding map ( coding , indexOfCoding ) -> {
			system: coding.system,
			code: coding.code,
			display: coding.display
		}
	},
	reasonReference: payload.reasonReference map ( reasonReference , indexOfReasonReference ) -> {
		reference: reasonReference.reference,
		display: reasonReference.display
	},
	priority: payload.priority,
	description: payload.description,
	start: payload.start,
	end: payload.end,
	comment: payload.comment,
	participant:
	[
		{
			actor: {
				reference: vars.practitionerId,
			}
		}
	]
}

You are now ready to send the transformed payload to Salesforce Health Cloud. Earlier in this lab you had added the FHIR R4 Appointment connector to this project. Click on that component in the Mule Palette. Now drag and drop the Create appointment component to the right of the Transform Message component.

In the Create appointment editor locate the Connector configuration property and click on the Add (green plus sign) icon. The FHIR R4 Appointment API Config window will appear.

Host: sfhc-appointment-sapi.us-e1.cloudhub.io
Port: 80
basePath: /api
Protocol: HTTP

Click on Test Connection. a Test connection successful message should appear.

Click on the OK button to clear the message, then click on the OK button on the FHIR R4 Appointment API Config window to close it.

Save your work by going to File→Save.

Right click anywhere on the canvas and select Run project <your project name>.

Anypoint Studio will initialize the test instance of the MuleSoft runtime and deploy your API to it. As soon as the Console shows a DEPLOYED status for your API, click on the Open Console button that is in the APIkit Consoles tab in the lower left of Studio.

The APIkit will open in your browser where you can now locally test your application. On the left side of the APIkit page, expand the /Appointment resource and click on POST.

Check the box to Show optional headers. Select application/fhir+json for the Accept, Content-Type, and Body content type variables.

Clear the content of the Body section, then copy and paste the Test Message below into the Body. Click on the Send button to initiate the test.

{
  "resourceType": "Appointment",
  "id": "example",
  "text": {
    "status": "generated",
    "div": "<div xmlns=\"http://www.w3.org/1999/xhtml\">Brian MRI results discussion</div>"
  },
  "status": "booked",
  "serviceCategory": [
    {
      "coding": [
        {
          "system": "http://example.org/service-category",
          "code": "gp",
          "display": "General Practice"
        }
      ]
    }
  ],
  "serviceType": [
    {
      "coding": [
        {
          "code": "52",
          "display": "General Discussion"
        }
      ]
    }
  ],
  "specialty": [
    {
      "coding": [
        {
          "system": "http://snomed.info/sct",
          "code": "394814009",
          "display": "General practice"
        }
      ]
    }
  ],
  "appointmentType": {
    "coding": [
      {
        "system": "http://terminology.hl7.org/CodeSystem/v2-0276",
        "code": "FOLLOWUP",
        "display": "A follow up visit from a previous appointment"
      }
    ]
  },
  "reasonReference": [
    {
      "reference": "Condition/example",
      "display": "Severe burn of left ear"
    }
  ],
  "priority": 5,
  "description": "Discussion on the results of your recent MRI",
  "start": "2013-12-10T09:00:00Z",
  "end": "2013-12-10T11:00:00Z",
  "created": "2013-10-10",
  "comment": "Further expand on the results of the MRI and determine the next actions that may be appropriate.",
  "basedOn": [
    {
      "reference": "ServiceRequest/myringotomy"
    }
  ],
  "participant": [
    {
      "actor": {
        "reference": "Patient/0015e00000CtBhcAAF",
        "display": "Peter James Chalmers"
      },
      "required": "required",
      "status": "accepted"
    },
    {
      "type": [
        {
          "coding": [
            {
              "system": "http://terminology.hl7.org/CodeSystem/v3-ParticipationType",
              "code": "ATND"
            }
          ]
        }
      ],
      "actor": {
        "reference": "Practitioner/0035e000009jpKjAAI",
        "display": "Dr Adam Careful"
      },
      "required": "required",
      "status": "accepted"
    },
    {
      "actor": {
        "reference": "Location/1",
        "display": "South Wing, second floor"
      },
      "required": "required",
      "status": "accepted"
    }
  ]
}

Scroll the browser window down to view the HTTP response. You should see 200 OK as the HTTP response code and a payload returned to the browser.