CognitiveConcierge

CognitiveConcierge is an end-to-end Swift application sample with an iOS front end and a Kitura web framework back end. This application also demonstrates how to pull in a number of different Watson services to your Swift client and server side apps via the Watson Developer Cloud's Swift SDK, including Conversation, Text to Speech, Speech to Text, and the Natural Language Understanding service.

Included Components

Watson Conversation service
Watson Text to Speech service
Watson Speech to Text service
Watson Natural Language Understanding service
Google Places API

Application Workflow Diagram

The user deploys the server application to IBM Cloud.
The user interacts with the IOS application.
When the user performs any action, IOS application the server application API which uses the Watson services and Google Places API to provide the user recommendations.

Prerequisite

Obtain a Google Places API Key for Web: For this project, you'll need an API Key from Google Places, so that app can have access to review text which will be sent to the Natural Language Understanding service for analysis. Instructions for obtaining a key can be found here. Once you have an API Key, go to the Google Developer's Console, create a project, add your API key and enable the Google Places API for iOS as well. Make note of the API key for later use in your server and iOS applications.

If you haven't so yet, you also need to download and install the following:

Steps

Use the following steps to deploy the application

Deploy the Server Application
Update Conversation Service
Run the IOS Application

1. Deploy the Server Application

You can deploy the server application using any one of the following ways:

Deploy to Bluemix button
IBM Cloud Application Tools (ICAT)
Bluemix command line

a) Using the Deploy to Bluemix button

Clicking on the button below creates a IBM Cloud DevOps Toolchain and deploys this application to the IBM Cloud. The manifest.yml file [included in the repo] is parsed to obtain the name of the application, configuration details, and the list of services that should be provisioned. For further details on the structure of the manifest.yml file, see the Cloud Foundry documentation.

Once deployment to the IBM Cloud is completed, you can view the deployed application and services from your IBM Cloud account.

b) Using IBM Cloud Application Tools (ICAT)

Install IBM Cloud Application Tools for MacOS.
Once you've installed the application, you can open it to get started.
Click the Create (+) button to set up a new project, and then select the Cognitive Concierge Sample Application.
Click Save Files to Local Computer to clone the project.
Once the project is cloned, open up the .xcodeproj file that was created for you in ICAT under Local Server Repository. Edit the Sources/restaurant-recommendations/main.swift file's Constants struct with your own Google API Key for Web.
Finally, you can use ICAT to deploy the updated server to Bluemix. Click Provision and Deploy Sample Server on Bluemix under Cloud Runtimes.
Give your Cloud Runtime a unique name, and click Next. This deployment to Bluemix may take a few minutes.
In ICAT, ensure that the Connected to: field in the Client application is pointed to your server instance running on Bluemix. You can also point to your localhost for local testing, but you need to be running a local instance of the server application for this to work.

c) Using the Bluemix command line

You can also manually deploy the Server Application to the IBM Cloud. Though not as magical as using the Bluemix button above, manually deploying the app gives you some insights about what is happening behind the scenes. Remember that you'd need the Bluemix command line installed on your system to deploy the app to the IBM Cloud.

Execute the following command to clone the Git repository:

git clone https://github.com/IBM/CognitiveConcierge.git

Go to the project's root folder on your system and execute the Cloud-Scripts/services/services.sh script to create the services CognitiveConcierge depends on. Please note that you should have logged on to the IBM Cloud before attempting to execute this script. For information on how to log in, see the IBM Cloud documentation.

Executing the Cloud-Scripts/services/services.sh script:

$ Cloud-Scripts/cloud-foundry/services.sh
Creating services...
Invoking 'cf create-service conversation free CognitiveConcierge-Conversation'...

Creating service instance CognitiveConcierge-Conversation in org [email protected] / space dev as [email protected]...
OK
Invoking 'cf create-service speech_to_text standard CognitiveConcierge-Speech-To-Text'...

Creating service instance CognitiveConcierge-Speech-To-Text in org [email protected] / space dev as [email protected]...
OK

Attention: The plan `standard` of service `speech_to_text` is not free.  The instance `CognitiveConcierge-Speech-To-Text` will incur a cost.  Contact your administrator if you think this is in error.

Invoking 'cf create-service text_to_speech standard CognitiveConcierge-Text-To-Speech'...

Creating service instance CognitiveConcierge-Text-To-Speech in org [email protected] / space dev as [email protected]...
OK

Attention: The plan `standard` of service `text_to_speech` is not free.  The instance `CognitiveConcierge-Text-To-Speech` will incur a cost.  Contact your administrator if you think this is in error.

Invoking 'cf create-service natural-language-understanding free CognitiveConcierge-NLU'...

Creating service instance CognitiveConcierge-NLU in org [email protected] / space dev as [email protected]...
OK
Services created.

After the services are created, you can issue the bx app push command from the project's root folder to deploy the server application to IBM Cloud.

Once the application is running on the IBM Cloud, you can access your application assigned URL (i.e. route). To find the route, you can log on to your IBM Cloud account, or you can inspect the output from the execution of the bluemix app push or bx app show <application name> commands. The string value shown next to the urls field contains the assigned route. Use that route as the URL to access the sample server using the browser of your choice.

$ bx app show CognitiveConcierge
Invoking 'cf app CognitiveConcierge'...

