Python Live Data API

This API allows you to create Python applications that utilise ARDI live data.

from ardi.api.ardiapi as ardiapi

#Connnect to the ARDI server
ardi = ardiapi.Server("<ardiserverurl>","<ardisitename>")
if ardi.Connect() == False:
   sys.exit(0)

#Create a live-data session
sub = ardiapi.Session(ardi)
mychannel = sub.AddChannel('<asset name>','<property name>')

def ProcessNewValues(values):
    echo str(values)

sub.Callback(ProcessNewValues)
sub.Start()
sub.End()

This is a quick breakdown of how the above example operates.

1) Create an ARDICore object and connect it to the ARDI server.

ardi = ardiapi.Server("<ardiserverurl>","<ardisitename>")
if ardi.Connect() == False:
   sys.exit(0)

This loads all of the settings of your ARDI server, including the location of the consolidator.

2) Create an ARDISubscription for live data, passing the ARDICore we just created.

sub = ardiapi.Session(ardi)

This prepares the subscription by telling it the consolidator details. It does not connect to the server yet.

3) Add all of the data points you want to subscribe to.

mychannel = sub.AddChannel('<asset name>','<property name>')

You can add as many data points as you need.

The function returns a Channel object, which includes both the current value and some metadata about the point.

There are also other mechanisms for subscribing to data points. You can use AddChannels(aqlquery) to add a number of different channels using an AQL query, or call AddChannel with only the one parameter to add properties via data points.

Finally, there is also the AddBulkChannels function which takes an array of point names in asset.property format (ie. Main Oven.Temperature). This function uses a dedicated lookup system in ARDI rather than sending many individual AQL queries, which is much faster when you want to add a large number of points to your subscription.

4) Define and Set the Callback Function

def ProcessNewValues(values):
    echo str(values)

sub.Callback(ProcessNewValues)

The callback function is invoked every time a value changes. Create the function (it can be a class member if you're implementing this in a class) and call Callback on the Session.

Note that the array returned in the parameter to ProcessNewValues contains only the channels that have changed since the last update - except for the first time the function is called, where it contains the value for every channel.

5) Start the Subscription

sub.Start()

Note that the Connect function is a blocking function - it will stop the code from proceeding.

If you'd like to continue with other tasks, we suggest running this function in a separate thread.

The subscription can be stopped with sub.Stop().