advanced-filter-convert-publish

README

Example Advanced App Functions Service

This advanced-filter-convert-publish Application Service depends on the new Device Virtual Go device service to be generating random number events. It uses the following functions in its pipeline:

• Built in Filter by Value Descriptor function to filter just for the random float32 & float64 values.
• Custom Value Converter function which converts the encoded float values to human readable string values.
• Custom Print to Console function which simply prints the human readable strings to the console.
• Custom Publish function which prepares the modified event with converted values and outputs it to be published back to the message bus using the configured publish host/topic.

The end result from this application service is random float values in human readable format are published back to the message bus for another App Service to consume.

End to end Edgex integration proof point for App Functions

Using the following setup, this example advanced App Functions Service can be used to demonstrate an EdgeX end to end proof point with App Functions.

1. Start EdgeX Mongo

   • clone developer-scripts repo
   • cd compose-files folder
   • run "docker-compose up mongo"
     • This uses the default compose file to start the EdgeX Mongo service which exposes it's port to apps running on localhost

2. Run EdgeX Core Services

   • Clone edgex-go repo
   • run "make build"
   • run "make run"
     • This starts all the required Edgex core, export and support services

3. Run Advanced App Functions example

   • Clone app-functions-sdk-go repo
   • run "make build"
   • cd to examples/advanced-filter-convert-publish folder
   • run "./advanced-filter-convert-publish"

4. Configure and Run Simple Filter XML App Functions example

   • cd examples/simple-filter-xml folder

   • edit res/configuration.toml so the MessageBus and Binding sections are as follows:

             [MessageBus]
             Type = 'zero'
                 [MessageBus.PublishHost]
                     Host = '*'
                     Port = 5565
                     Protocol = 'tcp'
                 [MessageBus.SubscribeHost]
                     Host = 'localhost'
                     Port = 5564
                     Protocol = 'tcp'
     
             [Binding]
             Type="messagebus"
             SubscribeTopic="converted"
             PublishTopic="xml"
     

   • Run "./simple-filter-xml"

5. Run Device Virtual service

   • Clone https://github.com/edgexfoundry/device-virtual-go repo

   • run "make build"

   • cd to cmd/res folder

   • Edit the device.virtual.float.yaml file

     This app functions example expects the float encoding for all random floats to be Base64 and needs to restrict the range of values that are generated so they are easy to read. By default the device-virtual is using Base64 for Float32 & eNotation for Float64 and doesn't set any range limits. Make the following changes to the deviceResources section to meet these needs.

     For RandomValue_Float32 change the value property to:

     { type: "Float32", readWrite: "R", defaultValue: "0", floatEncoding: "Base64", minimum: "1.0", maximum: "1.9" }
     

     For RandomValue_Float64 change the value property to:

     { type: "Float64", readWrite: "R", defaultValue: "0", floatEncoding: "Base64", minimum: "2.0", maximum: "2.9" }
     

   • If you previously ran the Device Virtual service, run the follow curl commands to clear the old profile so that the new changes are used when the device service is started.

     curl -X DELETE http://localhost:48081/api/v1/deviceservice/name/device-virtual
     curl -X DELETE http://localhost:48081/api/v1/deviceprofile/name/Random-Float-Device
     

   • cd to cmd folder

   • run "./device-virtual"

     first time this is run, the output will have these messages :

     level=INFO ts=2019-04-17T22:42:08.238390389Z app=device-virtual source=service.go:138 msg="**Device Service  doesn't exist, creating a new one**"
     level=INFO ts=2019-04-17T22:42:08.277064025Z app=device-virtual source=service.go:196 msg="**Addressable device-virtual doesn't exist, creating a new one**"
     

     One subsequent runs you will see this message:

     level=INFO ts=2019-04-18T17:37:18.304805374Z app=device-virtual source=service.go:145 msg="Device Service device-virtual exists"
     

6. Now data will be flowing due to the auto-events configured in Device Virtual Go.

   • In the terminal that you ran advanced-filter-convert-publish you will see the random float values printed.

     RandomValue_Float64 readable value from Random-Float-Device is '2.1742
     RandomValue_Float32 readable value from Random-Float-Device is '1.3577
     

   • In the terminal that you ran simple-filter-xml you will see the xml representation of the events printed. Note the human readable float values in the event XML.

     <Event><ID></ID><Pushed>0</Pushed><Device>Random-Float-Device</Device><Created>0</Created><Modified>0</Modified><Origin>1555609767442</Origin><Readings><Id>835c5541-d4d2-42a8-8937-8b24b4308d3f</Id><Pushed>0</Pushed><Created>0</Created><Origin>1555609767411</Origin><Modified>0</Modified><Device>Random-Float-Device</Device><Name>RandomValue_Float64</Name><Value>2.1742</Value><BinaryValue></BinaryValue></Readings></Event>
     <Event><ID></ID><Pushed>0</Pushed><Device>Random-Float-Device</Device><Created>0</Created><Modified>0</Modified><Origin>1555609797452</Origin><Readings><Id>21c8ccdc-3438-4baa-8fab-23a63bf4fa18</Id><Pushed>0</Pushed><Created>0</Created><Origin>1555609797419</Origin><Modified>0</Modified><Device>Random-Float-Device</Device><Name>RandomValue_Float32</Name><Value>1.3577</Value><BinaryValue></BinaryValue></Readings></Event>