Showing health and status for app CognitiveConcierge in org [email protected] / space dev as [email protected]...
OK

requested state: started
instances: 1/1
usage: 512M x 1 instances
urls: cognitiveconcierge-lazarlike-archaizer.mybluemix.net
last uploaded: Mon Jun 5 18:01:42 UTC 2017
stack: cflinuxfs2
buildpack: swift_buildpack

     state     since                    cpu    memory         disk           details
#0   running   2017-06-05 11:05:41 AM   0.3%   6.4M of 512M   269.8M of 1G

2. Update Conversation Service

Conversation service enables you to add a natural language interface to your applications. While you could create a conversation tree manually, ICAT has run some setup scripts (found in the Cloud-Scripts/conversation folder) to add a populated workspace to your conversation service.
If you are not using ICAT, go to the IBM Cloud dashboard and launch the Conversation service. Now manually populate the workspace by uploading the JSON found in Resources/conversationWorkspace.json. Make note of the workspace id for later use in running the iOS application.

3. Run the iOS Application

Install the necessary dependencies

From Terminal, change directories into the YourProjectName/CognitiveConcierge-iOS folder and run the following command to install the necessary dependencies (This may take some time):

carthage update --platform iOS
pod install

Update configuration for iOS app

Open the CognitiveConcierge.xcworkspace file in Xcode 8.3 either from ICAT or from your terminal using open CognitiveConcierge.xcworkspace
Update CognitiveConcierge.plist file: One way to persist data in Swift is through the property list or .plist file. ICAT has run some set up scripts to generate and populate the CognitiveConcierge-iOS/CognitiveConcierge/CognitiveConcierge.plist file. You will need to open this file and add your Google API Key. If you are not using ICAT, then manually update the credentials for all the services. You can get the credentials for services either from the environment variables section present in the runtime tab from your IBM Cloud dashboard or using the command bx app env CognitiveConcierge. ConversationWorkspaceID is the workspace id of the conversation service.
Update bluemix.plist file:
- You should set the isLocal value to YES if you'd like to use a locally running server; if you set the value to NO, then you will be accessing the server instance running on the IBM Cloud.
- To get the appRouteRemote value, you should go to your application's page on the IBM Cloud. There, you will find a View App button near the top right. Clicking on it should open up your app in a new tab, the url for this page is your route which maps to the appRouteRemote key in the plist. Make sure to include the http:// protocol in your appRouteRemote and to exclude a forward slash at the end of the url.
- You can also use the command 'bx app env CognitiveConcierge' where appRouteRemote is uris, bluemixAppGUID is application_id and bluemixAppRegion is your IBM Cloud region for eg: us-south.
```
 {
  "VCAP_APPLICATION": {
   "application_id": "3d06c0e7-1fff-4dbf-b0cb-b289770eccfe",
   "application_name": "CognitiveConcierge",
   "application_uris": [
    "cognitiveconcierge-lazarlike-archaizer.mybluemix.net"
   ],
   "application_version": "3ef63168-35f5-4517-84e9-e8f19c8f34b4",
   "limits": {
    "disk": 1024,
    "fds": 16384,
    "mem": 512
   },
   "name": "CognitiveConcierge",
   "space_id": "2b3083b9-7ef9-4d55-9741-34433be4cea1",
   "space_name": "dev",
   "uris": [
    "cognitiveconcierge-lazarlike-archaizer.mybluemix.net"
   ],
   "users": null,
   "version": "3ef63168-35f5-4517-84e9-e8f19c8f34b4"
  }
 }

 Running Environment Variable Groups:
 BLUEMIX_REGION: ibm:yp:us-south

 Staging Environment Variable Groups:
 BLUEMIX_REGION: ibm:yp:us-south
```
- We need to get the value for bluemixAppRegion, which can be one of three options currently:

REGION US SOUTH	REGION UK	REGION SYDNEY
`.ng.bluemix.net`	`.eu-gb.bluemix.net`	`.au-syd.bluemix.net`

Running the application

Press the Play button in Xcode to build and run the project in the simulator or on your iPhone!

Running the Kitura-based server locally

You can build the CognitiveConcierge-Server by going to the CognitiveConcierge-Server directory of the cloned repository and running swift build. To start the Kitura-based server for the CognitiveConcierge app on your local system, go to the CognitiveConcierge-Server directory of the cloned repository and run .build/debug/CognitiveConcierge. You should also update the bluemix.plist and CognitiveConcierge.plist file in the Xcode project in order to have the iOS app connect to this local server. See the Update configuration for iOS app section for details.

Privacy Notice

This Swift application includes code to track deployments to IBM Cloud and other Cloud Foundry platforms. The following information is sent to a Deployment Tracker service on each deployment:

Swift project code version (if provided)
Swift project repository URL
Application Name (application_name)
Space ID (space_id)
Application Version (application_version)
Application URIs (application_uris)
Labels of bound services
Number of instances for each bound service and associated plan information

This data is collected from the parameters of the CloudFoundryDeploymentTracker, the VCAP_APPLICATION and VCAP_SERVICES environment variables in the IBM Cloud and other Cloud Foundry platforms. This data is used by IBM to track metrics around deployments of sample applications to the IBM Cloud to measure the usefulness of our examples, so that we can continuously improve the content we offer to you. Only deployments of sample applications that include code to ping the Deployment Tracker service will be tracked.

Disabling Deployment Tracking

Deployment tracking can be disabled by removing the following line from main.swift:

CloudFoundryDeploymentTracker(repositoryURL: "https://github.com/IBM-MIL/CognitiveConcierge/", codeVersion: nil).track()