Filter in compareToReference.py does not work for all data type.

[Yuan Chen]

The function "filter" in the file "compareToReference.py" works properly for the data type Int. The files "campaignConfiguration.py", "config.py" and "compareToReference.py" in the attachment are prepared for the test.

# python compareToReference.py
Reading lte.total.window.incoming.bitThroughput_UT_PDF
{'full': 1}
Reading lte.total.window.aggregated.bitThroughput_UT_PDF
{'full': 1}

However, it does not work for the other data type such as Bool, Float and String. In order to test the bug, the line in "compareToReference.py"

if scen.parameterSet.params["full"] == 1:

is replaced by the following line

if scen.parameterSet.params["fullPLCompensation"] == True:

After starting the script in a shell, the following error message is given.

# python compareToReference.py
Reading lte.total.window.incoming.bitThroughput_UT_PDF
Traceback (most recent call last):
  File "compareToReference.py", line 40, in <module>
    cdf = api.getCDFs(name, campaign, forScenarios = filter(allscen), agg="AVG")
  File "compareToReference.py", line 11, in filter
    if scen.parameterSet.params["fullPLCompensation"] == True:
KeyError: 'fullPLCompensation'

If the above mentioned line is substituted with

if scen.parameterSet.params["alpha"] == 1.0:

or

if scen.parameterSet.params["pNull"] == "-106 dBm":

similar errors are reported.

Can you please try to make it work for all data type?

Best Regards,

Yuan