rtplib3

rtplib3 offers a high level Python drop module for rtplib2 that is compatible with HIL API MAPort Python and XIL API .NET MAPort. It does this by hiding the rtplib2 to XIL API migration guide behind a layer of abstraction.

Why?

Executive Summary: Combination of Windows, dSpace and Matlab compatibilities are going to make a lot of current running code obsolete. Any scripts using rtplib2 or HIL API MAPort interfaces might no longer work once machines are upgraded to Windows 10.

Microsoft's main stream support for Windows 7 ended January 13, 2015 and extended support ends January 14, 2020..

dSpace doesn't actually support Windows 10 yet:

With dSPACE Release 2016-B TargetLink and Model Compare will support Windows 10.

Full qualified support of Windows 10 for the complete dSPACE Release 2016-B will be granted by means of a service pack available mid 2017.

Beginning with dSPACE Release 2017-A dSPACE products will generally support Windows 10. dSPACE will support only the 64-bit Professional and Enterprise editions of Windows 10. - Windows 10 Support of dSPACE Products

In the past dSpace has exposed multiple programming interfaces for automation, rtplib2 for Python and mlib and Matlablib2 for MATLAB. rtplib2 is discontinued and will be distributed for the last time with dSPACE Release 2016-A.. dSPACE Release 2013-B was the last release comprising MLIB.

There is a rtplib2 to XIL API migration guide however it is non-trivial and there is a large existing code base of rtplib2 Python scripts running dSpace anywhere dSpace is sold.

Examples:

Creating and Configuring MAPorts

rtplib2:

import rtplib2

platformIdentifier = "ds1005"
applicationPath = r"C:\Path\To\MyApplication.sdf"
appl = rtplib2.Appl(applicationPath, platformIdentifier)

rtplib3:

import rtplib3 as rtplib2

platformIdentifier = "ds1005"
applicationPath = r"C:\Path\To\MyApplication.sdf"
appl = rtplib2.Appl(applicationPath, platformIdentifier)

XIL API .NET:

# import Python for .NET
import clr
# make some collections and data types of the .NET framework available in Python
from System import Array
from System.Collections.Generic import Dictionary
# load ASAM assemblies from the global assembly cache (GAC)
clr.AddReference("ASAM.XIL.Implementation.TestbenchFactory, Version=2.0.1.0,
Culture=neutral, PublicKeyToken=fc9d65855b27d387")
clr.AddReference("ASAM.XIL.Interfaces, Version=2.0.1.0, Culture=neutral,
PublicKeyToken=bf471dff114ae984")
# import XIL API .NET classes from the .NET assemblies
from ASAM.XIL.Implementation.TestbenchFactory.Testbench import TestbenchFactory
from ASAM.XIL.Interfaces.Testbench.Common.Error import TestbenchPortException 

testbenchFactory = TestbenchFactory()
testbench = testbenchFactory.CreateVendorSpecificTestbench("dSPACE GmbH", "XIL API", "2015-B");

maPortConfigFilePath = r"..\MAPortConfiguration.xml"
maPort = testbench.MAPortFactory.CreateMAPort("DemoMAPort")
maPortConfig = maPort.LoadConfiguration(maPortConfigFilePath)
maPort.Configure(maPortConfig, False)

Reading and Writing Values

rptlib2:

turnSignalLever = "Model Root/TurnSignalLever[-1..1]/Value"
turnSignalLeverVar = appl.Variable(turnSignalLever)
readValue = turnSignalLeverVar.Read()
print readValue
turnSignalLeverVar.Write(1.0)

rptlib3:

turnSignalLever = "Model Root/TurnSignalLever[-1..1]/Value"
turnSignalLeverVar = appl.Variable(turnSignalLever)
readValue = turnSignalLeverVar.Read()
print readValue
turnSignalLeverVar.Write(1.0)

rptlib3 w/syntactic sugar:

Since rtplib3 is an abstraction, you can use syntactic sugar to simplify new code bases.

turnSignalLever = "Model Root/TurnSignalLever[-1..1]/Value"
turnSignalLeverVar = appl.Variable(turnSignalLever)
print turnSignalLeverVar
turnSignalLeverVar = 1.0

XIL API .NET:

(Since XIL API is a .NET module, all numbers must be cast to .NET types before use)

turnSignalLever = "Model Root/TurnSignalLever[-1..1]/Value"
readValue = maPort.Read(turnSignalLever)
print readValue.Value
maPort.Write(turnSignalLever, testbench.ValueFactory.CreateFloatValue(1.0))

Working with PyCANape

Fully compatible with PyCANape. This allows you to run data analysis on your Vector CANape and dSpace data concurrently. Use any Python testing framework to automate any hardware or software in the loop testing.

# Imports
import CANape
import rtplib3 as rtplib2

# Setup dSpace
platformIdentifier = "ds1005"
applicationPath = r"C:\Path\To\MyApplication.sdf"
appl = rtplib2.Appl(applicationPath, platformIdentifier)

# Setup CANape
canape = CANape.CANape()

# Set dSpace Value
turnSignalLever = "Model Root/TurnSignalLever[-1..1]/Value"
turnSignalLeverVar = 0

# Start CAnape 
canape.start_measurement()

# Test to make sure they are equal.
# Test that the value was set by reading it back
assert(0, turnSignalLeverVar)
# Test that CANape read the change
assert(0, canape.read_calibration_object(0, 'TurnSignalLever', 1))

# Generate Sinewave in Python for swept sine analysis.
t1 = t2 = time()
amplitude = 1
offset = 1
frequency = 1
while t2-t1<10:
    turnSignalLeverVar = amplitude * np.sin(t2*(2*np.pi*frequency)) + offset
    sleep(0.5)
    t2 = time()