IoT: Displaying the temperature from the BeagleBone Black on my iPhone

In the first post of this series I used Javascript to read a TMP36GZ temperature sensor and write the current temperature to the console.  In the second post I used the Restify library with Javascript to create a rest based web service that would send the current temperature to a remote client upon request.  In this post I will access that web service to display the temperature on my iPhone using the RSNetworking library.

Let begin by reviewing how we wired up the TMP36GZ temperature sensor to the BeagleBone Black.  Here is the wiring diagram:

Next lets review how we developed the REST Web Service that will respond to client requests for the current temperature.  Note:  I am making one change to the code, from my last post, so the REST service returns a valid JSON object. 

The first thing we need to do is to setup a directory structure that contains the node.js modules needed to run the service.  I used the following steps to get the structure/modules set up (this is based off of the latest Debian image 2014-05-14):

1.  Start off within our home directory.  In Linux the ~ directory is the users home directory.

cd ~

2.  Create a work directory and change to that directory

mkdir temperature

cd temperature

3.  Install the restify module

npm install restify

4.  Copy the bonescript module to our working structure

mkdir node_modules/bonescript

cp /var/lib/cloud9/static/bonescript.js node_modules/bonescript/

At this point you should still be in the temperature directory and both the restify and bonescript modules should be located in the ~/temperature/node_modules directory.   This will let our application use these two modules.

From the ~/temperature directory, create a file called tempServer.js and put the following code in it.

var b = require('bonescript');

var rest = require('restify');

var ip = '0.0.0.0'

var port = '18080'

var path = 'temperature'

var tempPin = 'P9_38';

var currentTemp = 0.0;

b.analogRead(tempPin, readTemperature);

setInterval(function() {b.analogRead(tempPin, readTemperature)},30000);

var server = rest.createServer({

   name : "Temperature"

});

server.get({path : path , version : '0.1'}, getTemperature);

server.listen(port,ip, function() {

   console.log('%s listening at %s ',server.name, server.url);

});

function getTemperature(req, res, next) {

   var key = "temperature";

   var tempResponse = '{"temperature":"' + currentTemp + '"}';

   res.send(200, JSON.parse(tempResponse));

}

function readTemperature(aRead) {

   console.log("Geting Temp");

       var x = (aRead.value * 1800/1024);

       var cel = 100*x -50;

       var fah = (cel *9/5)+32;

   currentTemp = fah;

}

This code begins by loading both the bonescript and restify modules that are needed for this application.  We then set the following variables:

ip:  The IP address of the interface to bind too.  By using 0.0.0.0 the server will bind to all available interfaces (this is what we want).

port:  The port to bind too.  Typically web servers bind to port 80 but we do not want to take up that privileged port (and it also requires root access to bind to ports below 1024) so we will use 18080 for our service.

path:  The URL path for our service. 

tempPin:  The pin that will be connected to the TMP36GZ temperature sensor.

CurrentTemp:  Will contain the current temperature.  This will be updated every 30 seconds.

After we set the variables, we then read the temperature using the analogRead function from the bonescript module.  This function will read the voltage from the tempPin and then call the readTemperature function when it has the voltage.  The readTemperature function calculates the current temperature based on the voltage of the pin and stores that temperature into the currentTemp variable.

We use the Javascript setInterval function to call the readTemperature function every 30 seconds to update the currentTemp variable.

Now that we have the temperature and updating it every 30 seconds, we need to create our web service that will respond to our requests.  We start off by creating a server object using restify’s createServer function. 

Next we define what services we wish to offer though this server object.  In this case we only have one service.  This service will respond to HTTP GET requests so the get function from our server object is used to define the service.  This service will listen on the path defined in our path variable and when a request comes in it will call the getTemperature function.

Finally we till the server to listen on the port and interface that we defined in the variables earlier.

The getTemperature function simply creates a JSON object that contains the current temperature and uses the send function from the res response object to send the object back to the client that requested it.  This is where I made the code change from my previous post.  To create the JSON object, I first create a string that contains the response that I want to send back to the client and I then use JSON.parse() function to create a valid JSON object.

Now lets look at how we would write the iOS client application to retrieve the temperature from the BeagleBone Black’s web service.  We will be writing the client app in Swift (Apple’s new development language) and will use the RSNetworking library.  RSNetworking is a network library written entirely in the Swift programming language.  RSNetworking is built using Apple’s powerful URL Loading System.  The main design goal of RSNetworking is to make it easy and quick for developers to add powerful asynchronous networking requests, to their applications written in Swift.  As a disclaimer, I created and maintain the RSNetworking library. 

The first thing we will need to do is to download the RSNetworking library from Github and include the library in our project. 

In the MainStoryboard of our project we will add a UILabel (to display the temperature) and a UIButton (to get the temperature from the BeagleBone Black).  Here is a screen shot of how UI is laid out.

Now lets look at the code that will retrieve the temperature from the BeagleBone Black’s Web Service and display it on the screen when the button is pressed.

import UIKit

class ViewController: UIViewController {

   

    @IBOutlet var tempDisplay : UILabel!

                           

    override func viewDidLoad() {

        super.viewDidLoad()

       }

    override func didReceiveMemoryWarning() {

        super.didReceiveMemoryWarning()

       }

    

    @IBAction func getTemp(AnyObject) {

        var rsRequest : RSURLRequest = RSURLRequest();

        var url : NSURL = NSURL.URLWithString("http://10.0.1.30:18080/temperature");

        rsRequest.dictionaryFromJsonURL(url, completionHandler:{(response : NSURLResponse!, responseDictionary: NSDictionary!, error: NSError!) -> Void in

            if error == nil {

                println(responseDictionary);

                var temp: Double! = responseDictionary["temperature"]?.doubleValue!

                let displayStr : NSString = "".stringByAppendingFormat("%.2f", temp!)

                self.tempDisplay.text = displayStr

               

            } else {

                //If there was an error, log it

                println("Error : \(error)")

            }

        })

    }

}

The getTemp() function is tied to the UIButton.  When the button is pressed the applications accesses the Web Service on the BeagleBone Black and retrieves the temperature.  If there were no errors it displays the temperature in the UILabel.  So lets look at the getTemp() function.

We begin by creating an instance of the RSURLRequest class.  We then create an instance of NSURL using the URL for our Web Service.  We then use the dictionaryFromJsonURL() function from the RSURLRequest class to send our request to the BeagleBone Black. 

We pass a block (of code) to the dictionaryFromJsonURL() function.  This block of code verifies that we did not receive any errors and if no errors were received it converts the temperature that was received to a Double values and displays the temperature, with the precision that we want, in the UILabel.  If there were errors we simply log them.

That is all there is to creating our first IoT service with the BeagleBone Black.